Ecology and Natural History of a Neotropical Savanna - LERF - USP

00 oliveira fm 7/31/02 8:11 AM Page i 

The Cerrados of Brazil

00 oliveira fm 7/31/02 8:11 AM Page ii

00 oliveira fm 7/31/02 8:11 AM Page iii 

The Cerrados of Brazil 

Ecology and Natural History 

of a Neotropical Savanna 

Editors 

Paulo S. Oliveira 

Robert J. Marquis 

Columbia University Press 

New York

00 oliveira fm 7/31/02 8:11 AM Page iv 

Columbia University Press 

Publishers Since 1893 

New York Chichester, West Sussex 

© 2002 Columbia University Press 

All rights reserved 

Library of Congress Cataloging-in-Publication Data 

The cerrados of Brazil : ecology and natural history of a neotropical 

savanna / Paulo S. Oliveira and Robert J. Marquis. 

p. cm. 

Includes bibliographical references. 

ISBN 0-231-12042-7 (cloth : alk. paper)—ISBN 0-231-12043-5 

(pbk. : alk. paper) 

1. Cerrado ecology—Brazil. I. Oliveira, Paulo S., 1957– 

II. Marquis, Robert J., 1953– 

QH117 .C52 2002 

577.4'8'0981—dc21 2002022739 

Columbia University Press books are printed on permanent 

and durable acid-free paper. 

Printed in the United States of America 

c 10 9 8 7 6 5 4 3 2 1 

p 10 9 8 7 6 5 4 3 2 1

00 oliveira fm 7/31/02 8:11 AM Page v 

Contents 

Preface vii 

1 Introduction: Development of Research in the Cerrados 1 

Paulo S. Oliveira and Robert J. Marquis 

I 

Historical Framework and the Abiotic Environment 

2 Relation of Soils and Geomorphic Surfaces in the 

Brazilian Cerrado 13 

Paulo E. F. Motta, Nilton Curi, and Donald P. Franzmeier 

3 Late Quaternary History and Evolution of the Cerrados 

as Revealed by Palynological Records 33 

Marie-Pierre Ledru 

4 The Fire Factor 51 

Heloisa S. Miranda, Mercedes M. C. Bustamante, 

and Antonio C. Miranda 

5 Past and Current Human Occupation, and Land Use 69 

Carlos A. Klink and Adriana G. Moreira 

II 

The Plant Community: Composition, 

Dynamics, and Life History 

6 Vegetation Physiognomies and Woody Flora of the 

Cerrado Biome 91 

Ary T. Oliveira-Filho and James A. Ratter 

7 Herbaceous Plant Communities 121 

Tarciso S. Filgueiras 

8 Patterns and Dynamics of Plant Populations 140 

Raimundo P. B. Henriques and John D. Hay 

9 The Role of Fire in Population Dynamics of Woody Plants 159 

William A. Hoffmann and Adriana G. Moreira 

v

00 oliveira fm 7/31/02 8:11 AM Page vi 

vi contents 

10 Ecophysiology of Woody Plants 178 

Augusto C. Franco 

III 

The Animal Community: Diversity and Biogeography 

11 Lepidoptera in the Cerrado Landscape and the Conservation 

of Vegetation, Soil, and Topographical Mosaics 201 

Keith S. Brown Jr. and David R. Gifford 

12 The Character and Dynamics of the Cerrado Herpetofauna 223 

Guarino R. Colli, Rogério P. Bastos, and Alexandre F. B. Araujo 

13 The Avifauna: Ecology, Biogeography, and Behavior 242 

Regina H. F. Macedo 

14 The Cerrado Mammals: Diversity, Ecology, and Natural 

History 266 

Jader Marinho-Filho, Flávio H. G. Rodrigues, 

and Keila M. Juarez 

IV 

Insect-Plant Interactions 

15 Ant Foraging on Plant Foliage: Contrasting Effects 

on the Behavioral Ecology of Insect Herbivores 287 

Paulo S. Oliveira, André V. L. Freitas, and Kleber Del-Claro 

16 Interactions Among Cerrado Plants and Their Herbivores: 

Unique or Typical? 306 

Robert J. Marquis, Helena C. Morais, and Ivone R. Diniz 

17 Pollination and Reproductive Biology in Cerrado 

Plant Communities 329 

Paulo E. Oliveira and Peter E. Gibbs 

V 

The Conservation of the Cerrados 

18 Biodiversity and Conservation Priorities in the 

Cerrado Region 351 

Roberto B. Cavalcanti and Carlos A. Joly 

Contributors 369 

Index 373

00 oliveira fm 7/31/02 8:11 AM Page vii 

Preface 

This is a book about the Cerrado Biome, a major Brazilian 

savanna-like ecosystem for which no such summary exists. Biologists 

outside Brazil know little about the cerrados, despite the fact that the 

biome covers approximately 22% of the country’s surface area, or 2 million 

km 2 . Even though much of the attention of conservationists has 

focused on rainforests such as the Amazon and Atlantic forests, the cerrados 

are currently one the most threatened biomes of South America due 

to the rapid expansion of agriculture. Nearly 50% of the cerrado region 

is currently under direct human use, and about 35% of its total natural 

cover has been converted into planted pastures and crops. The average 

annual rate of land clearing in the cerrados during 1970–1975 was nearly 

twice the estimated deforestation rate of the Amazon forest during 

1978–1988. Overall biodiversity for the Cerrado Biome, including all its 

physiognomic forms, is estimated at 160,000 species of plants, animals, 

and fungi. Endemicity of cerrado higher plants has recently been estimated 

at 4,400 species, representing 1.5% of the world’s total vascular 

plant species. Endemic vertebrates range from 3% (birds) to 28% (amphibians) 

of the species recorded. The cerrados are also unique in that they 

serve as corridors for species inhabiting neighboring biomes such as the 

Amazonian and Atlantic rainforests. For example, although endemicity is 

low among birds, 90% of the species breed in the cerrado region. Given 

their geographic extent, it is surprising that the cerrados remain largely 

ignored at the international level. Because of the threatened status and 

rich biodiversity of this Neotropical savanna, and the lack of familiarity 

with cerrado ecosystems at the international level, a volume that compiles 

the known natural history, ecology, and biogeography of this biome is 

extremely timely. 

This is perhaps the first volume in English covering a tropical ecosystem 

in which the vast majority of the contributors are from the region in 

question. The foreign exceptions include scientists that are very familiar 

with the cerrados and have long-lasting collaborations with Brazilian 

researchers. The volume is broad in scope and raises relevant ecological 

questions from a diversity of fields, indicating areas in which additional 

vii

00 oliveira fm 7/31/02 8:11 AM Page viii 

viii preface 

research is needed. Such a wide thematic approach should provide the 

international audience with a broad ecological framework for understanding 

the cerrado savanna. The editors hope that such a book will 

make an important contribution for ecology, and for tropical biology in 

particular, stimulating future research in the cerrados. 

The idea of preparing a book summarizing research on cerrado biology 

arose in 1997 in San José, Costa Rica, during a most exciting meeting 

of the Association for Tropical Biology. As the book project developed, 

a number of people helped us shape the scope of the volume, 

establishing the main research areas to be covered, adjusting chapter contents, 

and writing the book proposal. At the early stages we have benefited 

greatly from the encouragement as well as the technical and editorial 

experience of Susan E. Abrams of the University of Chicago Press and 

Peter W. Price of Northern Arizona University. Helpful suggestions were 

also given by Keith S. Brown, William A. Hoffmann, Regina Macedo, Ary 

T. Oliveira-Filho, and Guy Theraulaz. Humberto Dutra helped with the 

preparation of the book index, and Glauco Machado and André Freitas 

helped with the scanning and printing of the figures. Mailing costs were 

covered in part by the Ecology Graduate Program of the Universidade 

Estadual de Campinas. 

Each chapter was substantially improved by the comments and suggestions 

of external reviewers. They include Steve Archer, John A. Barone, 

Kamaljit S. Bawa, John G. Blake, Keith S. Brown, Ray B. Bryant, Phyllis 

D. Coley, Philip J. DeVries, Peter E. Gibbs, Guillermo Goldstein, Gary S. 

Hartshorn, W. Ronald Heyer, Peter Kershaw, W. John Kress, Thomas H. 

Kunz, Diana Lieberman, Arício X. Linhares, Vera Markgraf, Ernesto 

Medina, Daniel C. Nepstad, Ary T. Oliveira-Filho, James L. Patton, A. 

Townsend Peterson, Ghillean T. Prance, Peter W. Price, James A. Ratter, 

José F. Ribeiro, Juan F. Silva, Robert B. Srygley, and Laurie J. Vitt. We 

appreciate the time they took to give critical reviews. 

Finally, we thank Science Editor Holly Hodder and Assistant Editor 

Jonathan Slutsky, formerly of Columbia University Press, for their initial 

encouragement and advice on the development of this project. Current 

Assistant Editor Alessandro Angelini helped at the final stage of the editing 

process, and Diana Senechal copyedited the entire manuscript. We are 

especially grateful to Julie S. Denslow and Lucinda A. McDade, reviewers 

of the book proposal for Columbia University Press, for their careful 

and constructive suggestions concerning the initial book project. 

Paulo S. Oliveira 

Robert J. Marquis

00 oliveira fm 7/31/02 8:11 AM Page ix 

The Cerrados of Brazil

00 oliveira fm 7/31/02 8:11 AM Page x

01 oliveira ch 1-2 7/31/02 8:12 AM Page 1 

1 

Introduction: Development 

of Research in the Cerrados 

Paulo S. Oliveira and Robert J. Marquis 

The first detailed account of the Brazilian cerrados 

was provided by Danish botanist Eugene Warming (1892) in the book 

Lagoa Santa, in which he describes the main features of the cerrado vegetation 

in the state of Minas Gerais. Since the publication of Warming’s 

book a number of descriptive studies from several cerrado regions in 

Brazil have been published. The vast majority of this literature is in Portuguese 

and oriented mostly toward botanical aspects of the cerrado. The 

studies can be roughly categorized into two major groups: (1) Surveys of 

woody floras, frequently providing also the general physiognomic characteristics 

of the vegetation (thorough reviews of this literature are given 

by Eiten 1972; Goodland and Ferri 1979). (2) Studies on plant ecophysiology 

focusing particularly on mineral nutrition, fire, and water economy 

at the plant-soil and plant-atmosphere levels; and on how these factors 

can account for the characteristic xeromorphic aspect of cerrado woody 

plants (extensive lists of these studies are given by Labouriau 1966; Ferri 

1977; Goodland and Ferri 1979). 

The cerrados gained international attention in the early 1970s after 

the influential works of Goodland (1971), Eiten (1972), and Ratter et al. 

(1973). These studies established quantitative parameters (i.e., canopy 

and ground cover, tree density, species richness) to characterize the several 

physiognomic forms of the cerrado vegetation; provided quantitative 

and comparative data toward the analyses of shifts in floristic composition 

along intergrading physiognomic communities (both over geographical 

and local scales); and enhanced the notion that the cerrado complex 

is the interactive product of climatic, topographic, and edaphic factors. 

1


2 introduction 

One may say with justice that these works have set the very basic grounds 

for modern ecological research in the cerrados. 

PATTERNS OF RESEARCH PRODUCTIVITY 

To understand the development and scope of scientific research in the cerrados, 

we have analyzed the bibliography in the form of journal articles 

appearing in the citation databases of the Institute of Scientific Information 

(ISI). We compiled the list by using cerrado and cerrados as “Topic 

Search’’ terms. Our goal was to detect changes in the quantity of published 

research papers over time, as well as in the subject matter treated. First 

we examined the general research productivity from 1966 to 1999, and 

assigned each study to one of seven major subject areas, as indicated in 

table 1.1. We treated zoology, botany, and mycology as separate areas to 

illustrate the allocation of research effort toward studies of animals, 

plants, and fungi. 

In a second phase of the analysis we assigned each article in the ecology 

category to one of six main research areas, in accordance with the 

thematic scheme employed by McDade and Bawa (1994), as summarized 

in table 1.2. Studies linked with agriculture, cattle, and wood industry, 

however, are not placed under the ecology category, because their research 

Table 1.1 Major Subject Categories Used to Analyze Patterns 

of Research Productivity in the Brazilian Cerrados 

Major Subject Areas Fields of Research 

Ecology General ecology, interspecific ecology, community 

ecology, physiological ecology, ecosystem 

ecology, applied ecology, and conservation 

Zoology, botany, and Species descriptions, species lists, systematic 

mycology biology, anatomy, morphology, physiology, 

genetics, and chemistry of organisms 

Soils Chemical and physical properties of soils, geology 

and geomorphology, and soil microbiology 

Agriculture, sylviculture, Any research linked with the use of cerrado areas 

and livestock for the raising of crops, commercial trees (timber 

industry), or cattle 

Gas emission and landsat Satellite sensing of fires, smoke, and regional 

mapping energy budgets, gas emission, climate, and landsat 

mapping of vegetation


introduction 3 

Table 1.2 Main Research Areas Used to Analyze Patterns 

of Ecological Research in the Brazilian Cerrados 

Areas of 

Ecological Research Specific Types of Research 

General ecology Biology of individual species, life history, 

demography, and behavior 

Interspecific ecology Interactions between species, including pollination, 

frugivory, herbivory, parasitism, and predation 

Community ecology Structure, dynamics, and organization in space and 

time of plant and animal communities 

Physiological ecology Physiological adaptations of organisms to the 

abiotic environment 

Ecosystem ecology Nutrient cycling, energy flow, and physical features 

of the habitat 

Applied ecology and Conservation of natural resources, biodiversity 

conservation 

Source: Based partially on McDade and Bawa (1994). 

approach and goals were generally not related to ecological issues 

(although the results could have a major ecological impact in the environment; 

see below). 

Although such thematic divisions are widely used in ecology textbooks 

and professional journals, obviously there are other ways of arranging 

research papers, as well as other recognizable thematic categories. 

In fact, as McDade and Bawa (1994) stress, the distinctions between such 

ecological thematic categories are sometimes arbitrary, and a given paper 

could probably be assigned to more than one category. In general, however, 

the assignment of papers to a category was quite easy. 

A final note on the accuracy of this bibliographic analysis. The assembled 

literature is of course incomplete, because it does not include several 

of the Brazilian publications which are not compiled by the ISI, including 

local journals, books, and symposium volumes. We believe, however, that 

such a compilation of articles does provide a general pattern of research 

productivity in the cerrados. 

The results show that research on the cerrados has increased markedly 

over the last two decades, especially over the past ten years (see fig. 1.1A). 

Studies linked with the use of cerrado areas for agriculture and pasture 

accounted for 24% of the papers (see fig. 1.1B). The ever-increasing


Figure 1.1 Research productivity in the Brazilian cerrados as compiled by the 

citation databases of the Institute of Scientific Information (ISI), using cerrado 

and cerrados as topic search terms. (A) General research over time. (B) Distribution 

of research articles by major thematic categories.



exploitation of natural cerrado areas for growing crops, trees (Pinus and 

Eucalyptus), and cattle, and the clearings caused by these practices, has 

urged the necessity of satellite measurements of gas emission and vegetation 

cover within the cerrado region in the late 1990s (fig. 1.1B). Research 

on soil properties and soil microbiology comprised 14% of the papers 

compiled. 

Studies on ecology, zoology, botany, and mycology comprised 54% 

of all publications assembled, ranging from less than five papers in 1990 

to about 35 papers per year in the late 1990s (see fig. 1.2A). This burst of 

biological research on the cerrados results from the founding of the first 

ecologically oriented graduate programs in Brazil in the 1970s. Some of 

these programs included field courses of 4–5 weeks in natural reserves 

where students developed field projects, some of which eventually led to 

theses. Such initiatives have resulted in the remarkable development of 

natural history and ecological research in a number of Brazilian ecosystems, 

including the cerrados. Originating mostly from the graduate programs 

of the public Universities in São Paulo (Southeast Brazil) and 

Brasília (Central Brazil), numerous student theses were developed in the 

cerrado savanna. In the state of São Paulo, 203 university theses were produced 

between 1966 and 1999. In the University of Brasília (UnB), located 

at the very core of the cerrado distribution, 62 theses were produced 

between 1997 and 1999. (Data assessed through the library databases of 

the public Universities of São Paulo, and the University of Brasília; compiled 

by using cerrado and cerrados as search terms.) 

Ecological research in cerrado has concentrated mostly in the three 

major fields of community ecology, general ecology, and interspecific ecology 

(see fig. 1.2B), which are also among the main ecological research areas 

investigated in Central American tropical forests (McDade and Bawa 

1994; Nadkarni 2000). Perhaps for historical reasons, studies on community 

ecology have been plant-oriented and have focused mainly on vegetation 

structure and dynamics, including paleoecology. Ecological studies on 

vertebrates were usually grouped under general ecology and, to a lesser 

extent, community ecology. They have been mostly oriented toward mammals, 

birds, and lizards, and generally have dealt with patterns of space 

use, feeding behavior, guild structure, and biogeography. Invertebrate 

research, generally incorporated into interspecific ecology, comprises studies 

on insect-plant interactions, in particular herbivory, pollination, and 

multitrophic associations. A comparatively small number of studies have 

reported results on physiological ecology (mostly plants), ecosystem ecology 

(nutrient cycling, fire ecology), and conservation (biodiversity inventories). 

Research areas that are clearly poorly represented include animal


Figure 1.2 Ecology and natural history research in cerrados, as compiled 

by the citation databases of the Institute of Scientific Information (ISI), using 

cerrado and cerrados as topic search terms. (A) Number of articles in ecology, 

zoology, botany, and mycology over time. (B) Distribution of ecological 

research by subject matter.



ecophysiology, chemical ecology, invertebrates (except butterflies, and 

social insects), large mammals, wildlife management, aquatic biology and 

hydrology, and landscape ecology. 

SCOPE AND ORGANIZATION OF THE BOOK 

The purpose of this book is to provide a picture of the Cerrado Biome 

based on broad synthetic treatments by experts from a diversity of 

research areas. Although the book has chapters whose approach is by 

necessity mostly descriptive, it also focuses on basic conceptual issues in 

evolutionary ecology and ecosystem functioning, and points toward 

future research avenues. Authors were instructed to write for an interdisciplinary 

audience, giving broad synthetic views within their specialties 

and making the text palatable enough to attract the interest of nonexperts 

as well as graduate students. As such, it is intended to provide an in-depth 

summary of current understanding for researchers versed in the field, as 

well as an introduction to cerrado biology for the mostly uninitiated international 

community. The book also provides a synthesis of the extensive 

cerrado literature in Portuguese, generally not easily accessible by the 

international audience. Similar volumes exist for African savannas alone 

(Sinclair and Norton-Griffiths 1979; Sinclair and Arcese 1995), and for 

Australian and African savannas (Werner 1991), but there is no equivalent 

for Brazilian cerrados. Most of the literature on neotropical savannas 

emphasizes the savannas of the northern parts of the South and Central 

Americas (see Sarmiento 1984), which do not have the extension and 

the rich biodiversity of the savannas of central Brazil (Dias 1992; Myers 

et al. 1999). Moreover, most studies on neotropical savannas have focused 

mainly on vegetation-related processes. A recent attempt toward a more 

multidisciplinary approach can be found in Solbrig et al. (1995). 

This volume treats the historical origins and physical setting, the role 

of fire, major biotic taxa, insect-plant interactions, and functional 

processes at different levels of organization (population and community) 

and scale (local and landscape). The book is organized in five sections, as 

follows: 

Part I provides the historical background and presents the main abiotic 

properties of the cerrado region. Geology, geomorphology, climatic 

influence, palynology, fire ecology, and history of human influence are 

treated in chapters 2–5. 

Part II focuses on the plant community and begins with the description 

of the vegetation physiognomies and the origins of the cerrado biome 

(chapter 6), followed by the main attributes of the herbaceous layer



(chapter 7). Population characteristics of trees in the absence and presence 

of fire, including spatial patterns and growth and mortality rates, are 

treated in chapters 8 and 9. The section concludes with the ecophysiological 

strategies of cerrado woody plants in chapter 10. 

Part III gives a general picture of the animal community, focusing on 

what are probably the five best-known animal taxa of the cerrados. Chapter 

11 examines the communities of plant-feeding Lepidoptera (bestknown 

invertebrate group) in conjunction with the complex landscape 

mosaics in the cerrado region. The diversity, biogeography, and natural 

history of the four best-known major vertebrate groups (amphibians, reptiles, 

birds, and mammals) are treated in chapters 12–14. 

Part IV covers those species interactions in the cerrado that are currently 

best documented: namely, insect-plant systems. Chapters 15 and 16 

deal with herbivorous insects, and chapter 17 treats the flowering plant 

pollination systems. 

Chapter 18 of Part V closes the book by examining the state of preservation 

of the cerrado ecosystem, the current threats to its biodiversity, and 

the appropriate strategies to be adopted based on the identification of priority 

areas deserving immediate conservation actions. 

We would like to comment briefly on a nomenclatural norm to be followed 

throughout the book. The Portuguese word cerrado means “halfclosed,’’ 

“closed,’’ or “dense,’’ and the name is particularly appropriate 

because this vegetation is neither open nor closed (Eiten 1972). The whole 

biome is characterized by an extremely variable physiognomy, ranging 

from open grassland to forest with a discontinuous grass layer. Between 

these two extremes lies a continuum of savanna formations spanning the 

entire range of woody plant density, referred to collectively as the cerrados. 

As we shall see in chapter 6, there are several physiognomic “types’’ 

of cerrado vegetation that can be recognized along this gradient (Goodland 

1971) and that are commonly designated by Portuguese terms. For 

instance, dry grassland without shrubs or trees is called campo limpo 

(“clean field’’); grassland with a scattering of shrubs and small trees is 

known as campo sujo (“dirty field’’). Where there are scattered trees and 

shrubs and a large proportion of grassland, the vegetation is termed 

campo cerrado (“closed field’’). The next stage is known as cerrado (sensu 

stricto) and consists of a vegetation dominated by 3–8-m-tall trees and 

shrubs with more than 30% crown cover but with still a fair amount of 

herbaceous vegetation between them. The last stage is an almost closed 

woodland with crown cover of 50% to 90%, made up of 8–12-m-tall 

trees casting considerable shade so that the ground layer is much reduced. 

This form is called cerradão. Clearly, the dividing line between these



physiognomies is somewhat arbitrary, but researchers usually agree surprisingly 

well on the classification. Other formations commonly associated 

with the cerrado landscape will be referred to by their local names 

(e.g., veredas, campo de murundus). The Brazilian nomenclature will be 

used throughout the book because it is currently well accepted internationally, 

unambiguous, and appropriate. As a general rule, whenever a 

given “type’’ of vegetation physiognomy is referred to by its Brazilian 

name in some part of the book, the reader will be directed to chapter 6 

for a detailed description of that particular physiognomy. 

ACKNOWLEDGMENTS 

We are grateful to Ana Rabetti and Ana Carvalho, from the Biology Library 

of the Universidade Estadual da Campinas, for their most valuable help 

with literature compilation. Augusto C. Franco, William A. Hoffmann, and 

Ary T. Oliveira-Filho offered useful suggestions on the manuscript. 

REFERENCES 

Dias, B. F. S. 1992. Cerrados: Uma caracterização. In B. F. S. Dias, ed., Alternativas 

de Desenvolvimento dos Cerrados: Manejo e Conservação dos 

Recursos Naturais Renováveis, pp. 11–25. Brasília: Fundação Pró- 

Natureza. 

Eiten, G. 1972. The cerrado vegetation of Brazil. Bot. Rev. 38:201–341. 

Ferri, M. G. 1977. Ecologia dos cerrados. In M. G. Ferri, ed., IV Simpósio 

sobre o Cerrado, pp. 15–36. São Paulo: Editora da Universidade de São 

Paulo. 

Goodland, R. 1971. A physiognomic analysis of the “cerrado’’ vegetation of 

central Brazil. J. Ecol. 59:411–419. 

Goodland, R. and M. G. Ferri 1979. Ecologia do Cerrado. São Paulo: Editora 

da Universidade de São Paulo. 

Labouriau, L. G. 1966. Revisão da situação da ecologia vegetal nos cerrados. 

An. Acad. Bras. Ciênc. 38:5–38. 

McDade, L. A. and K. S. Bawa. 1994. Appendix I: Patterns of research productivity, 

1951–1991. In L. A. McDade, K. S. Bawa, H. A. Hespenheide, 

and G. S. Hartshorn, eds., La Selva: Ecology and Natural History of a 

Neotropical Rain Forest, pp. 341–344. Chicago: University of Chicago 

Press. 

Myers, N., R. A. Mittermeier, C. G. Mittermeier, G. A. B. Fonseca, and J. 

Kent. 2000. Biodiversity hotspots for conservation priorities. Nature 

403:853–858.



Nadkarni, N. M. 2000. Scope of past work. In N. M. Nadkarni and N. T. 

Wheelwright, eds., Monteverde: Ecology and Conservation of a Tropical 

Cloud Forest, pp. 11–13. Oxford: Oxford University Press. 

Ratter, J. A., P. W. Richards, G. Argent, and D. R. Gifford. 1973. Observations 

on the vegetation of northeast Mato Grosso: I. The woody vegetation 

types of the Xavantina-Cachimbo Expedition area. Phil. Trans. 

Royal Soc. London B 266:499–492. 

Sarmiento, G. 1984. The Ecology of Neotropical Savannas. Cambridge, MA: 

Harvard University Press. 

Sinclair, A. R. E. and P. Arcese, eds. 1995. Serengeti II: Dynamics, Management, 

and Conservation of an Ecosystem. Chicago: University of 

Chicago Press. 

Sinclair, A. R. E. and M. Norton-Griffiths, eds. 1979. Serengeti: Dynamics of 

an Ecosystem. Chicago: University of Chicago Press. 

Solbrig, O. T., E. Medina, and J. F. Silva, eds. 1996. Biodiversity and Savanna 

Ecosystems Processes: A Global Perspective. Berlin: Springer-Verlag. 

Warming, E. 1892. Lagoa Santa: Et bidrag til den biologiske plantegeographi. 

Copenhagen: K. danske vidensk Selsk., 6. 

Werner, P. A., ed. 1991. Savanna Ecology and Management. Oxford: Blackwell 

Scientific.


2 

Relation of Soils and Geomorphic 

Surfaces in the Brazilian Cerrado 

Paulo E. F. Motta, Nilton Curi, and Donald P. Franzmeier 

The cerrado region is located between the equatorial 

zone and 23° south latitude. It is bordered by the Amazon forest to the 

north, by the Atlantic forest to the south and southeast, and by the 

caatinga (deciduous xerophytic vegetation) of the semiarid region to 

the northeast. Also included in the cerrado region is the nonflooded part 

of the western pantanal (wet plains; see chapter 6). During its evolutional 

process, the areal extent of the cerrado expanded and contracted in 

response to climatic fluctuations. During dry periods, the cerrado 

expanded at the expense of forest (Ab’Saber 1963). During wet periods, 

forest expanded at the expense of cerrado except in places that were 

depleted of plant nutrients and that presented some water deficiency 

(Resende 1976). Once established, the cerrado tends to maintain itself 

with more tenacity than other vegetation formations because the climate 

and soil factors that favor it are not extreme (Ker and Resende 1996). In 

contrast, other vegetation types are favored by more severe conditions. 

For example, the xerophytic caatinga is maintained by the very pronounced 

water deficiency in a semiarid climate. The pantanal, an extensive, 

low-lying waterlogged plain with hydrophytic grassland in the central-western 

region, is maintained by a severe oxygen deficiency. The 

cerrado region has great climatic diversity because of its wide latitudinal 

and altitudinal ranges. In addition to its 15° range in latitude, the cerrado 

varies in altitude from 100 m in the pantanal to 1,500 m in some of the 

more elevated tablelands of the Central Plateau. 

13


14 historical framework and the abiotic environment 

SOIL FORMATION PROCESSES AND TROPICAL SOILS 

In this chapter we present soil characterization data, classify the soils 

according to the Brazilian soil classification system (Embrapa 1999) and 

U.S. Soil Taxonomy (Soil Survey Staff 1999), and discuss how soil properties 

affect plant growth. The next section provides background for the 

subsequent sections of the chapter. 

Soil Formation Processes 

The relationship of soils to their environment is explained by the equation, 

s = f(cl, o, r, p, t, . . .), 

which shows that any soil property (s) is a function of regional climate 

(cl), organisms (o), landscape position or relief (r), geologic parent material 

(p), time (t), and possibly additional factors (. . .). Many soils of the 

cerrado region formed from weatherable minerals (p) on old (t) land surfaces 

conducive to leaching because of their landscape position (r) in a 

warm climate (cl) where organisms (o) were very active. Together, the 

individual factors all contribute to the formation of highly weathered 

tropical soils in much, but not all, of the cerrado. They are called Latosols 

in the Brazilian soil classification system, and Oxisols in the U.S. (comprehensive) 

system. 

The 10 most abundant elements in the earth’s crust are O > Si > Al > 

Fe > Ca > Mg = Na > K > Ti > P (Sposito 1989). The fate of these elements 

during soil formation provides an overview of soil formation. Minerals 

and rocks from which soils form are made up mainly of the first eight elements 

of the list, and clay minerals are composed mainly of the first three. 

Oxygen is unique among the 10 elements. It has a negative charge and is 

much larger than the others—so large that most of the other, positively 

charged, elements fit within a “stack’’ of Os and balance their negative 

charge. 

During soil formation, parent rocks weather and release weathering 

products that are leached from the soil or remain in the soil and combine 

to form clay minerals, many with a negative charge. Most base cations 

(Ca, Mg, Na, and K) are leached from the soil if they are not held by negative 

charges on clay minerals. Some of the Si released in weathering is 

leached, and some remains to form clay minerals. Al weathering products 

are mainly insoluble and remain in the soil. In freely drained soils, Fe also 

tends to remain in the soil as Fe-oxide minerals such as goethite and


Relation of Soils and Geomorphic Surfaces 15 

hematite. In wet soils, Fe-oxides are reduced and dissolved, and soluble 

Fe 2+ is leached. In summary, the mobility of elements in the soil follows 

the sequence, Ca > Na > Mg > K >> Si >> Fe > Al. The elements at the 

beginning of the sequence are major plant nutrients and are subject to 

leaching. Because they are so highly weathered, Latosols tend to be infertile 

and rich in Al and Fe. By this process, Fe-oxides accumulate in soils 

because other materials are lost, which could be called a passive accumulation 

of Fe. 

Iron can also accumulate in soils by active processes. When the water 

table is high and soils are saturated, oxygen is not available to accept electrons 

produced by microbial respiration, so they are accepted by Fe 3+ , 

resulting in reduction to Fe 2+ which can move within the soil profile and 

landscape. When the water table is low, oxygen becomes available, and 

Fe 2+ is oxidized to Fe 3+ and precipitates as iron oxide minerals to form 

Fe-rich soil materials in subsurface horizons. When first formed in soils, 

this material is soft. When it dries, it hardens irreversibly, meaning that 

it does not soften up when the soil becomes moist. Previously, both the 

hard and soft materials were called laterite. In order to distinguish 

between the two forms, the soft material was called plinthite, and the hard 

material was called ironstone in early versions of Soil Taxonomy (Soil 

Survey Staff 1999). In the Brazilian Soil Classification (Embrapa 1999), 

these materials are called plinthite and petroplinthite, respectively. Adjectival 

forms of these words are used in the names of many soil classes. 

Depending on the size of the original Fe concentrations in the soil, 

plinthite may harden into small (sand- and gravel-size), large (gravel and 

cobbles), or even continuous masses of petroplinthite when the soil dries. 

Various kinds of clay minerals form during soil formation. They are 

made up of sheets composed of Si and O and of Al and O. One way to 

describe different clay minerals is by the number of Si sheets and Al sheets 

in their structure. Thus, 2:1 clay minerals have two Si sheets and one Al 

sheet. Examples are mica, smectite, vermiculite, and illite. In the structure 

of these minerals, Al 3+ may substitute for Si 4+ , which leaves an extra negative 

charge on the clay surface to which cations such as Ca 2+ are attracted. 

This Ca is called exchangeable Ca, because it can exchange with 

other cations in the soil solution, and the total charge on the mineral is 

called cation exchange capacity, CEC. Soils on young land surfaces tend 

to be rich in 2:1 clay minerals. 

Kaolinite, a 1:1 clay mineral, and gibbsite (Al(OH) 3 ), a 0:1 clay mineral 

with no Si and little or no CEC, are abundant in Latosols, especially 

kaolinite. In the course of soil formation, base cations are leached and 

clays lose CEC. The two processes complement each other with the result



that Latosols have very low contents of exchangeable base cations and are 

thus infertile. When base cations are removed from negative sites, they are 

first replaced with H + which makes the soil acid, but later acid Al-compounds 

replace H + . 

Soil Characterization 

Tables 2.1–2.3 present characterization data for the main soils of the cerrado. 

The discussion below explains how the properties reported in these 

tables relate to soil formation processes, soil classification, and soil fertility. 

Color. Three attributes of color are represented in a Munsell designation 

such as 5YR 4/8. Hue (5YR) represents the spectral colors (Y = yellow, 

R = red). Soil hues grade from yellowish to reddish in the 

sequence 2.5Y, 10YR, 7.5YR, 5YR, 2.5YR, 10R. Value (4) represents 

the relative darkness, from black ≈ 2, to light or pale ≈ 8. 

Chroma (8) represents the purity of the hue. Chroma = 0 is a blackwhite 

transition, and chroma ≈ 8 is relatively pure red, yellow, etc. 

Soil color has several important interpretations. Low chroma (≤ 2) 

indicates soil wetness and lack of Fe-oxides. Hue, with higher chromas, 

indicates the kind of Fe-oxide minerals present and is used to 

subdivide Latosols. Hematite is reddish, and goethite is yellowish. 

Yellow Latosols have 10YR and 7.5YR hues, and goethite is dominant. 

Red-Yellow Latosols have 5YR hue, and neither mineral dominates 

the color. Red Latosols have 2.5YR and redder hues, and 

hematite is dominant. 

s, r (silt, clay). Represents soil texture, the relative contents of sand, silt, 

and clay. Sand = 1,000 – s – r. Other factors being similar, more 

weathered soils contain more clay than less weathered ones. Most 

Latosols are rich in clay. 

C (organic carbon). C oxidizes readily in tropical soils, but the C content 

in subsoils is high relative to well-drained soils of temperate areas, 

probably because of ant and termite activity. 

pH. Soil pH is a measure of soil acidity. pH is measured in both water and 

KCl solution. In KCl, K + replaces H + and other cations, and Cl – 

replaces mainly OH – . If the soil has more cation exchange capacity, 

CEC, than anion exchange, AEC, more H + is replaced than OH – , 

and the pH is lower in KCl than in water. Then, ∆pH, pH H2O – pH KCl 

is positive. On the other hand, a negative ∆pH indicates that AEC is 

larger than CEC and that the soil has a net positive charge. Such a 

soil could adsorb more NO 3 – than K + or NH4 + , for example.



T (cation-exchange-capacity, CEC). Total negative charge in soil measured 

at pH 7. It originates mainly in clay particles and organic matter. 

S (sum of bases). Amount of CEC that is balanced by base cations (Ca 2+ , 

Mg 2+ , K + , Na + ). T – S = acidity (H + or Al-compounds) on exchange 

sites. Soils lose base cations and soil fertility during weathering. 

V (base saturation). V = (S/T) × 100. The lower the value, the more 

leached (and weathered) the soil and the less its supply of plant-available 

Ca, Mg, and K. For reference, V ranges up to 100%. 

m (Al saturation). The percentage of negative sites balanced by positively 

charged Al-compounds. Soils are considered to be high in Al (allic) 

if the extractable Al content is > 0.5 cmol c /kg soil and m ≥ 50%. Al 

may be toxic to some plant roots growing in these soils. If roots are 

stunted they are limited in their ability to take up water, so plants 

may show drought symptoms. 

Fe 2 O 3 (content of Fe-oxides, mostly as goethite and hematite). These minerals 

may also be a source of positive charge in soils. 

TiO 2 (Ti-containing minerals are very resistant to weathering). Generally, 

the higher the content, the more weathered the soil. 

K i (molar SiO 2 /Al 2 O 3 ratio of the clay fraction). K i decreases with the 

degree of weathering of the soil. Latosols must have K i < 2.2 and usually 

< 2.0. 

K r (molar SiO 2 /(Al 2 O 3 + Fe 2 O 3 ) ratio of the clay fraction). K r > 0.75 indicates 

that the clay fraction has significant kaolinite content, and K r 

< 0.75 indicates that it consists mainly of oxides. 

Plant-Soil Relations 

Latosols tend to have good physical but poor chemical properties relative 

to plant growth. The good physical properties are mainly due to high 

aggregate stability. Aggregates of clay (largely kaolinite and gibbsite) are 

stabilized by high contents of Fe- and Al-oxides, by organic matter, or 

both. Strong aggregate stability allows water and air to move through the 

soil readily and permits roots to penetrate with little resistance. Stable 

aggregates are also less subject to erosion than unstable ones. 

Latosols are low in plant nutrients, especially P and Ca, and many are 

low in micronutrients. In many cases the Al content is so high that it is 

toxic to plant roots. Large applications of lime and P fertilizer are needed 

to make these soils productive for agricultural crops. Lime (CaCO 3 ) neutralizes 

some of the acidity, decreases available Al levels, and increases the 

amount of Ca 2+ on exchange sites and thus available to plants. Large 

applications of P are required because much of the fertilizer P is tied up

Table 2.1 Color, Physical, and Chemical Attributes of Selected Horizons of the Soils of the First Geomorphic Surface Horizon 

s r Org. C pH S T Fe 2 O 3 TiO 2 

Depth Color V m 

Horizon a (cm) moist (g/kg) H 2 O KCl (cmol c /kg) (%) (%) (g/kg) K i K r 

Typic Acric RED LATOSOL 

Ap 0–17 2.5YR 3/5 100 690 19.5 4.7 4.6 2.1 9.3 23 13 130 9.5 0.72 0.54 

Bw2 106–150 2.5YR 4/6 100 730 5.8 4.9 6.0 0.4 2.2 18 0 133 11.2 0.72 0.55 

Typic Dystrophic RED-YELLOW LATOSOL 

A 0–18 5YR 4/4 110 820 15.2 4.7 3.9 1.0 12.4 8 47 96 12.0 0.60 0.51 

Bw 70–100 5YR 5/8 100 790 11.4 5.0 4.7 0.9 6.6 14 10 103 12.6 0.61 0.51 

Typic Acric RED-YELLOW LATOSOL 

A 0–17 5YR 3/3 130 750 22.7 4.5 4.4 0.5 9.2 5 71 114 10.8 0.53 0.43 

Bw2 93–170 5YR 4/8 80 820 9.2 5.3 5.8 0.3 3.4 9 0 120 13.3 0.54 0.43 

Endopetroplinthic Dystrophic YELLOW LATOSOL 

Ap 0–19 10YR 4/3 150 660 17.5 5.2 4.1 4.0 7.8 51 0 122 14.3 0.62 0.48 

Bw2 119–155 7.5YR 5/6 90 790 6.5 5.0 4.1 0.4 2.7 15 0 130 17.8 0.63 0.48 

Bwcf 155–189 7.5YR 5/6 100 770 5.7 4.7 4.6 0.4 2.2 18 0 119 17.5 0.67 0.53 

Petroplinthic Acric YELLOW LATOSOL 

Acf 0–25 7.5YR 4/2 70 310 13.7 4.6 4.1 0.6 7.2 8 67 156 10.6 0.71 0.42 

Bwcf1 78–135 7.5YR 5/6 90 620 8.0 4.7 4.6 0.4 4.0 10 33 156 11.0 0.71 0.50 

Bwcf2 135–220 7.5YR 5/8 120 610 4.8 4.1 5.6 0.3 2.3 13 0 174 11.0 0.73 0.49 

Typic Dystrophic HAPLIC PLINTHOSOL 

AB 5–17 10YR 7/2 130 700 19.0 5.2 5.4 0.5 4.8 10 0 54 13.3 0.31 0.28 

Bf 17–38 2.5Y 7/4 150 720 9.7 5.4 6.6 0.5 2.4 21 0 107 14.9 0.35 0.29 

Bgf2 75–120 2.5Y 7/4 160 690 3.4 6.1 7.3 0.3 0.9 33 0 118 15.3 0.40 0.33 

Source: Embrapa (2001). 

Abbreviations: s = silt; r = clay; S = sum of bases; T = cation-exchange-capacity; V = base saturation; m = Al saturation. 

a p = pedoturbation, w = intensive weathering, c = indurated concretions, f = plinthite, g = gley (Embrapa, 1988). 

01 oliveira ch 1-2 7/31/02 8:12 AM Page 18

Table 2.2 Color, Physical, and Chemical Attributes of Selected Horizons of the Soils 

of the Second Geomorphic Surface Horizon 




Typic Acric RED LATOSOL 

A 0–22 5YR 4/4 150 390 11.1 4.8 4.5 0.4 3.9 10 33 57 5.2 0.64 0.53 

Bw2 105–160 2.5YR 4/6 160 440 3.4 5.6 5.8 0.4 1.4 29 0 68 6.0 0.62 0.51 

Typic Acriferric RED LATOSOL 

A 0–15 2.5YR 3/6 190 590 21.5 4.8 4.3 0.5 9.5 5 44 255 25.9 0.63 0.35 

Bw 65–100 10R 3/6 180 610 10.9 5.1 5.2 0.4 4.9 8 0 263 26.5 0.58 0.33 

Typic Dystrophic RED-YELLOW LATOSOL 

Ap2 8–24 7.5YR 4/4 90 410 8.7 5.0 3.8 0.6 4.1 15 33 46 4.7 0.60 0.52 

Bw1 56–140 5YR 5/8 110 450 3.1 5.2 5.1 0.3 1.6 19 0 60 5.5 0.65 0.55 

Endopetroplinthic Acriferric YELLOW LATOSOL 

Ap 0–24 10YR 4/4 90 570 14.4 4.7 4.5 0.4 6.8 6 60 181 21.3 0.61 0.40 

Bw2 88–119 7.5YR 5/6 90 650 6.4 5.5 5.1 0.3 2.3 13 0 183 21.9 0.59 0.40 

Bwcf 119–160 7.5YR 5/6 100 710 4.7 5.3 6.2 0.4 1.5 27 0 181 18.3 0.61 0.42 


Abbreviations: s = silt; r = clay; S = sum of bases; T = cation-exchange-capacity; V = base saturation; m = Al saturation. 

a p = pedoturbation, w = intensive weathering, c = indurated concretions, f = plinthite (Embrapa, 1988). 


Table 2.3 Color, Physical, and Chemical Attributes of Selected Horizons of the Soils of the Third Geomorphic Surface Horizon 




Typic Orthic ARGILUVIC CHERNOSOL 

Ap 0–25 5YR 3/2 250 480 19.8 5.8 5.0 18.2 25.5 71 0 160 37.0 1.64 0.99 

Bt 37–90 2.5YR 3/6 190 600 15.0 6.1 5.5 14.2 17.6 81 0 157 27.0 1.53 1.02 

Typic Eutrophic RED ARGISOL 

A1 0–10 10YR 2/1 290 350 47.3 5.7 4.9 16.1 24.7 65 0 139 35.6 2.00 1.09 

Bt2 70–115 3.5YR 2.5/4 190 540 8.2 6.7 5.6 8.6 10.9 79 0 146 26.7 1.68 1.11 

Typic Eutrophic RED-YELLOW ARGISOL 

A 0–15 10YR 3/4 450 370 15.6 5.8 4.6 6.1 10.7 57 0 65 7.0 1.94 1.51 

Bt 50–70 5YR 5/6 340 530 4.9 5.8 4.9 4.2 6.6 64 0 95 7.0 1.78 1.36 

Typic Tb Dystrophic HAPLIC CAMBISOL 

A 0–10 10YR 3/3 250 340 15.9 4.6 3.9 1.8 7.8 23 63 55 4.2 2.13 1.54 

Bi 35–65 7.5YR 6/6 270 440 5.6 4.7 4.0 0.6 5.2 12 84 72 4.0 2.01 1.51 

Typic Tb Eutrophic HAPLIC CAMBISOL 

A 0–12 10YR 3/2 350 210 38.6 5.6 5.2 13.3 19.5 68 0 83 11.1 2.02 1.59 

Bi 31–55 5YR 5/6 360 160 3.0 5.7 4.8 2.2 3.9 56 0 94 11.6 1.86 1.62 

Typic Dystrophic LITHOLIC NEOSOL 

A 0–20 10YR 4/4 360 420 15.0 4.6 3.9 1.0 7.3 14 75 75 12.0 2.04 1.46 

Typic Tb Dystrophic FLUVIC NEOSOL 

A 0–20 10YR 3/2.5 410 450 29.9 4.9 4.2 4.8 14.9 32 25 58 6.4 1.68 1.42 

C3 70–120 Variegated 220 260 2.9 5.5 4.1 1.1 3.6 31 48 38 4.5 1.73 1.42 

Typic Tb Eutrophic FLUVIC NEOSOL 

A 0–20 10YR 4/2 620 250 14.5 4.9 4.5 5.2 6.8 76 19 37 3.7 1.86 1.59 

C2 40–60 10YR 4/2 390 140 5.7 5.5 4.8 3.0 4.9 61 0 21 2.5 1.90 1.64 


Abbreviations: s = silt; r = clay; S = sum of bases; T = cation-exchange capacity; V = base saturation; m = Al saturation. 

a p = pedoturbation, t = clay accumulation, i = incipient development (Embrapa, 1988). 




by Fe- and Al-oxides. Organic matter helps to hold the meager supply of 

plant nutrients in Latosols. 

GEOMORPHIC SURFACES AND SOILS 

The Central Brazil region constitutes a classic example of polycyclic landscape 

evolution, with both young (Pleistocene) forms and well-preserved 

remnants of much older surfaces (Lepsch and Buol 1988). Overall, three 

major geomorphic surfaces have been identified by Feuer (1956) in the 

area of the Federal District (FD). He called them the first, second, and 

third surfaces (see figs. 2.1, 2.2). A geomorphic surface is a portion of the 

landscape specifically defined in space and time (Ruhe 1969). The geomorphic 

surfaces consist of plains, the generally level or rolling surfaces, 

and bevels, erosion surfaces that cut and descend from a plain (Bates and 

Jackson 1987). 

First Geomorphic Surface 

The first surface (Surface I) corresponds to the peneplane formed during 

the arid South American erosion cycle (Braun 1971), and is often called 

the South American Surface. This cycle lasted long enough to affect 

almost all of the Brazilian landscape (King 1956; Suguio and Bigarella 

1979). Subsequent moister climatic conditions propitiated the deepening 

of the weathering mantle. After the epirogenic upliftings of the Medium 

Tertiary (King 1956), and the consequent lowering of the base level of erosion, 

dissection of this surface was initiated. 

In the region south of the Federal District, the high tablelands (900 

to 1,100 m altitude) with slopes of less than 3% are remnants of the South 

American surface. This surface is covered with a thick layer of Tertiary 

sediments (Radambrasil 1983). We know little about the origin and mode 

of deposition of these sediments. In part of the region, the edges of the 

remnants of this surface are covered by a thick layer of hard iron-rich fragments 

(petroplinthite). Highly resistant to erosion, this layer effectively 

protects and maintains the remnants of said surface. In other places, the 

plateau is protected by quartzitic mountain crests. Where there is no such 

protection, the tablelands are eroded rapidly by parallel slope retreat. 

Soils on Surface I 

The distribution of the soils on Surface I depends on the size of the 

tableland remnants. In wide remnants, the soil distribution is similar to



Figure 2.1 Schematic representation of the soils on Geomorphic Surfaces 

I, II, and III, south of the Federal District. RL = Red Latosol; RYL = Red- 

Yellow Latosol; PYL = Petroplinthic Yellow Latosol; EYL = Endopetroplinthic 

Yellow Latosol; ARL = Acriferric Red Latosol; RN = Red Nitosol; 

RA = Red Argisol; FN = Fluvic Neosol; C = Cambisol. 

that described by Macedo and Bryant (1987) for areas of the Federal District. 

From the center to the border there is a sequential occurrence of: 

Red Latosols (RL), Red-Yellow Latosols (RYL), and Petroplinthic Yellow 

Latosols (PYL) (see figs. 2.1 and 2.3a). They are all in the Oxisol order in 

Soil Taxonomy (Soil Survey Staff 1999). The PYL soil constitutes the 

major part of the borders with the second geomorphic surface. In the 

narrower remnants, RL does not occur, and the PYL soil, rich in ironstone 

fragments, is more widespread and in some places is the main soil 

of the area. 

These soils developed from fine sediments of unknown origin apparently 

not related to the underlying strata (Brasil 1962; Braun 1971). The 

soils differ from each other in the hydric regime or natural soil drainage 

along the gentle slopes. The Red Latosols on the higher areas near the center 

of the tableland remnant have good internal drainage as shown by 

their red color. The red color is due to hematite, an iron oxide mineral in 

the clay fraction which indicates an oxidizing environment (Cornell and



Figure 2.2 Schematic representation of the soils on Geomorphic Surfaces I, 

II, and III in northwest Minas Gerais state. RYL = Red-Yellow Latosol; RL = 

Red Latosol; C = Cambisol; QN = Quartzarenic Neosol; LN = Litholic 

Neosol. 

Schwertmann 1996). The Red-Yellow Latosols and the Petroplinthic Yellow 

Latosols both have the red-yellow colors of the iron oxide mineral 

goethite, which indicates moister soil conditions than in redder soils. The 

red-yellow and yellow soils are adjacent to soils with seepage zones and 

iron oxide concretions on the edges of the high tablelands (Macedo and 

Bryant 1987). Apparently Fe was reduced in the red-yellow soils, transported 

in solution to the edge of the tableland remnant, and oxidized there 

to form Fe-rich concretions. 

Hydromorphic soils occur near the borders of the tablelands where 

there are occasional springs associated with a drainage net. Hydromorphic 

soils are also common in scattered low-lying areas in the interior of 

the tablelands. These areas are generally flat with many microelevations 

or swells averaging 1 to 1.5 m high and 5 to 10 m in diameter (see fig. 

2.4A). According to Corrêa (1989) these swells are paleotermite mounds. 

Between the swells there are gently concave swales through which water 

flows much of the year. This topography, regionally called covoal (termite 

mounds), or murundu field (chapter 6), has been described in various



Figure 2.3 Map of Brazil, showing the schematic localization of the area 

(a) south of the Federal District (FD) and (b) northwest of the state of Minas 

Gerais. 

regions of the Central Plateau (Embrapa 1982; Corrêa 1989; Motta and 

Kämpf 1992; Resende et al. 1999). The depth of the water table in the 

soils fluctuates seasonally, resulting in the reduction, transport, and oxidation 

of Fe to form plinthite. The main soils in this landscape are 

Plinthosols (Aquox) and, in smaller proportion, Plinthic Red-Yellow 

Latosols. Both formed under hydrophytic grassland. Hydromorphic soils 

also occur in the veredas, small valleys with distinctive hydrophytic vegetation 

(Lima 1996) characterized by a tree-shrub set in which the buriti 

palm (Mauritia vinifera Mart.) predominates, and a grass zone in areas in 

which water seeps to the surface (see chapter 6). 

In part of the region, Surface I has a prominent border in which the 

soils contain much gravel, cobbles, and boulders of petroplinthite (see 

figs. 2.4B and 2.4C). Except for the high content of coarse fragments and 

the fact that clay contents increase with depth (table 2.1 and field observations), 

the B horizons are similar to latosollic B horizons of other soils 

of the high tablelands. In general, the surface on which these soils occur 

is relatively high in the tableland landscape. Apparently, nearby surfaces


Figure 2.4 (A) Undulating topography on Surface I due to covoal, or murundus 

(termite mounds). (B) Surface II in the foreground and surface I in the 

background, south of the Federal District. (C) Petroplinthite (ironstone) on 

the soil surface near the border of Surface I. (D) View of the region south of 

the Federal District, showing the bevel and escarpment between Surface I and 

Surface II. (E) Isolated elevation on a gentle undulated area of the second geomorphic 

surface. (F) Gully erosion on the borders of Surface II. (G) Red 

Nitosols and Red Argisols on Surface III. (H) Aspect of relief and vegetation 

on the erosion segment of the third geomorphic surface.



were lowered by erosion because the soils on them lacked the protective 

cover of petroplinthite. In other parts of the region, the bevel leading to 

the second surface consists of escarpments 100 to 200 m high (see fig. 

2.4D), probably due to the effect of tectonic movement of small geographic 

expression (Cline and Buol 1973). 

In the region of the Minas Gerais Triangle, the Surface I is 300 to 400 

m lower than the corresponding surface in the Federal District, and the 

distribution of soils does not follow a definite pattern. In the Triangle, 

there is a predominance of high-clay Red Latosols and Red-Yellow 

Latosols on flat and gently undulating topography. 

The soils of Surface I are very highly weathered and have very low 

natural fertility and a limited reserve of nutrients (table 2.1). The Ki ratio 

for all horizons is well below 2.0, the upper limit for Latosols, and the Kr ratio of most horizons is less than 0.75, which suggests that the clay fraction 

consists mainly of Fe- and Al-oxide minerals. This mineralogy is confirmed 

by the low or negative ∆pH (pHH2O – pHKCl ) values in most horizons. 

The rest of the clay is probably kaolinite. In addition, some soils 

have more than 50% aluminum saturation in the surface horizon (table 

2.1). The low nutrient supply and reserve capacity in these soils is illustrated 

by sum of bases, S, in table 2.1. In many subsoil horizons, the sum 

of Ca, Mg, and K is only 0.3 to 0.5 cmolckg –1 , a negligible quantity of 

these plant nutrients. 

Semideciduous tropical cerrado is the main form of native vegetation 

in the flat segment as well as in the borders of the first geomorphic surface, 

although there is occurrence of semiperennial tropical cerrado, and, 

in more restricted areas, semideciduous and semiperennial tropical forest, 

beyond hydrophytic grassland. 

Second Geomorphic Surface 

A cycle of pediplanation in the mid-Tertiary, called the “Velhas” erosion 

cycle by King (1956), initiated the dissection of the Surface I and the formation 

of the second geomorphic surface (Surface II). South of the Federal 

District, this surface is a plain that slopes downward in the main direction 

of water flow (see fig. 2.4E) from the borders of the first surface. 

Soils on Surface II 

Soils on the Surface II show more influence of the underlying bedrock than 

soils on Surface I. South of the Federal District, soils over Precambrian 

schists (Araxá Group) are clayey, those over mafic granulite (Anápolis-



Itauçu Granulitic Complex) are very clayey, and soils on quartzite (Araxá 

Group) are medium-textured. Soils formed largely from Tertiary sediments 

from the first surface are also very clayey. 

On Surface II, the relief is mainly gently undulating with small flat 

areas. The native vegetation is again cerrado, with small areas of semideciduous 

forest. In the region of the Minas Gerais Triangle, there is a predominance 

of cerradão (woodland savanna; chapter 6). In the Central 

Plateau, the predominant soils are Red Latosols and Red-Yellow Latosols, 

but there are small areas of Acriferric Red Latosols (ARL) mainly south 

of the Federal District (Cline and Buol 1973; Mothci 1977; Embrapa 

2001), where they formed from mafic rocks (Anápolis-Itauçu Granulitic 

Complex). On the higher parts of this surface, just below the borders of 

Surface I, some soils have strata rich in concretions (concentration masses) 

of petroplinthite. These concretions show no edges and are mixed with 

the latosollic (oxic) material at variable depths in profiles of Endopetroplinthic 

Yellow Latosols (EYL) (fig. 2.1 and table 2.2). The deposition of 

these concretions as stonelines, their mixing with the latosollic material, 

and the absence of morphological evidence of reducing conditions necessary 

for the formation of ironstone concretions suggest that these concretionary 

deposits originated from the borders of the first surface. The 

EYL occur in a zone adjacent to the escarpment base, while the Red 

Latosols or Acriferric Red Latosols are in a lower position, hundreds of 

meters from the EYL zone. 

In the northwest of the state of Minas Gerais (fig. 2.2), this surface 

cut Cretaceous sandstones (Areado Formation). The relief is dominantly 

gently undulating, and the soils are medium-textured Red-Yellow Latosols, 

and Cambisols (Inceptisols) or sandy-textured Neosols (Quartzarenic 

Neosols, QN, Entisols). Here the drainage net is more dense 

and the veredas more common. In the Minas Gerais Triangle, mediumtextured 

Red Latosols predominate, also developed from Cretaceous 

sandstones (Bauru Group) (Embrapa 1982). 

The mixture of this preweathered material with the less altered material 

that originated from weathering of the subjacent rocks constitutes the 

parent material of the soils of the Surface II. The relative proportion of 

these two materials influences the distribution and attributes of the soils. 

Surface II is being dissected by contemporary erosion processes creating 

the third surface; in many places the two surfaces are intertwined. 

In the hilly borders with many remnants of Surface II, the soils are very 

susceptible to gully erosion. Where native vegetation has been removed 

to facilitate farming, grazing, or mining projects, gully erosions may 

progress rapidly (see fig. 2.4F).



Although Surface I is much older than Surface II, the K i index, which 

expresses the degree of loss of silica from soils and their degree of weathering, 

did not show substantial differences (tables 2.1, 2.2), in agreement 

with the observations of Cline and Buol (1973). The index might even be 

smaller for soils on the Surface II; this might be explained, beyond the contribution 

of the preweathered material from Surface I, by the more freely 

drained conditions on Surface II, which accelerate the processes of weathering 

and leaching. The K r ratios are also small, and ∆pH is small or negative 

in several horizons of soils on Surface II (table 2.2). Similar to soils 

on Surface I, soils on Surface II have low natural fertility, and all of them 

have medium (20% to 40%) to high (>40%) levels of Al saturation in the 

plow layer (table 2.2). In subsoils, the cation exchange capacity is so low 

that the soil can hold neither base cations, which are needed by plants, or 

Al, which is detrimental to plants. 

Third Geomorphic Surface 

The third surface (Surface III) was formed when geologic erosion cut 

through Surfaces I and II, forming a new surface including an erosion segment 

with sloping relief and a depositional segment with gentle relief. 

South of the Federal District, this surface is still in the initial stage of development 

and includes little but the erosion segment. In other areas, however, 

the depositional segment is more extensive. 

In the northwest of the state of Minas Gerais (see figs. 2.2 and 2.3B), 

removal of sandy material from Surface II exposed Precambrian pelitic 

(clay-rich) rocks of the Bambuí Group. The weathering products of these 

rocks were subsequently reworked. Where the process of planation of the 

former surface is extensive, the new surface is called an “Exhumed Pre- 

Cretaceous Surface” (Cetec 1981). In the major part of the region, however, 

removal of the former surface is less extensive, and there is predominance 

of well-dissected landforms, resulting in many residual hills within 

the deposition zone. 

The depositional segment of Surface III is most extensive along the 

valleys of the main rivers, which start in the Central Plateau. These rivers 

are in the Amazon, São Francisco and Paraná basins. This segment is well 

expressed in the São Francisco Depression, where the cover deposits lie 

between 400 and 600 m altitude and were dated as Pleistocene by Penteado 

and Ranzani (1973). The deposits vary from a few centimeters to 

several meters thick. Local textural variations are linked to the erosive 

reworking of Pleistocene deposits and to a greater or lesser contribution 

of fluvial material. Regional variations are related to the proportions of



sandy detritus from Cretaceous formations and clayey detritus from Precambrian 

formations. Sandier covers occur in areas where the escarpments 

which border the depression are located mainly in sandstones 

(Areado Formation). Clayey covers are observed in areas where the 

escarpments are in Precambrian pelitic rocks (Bambuí Group). 

Soils on Surface III 

Shallow soils predominate on the erosional segment of Surface III, mainly 

Cambisols (C, Inceptisols) and, in smaller proportions, Red-Yellow 

Argisols (RYA), Red Argisols (RA) and Red Nitosols (RN), the last three 

occurring in the Alfisol order in Soil Taxonomy (Soil Survey Staff 1999), 

beyond the Litholic Neosols (LN, Entisols). The RN soils occur in association 

with mafic rocks outcroppings (see fig. 2.4G). Because of the close 

relationship between soils and geology, the soils of the erosional segment 

of Surface III are much more variable than those of the other surfaces, especially 

in relation to base saturation and texture (table 2.3). For example, 

soils may be allic (low base saturation, Al-rich), dystrophic (low base saturation), 

or eutrophic (high base saturation), and texture varies from 

medium to clayey. In general, soils that have a textural B horizon (argillic 

horizon) also have high clay contents, > 450 g kg –1 , and medium to high 

base saturation (table 2.3). The majority of Cambisols and Litholic 

Neosols, on the other hand, are characterized by low natural fertility and 

high aluminum saturation. Cambisols generally have a high content of 

gravels and cobbles. Mainly, they support campo cerrado (open cerrado). 

On the depositional segment in the São Francisco Basin the main soils 

are clayey Red Latosols and Red-Yellow Latosols, although Cambisols 

and Litholic Neosols are important in residual elevations. Vegetation follows 

the great variability of soils. The native vegetation includes forest, 

cerradão, cerrado, campo cerrado, and tropical grassland (physiognomic 

descriptions in chapter 6). The RN, RA, and RYA soils have relatively 

good natural fertility, and the native vegetation is forest (see fig. 2.4H). 

Riparian forests occur in the fluvial plains over Fluvic Neosols (FN, Fluvents) 

in the wider valleys, or over Hydromorphic soils along the smaller 

water flows. The relief on Surface III varies from gently undulating to 

mountainous. 

In contrast to soils on older surfaces, soils on Surface III do not exhibit 

net positive charge. Additionally, the K i indices, although low enough to 

be characteristic of low-activity clays, tend to be higher than those for 

soils on the other surfaces. K r ratios are also higher. All these trends confirm 

that these soils are less weathered than those on older surfaces.



CONCLUDING REMARKS 

The Brazilian cerrado region consists of three main geomorphic surfaces, 

each having a set of soils with distinctive attributes. On the oldest surface 

(Surface I), the much longer exposure of the Tertiary sediments to weathering 

and leaching has overcome the influence of the subjacent rocks on 

the kinds and distribution of soils. On that surface the main soil differences 

are due to the water regime. The seasonal saturation of some soils 

results in redistribution of Fe in the soil profile and landscape, and formation 

of petroplinthite. This rocklike material retards erosion, especially 

at the edge of tablelands, thereby helping keep the tablelands intact. Thus, 

petroplinthite has significance for both soils and geomorphology. The 

index values based on laboratory data, K i , K r , and ∆pH, confirm that 

these soils are highly weathered. On the youngest surface (Surface III), the 

relationship between the distribution and properties of soils and their parent 

materials is more evident. On the intermediate surface (Surface II), the 

distribution of soils as well as their attributes are related to the degree of 

mixing of the Tertiary material from Surface I and the products of decomposition 

of the subjacent rocky material. The index values are similar to 

those on Surface I, however, showing that these soils are also highly 

weathered. 

REFERENCES 

Ab’Saber, A. N. 1963. Contribuição a geomorfologia da área dos cerrados. 

In M.G. Ferri, ed., Simpósio sobre o Cerrado, pp. 119–128. São Paulo: 

Editora da Universidade de São Paulo. 

Bates, R. L. and J. A. Jackson. 1987. Glossary of Geology. 3rd edition. Am. 

Geol. Institute, Alexandria, VA. 

Brasil. 1962. Levantamento de Reconhecimento de Solos da Região sob 

Influência do Reservatório de Furnas. Rio de Janeiro: Serviço Nacional 

de Pesquisas Agronômicas, Ministério da Agricultura. 

Braun, O. P. G. 1971. Contribuição à geomorfologia do Brasil Central. Rev. 

Bras. Geogr. 32:3–39. 

Cetec. 1981. Segundo Plano de Desenvolvimento Integrado do Noroeste 

Mineiro: Recursos Naturais. Belo Horizonte: Fundação Centro Tecnológico 

de Minas Gerais. 

Cline, M. G. and S. W. Buol. 1973. Solos do Planalto Central do Brasil. 

Ithaca: Cornell University Press. 

Cornell, R. M. and U. Schwertmann. 1996. The Iron Oxides: Structure, Properties, 

Reactions, Occurrence and Uses. Weinheim: VCH. 

Corrêa, G. F. 1989. Les Microreliefs “Murundus’‘ et Leur Environnement



Pédologique dans l’Ouest du Minas Gerais, Région du Plateau Central 

Brésilien. Thèse de doctorat, Université de Nancy, Vandoeuvres-les- 

Nancy, France. 

Embrapa. 1982. Levantamento de Reconhecimento dos Solos e Aptidão Agrícola 

das Terras do Triângulo Mineiro. Rio de Janeiro: Embrapa, Serviço 

Nacional de Levantamento e Conservação de Solos. 

Embrapa. 1988. Definição e Notação de Horizontes e Camadas do Solo. 

Segunda edição. Rio de Janeiro: Embrapa, Serviço Nacional de Levantamento 

e Conservação de Solos. 

Embrapa. 1999. Sistema Brasileiro de Classificação de Solos. Rio de Janeiro: 

Embrapa, Centro Nacional de Pesquisa de Solos. 

Embrapa. 2001. In press. Levantamento de Reconhecimento de Alta Intensidade 

dos Solos do Município de Silvânia, GO. Rio de Janeiro: Embrapa, 

Serviço Nacional de Levantamento e Conservação de Solos. 

Feuer, R. 1956. “An Exploratory Investigation of the Soils and Agricultural 

Potential of the Soils of the Future Federal District in the Central Plateau 

of Brazil.’’ Ph.D. thesis, Cornell University, Ithaca, USA. 

Ker, J. C. and M. Resende. 1996. Recursos edáficos dos cerrados: Ocorrência 

e potencial. In: Biodiversidade e produção sustentável de alimentos e 

fibras nos cerrados. In R.C. Pereira and L.C.B. Nasser, eds., Anais do 

VIII Simpósio Sobre o Cerrado, pp. 15–19. Brasília: Embrapa, Centro 

de Pesquisa Agropecuária dos Cerrados. 

King, L. C. 1956. A geomorfologia do Brasil Oriental. Rev. Bras. Geogr. 

18:147–265. 

Lepsch, I. F. and S. W. Buol. 1988. Oxisol-landscape relationships in Brazil. 

In F. H. Beinroth, M. N. Camargo, and H. Eswaran, eds., Proceedings of 

the International Soil Classification Workshop, 8, pp. 174–189. 

Lima, S. C. 1996. “As Veredas do Ribeirão Panga no Triângulo Mineiro e a 

Evolução da Paisagem.’’ Doctor in Science thesis, Universidade de São 

Paulo, São Paulo, Brasil. 

Macedo, J. and R. B. Bryant. 1987. Morphology, mineralogy and genesis of 

a hydrosequence of Oxisols in Brazil. Soil Sci. Soc. Am. J. 51:690–698. 

Mothci, E. P. 1977. “Características e Gênese de um Seqüência de Oxisols no 

Planalto Central Brasileiro.’’ Master’s thesis, Universidade Federal do 

Rio Grande do Sul, Porto Alegre, Brasil. 

Motta, P. E. F. and N. Kämpf. 1992. Iron oxide properties as support to soil 

morphological features for prediction of moisture regimes in Oxisols of 

Central Brazil. Z. Pflanzenernähr 155:385–390. 

Penteado, M. M. and G. Ranzani. 1973. Relatório de viagem ao Vale do Rio 

São Francisco: Geomorfologia. São Paulo: Editora da Universidade de 

São Paulo. 

Radambrasil. 1983. Folha SE 22 Goiânia: Geologia, Geomorfologia, Pedologia, 

Vegetação e Uso Potencial da Terra. Rio de Janeiro: Ministério das 

Minas e Energia. 

Resende, M. 1976. Mineralogy, Chemistry, Morphology and Geomorphology



of Some Soils of the Central Plateau of Brazil. Ph.D. thesis, Purdue University, 

West Lafayette, USA. 

Resende, M., N. Curi, S. B. Rezende, and G. F. Corrêa. 1999. Pedologia: Base 

para Distinção de Ambientes. Viçosa, Brasil: Núcleo de Estudos de Planejamento 

de Uso da Terra. 

Ruhe, R. 1969. Quaternary Landscapes in Iowa. Ames: Iowa State University 

Press. 

Soil Survey Staff. 1999. Soil Taxonomy. 2nd edition. U.S. Department of 

Agriculture Handbook 436. Washington, D.C.: U.S. Government Printing 

Office. 

Sposito, G. 1989. The Chemistry of Soils. New York: Oxford University 

Press. 

Suguio, K. and J. J. Bigarella. 1979. Ambiente fluvial. In J. J. Bigarella, K. Suguio, 

and R. D. Becker, eds., Ambientes de Sedimentação: Sua Interpretação 

e Importância. Curitiba: Editora Universidade do Paraná.


3 

Late Quaternary History 

and Evolution of the Cerrados as 

Revealed by Palynological Records 

Marie-Pierre Ledru 

Whether cerrados are anthropogenic or natural formations 

has been a matter of strong debate over the last century. The fact is 

that cerrados and forests can occur in the same region, at the same latitude, 

under the same climatic conditions (chapters 2, 6). These observations 

generated two types of hypotheses about the origin of the cerrado. 

The first favors the human-induced origin of the vegetation and is based 

on observations of fire-adapted species, which suggest that cerrados 

would result from the development of dry forests under the influence of 

fire (Lund 1835; Loefgren 1897; Aubréville 1961; Schnell 1961; Eiten 

1972). The second hypothesis supports the natural origin of the cerrados 

based on the occurrence of cerrados in areas only recently colonized by 

humans, and on the discovery of extinct giant mammals living in open 

forest landscapes at the end of the Pleistocene (Warming 1918 in Warming 

and Ferri 1973; Azevedo 1965; Cartelle 1991; Guerin 1991; Vilhena 

Vialou et al. 1995). Palynological research started in Brazil at the end of 

the 1980s and has provided an excellent tool for the elucidation of patterns 

and causes of vegetation and climate change. Pollen grains accumulated 

in lakes or bogs are well preserved in sedimentary deposits and can 

show how the vegetation changed through time and space. The construction 

of pollen reference collections and publication of a pollen atlas for 

the cerrados (Salgado-Labouriau 1973) has facilitated the research. In this 

chapter I will first give a definition of the cerrado in terms of modern 

pollen rain and subsequently review the published fossil pollen diagrams 

33



as a basis for examination of the evolution of the cerrados during the Late 

Quaternary. 

POLLEN ANALYTICAL METHODS 

Samples for pollen analysis derived from recent surface or core sediments 

are treated according to the standard palynological techniques defined in 

Faegri and Iversen (1989). These techniques involve concentrating the 

pollen and spores content of each sample. The inorganic component is 

removed with acids; humic acids with KOH and palynomorphs are separated 

from the sediment residue using a solution of density 2 and mounted 

in glycerine for light microscopical analysis. Pollen and spores are identified 

by comparison with pollen reference collections and published pollen 

atlases and are counted to calculate their frequencies. Pollen taxa are 

grouped and expressed as percentages of a total pollen sum. Aquatics, wetland 

taxa, and spores are usually excluded from the total pollen sum and 

expressed as percentages thereof. Pollen diagrams represent the counts for 

each taxon along a depth scale for fossil samples. Counts can be also 

expressed as a ratio of Arboreal Pollen/Non-Arboreal Pollen (AP/NAP) or 

as different ecological groups such as cerrado elements or semideciduous 

forest elements. Changes in fossil pollen frequencies are interpreted to 

reflect environmental and climatic changes. Charcoal particles deposited 

in the sediments can also be counted. They are expressed either as a total 

of the pollen sum or as particle concentration per unit volume or weight 

of sample. Radiocarbon dates obtained from the same sediment core and 

lithology of the core are also reported on fossil pollen diagrams. 

Detailed pollen counts of several pollen records used in this review 

are archived in the NOAA–World Data Center A for Paleoclimatology 

(Latin American Pollen Database). 

MODERN POLLEN RAIN AND CERRADOS INDICATORS 

Pollen taxa that are representative for a specific vegetation type are called 

indicator taxa. Pollen indicator taxa must show good preservation, dispersion, 

and production to have any chance of being found in fossil sediments. 

In addition, the source plants must be well represented in the 

vegetation, and a comparison of modern pollen rain with phytosociological 

surveys is often needed to define their value. Once an association of 

indicator species is defined, they can be related to present-day climatic 

parameters such as the length of the dry season or mean winter tempera-


Late Quaternary History as Revealed by Palynological Records 35 

ture. This approach can be used to provide a quantitative climatic interpretation 

of the fossil pollen spectra (see table 3.1). The problem with the 

cerrados is that, apart from the wind-pollinated Poaceae and Cyperaceae, 

most of the pollen taxa recorded are insect-pollinated. 

The first study of modern pollen rain for the cerrados was published 

by Salgado-Labouriau (1973). Pollen traps were collected monthly in a 

cerrado located in the state of Goiás (Central Plateau). Results showed 

mainly a dominance of Poaceae (Gramineae) with 74%, followed by the 

Fabaceae-Caesalpiniaceae-Mimosaceae (Leguminosae) with 3.2%. Ledru 

(1991) collected surface samples in six different types of cerrado in Mato 

Grosso, Distrito Federal, and São Paulo (see figs. 3.1–3.3; table 3.2). 

Pollen counts were undertaken in relation to phytosociological data in 

order to define cerrados indicators well represented in both pollen rain 

and vegetation. Two taxa can be distinguished and considered good cerrado 

indicators, Byrsonima and Didymopanax. Well represented as a tree, 

Curatella is often considered a cerrado indicator but does not produce 

much pollen. Byrsonima and Didymopanax are actually overrepresented 

in terms of pollen proportions (i.e., percent of pollen is greater than percent 

of trees; Byrsonima occurs as small trees with a Diameter at Breast 

Height (DBH) < 10 cm and therefore is not counted in the botanical surveys). 

Byrsonima is present in different proportions in the six types of cerrado 

studied, with frequencies of less than 5%. Qualea and Caryocar are 

underrepresented in the pollen counts compared to the associated basal 

area of these trees (fig. 3.3). Grasses are not taken into account in the phytosociological 

surveys but are also abundant and representative of the 

cerrados (chapter 7), with ca. 50% Poaceae in campo cerrado, and less 

than 20% Poaceae in cerradão (see also chapter 6). Vellozia (Velloziaceae) 

is a good indicator for campo cerrado; Borreria (Rubiaceae) is 

also well represented although not recorded in the cerrado of Cuiabá 

Salgadeira (Mato Grosso), and the cerradão of Bauru (São Paulo). The 

differences in taxa distribution complicate the definition of a quantified 

Table 3.1 Relation Between Climatic Parameters 

and Vegetation in Central Brazil 

Climate 

Vegetation Dry season Mean winter temperature 

Cerrado 4–5 months ≥ 15˚C 

Semideciduous forest 2–3 months > 10˚C and < 15˚C 

Araucaria forest None ≤ 10˚C



Figure 3.1 Map of Brazil showing location of sites cited in the text: modern 

and paleoenvironmental records. 1. Cuiabá Salgadeira; 2. Cuiabá Rio 

Claro; 3. Brasília and Aguas Emendadas; 4. Bauru; 5 Lagoa do Caço; 6. Lagoa 

dos Olhos; 7. Cromínia; 8. Carajás; 9. Lagoa do Pires; 10. Icatu River Valley. 

pattern of indicator taxa. In lakes or in the veredas (palm swamp forest; 

chapter 6), where cores for paleoenvironmental studies are generally 

recovered, surface samples show high levels of Mauritia (Buriti palm) and 

Cyperaceae (Barberi 1994; Salgado-Labouriau et al. 1997). This further 

complicates the possibility of recognizing the regional vegetation. 

PALEORECORDS LOCATED IN THE CERRADO AREA: 

POLLEN AND CHARCOAL RESULTS 

Pollen records obtained from lake or peat bog sediments provide a 

detailed paleoenvironmental reconstruction of the changes that affected 

the cerrados during the Late Quaternary.

Figure 3.2 Modern pollen rain in the cerrado environment and other Brazilian forests. The cerrado pollen counts were 

collected in different types of cerrado and are presented in the following order: samples 1 to 21 come from Brasília (Distrito 

Federal, DF); 22 to 24 from Cuiabá Salgadeira (Mato Grosso, MT); 25 to 29 from Cuiabá Rio Claro (Mato Grosso, 

MT); and 30 to 37 from Bauru (São Paulo, SP). See also table 3.2. 



Figure 3.3 Relation between tree basal area (Diameter at Breast Height Percent) 

and pollen percent for four cerrado taxa. The sites are in the same order 

as in figure 3.2. Every pollen sample represents a mean value of the pollen 

counts presented in figure 3.2 to allow comparison with tree basal area.



Table 3.2 Location and Structural Vegetation Attributes of Surface 

Sample Sites Located in Different Cerrado Areas of Brazil 

Basal 

Elevation Density area 

Sampling site (m) Latitude (trees/ha) (m 2 /ha) Source a 

Brasília (DF) 

cerradão 1030 15˚35' S 2231 20.9214 1 

Cerrado sensu stricto 1125 15˚35' S 911 9.65 1 

Campo cerrado 1175 15˚35' S 203 1.6686 1 

Cuiabá (MT) 

cerradão 350 15˚21' S 1546 16.116 2, 3 

Cerrado sensu stricto 350 15˚21' S 1888 21.044 2, 3 

Bauru (SP): 

Cerrado sensu stricto 570 22˚19' S 8198 40.8793 4 

a Key to sources: 1 = Ribeiro et al. 1985; 2 = Oliveira-Filho 1984; 3 = Oliveira-Filho and 

Martins 1986; 4 = Cavassan and Martins 1984. 

Note: See chapter 6 for descriptions of cerrado physiognomies; and figure 3.2. 

Four palynological records show changes in the vegetation composition: 

Aguas Emendadas (15° S 47°35' W, 1,040 m elevation; Barberi 

1994; Barberi et al. 1995); Cromínia (17°17' S 49°25' W, 730 m elevation; 

Vicentini 1993; Salgado-Labouriau et al. 1997); Lagoa do Caço 

(25°8' S, 43°25' W, 80 m elevation; Sifeddine et al. 1999; Ledru et al. 

2001); and Lagoa dos Olhos (19°38' S 43°54' W, 730 m elevation; De 

Oliveira 1992; see figs. 3.1, 3.4A–D). 

Pollen records from the cerrados date back to about 32,000 YBP 

(years Before Present). The indicator taxa Byrsonima, Didymopanax, and 

Curatella appear throughout the fossil spectra with frequencies of less 

than 5%, which is far lower than in the modern spectra for the latter two 

genera. In the Cromínia record (ca. 32,000YBP), Byrsonima and Mauritia 

are recorded at frequencies of less than 2%, and between 10% and 

30%, respectively. Arboreal Pollen (AP) frequency is high and the climate 

can be defined as moist and warm, probably with seasonality in precipitation. 

Microscopic charcoal particles are also abundant. This observation 

reveals that fires had occurred in the cerrados without human 

influence, for the presence of people in South America is not confirmed 

before 12,000 YBP (Cooke 1998; see also chapter 4). This pre-full-glacial 

moist and warm phase is found in several other tropical South American 

records (De Oliveira 1992; Ledru et al. 1996). In tropical northeastern 

Australia high microscopic charcoal frequencies are recorded from ca. 

38,000 YBP, when moist forest was replaced by sclerophyll vegetation.


(A) 

Figure 3.4 Summary pollen diagrams of the records located within the 

cerrado region, (A) Aguas Emendadas; (B) Cromínia; (C) Lagoa dos Olhos; 

(D) Lagoa do Caçó (data from Barberi et al. 1995; Salgado-Labouriau et al. 

1997; De Oliveira 1992; Ledru, in preparation). Soil codes under (C).


(B)


(C)


(D)



This was probably related to human influence that induced the expansion 

of the fire-adapted vegetation type (Kershaw 1986; Kershaw et al. 1997). 

Brazilian records instead show that fire-adapted vegetation can be the 

result of climate influence alone, although older records are needed in lowland 

South America to better compare with Australian vegetation history. 

A particularly dry and cold climatic phase is recorded in the tropics 

between 20,000 and 18,000 YBP, corresponding to the last glacial maximum. 

It was detected in light of the absence of sediment accumulation or 

palynologically sterile sediment during this period of time (Ledru et al. 

1998a). Where pollen is present, this cold and dry phase started at ca. 

25,000 YBP and is characterized by a decrease in swamp forest taxa, 

absence of Mauritia, low frequencies of Arboreal Pollen, and dominance 

of Poaceae, Asteraceae, and Cyperaceae. The lakes were replaced by 

marshes. Absence of cerrado indicators and microscopic charcoal particles 

show that climatic conditions did not allow the development of the 

cerrado at this time. 

When sedimentation rates increased after ca. 18,000 YBP at the onset 

of the late glacial, the absence of assemblages with modern analogues does 

not allow clear definition of the environment. At Lagoa dos Olhos an 

association of Podocarpus and Caryocar is recorded, while at Lagoa do 

Caçó Podocarpus is recorded along with high frequencies of Byrsonima. 

These pollen assemblages characterize cool climatic conditions if compared 

with modern Podocarpus associations in the montane region of 

Minas Gerais (A.T. Oliveira-Filho, personal communication). The absence 

of charcoal particles or Mauritia pollen grains during the whole lateglacial 

time period also suggests dry conditions. 

After ca. 10,000 YBP and until 7,000 YBP the landscape was more 

open. The virtual absence of Mauritia from the records suggests that palm 

swamp forests remain restricted, and a gap in sedimentation is often 

recorded attesting to a dry climate without seasonality. This could be due 

to changes in orbital parameters that attenuated the “monsoon effect’’ 

and reduced the overall precipitation (Martin et al. 1997; Ledru et al. 

1998b). 

The cerrado of Lagoa do Caçó indicates a different pattern of evolution. 

Byrsonima is recorded during the whole sequence, indicating that 

cerrado-like vegetation, probably as a campo cerrado type, was maintained. 

No microscopic charcoal fragments are recorded before the beginning 

of the Holocene, when cerrado tree frequencies started to increase, 

and the presence of Mauritia attests to the establishment of a seasonal climate 

and increase in temperatures. These differences are probably due to 

the northern location of the lake, close to the Amazon Basin, and to the



influence of the Meteorological Equator. In other cerrado pollen records, 

when sedimentation restarts after 7,000 YBP, the palm swamp forest is 

well established, and the climate is warm (mean winter temperatures 

above 15°C) and seasonal (4 to 5 months dry season). Microscopic charcoal 

particles are abundant, Poaceae increases, and Asteraceae decreases. 

This indicates influence of both human and climate on the landscape. 

A dry early Holocene is also documented in the Colombian savannas, 

where an expansion of grassland taxa is recorded. After 6,000 YBP the 

expansion of gallery forest taxa, mainly the swamp forest taxa Mauritia 

and Mauritiella, and the abundance of microscopic charcoal particles, 

indicate wetter climatic conditions for the Late Holocene and full development 

of the savanna (Behling and Hooghiemstra 1998; Behling and 

Hooghiemstra 1999). 

Biomass was sufficiently high to induce repeated fires during the Late 

Holocene in Central Brazil. The increase in seasonality that started at ca. 

7,000 YBP was necessary to allow cerrado vegetation to grow on the Central 

Plateau. It started at different times during the Holocene according to 

the latitude of the respective record. 

PALEORECORDS LOCATED OUTSIDE THE CERRADO 

AREA SHOWING CERRADO INDICATORS 

In other records located in tropical forest areas (Amazonia, montane forest, 

Atlantic forest), a local increase of cerrado-type indicators confirms 

that these regions were linked with central regions during extreme climatic 

conditions. 

A pollen record located within the Amazonian Basin (Carajás, 50°25' 

W 6°20' S, 700–800 m elevation), shows three main changes in vegetation 

composition during the last 60,000 YBP (Absy et al. 1991; Sifeddine et al. 

2001; see fig. 3.5). Cerrado taxa such as Byrsonima and Ilex are present 

today on the plateau around the lakes. These taxa are well represented, 

together with high frequencies of rainforest taxa during past moist climatic 

phases. Open landscape elements associated with campo cerrado vegetation, 

Poaceae, Borreria, and Cuphea, increase in frequency during dry climatic 

periods when moisture levels were insufficient to maintain the 

forests. Podocarpus frequencies increase together with the campo cerrado 

species, attesting to an open landscape with Podocarpus on the lake margins 

similar to the situation found in the record from the Maranhão region 

during the late-glacial period (Ledru et al. 2001). A cerrado-type vegetation 

could grow on the southern and eastern periphery of the Amazonian


Figure 3.5 Summary pollen diagram of the Carajás record (Amazonia) 

(data from Absy 1991).



basin, and within the dry corridor during drier and cooler climates (Bush 

1994; van der Hammen and Absy 1994). This cerrado vegetation is maintained 

today on the Plateau of the Serra do Carajás due to specific edaphic 

conditions. 

In the Lagoa do Pires record (17°57' S, 42°13' W, 390 m elevation), 

located in a semi-deciduous forest region, an increase of cerrado elements 

(mainly represented by Curatella) is registered at the beginning of the 

Holocene, when the climate was drier in this region (Behling 1995). 

A pollen record in the Icatu River valley (10°24' S 43°13' W, ca 400 

m elevation) shows an increase of montane and moist forest elements at 

the beginning of the Holocene, preceding the development of cerradocaatinga 

vegetation after 9,000 YBP (De Oliveira et al 1999). This moist 

early Holocene also detected in the Salitre record was interpreted as 

related to stronger polar advections (Ledru 1993). Fluctuations in Mauritia 

pollen frequencies are detected throughout the Holocene. 

CONCLUSION 

The floristic composition of the cerrado vegetation varied through time 

as it does today according to location (chapter 6). The earliest record of 

cerrado-type vegetation dates back to 32,000 YBP and is located on the 

Central Brazilian Plateau. Vegetation resembling present-day cerrados 

does not occur prior to 7,000 YBP in central Brazil and 10,000 YBP in 

northern Brazil. Their presence in both regions is most likely a consequence 

of a progressive increase of seasonality, concomitant with an 

increase in temperature. Both factors, in addition to man’s influence, contributed 

to increased fire frequencies, although cerrado vegetation was 

probably fire-adapted before people arrived in South America. Local 

increases of cerrado-type vegetation are recorded in moist forest areas 

when climatic conditions became drier and/or colder. During these periods 

forest taxa remained connected through a network of gallery forests. 

This contributed to the differences in physiognomy observed today (chapter 

6). Cerrados recognized in Amazonia or in São Paulo might be the 

result of this ancient connection. 

REFERENCES 

Absy, M. L., A. Cleef, M. Fournier, L. Martin, M. Servant, A. Sifeddine, M. Ferreira 

da Silva, F. Soubiès, K. Suguio, B. Turcq, and T. van der Hammen.



1991. Mise en évidence de quatre phases d’ouverture de la forêt amazonienne 

dans le sud-est de l’Amazonie au cours des 60000 dernières années: 

Première comparaison avec d’autres régions tropicales. C. R. Acad. Sc. 

Paris 312:673–678. 

Aubréville, A. 1961. Etude Ecologique des Principales Formations Végétales 

du Brésil et Contribution à la Connaissance des Forêts de l’Amazonie 

Brésilienne. Nogent: Centre Technique Forestier Tropical. 

Barberi, M. 1994. “Paleovegetação e Paleoclima no Quaternario Tardio da 

Vereda de Aguas Emendadas, DF.’’ Master’s thesis, Universidade de 

Brasília, Brasília, Brasil. 

Barberi, M., M. L. Salgado-Labouriau, K. Suguio, L. Martin, B. Turcq., and 

J. M. Flexor. 1995. Análise palinológica da vereda de Águas Emendadas 

(DF). Vth Congresso da Associação Brasileira de Estudos do Quaternário, 

Universidade Federal Fluminense, Niteroí, Brasil. 

Behling, H. 1995. A high resolution Holocene pollen record from Lago do 

Pires, SE Brazil: Vegetation, climate and fire history. J. Paleolimnol. 

14:253–268. 

Behling, H. and H. Hooghiemstra. 1998. Late Quaternary palaeoecology and 

palaeoclimatology from pollen records of the savannas of the Llanos 

Orientales in Colombia. Palaeogeogr. Palaeoclimatol. Palaeoecol. 139: 

251–267. 

Behling, H. and H. Hooghiemstra. 1999. Environmental history of the 

Colombian savannas of the Llanos Orientales since the Last Glacial 

Maximum from lake records El Pinal and Carimagua. J. Paleolimnol. 

21:461–476. 

Bush, M. 1994. Amazonian speciation: A necessarily complex model. J. Biogeogr. 

21:5–17. 

Cartelle, C. 1991. Um novo Mylodontinae (Edentata, Xenarthra) do pleistoceno 

final da região intertropical brasileira. An. Acad. Bras. Ciênc. 

63:161–170. 

Cavassan, O., O. Cesar, and F. R. Martins. 1984. Fitossociologia da vegetação 

arbórea da reserva Estadual de Bauru, Estado de São Paulo. Rev. Bras. 

Bot. 7:91–106. 

Cooke, R. 1998. Human settlement of central America and northernmost 

South America (14,000–8000 BP). Quat. Int. 49/5:177–190. 

De Oliveira, P. E. 1992. “A Palynological Record of Late Quaternary Vegetational 

and Climatic Change in Southeastern Brazil.’’ Ph.D. thesis, Ohio 

State University, Columbus, Ohio, U.S.A. 

De Oliveira, P.E., A. M. F. Barreto, and K. Suguio. 1999. Late Pleistocene/ 

Holocene climatic and vegetational history of the Brazilian caatinga: The 

fossil dunes of the middle São Francisco River. Palaeogeogr. Palaeoclimatol. 

Palaeoecol. 152:319–337. 


Faegri, K. and J. Iversen. 1989. Textbook of Pollen Analysis. 4th edition. 

Chichester: John Wiley and Sons.



Guérin, C. 1991. La faune de vertébrés du Pléistocène supérieur de l’aire 

archéologique de São Raimundo Nonato (Piauí, Brésil). C. R. Acad. Sc. 

Paris 313:567–572. 

Kershaw, A. P. 1986. Climatic change and Aboriginal burning in north-east 

Queensland during the last two glacial/interglacial cycles. Nature 

322:47–49. 

Kershaw, A. P., M. B. Bush, G. S. Hope, K.-F. Weiss, J. G. Goldammer, and 

R. Sanford. 1997. The contribution of humans to past biomass burning 

in the tropics. In J. S. Clark, H. Cachier, J. G. Goldammer, and B. Stocks, 

eds., Sediment Records of Biomass Burning and Global Change, pp. 

413–442. Berlin: Springer-Verlag. 

Ledru, M.-P. 1991. “Etude de la Pluie Pollinique Actuelle des Forêts du Brésil 

Central: Climat, Végétation, Application à l’Etude de l’Evolution Paléoclimatique 

des 30000 Dernières Années.’’ Ph.D. thesis, Musée National 

d’Histoire Naturelle, Paris, France. 

Ledru, M.-P. 1993. Late Quaternary environmental and climatic changes in 

central Brazil. Quat. Res. 39:90–98. 

Ledru, M-P., P. I. Soares Braga, F. Soubiès, M. Fournier, L. Martin, K. Suguio, 

and B. Turcq. 1996. The last 50,000 years in the Neotropics (southern 

Brazil): Evolution of vegetation and climate. Palaeogeogr. Palaeoclimatol. 

Palaeoecol. 123:239–257. 

Ledru, M.-P., J. Bertaux, A. Sifeddine, and K. Suguio. 1998a. Absence of Last 

Glacial Maximum records in lowland tropical forest. Quat. Res. 49: 

233–237. 

Ledru, M.-P., M. L. Salgado Labouriau, and M. L. Lorscheitter. 1998b. Vegetation 

dynamics in Southern and Central Brazil during the last 10,000 

yr B.P. Rev. Palaeobot. Palynol. 99:131–142. 

Ledru M.-P., R. Campello Cordeiro, J. M. D. Landim, L. Martin, P. Mourguiart, 

A. Sifeddine, and B. Turcq. 2001. Late-glacial cooling in Amazonia 

inferred from pollen at Lago do Caço, northern Brazil, Quat. Res. 

55:47–56. 

Loefgren, A. and G. Edwall. 1897–1905. Flora Paulista. Bol. Com. Geogr. 

Geolog. São Paulo 1–4:12–15. 

Lund, P. W. 1835. Bemaerkninger over Vegetationen paa de indre Hogsletter 

af Brasilien, isaer I plantehistorike henseende. Kgl. Danske Videnskab. 

Selsk. Skrifer. 6:145–188. 

Martin, L., J. Bertaux, T. Correge, M.-P. Ledru, P. Mourguiart, A. Sifeddine, 

F. Soubies, D. Wirrmann, K. Suguio, and B. Turcq. 1997. Astronomical 

forcing of contrasting rainfall changes in tropical South America between 

12,400 and 8800 cal yr B.P. Quat. Res. 47:117–122. 

Oliveira-Filho A. T. 1984. “Estudo Florístico e Fitossociológico em um Cerrado 

na Chapada dos Guimarães, Mato Grosso: Uma Análise de Gradientes.’’ 

Master’s thesis, Universidade Estadual de Campinas, Campinas, 

Brazil. 

Oliveira-Filho A. T. and F. R. Martins. 1986. Distribuição, caracterização e



composição florística das formações vegetais da região da Salgadeira na 

Chapada dos Guimarães (MT). Rev. Bras. Bot. 9:207–223. 

Ribeiro, J. F., J. C. Souza Silva, and G. J. Batmanian. 1985. Fitossociologia 

de tipos fisionômicos de cerrado em Planaltina DF. Rev. Bras. Bot. 

8:131–142. 

Salgado-Labouriau, M. L. 1973. Contribuição à Palinologia dos Cerrados. 

Acadêmia Brasileira de Ciências: Rio de Janeiro. 

Salgado-Labouriau, M. L., V. Casseti, K. R. C. F. Vicentini, L. Martin, F. Soubies, 

K. Suguio, and B. Turcq. 1997. Late Quaternary vegetational and 

climatic changes in cerrado and palm swamp from central Brazil Palaeogeogr. 

Palaeoclimatol. Palaeoecol. 128:215–226. 

Schnell, R. 1961. Le problème des homologies phytogéographiques entre 

l’Afrique et l’Amérique tropicale. Mem. Mus. Hist. Nat. 11:137–241. 

Sifeddine, A., R. C. Cordeiro, M.-P. Ledru, B. Turcq, A. L. S Albuquerque, L. 

Martin, J. M. Landim Dominguez, H. Pasenau, G. Ceccantini, and J. J. 

Abrão. 1999. Late glacial multi-proxy paleo-environmental reconstruction 

in Caço Lake (Maranhão state, Brazil). INQUA Congress, Durban, 

South Africa, Abstract p. 167. 

Sifeddine, A., L. Martin, B. Turcq, C. Volkemer-Ribeiro, F. Soubies, R. C. 

Cordeiro, and K. Suguio. 2001. Variations of the Amazonian rainforest 

environment: A sedimentological record covering 30,000 years. Palaeogeogr. 

Palaeoclimatol. Palaeoecol. 168:221–235. 

Van der Hammen, T. and M. L. Absy. 1994. Amazonia during the last glacial. 

Palaeogeogr. Palaeoclimatol. Palaeoecol. 109:247–261. 

Vicentini, K. R. C. F. 1993. “Análise Palinológica de uma Vereda em Cromínia, 

Goiás.’’ Master’s thesis, Universidade de Brasília, Brasília, Brazil. 

Vilhena Vialou A., T. Aubry, M. Benabdelhadi, C. Cartelle, L. Figuti, M. 

Fontugne, M. E. Solari, and D. Vialou. 1995. Découverte de Mylodontinae 

dans un habitat préhistorique daté du Mato Grosso (Brésil): L’ abri 

rupestre de Santa Elina. C. R. Acad. Sc. Paris 320:655–661. 

Warming, E. and M. G. Ferri. 1973. Lagoa Santa e a Vegetação de Cerrados 

Brasileiros. : São Paulo e Belo Horizonte: Editora da Universidade São 

Paulo e Editora Itatiaia.


4 

The Fire Factor 

Heloisa S. Miranda, Mercedes M. C. Bustamante, 

and Antonio C. Miranda 

Fire is a common feature of the cerrados, as it is for most 

savanna ecosystems. Fires set by man or lightning are common and have 

been for thousands of years. Vicentini (1993), in a paleoclimatic and paleovegetational 

study, has registered the occurrence of fire 32,400 Years 

Before Present (YBP) in the region of Cromínia (GO); De Oliveira (1992) 

registered the presence of charcoal particles dated from 13,700 YBP in lake 

sediments in cerrado of the southeastern Brazil; and Coutinho (1981) has 

reported the occurrence of charcoal pieces dated from 8,600 YBP from a 

campo cerrado soil horizon lying at 2 m depth. Although Gidon and Delibrias 

(1986) date the presence of man in Brazil to 32,000 YBP, according to 

Prous (1992) and Cooke (1998) there is no evidence of human presence in 

central Brazil before 12,000–11,000 YBP. Therefore, the particles of charcoal 

and burned wood dated from 11,000 YBP could, at least in part, be 

caused by the early inhabitants of the cerrado region (Salgado-Laboriau 

and Vicentini 1994). The indigenous people of the cerrado region used fire 

for hunting, stimulation of fruit production, control of undesirable species, 

and tribal war (Coutinho 1990a; Mistry 1998). Nowadays, the principal 

cause of fire in the cerrado is agricultural, its purpose either to transform 

cerrado into crop fields or to manage natural (more open cerrado forms) 

or planted pasture (Coutinho 1990a; see chapter 5). 

Although fire is considered one of the determinants of the cerrado 

vegetation, the rapid occupation of the cerrado region has changed the 

natural fire regime (season and frequency of burning) with consequences 

for the vegetation structure and composition. 

In this chapter we present a review of cerrado fire ecology, with 

emphasis on fuel dynamics, fire behavior, nutrient fluxes, and changes in 

51



the structure and composition of the vegetation. A review of fire effects 

on population dynamics of woody plants is presented in chapter 9. 

THE CERRADO FIRES 

Cerrado fires, like most savanna fires, are characterized as surface fires, 

which consume the fine fuel of the herbaceous layer. Luke and McArthur 

(1978) define fine fuel as live and dead grasses, and leaves and stems with 

diameter smaller than 6 mm. Depending upon the physiognomic form 

(chapter 6) and the time since the last fire, the total fine fuel load, up to 

a height of 2 m, may vary from 0.6 kg/m 2 to 1.2 kg/m 2 . The fine fuel of 

the herbaceous layer represents 97% of the fuel load for campo sujo, 90% 

for cerrado sensu stricto, and 85% for cerradão (Miranda 2000; see 

descriptions of physiognomies in chapter 6). These values are similar to 

those presented by Castro and Kauffman (1998) for cerrado sensu stricto, 

by San José and Medina (1977) for other South American savannas, and 

by Kelmann et al. (1987) for African savannas. 

The vegetation of the herbaceous layer represents 94% of the fuel 

consumed during the fires. Most of the fine fuel in the woody layer is not 

consumed during the fires (see table 4.1). This may be a consequence of 

the high water content of the live fuel, of the fast rate of spread of the fire 

front (Kauffman et al. 1994; Miranda et al. 1996a; Castro and Kauffman 

1998), and of the height of the flames during the fires. Flame height for 

savanna fires ranges from 0.8 m to 2.8 m (Frost and Robertson 1987). 

Castro Neves (unpublished data) determined a reduction of 4% in the 

canopy cover of a cerradão immediately after a prescribed fire. The abscission 

of the damaged leaves resulted in a reduction of 38% in the canopy 

cover in the 15 days after the fire, suggesting that most of the live leaves 

are not consumed but damaged by the hot air flow during the fire. 

The fuel consumption in the different physiognomic forms (table 4.1) 

reflects the difference in fine fuel composition and in fire regime. In the 

campo sujo most of the fine fuel is composed of grasses (live and dead, or 

dormant) that are not in close contact with the wet soil surface and are 

well exposed to the wind and solar radiation, quickly losing moisture to 

the environment. 

In the denser forms of cerrado (i.e., cerradão) the composition of the 

fine fuel reflects the fire regime much more than the open form, with the 

dead leaves of the litter layer representing most of the fuel load after long 

periods of protection from fire. In this case the dead fuel is in close contact 

with the soil surface. The microclimate may affect the rate of fuel moisture


The Fire Factor 53 

Table 4.1 Range of Fine Fuel Load and Fuel Consumption 

During Prescribed Fires in the Cerrado Vegetation 

at the Reserva Ecológica do IBGE, Brasília, DF 

Campo sujo Cerrado sensu stricto Cerradão 

Before fire (kg/m 2 ) 

Herbaceous layer 0.64–0.96 0.59–1.11 0.50–0.75 

Woody layer 0.03–0.03 0.06–0.14 0.08–0.14 

Total 0.67–0.99 0.65–1.25 0.58–0.89 

Dead fuel (%) 69–75 76–61 50 – 69 

Live fuel (%) 31–25 24–39 50 – 31 

After fire (kg/m 2 ) 

Herbaceous layer 0.05–0.06 0.09–0.00 0.08–0.04 

Woody layer 0.01–0.01 0.06–0.04 0.06–0.14 

Total 0.06–0.07 0.15–0.04 0.14–0.18 

Fuel consumption (%) 91–93 77–97 75–76 

Source: H. S. Miranda, unpublished data. 

Note: See chapter 6 for descriptions of cerrado physiognomies. 

loss due to shading of the fine fuel by the trees and shrubs, leaving patches 

of vegetation unburned (Miranda et al. 1993; Kauffman et al. 1994). 

As the fire front advances, air temperatures rise sharply. Miranda et 

al. (1993, 1996a), in studies of vertical distribution of temperatures (1 cm, 

60 cm and 160 cm height) during cerrado fires, have registered maximum 

values in the range of 85°C to 840°C (regardless of the physiognomic 

form), with the highest temperatures occurring most of the time at 60 cm 

above ground. The range of temperatures reported is similar to that of 

savanna fires, where a considerable range of temperatures has been 

recorded, from 70–800°C at ground level, or just above it, to 200–800°C 

at about 1 m (Frost and Robertson 1987). The great variability in the temperature 

may reflect the varying compositions and spatial distributions of 

the fine fuel; fuel water content; days since last rain; and weather conditions 

at the moment of the fire (Miranda et al. 1993). The duration of 

temperatures above 60°C varied from 90 to 270 seconds at 1 cm height, 

from 90 to 200 seconds at 60 cm height, and from 20 to 70 seconds at 

160 cm height. Although 60°C is considered the lethal temperature for 

plant tissue, Kayll (1968) has shown that lethal temperature varies in relation 

to the exposure time, with leaves withstanding 49°C for 2 hours, 

60°C for 31 seconds, or 64°C for 3 seconds. The duration of temperatures 

above 60°C in cerrado fires was long enough to kill the leaves exposed to 

the hot air flow.



Wright (1970) shows that the death of plant tissue depends primarily 

on moisture content and is an exponential function of temperature and 

time. Consequently, the heat tolerance of a tree (the ability of a tree’s 

organs to withstand high temperatures) along with its fire resistance 

(mainly determined by its size, bark thickness, and foliage distribution) 

may vary with season as a result of the seasonal changes in plant water 

content. Guedes (1993) and Rocha e Silva and Miranda (1996) have 

shown that the short duration of the heat pulse and the good insulating 

effect of thick bark, characteristic of the cerrado trees, provide protection 

to the cambium so that the increase in cambium temperature during the 

fires is small. These authors determined a minimum bark thickness of 

6–8 mm for effective protection of the cambium tissue. However, in the 

lower branches, where bark is not thick enough to produce an effective 

insulation, the cambium may reach high temperatures, remaining over 

60°C long enough to cause the death of the cambium tissue, and consequently 

the death of the branches, altering the structure of the tree canopy. 

As a consequence of the short time of residence of the fire front, the 

increase in soil temperature is small. At 1 cm depth the highest temperatures 

range from 29°C to 55°C. Soil temperature changes are negligible 

at and below 5 cm depth, with a maximum increase of 3°C regardless of 

the physiognomic form of cerrado being burned. The maximum temperatures 

are registered 10 min after the fire at 1 cm depth and after 4 h at 

10 cm depth (Coutinho 1978; Castro Neves and Miranda 1996). The rise 

in soil temperature during cerrado fires may have little effect on soil 

organic matter, microbial population, and buried seeds, and also is likely 

to have little effect on the loss of nutrients from the soil pool. 

The reduction of the vegetation cover and the deposition of an ash 

layer over the soil surface result in a postfire alteration in soil microclimate. 

Castro Neves and Miranda (1996) found that the albedo (ρ) between 

10:00 a.m. and 2:30 p.m. is reduced from 0.11 to 0.03 after a 

campo sujo fire, where 94% of the vegetation was consumed. This decrease 

in ρ represents a 10% increase in the energy absorbed. One month 

after the fire, ρ returned to 54% of the prefire value. Soil heat flux (G) 

changed from 55 W/m 2 before the fire to 75 W/m 2 after the fire, representing 

7% of the incident solar radiation (Castro Neves and Miranda 

1996). The alteration in ρ and G results in an increase in the amplitude 

of soil temperature after the fire on the order of 30°C at 1 cm depth, 

and 10°C at 5 cm depth, with no alteration at 10 cm depth (Dias 1994; 

Castro Neves and Miranda 1996). Although these alterations in soil microclimate 

may have some effect on plant colonization and soil microorganisms 

(Frost and Robertson 1987), they are of short duration as a



consequence of the fast recovery of the vegetation of the herbaceous layer. 

Andrade (1998) has shown that 80% of the fuel load of the herbaceous 

layer of a campo sujo is recovered one year after a fire. Neto et al. (1998) 

have determined that after 2 years the biomass of the herbaceous layer of 

campo sujo has completely recovered from fire. 

Bustamante et al. (1998) showed that soil water content (0–10 cm 

depth) was lower in a burned cerrado sensu stricto area than in an adjacent 

unburned area. The difference in soil water content between the two 

areas lasted for 15 months, probably as a consequence of plant cover 

removal and alteration of the vegetation composition. The frequent fires 

in the area reduced the woody plant density, favoring the recolonization 

of grasses that used water of the superficial soil layer. 

NUTRIENT CYCLING 

Nutrient cycling is a very important aspect of fire ecology, especially in 

the cerrado region where the native vegetation presents a low nutrient 

content (Haridasan 2000), with low decomposition rate of the litter (Silva 

1983), and the soils are poor (see chapter 2). As a consequence of the 

many different physiognomic forms, floristic composition, soil characteristics, 

differences in fire regime, and differences in the sampling methods 

used, there is a great variability in the data regarding the nutrient content 

of the cerrado vegetation (André-Alvarez 1979; Batmanian 1983; Pivello 

and Coutinho 1992; Kauffman et al. 1994; Castro 1996). 

During a fire the nutrients may be lost by volatilization or as particles 

deposited in the soil as ash, or remain in the unburned vegetation. 

The studies of loss of nutrients from the cerrado vegetation during fires 

suggest that the loss is greatest in campo limpo and declines along the gradient 

from campo sujo to cerradão (see table 4.2; chapter 6). Silva (1990) 

studied the partition of biomass and nutrients in the tree layer of a cerrado 

sensu stricto. The woody components of the vegetation are the major 

pool of nutrients (see table 4.3). In general, the large woody parts of the 

vegetation do not burn during cerrado fires. The nutrient stock in the 

leaves of cerrado trees (Silva 1990) is smaller than the stock of the green 

vegetation of the herbaceous layer (Batmanian 1983, see table 4.4). The 

difference in nutrient stock and the higher fuel consumption during fires 

in more open areas of cerrado may explain the decline in nutrient loss with 

the increase in the density of woody plants. 

In general 300 to 400 kg/ha of ash are deposited on the soil surface 

after a cerrado fire (Coutinho 1990b). Some of the nutrients deposited in



Table 4.2 Loss of Nutrients from Vegetation During Fires 

in Different Physiognomic Forms of Cerrado 

Loss of nutrients (%) 

Physiognomy N P K Ca Mg S Source 

Campo limpo 97 50 60 58 — 16 Kauffman et 

al. (1994) 

Campo limpo 85 — — — — 88 Castro 

(1996) 

Campo limpo 82 72 50 40 — 17 Kauffman et 

al. (1994) 

Campo sujo 81 — — — — 51 Castro 

(1996) 

Campo cerrado 93–97 45–61 29–62 22–71 19–62 43–81 Pivello and 

Coutinho 

(1992) 

Campo cerrado 66 47 53 67 — 35 Kauffman et 

al. (1994) 

Cerrado sensu 49 47 46 60 — 34 Kauffman et 

stricto al. (1994) 

Cerrado sensu 25 — — — — 44 Castro 

stricto (1996) 

Cerradão 18 — — — — 30 Castro 

(1996) 

Note: See chapter 6 for descriptions of cerrado physiognomies. 

the soil are quickly absorbed. Cavalcanti (1978) observed that, immediately 

after a fire, there was an increase in the concentration of nutrients 

to a depth of 5 cm, with a significant reduction in the next 3 months. The 

author observed little alteration in nutrient concentration at greater 

depths. After a fire in cerrado sensu stricto, Batmanian (1983) measured 

an increase in the concentration of K, Na, Ca and Mg to a depth of 60 cm; 

no alteration in the concentrations of N and P was observed. The high 

concentrations of K, Na, Ca, and Mg lasted for 3 months. Batmanian 

(1983) and Cavalcanti (1978) suggest that most of the nutrients liberated 

during the fires are absorbed by the superficial roots of the plants of the 

herbaceous layer. In fact, Dunin et al. (1997), in a study comparing the 

evapotranspiration of a burned campo sujo (3 months after the fire) with 

an unburned campo sujo (1 year since last fire), concluded that almost all 

the water used by the vegetation of the burned area is removed from the 

first 1.5 m of the soil. According to Coutinho (1990a), this is the region 

where the alteration of postfire nutrient concentration has been observed 

and where the greatest concentration of fine roots is found (Castro 1996).



Table 4.3 Partition of Above-Ground Biomass and Nutrients 

in Different Components of the Tree Layer of Cerrado Sensu Stricto 

Partition of biomass and nutrients 

Trunk Branches Stems Leaves Fruits Total 

(kg/ha) 

Biomass (kg/ha) 6591 4280 9416 1049 64 21,400 

Biomass (%) 30.8 20.0 44.0 4.9 0.3 

Nutrient (kg/ha) 

P 1.3 0.8 1.3 0.8 0.3 4.5 

K 6.7 4.8 8.9 6.0 3.3 29.7 

Ca 6.5 3.7 9.3 2.9 0.5 22.9 

Mg 3.6 1.9 4.0 1.2 0.3 11.0 

Al 4.9 2.9 5.9 2.0 0.4 

Source: Silva 1990. 

Note: Values represent the mean for 35 woody species and a tree density of 1333 trees/ha. 

As discussed before, the rise in soil temperature is small during cerrado 

fires and is restricted to the first centimeters. Therefore, it may have 

little effect on the loss of nutrients from the soil pool. Raison (1979) 

reported a loss of 25% of N in the soil after 2 h at 200°C. No alteration 

in the N concentration in the 0–20 cm soil layer was recorded by Kauffman 

et al. (1994) and Kozovits et al. (1996). Further studies of N and S 

concentration, to a depth of 2 m (Castro 1996), likewise showed no 

change after cerrado fires. Similar results were reported by Montes and 

San José (1993) for another neotropical savanna. The loss of nutrients in 

Table 4.4 Nutrient Content of the Herbaceous 

Layer During the Wet and Dry Seasons in an 

Unburned Cerrado Sensu Stricto 

Nutrient content (kg/ha) 

Grasses Nongrasses 

Nutrient Wet season Dry season Wet season Dry season 

P 0.9 0.6 0.5 0.3 

K 5.9 3.8 6.0 2.5 

Ca 0.5 0.5 2.5 1.8 

Mg 0.9 1.2 1.1 0.7 

Al 3.7 3.0 0.6 0.5 

Source: Batmanian 1983.



the system pool (vegetation + soil) is therefore a consequence of the burning 

of the above-ground biomass representing 3.8% of the system pool 

(Kauffman et al. 1994). 

During cerrado fires the maximum temperatures are around 800°C 

(Miranda et al. 1993, 1996a), and most of the nutrients are lost by 

volatilization. Considering that Ca and Mg have high volatilization temperatures, 

1240°C and 1107°C, respectively (Wright and Bailey 1982), 

Coutinho (1990a) assumes that they are lost by particle transport. Kauffman 

et al. (1994) estimated that about 33% of N, 22% of P and 74% of 

S are lost by volatilization during cerrado fires. Castro (1996) presented 

similar values for N (35%) and S (91%). 

Although a large proportion of the nutrients is lost from a determined 

area during a fire, some will return to the ecosystem as dry or wet deposition. 

Coutinho (1979) reported that for a cerrado area there is an annual 

total deposition of 2.5 kg/ha of K, 3.4 kg/ha of Na, 5.6 kg/ha of Ca, 

0.9 kg/ha of Mg, and 2.8 kg/ha of PO 4 . Considering Coutinho’s (1979) 

data on the input of nutrients, Pivello and Coutinho (1992) estimated that 

the replacement time for P and S lost during burning was far less than 

1 year; in the range of 1 to 3.4 years for Ca; 1.6 to 4.1 years for K; and 

1 to 5.3 years for Mg. They concluded that an interval of 3 years between 

burnings was initially considered adequate to stimulate the recycling of 

the elements retained in the dead plant material and to avoid a critical 

nutrient impoverishment in the ecosystem. The time interval between 

burnings suggested by Pivello and Coutinho (1992) was confirmed by 

Kauffman et al. (1994). 

In addition to the return of nutrients through dry and wet deposition, 

one has also to consider the transfer of nutrients through the decomposition 

of the scorched leaves that are prematurely dropped after the fire. 

Silva (1983) determined that in the litter accumulated during 1 year in a 

cerrado sensu stricto area there is 4.8 kg/ha of K, 3.6 kg/ha of Ca, 

3.0 kg/ha of Mg, and 0.8 kg/ha of P, and that after 300 days of decomposition 

there is a reduction of 70%, 55%, and 35% in the initial concentrations 

of K, Mg, and Ca, respectively. Considering that senescent 

leaves have a lower nutrient concentration than mature green leaves, the 

premature drop of scorched leaves may play an important role in the recycling 

of nutrients caused by fires, even considering the low decomposition 

rate for the cerrado litter. 

Alterations in the carbon cycle have also been observed for cerrado 

areas submitted to prescribed fires. Burned areas present higher soil CO 2 

fluxes than unburned areas, and this effect lasts several months after the 

fire (Poth et al. 1995). The higher fluxes might be due to the increase of



soil organic matter availability in response to the increase of soil pH. The 

CO 2 fluxes to the atmosphere over campo sujo areas under different fire 

regimes were also studied (Santos 1999; Silva 1999). From June to August 

the campo sujo fixed more CO 2 than was released through respiration. 

Maximum assimilation rates varied from 2.5 to 0.03 µmol CO 2 m –2 s –1 . 

In September the campo sujo became a source of CO 2 to the atmosphere, 

with a maximum emission rate of 1.5 µmol CO 2 m –2 s –1 . A prescribed 

fire in late September resulted in an increase of the CO 2 emission to 

4.0 µmol CO 2 m –2 s –1 . In November and December the campo sujo again 

became a sink for CO 2 , with the assimilation rate increasing to 15.0 µmol 

CO 2 m –2 s –1 . During the seven months of measurements the campo sujo 

accumulated 0.55 t C ha –1 (25% of the amount of carbon accumulated 

in one year by the cerrado sensu stricto as determined by Miranda et al. 

1996b, 1997). 

Burned areas are also a source of trace gases to the atmosphere. Poth 

et al. (1995) measured soil fluxes of NO, N 2 O and CH 4 from cerrado sites 

that had been burned within the previous 2 days, 30 days, and 1 year, and 

from a control site last burned in 1976. NO and N 2 O fluxes responded 

to fire with the highest fluxes observed from newly burned sites after addition 

of water. NO fluxes immediately after burning are among the highest 

observed for any ecosystem studied to date. However, these rates 

declined with time after burning, returning to control levels 1 year after 

the fire. The authors concluded that cerrado is a minor source of N 2 O and 

a sink of atmospheric CH 4 . 

FIRE EFFECTS ON THE VEGETATION 

The flora of the herbaceous/undershrub stratum is highly resistant to fire 

(Coutinho 1990a). Some plants are annuals, growing in the rainy season, 

and many species exhibit subterranean organs such as rizomes, bulbs, and 

xylopodia (Rawitscher and Rachid 1946; chapter 7) that are well insulated 

by soil. A few days after the fire, the organs sprout with full vigor 

(Coutinho 1990a). 

Many plant species appear to depend upon fire for sexual reproduction. 

Intense flowering can be observed a few days or weeks after cerrado 

fires for many species of the herbaceous layer. Oliveira et al. (1996) 

observed 44 species of terrestrial orchids flowering after fires in areas of 

cerrado sensu stricto, campo sujo, and campo limpo, with some of the 

species flowering in the first two weeks after fire. Intense flowering of 

Habenaria armata was observed just after an accidental fire, while in the



four preceding years, when the vegetation was protected, no flowering 

individuals could be observed. Coutinho (1976) observed that a great 

number of species depend on fire to flower, responding with intense flowering 

to burns occurring any season of the year. In experiments with four 

species of the herbaceous/undershrub stratum (Lantana montevidensis, 

Stylosanthes capitata, Vernonia grandiflora and Wedelia glauca), 

Coutinho (1976) showed that burning, cutting the plants close to the soil, 

exposing them to a period of drought, or causing the death of their epigeous 

parts all resulted in a high percentage of flowering. He concluded 

that the effect of fire on the induction of flowering is not a result of thermal 

action or fertilization by the ashes. In a comparative study on the 

effects of fire and clipping on the flowering of 50 species of the flora of 

the herbaceous layer of a campo sujo, Cesar (1980) concluded that, for 

most of the species studied, flowering was independent of the season of 

burning and resulted in a similar phenological response in both treatments. 

Fire-induced flowering has been frequently reported, especially for 

grasses and geophytic lilies and orchids (Gill 1981). The causes of intense 

flowering may be related both to the increase in productivity after fire and 

to the damage caused by fire to the above-ground plant parts, possibly 

stimulating the production of flower primordia (Whelan 1995). 

Regeneration after fire for savanna vegetation through germination 

of soil stored seeds has been reviewed by Frost and Robertson (1987), but 

few studies report on seed dispersal in relation to fire (Whelan 1986). For 

the herbaceous vegetation of cerrado, Coutinho (1977) observed that 

Anemopaegma arvense, Jacaranda decurrens, Gomphrena macrocephala 

and Nautonia nummularia dispersed their seeds shortly after fire. This 

suggests that fire may be beneficial to such species, since it promotes or 

facilitates the dispersal of their anemocoric seeds. Seed germination of 

Echinolaena inflexa (a C3 grass common in all cerrado forms) was higher 

after a mid-dry season quadrennial fire in a campo sujo than in an area 

protected from fire for 21 years. However, recolonization by vegetative 

growth was higher than by seeds in both areas (Miranda 1996). After the 

fire, the density of E. inflexa was twice the density determined for the 

unburned area. Parron (1992) observed no difference in the density of E. 

inflexa in a campo sujo burned annually for 3 years, at the beginning of 

the dry season, and an adjacent area protected from fire for 3 years. These 

results may reflect the interaction between fire regime and the reproductive 

strategies of E. inflexa. San José and Farinas (1991), in a long-term 

monitoring of density and species composition in a Trachypogon savanna, 

showed that the dominant species, Trachypogon plumosus, was replaced 

by Axonopus canescens when fire was suppressed. The alteration in



species density was associated with differences in reproductive strategies: 

T. plumosus presented vegetative reproduction; A. canescens, sexual 

reproduction. For A. canescens, fire suppression may increase the probability 

of seedling survival. 

Most of the woody species of the cerrado present strong suberization 

of the trunk and branches, resulting in an effective thermal insulation of 

the internal living tissues of those organs during fires. Nevertheless, plants 

differ greatly in their tolerance to fire and in their capacity to recover subsequently. 

Most of the work on the response of savanna woody vegetation to 

fire is related to mortality, regeneration through seedlings, or resprouting 

from epicormic meristems or lignotubers (Frost and Robertson 1987). 

Landim and Hay (1996) reported that fire damaged 79% of the fruits of 

Kielmeyera coriacea, irrespective of tree height (1 to 3 m), but there were 

no differences in flower and bud initiation in the next reproductive period. 

Hoffmann (1998) observed that fruits and seeds of Miconia albicans, 

Myrsine guianensis, Roupala montana, Periandra mediterranea, Rourea 

induta, and Piptocarpha rotundifolia were damaged by a late dry season 

biennial fire, with a negative impact on sexual reproduction. All species 

but P. rotundifolia exhibited overall reductions in seed production in the 

years following fire (see chapter 9). An increase in the reproductive success 

for Byrsonima crassa after a mid-dry-season fire was reported by 

Silva et al. (1996). 

Although most of the cerrado trees are well insulated by thick bark, 

the small individuals may not have produced an effective insulation 

between fires, being more susceptible to the effects of the high temperature 

of the flames (Guedes 1993). Consequently, frequent fires reduce the 

density of woody vegetation through the mortality of the smaller individuals 

(Frost and Robertson 1987) and through the alteration of the 

regeneration rate of the woody species (Hoffmann 1998; Matos 1994; 

Miyanish and Kellman 1986). 

In a study of the effects of a biennial fire regime on the regeneration of 

Blepharocalyx salicifolius in cerrado sensu stricto, Matos (1994) found 

twice the number of individuals (seedlings and juveniles) in an area protected 

from fire for 18 years than in an area that was burned biennially. The 

mortality caused by the biennial fires was greater than 90% for seedlings 

and less than 50% for juveniles. The author estimated that the critical size 

for survival and resprout after fire was 50 cm in height with 0.6 cm in basal 

diameter. Hoffmann (1998), investigating the postburn reproduction of 

woody plants in areas subjected to biennial fires, observed that fire caused 

a high mortality in seedlings of Miconia albicans (100%), Myrsine



guianensis (86%), Roupala montana (64%), Periandra mediterranea 

(50%), and Rourea induta (33%). Root suckers of M. guianensis, R. montana, 

and R. induta had a higher survival rate, perhaps a consequence of 

their stem diameters (1.7 mm to 2.4 mm, two to four times greater than the 

seedlings) and connection with the mother plant (chapter 9). 

The effect of two annual fires on small individuals (from 20 cm to 

100 cm in height and diameter greater than 1.5 cm, at 30 cm from the 

soil) was investigated by Armando (1994) for nine woody species: Aspidosperma 

dasycarpon, Blepharocalyx salicifolius, Caryocar brasiliense, 

Dalbergia miscolobium, Hymenaea stigonocarpa, Stryphnodendron 

adstringens, Sclerolobium paniculatum, Siphoneugena densiflora, and 

Virola sebifera. The two consecutive fires resulted in a reduction of 10 cm 

in the mean height of the plant community and in a mortality of 4%. Only 

four species presented reduction in the number of individuals: D. miscolobium 

(12%), S. adstringens (14%), S. densiflora (14%), and S. paniculatum 

(15%). 

Sato (1996) determined that, for woody vegetation submitted to a 

biennial fire regime in the middle of the dry season, after 18 years of protection, 

the highest mortality rate occurred among the individuals with 

height between 0.3 m and 2.0 m. After the first fire they accounted for 

40% of the mortality, and for 72% after the second fire. Ramos (1990) 

observed that young trees and shrubs up to 128 cm tall, and with diameter 

smaller than 3 cm (measured at 30 cm above ground), are seriously 

damaged by biennial fires. 

Mortality rates related to fire season have been reported by Sato and 

Miranda (1996) and Sato et al. (1998) for cerrado vegetation. The authors 

considered only the individuals with stem diameter equal to or greater 

than 5.0 cm, at 30 cm from soil surface. After 18 years of protection from 

fire, three biennial fires at the middle of the dry season resulted in mortality 

rates of 12%, 6%, and 12%, with a final total reduction in the number 

of individuals of 27% after the third fire. In an experimental plot 

burned at the end of the dry season, the mortality rates were 12%, 13%, 

and 19%, with a reduction of 38% in the number of individuals. Williams 

(1995) presents similar values for mortality of tropical savanna trees in 

Australia, and higher rates are presented by Rutherford (1981) and Frost 

and Robertson (1987) for species of African savanna. Most of the trees 

that died in the second fire (≈60%) had suffered top kill during the first 

fire after a long period of protection. Some of the mortality following the 

second and third fires may be an indirect effect of fire. Cardinot (1998), 

studying the sprouting of Kielmeyera coriacea and Roupala montana after 

fires, in the same experimental plots, reported that the mortality of some



trees is a consequence of herbivory and nutrient shortage. The higher mortality 

rates for late dry season fires may be related to the phenology of 

many species of the cerrado vegetation that launch new leaves, flowers, 

and fruits during the dry season (Bucci 1997). 

Similar results have been reported by Rocha e Silva (1999) for campo 

sujo. After protection from fire for 18 years, three biennial fires, at the 

middle of the dry season, resulted in tree and shrub mortality rates of 5%, 

8% and 10%, reducing the number of individuals by 20%. In an experimental 

plot burned for four years at the middle of the dry season, mortality 

rates were 10% and 12%, with a reduction of 20% of the number 

of individuals, suggesting that two quadrennial burns produce the same 

mortality as three biennial burns. Of the 30 woody species present in the 

experimental plots, only 7 did not suffer alteration in the number of individuals 

after the fires: Byrsonima verbascifolia, Caryocar brasiliensis, 

Eremanthus mollis, Eriotheca pubescens, Qualea parviflora, Syagrus 

comosa, and Syagrus flexuosa. 

The alteration in the regeneration rates of woody species and the high 

mortality rate determined in these studies suggest that the biennial fire 

regime is changing the physiognomies of cerrado sensu stricto and campo 

sujo to an even more open form, with grasses as the major component of 

the herbaceous layer. This alteration, in turn, favors the occurrence of 

more intense and frequent fires. 

REFERENCES 

Andrade, S. M. A. 1998. “Dinâmica do Combustível Fino e Produção 

Primária do Estrato Rasteiro de Áreas de Campo Sujo de Cerrado Submetidas 

a Diferentes Regimes de Queima.” Master’s thesis, Universidade 

de Brasília, Brasília, Brazil. 

André-Alvarez, M. 1979. “Teor de Nutrientes Minerais na Fitomassa do 

Estrato Herbáceo Subarbustivo do Cerrado de Emas (Pirassununga, Est. 

de São Paulo).” Master’s thesis, Universidade de São Paulo, São Paulo, 

Brazil. 

Armando, M. S. 1994. “O Impacto do Fogo na Rebrota de Algumas Espécies 

de Árvores do Cerrado.” Master’s thesis, Universidade de Brasília, 

Brasília, Brazil. 

Batmanian, G. J. 1983. “Efeitos do Fogo na Produção Primária e a Acumulação 

de Nutrientes no Estrato Rasteiro de um Cerrado.” Master’s thesis, 

Universidade de Brasília, Brasília, Brazil. 

Bucci, F. F. B. 1997. “Floração de Algumas Espécies de Melastomataceae do 

Distrito Federal: Uso de Dados de Herbário para Obter Padrões Sazonais.” 

Master’s thesis, Universidade de Brasília, Brasília, Brazil.



Bustamante, M. M. C., F. B. Nardoto, A. A. Castro, C. R. Garofalo, G. B. 

Nardoto, and M. R. S. Silva. 1998. Effect of prescribed fires on the inorganic-N 

concentration in soil of cerrado areas and on the assimilation of 

inorganic-N by woody plants. In D. X. Viegas, ed., Proceedings of the 

14 th Conference on Fire and Forest Meteorology 2:1361–1379. Coimbra: 

University of Coimbra, Portugal. 

Cardinot, G. K. 1998. “Efeitos de Diferentes Regimes de Queima nos Padrões 

de Rebrotamento de Kielmeyera coriacea Mart. e Roupala montana 

Aubl:, Duas Espécies Típicas do Cerrado.” Master’s thesis, Universidade 


Castro, E. A. 1996. “Biomass, Nutrient Pools and Response to Fire in the Brazilian 

Cerrado.” Master’s thesis, Oregon State University, Corvallis, U.S. 

Castro, E. A. and J. B. Kauffman. 1998. Ecosystem structure in the Brazilian 

Cerrado: A vegetation gradient of aboveground biomass, root mass and 

consumption by fire. J. Trop. Ecol. 14:263–283. 

Castro Neves, B. M. and H. S. Miranda. 1996. Efeitos do fogo no regime térmico 

do solo de um campo sujo de Cerrado. In H. S. Miranda, C. H. 

Saito, and B. F. S. Dias, (eds.), Impactos de Queimadas em Áreas de Cerrado 

e Restinga, pp.20–30. Brasília: ECL/Universidade de Brasília. 

Cavalcanti, L. H. 1978. “Efeito das Cinzas Resultantes da Queimada sobre a 

Produtividade do Estrato Herbáceo Subarbustivo do Cerrado de Emas.” 

Ph.D. thesis, Universidade de São Paulo, São Paulo, Brazil. 

Cesar, L. 1980. “Efeitos da Queima e Corte sobre a Vegetação de um Campo 

Sujo na Fazenda Água Limpa, Brasília–DF.” Master’s thesis, Universidade 


Cooke, R. 1998. Human settlement of central America and northernmost 

South America (14,000–8000 BP). Quat. Int. 49/5:177–190. 

Coutinho, L. M. 1976. “Contribuição ao Conhecimento do Papel Ecológico 

das Queimadas na Floração de Espécies do Cerrado.” Livre Docente’s 

thesis, Universidade de São Paulo, São Paulo, Brazil. 

Coutinho, L. M. 1977. Aspectos ecológicos do fogo no cerrado: II. As 

queimadas e a dispersão de sementes de algumas espécies anemocóricas 

do estrato herbáceo subarbustivo. Bol. Bot.,USP 5:57–64. 

Coutinho, L. M. 1978. Aspectos ecológicos do fogo no cerrado: I. A temperatura 

do solo durante as queimadas. Rev. Bras. Bot. 1:93–97. 

Coutinho, L. M. 1979. Aspectos ecológicos do fogo no cerrado: III. A precipitação 

atmosférica de nutrientes minerais. Rev. Bras. Bot. 2:97–101. 

Coutinho, L. M. 1981. Aspectos ecológicos do fogo no cerrado: Notas sobre 

a ocorrência e datação de carvões vegetais encontrados no interior do 

solo, em Emas, Pirassunga, S.P. Rev. Bras. Bot. 4:115–117. 

Coutinho, L. M. 1990a. Fire in the ecology of the Brazilian Cerrado. In J. G. 

Goldammer, ed., Fire in the Tropical Biota: Ecosystem Processes and 

Global Challenges, pp.82–105. Berlin: Springer Verlag. 

Coutinho, L. M. 1990b. O Cerrado: A ecologia do fogo. Ciência Hoje 

12:22–30.


Fire Factor 65 

De Oliveira, P. E. 1992. “A palynological record of Late Quarternary vegetation 

and climatic change in southeastern Brazil.’’ Ph. D. thesis, Ohio 

State University, Columbus, Ohio, U.S.A. 



Recursos Naturais Renováveis. Brasília: IBAMA, Ministério do Meio 

Ambiente. 

Dias, I. F. O. 1994. “Efeitos da Queima no Regime Térmico do Solo e na Produção 

Primária de um Campo Sujo de Cerrado.” Master’s thesis, Universidade 


Dunin, F. X., H. S. Miranda, A. C. Miranda, and J. Lloyd. 1997. Evapotranspiration 

responses to burning of campo sujo savanna in Central Brazil. 

Proceedings of the Bushfire’97, pp. 146–151. Australia: CSIRO. 

Frost, P. H. G. and F. Robertson. 1987. The ecological effects of fire in savannas. 

In B. H. Walker, ed., Determinants of Tropical Savannas, pp. 

93–141. Oxford: IRL Press Limited. 

Gidon, N. and G. Delibrias. 1986. Carbon-14 dates point to man in the Americas 

32,000 years ago. Nature 321:769–771. 

Gill, A. M. 1981. Adaptative response of Australian vascular plant species to 

fires. In A. M. Gill, R. H. Groves, and I. Noble, eds., Fire and the Australian 

Biota, pp. 243–272. Canberra: Australian Academy of Science. 

Guedes, D. M. 1993. “Resistência das Árvores do Cerrado ao Fogo: Papel da 

Casca como Isolante Térmico.’’ Master’s thesis, Brasília, Universidade de 


Haridasan, M. 2000. Nutrição mineral de plantas nativas do Cerrado. Rev. 

Bras. Fisiol. Vegetal 12:54–64. 

Hoffmann, W. 1998. Post-burn reproduction of woody plants in a neotropical 

savanna: The relative importance of sexual and vegetative reproduction. 

J. Appl. Ecol. 35:422–433. 

Kauffman, J. B., D. L. Cummings, and D. E. Ward. 1994. Relationships of 

fire, biomass and nutrient dynamics along a vegetation gradient in the 

Brazilian cerrado. J. Ecol. 82:519–531. 

Kayll, A. J. 1968. Heat tolerance of tree seedlings. Proceedings of the Tall 

Timbers Fire Ecology Conference. 3:88–105. 

Kelmann, M., K. Miyanishi, and P. Hiebert. 1987. Nutrient retention by 

savanna ecosystem. J. Ecol. 73:953–962. 

Kozovits, A. R., M. M. C. Bustamante, L. F. Silva, G. T. Duarte, A. A. Castro, 

and J. R. Magalhães. 1996. Nitrato de amônio no solo e sua assimilação 

por espécies lenhosas em uma área de cerrado submetida a 

queimadas prescritas. In H. S. Miranda, C. H. Saito and B. F. S. Dias, 

eds., Impactos de Queimadas em Áreas de Cerrado e Restinga, pp. 

137–147. Brasília: ECL/Universidade de Brasília. 

Landim, M. F. and J. D. Hay. 1996. Impacto do fogo sobre alguns aspectos 

da biologia reprodutiva de Kielmeyera coriacea Mart. Rev. Bras. Biol. 

56:127–134.



Luke, R. H. and A. G McArthur. 1978. Bushfire in Australia. Canberra: Australian 

Government Publishing Service. 

Matos, R. B. M. 1994. “Efeito do Fogo sobre Regenerantes de Blepharocalyx 

salicifolius (H.B.K.) Berg. (Myrtaceae) em Cerrado Aberto, Brasília, DF.’’ 

Master’s thesis, Universidade de Brasília, Brasília, Brazil. 

Miranda, A. C., H. S Miranda, I. F. O. Dias, and B. F. S Dias. 1993. Soil and 

air temperatures during prescribed cerrado fires in Central Brazil. J. 

Trop. Ecol. 9:313–320. 

Miranda, A. C., H. S Miranda, J. Grace, J. Lloyd, J. McIntyre, P. Meier, P. 

Riggan, R. Lockwood, and J. Brass. 1996b. Fluxes of CO 2 over a Cerrado 

sensu stricto in Central Brazil. In J. H. Gash, C. A. Nobre, J. M. 

Roberts, and R. L. Victoria, eds., Amazonian Deforestation and Climate, 

pp. 353–363. Chichester: John Wiley and Sons. 

Miranda, A. C., H. S Miranda, J. Lloyd, J. Grace, J. A. Francey, J. McIntyre, 

P. Meier, P. Riggan, R. Lockwood, and J. Brass. 1997. Fluxes of carbon, 

water and energy over Brazilian cerrado: An analysis using eddy covariance 

and stable isotopes. Plant Cell Environ. 20:315–328. 

Miranda, H. S. 2000. Queimadas de Cerrado: Caracterização e impactos na 

vegetação. In Plano de Prevenção e Combate aos Incêndios Florestais do 

DF, pp.133–149. Brasília: Secretaria de Meio Ambiente e Recursos 

Hídricos. 

Miranda, H. S., E. P. Rocha e Silva, and A. C. Miranda. 1996a. Comportamento 

do fogo em queimadas de campo sujo. In H. S. Miranda, C. H. 

Saito and B. F. S. Dias, eds., Impactos de Queimadas em Áreas de Cerrado 

e Restinga, pp. 1–10. Brasília: ECL/Universidade de Brasília. 

Miranda, M. I. 1997. “Colonização de Campo Sujo de Cerrado por Echinolaena 

inflexa (Poaceae).’’ Master’s thesis, Universidade de Brasília, 


Mistry, J. 1998. Fire in the cerrado (savannas) of Brazil: An ecological review. 

Prog. Phys. Geog. 22:425–448. 

Miyanish, K. and M. Kellman. 1986. The role of fire in the recruitment of 

two neotropical savanna shrubs, Miconia albicans and Clidemia sericea. 

Biotropica 18:224–230. 

Montes, R. and J. J. San José. 1993. Ion movement in the well drained 

neotropical savanna. Curr. Top. Bot. Res. 1:391–418. 

Neto, W. N., S. M. A. Andrade, and H. S. Miranda. 1998. The dynamics of 

the herbaceous layer following prescribed burning: a four year study in 

the Brazilian savanna. In D. X. Viegas, ed., Proceedings of the 14th Conference 

on Fire and Forest Meteorology 2:1785–1792. Coimbra: University 

of Coimbra, Portugal. 

Oliveira, R. S., J. A. N. Batista, C. E. B. Proença, and L. Bianchetti. 1996. 

Influência do fogo na floração de espécies de Orchidaceae em cerrado. 

In H. S. Miranda, C. H. Saito, and B. F. S. Dias, eds., Impactos de 

Queimadas em Áreas de Cerrado e Restinga, pp.61–67. Brasília: 

ECL/Universidade de Brasília.


Fire Factor 67 

Parron, L. M. 1992. “Dinâmica de Crescimento, Sobrevivência, Produção de 

Sementes, Repartição de Biomassa Aérea e Densidade das Gramíneas 

Echinolaena inflexa e Trachypogon folifolius, numa Comunidade de 

Campo Sujo, Com e Sem Queima.’’ Master’s thesis, Universidade de 

Brasília, Brasília, Brazil. 

Pivello, V. R. and L. M. Coutinho. 1992. Transfer of macro-nutrients to the 

atmosphere during experimental burnings in an open cerrado (Brazilian 

savanna). J. Trop. Ecol. 8:487–497. 

Poth, M., I. C. Anderson, H. S. Miranda, A. C. Miranda, and P. J. Riggan. 

1995. The magnitude and persistence of soil NO, N 2 O, CH 4 , and CO 2 

fluxes from burned tropical savanna in Brazil. Global Geoch. Cycles 

9:503–513. 

Prous, A. 1992. Arqueologia Brasileira. Brasília: Editora Universidade de 

Brasília. 

Raison, R. J. 1979. Modification of the soil environment by vegetation fires, 

with particular reference to nitrogen transformations: A review. Plant 

and Soil 51:73–108. 

Ramos, A. E. 1990. “Efeitos da Queima sobre a Vegetação Lenhosa do Cerrado.’’ 


Rawitscher, F. and M. Rachid. 1946. Tronco subterrâneos de plantas 

brasileiras. An. Acad. Bras. Ciênc. 18:261–280. 

Rocha e Silva, E. P. and H. S. Miranda. 1996. Temperatura do câmbio de espécies 

lenhosas do cerrado durante queimadas prescritas. In R. C. Pereira 

and L. C. B Nasser, eds., Anais do VII Simpósio sobre o Cerrado/1 st 

International Symposium on Tropical Savannas, pp.253–257. Brasília: 

Empresa Brasileira de Pesquisa Agropecuaria. 

Rocha e Silva, E. P. 1999. “Efeito do Regime de Queima na Taxa de Mortalidade 

e Estrutura da Vegetação Lenhosa de Campo Sujo de Cerrado.’’ 


Rutherford, M. C. 1981. Survival, regeneration and leaf biomass changes in 

a woody plants following spring burns in Burkea africana-Ochna pulchra 

savanna. Bothalia 13:531–552. 

San José, J. J. and M. R. Farinas. 1991. Temporal changes in the structure of 

a Trachypogon savanna protected for 25 years. Acta Oecol. 12:237–247. 

San José, J. J. and E. Medina. 1977. Produción de metéria organica en la 

sabana de Trachypogon, Calabozo, Venezuela. Bol. Soc. Ven.Cienc. Nat. 

134:75–100. 

Salgado-Laboriau, M. L. and K. R. F. C. Vicentini. 1994. Fire in the Cerrado 

32,000 years ago. Curr. Res. Pleist. 11:85–87. 

Santos, A. J. B. 1999. “Fluxos de Energia, Carbono e Água em Áreas de 

Campo Sujo.’’ Master’s thesis, Universidade de Brasília, Brasília, Brazil. 

Sato, M. N. 1996. “Taxa de Mortalidade da Vegetação Lenhosa do Cerrado 

Submetida a Diferentes Regimes de Queima.’’ Master’s thesis, Universidade 


Sato, M. N. and H. S. Miranda. 1996. Mortalidade de plantas lenhosas do



cerrado sensu stricto submetidas a diferentes regimes de queima. In H. S. 

Miranda, C. H. Saito, and B. F. S. Dias, eds., Impactos de Queimadas em 

Áreas de Cerrado e Restinga, pp.102–111. Brasília: ECL/Universidade de 

Brasília. 

Sato, M. N., A. A. Garda, and H. S. Miranda. 1998. Effects of fire on the 

mortality of woody vegetation in Central Brazil. In D. X. Viegas, ed., 

Proceedings of the 14th Conference on Fire and Forest Meteorology 

2:1777–1784. Coimbra: University of Coimbra, Portugal. 

Silva, D. M. S., J. D. Hay, and H. C. Morais. 1996. Sucesso reprodutivo de 

Byrsonima crassa (Malpighiaceae) após uma queimada em um cerrado 

de Brasília–DF. In H. S. Miranda, C. H. Saito, and B. F. S. Dias, eds., 

Impactos de Queimadas em Áreas de Cerrado e Restinga, pp. 122–127. 

Brasília: ECL/Universidade de Brasília. 

Silva, F. C. 1990. “Compartilhamento de Nutrientes em Diferentes Componentes 

da Biomassa Aérea em Espécies Arbóreas de um Cerrado.’’ Master’s 

thesis, Universidade de Brasília, Brasília, Brazil. 

Silva, G. T. 1999. “Fluxos de CO 2 em um Campo Sujo Submetido a 

Queimada Prescrita.’’ Master’s thesis, Universidade de Brasília, Brasília, 

Brazil. 

Silva, I. S. 1983. “Alguns Aspectos da Ciclagem de Nutrientes em uma Área 

de Cerrado (Brasília: DF): Chuva, Produção e Decomposição de Liter.” 


Vicentini, K. R. C. F. 1993. “Análise palinológica de uma vereda em Cromínia-GO.’’ 


Whelan, R. J. 1986. Seed dispersal in relation to fire. In D. R. Murray, ed., 

Seed Dispersal, pp. 237–271. Sydney: Academic Press. 

Whelan, R. J. 1995. The Ecology of Fire. Cambridge, Mass: Cambridge University 

Press. 

Williams, R. J. 1995. Tree mortality in relation to fire intensity in a tropical 

savanna of the Kakadu region, Northern Territory, Australia. CalmSci. 

Suppl. 4:77–82. 

Wright, S. J. 1970. A method to determine heat-caused mortality in bunchgrass. 

Ecology 51:582–587. 

Wright, S. J. and A. W. Bailey. 1982. Fire Ecology. New York: John Wiley and 

Sons.


5 

Past and Current Human 

Occupation, and Land Use 

Carlos A. Klink and Adriana G. Moreira 

BIT BY BIT, THE CERRADO LOSES GROUND 

“Drive south from Rondonópolis, and for mile after mile the flat tableland 

stretches away to the far horizon, a limitless green prairie carpeted 

with swelling crops. The monotony of the landscape is broken only by the 

artifacts of modern agribusiness: a crop-dusting plane swoops low over 

the prairie to release its chemical cloud, while the occasional farmhouses 

have giant harvesting machines lined up in the yard outside. It could be 

the mid-western United States. In fact, it is the very heart of tropical South 

America, its central watershed, in the Brazilian State of Mato Grosso.’’ 

That is how a major newspaper (The Economist 1999) has recently 

described the cerrado landscape in central Brazil. 

Over the past four decades, the Cerrado Biome has become Brazil’s 

largest source of soybeans and pastureland, and a significant producer of 

rice, corn, and cotton. In contrast to the small farms in other parts of 

Brazil, a very different kind of farming, capital-intensive, large-scale, 

mechanized, and scientific, has developed in the poor soils but cheap land 

of the cerrado. 

From a narrow, revenue generation perspective, the benefits of commercial 

agriculture in the cerrados are clear. Soybeans and soy products 

are among the largest of Brazil’s export commodities, and the cerrados 

support the largest cattle herd in the country. Even so, the development 

of modern agriculture in the cerrado region has exacerbated social 

69



inequality at a high environmental cost: landscape fragmentation, loss of 

biodiversity, biological invasion, soil erosion, water pollution, land degradation, 

and heavy use of chemicals (Klink et al. 1993, 1995; Davidson et 

al. 1995; Conservation International et al. 1999; see chapter 18). 

Transformation of the cerrado landscapes continues at a fast pace. 

Here we describe these changes and discuss their implications for sustainable 

use and conservation. We start with a brief historical overview 

of past and current human occupation and then explain the main driving 

forces behind agricultural expansion and the recent transformation of the 

cerrados. We end with a look at future land use in the cerrado region, 

offering recommendations for progressive management and conservation. 

HUMAN OCCUPATION 

Presettlement Inhabitants 

Archaeological evidence indicates that 9,000 YBP (Years Before Present) 

a hunter-gatherer culture flourished in the open habitats of the cerrado. 

Excavations near the town of Itaparica (state of Bahia), and later from 

other sites (especially in the town of Serranópolis, state of Goiás), revealed 

chipped stone tools and rock paintings from cave campsites (Schmitz 

1992; Barbosa and Nascimento 1994). Stratified prehistoric deposits 

uncovered in Serranópolis revealed cerrado tree fruits and wood, and faunal 

remains, to 8,800 and 6,500 YBP (Schmitz 1992). 

The first known cerrado humans, known as the Itaparica Tradition 

(Schmitz 1992), were foragers who used simple tools and subsisted on 

native plants and game, such as deer, armadillos, and lizards. Paintings on 

cave walls typically depict animals but rarely show human figures 

(Schmitz 1992). The existence of cerrado cave dwellers, and related cultures 

in the northeastern caatinga dry lands and in the Amazon rainforest, 

are changing our understanding of South American paleoindians 

(Guidon 1991; Roosevelt et al. 1996) 

The Itaparica Tradition persisted until 6,500 years YBP, when it was 

replaced by “specialized’’ hunters and gatherers subsisting primarily on 

small animals and terrestrial mollusks (Barbosa and Nascimento 1994). 

Why the Itaparica Tradition was replaced by what became the modern 

indigenous population is unknown. It has been hypothesized that the 

regional climate may have become wetter and permitted dispersion to 

other areas in Brazil (Barbosa and Nascimento 1994; see chapter 3). 

The last millennium has been characterized by sedentary indigenous


Past and Current Human Occupation 71 

populations that hunted and made use of utensils and agriculture until the 

arrival of the first Europeans. In all, 230 indigenous societies speaking 

170 distinct languages have been identified in Brazil. Culturally distinct 

societies have occupied the cerrado region, including the Bororo, Karajá, 

Parakanã, Kayapó, Canela, Krahô, Xavante and Xerente. 

At the time of Anglo-European colonization in the 16th century, the 

Xavante and the Xerente societies occupied extensive areas in the cerrado 

(Maybury-Lewis 1988). They foraged on native plant fruits and roots, 

fished, and hunted. They also used cerrado plants for home and utensil 

construction. The Xavante set fire to the native grasslands to facilitate 

hunting of cerrado game (Leeuwenberg and Salimon 1999). The number 

of indigenous people in the cerrado decreased dramatically since the contacts 

with the European colonizers. Initially they were decimated by disease 

or enslavement; more recently, the advance of the agricultural frontier 

has displaced many from their native lands. 

Postsettlement Occupation 

The idea of central Brazil as a region to be conquered and transformed 

has been embedded in the Brazilian society since colonial times. Cerrado 

exploration started with the Portuguese who searched for precious minerals 

and Indians for enslavement in the sixteenth century. The first permanent 

settlements were established in the early eighteenth century and 

were associated with gold mining. Some farming developed among these 

communities, and economic activity shifted to cattle production with the 

exhaustion of the mines (Klink et al. 1993, 1995). 

It was only after the Paraguay War (1864–1870) that cerrado occupation 

was promoted by the Brazilian authorities, concerned with the 

defense and maintenance of the border, challenged by low population density. 

The government encouraged occupation of the border province of 

southern Mato Grosso by providing incentives to grow tea (Almeida and 

Lima 1956). The existence of large areas of native grassland also made 

the area attractive for cattle raising and the development of large ranches. 

The occupation of the core area of the cerrado region was delayed 

until the 1900s. The first major economic boom in the cerrados came during 

the period 1920–1930, when coffee growing and processing industries 

were flourishing in the state of São Paulo, which hence became the major 

market for cerrado cattle (Hees et al. 1987). Later, the Getulio Vargas 

government (1930–1945) actively promoted a colonization of southern 

Goiás, providing land, subsidies, and technical assistance, thus encouraging 

farmers to settle on and clear fertile forested lands.



Infrastructure 

Both the distance of the cerrado from the major Brazilian coastal urban 

areas and the lack of a transport system have posed major obstacles for 

cerrado occupancy and development. Construction of the first railway, 

which linked São Paulo to Mato Grosso, was initiated in 1905 but not 

completed until 1947 (Lucarelli et al. 1989). After 1946, roads replaced 

railways as the main link between the Brazilian regions. In conjunction 

with the construction of Brazil’s new capital, Brasília, in the cerrado heartland 

in the late 1950s, roads crossing the cerrado were built to connect 

Brasília with São Paulo, Rio de Janeiro, and Belo Horizonte in the southeast, 

and Belém in the Amazon region (Lucarelli et al. 1989). The construction 

of Brasília and highways linking the new capital with the main 

Brazilian cities made way for the cerrado occupation that began in the 

mid-1960s. At the time of this writing, the construction of railways is 

again becoming fashionable. Most noticeable is Ferronorte, a newly built 

railway linking the cerrado to Brazil’s largest port, Santos, in the state of 

São Paulo. 

Population Growth and Urbanization 

The population of the cerrado region grew by 73% between 1950 and 

1960, mostly due to employment opportunities associated with the construction 

of Brasília (Lucarelli et al. 1989). The population growth is not 

limited to this decade, however, since from 1870 to 1960 the regional population 

grew at a rate twice that of Brazil, as a result of internal migration. 

Preliminary data of the national demographic survey done in 2000 

(IBGE 2001) indicate that the cerrado population may have reached 18 

million inhabitants in 2000 (see table 5.1) There has been a strong trend 

towards urbanization since the 1940s, particularly in the southern part of 

the cerrado. As of 2000, almost 30% of the cerrado inhabitants lived in 

eight cities: Brasília, Goiânia, Teresina, Campo Grande, Uberlândia, 

Cuiabá, Montes Claros, and Uberaba. Internal migration is the main 

cause of population growth in urban areas. The population of Palmas, the 

state capital of Tocantins, has grown 12.2% since 1996 (IBGE 2001), 

attracted by the ongoing construction of the city. 

Population growth and agricultural development had important 

implications for cerrado land use. Until the late 1950s, the contribution 

of the cerrados to Brazil’s agricultural output was very low, with the 

extent of farmland and agricultural output contributing



Table 5.1 The Population 

of Cerrado from 1872 to 2000 

Year Population 

1872 221,000 

1890 320,000 

1900 373,000 

1920 759,000 

1940 1,259,000 

1950 1,737,000 

1960 3,007,000 

1970 5,167,000 

1980 7,545,000 

1991 12,600,000 

2000 18,000,000 

Sources: Klink et al. 1993, 1995, and 

IBGE 2001, preliminary data. 

when the cerrados became Brazil’s major producer and exporter of important 

cash crops. 

LAND USE 

Changes in land use in the cerrado are a function of the technological 

innovations, capital investments, energy, and knowledge applied with the 

objective of promoting the expansion of intensive agriculture. Until 40 

years ago, the region was used primarily for extensive cattle raising. At 

the time of this writing it is estimated that 35% of its natural cover has 

been totally converted into planted pastures with African grasses and cash 

crops, mainly soybeans and corn (see table 5.2). It is estimated that 60% 

of the cerrado area is used by humans directly (Conservation International 

et al. 1999). In 1970, 202,000 km 2 of land (an area 2.2 times the size of 

Portugal) was used for intensive crops and planted pastures in the region. 

The area cleared had grown 3.3 times since then, and in 1996 it was equivalent 

to 672,000 km 2 (table 5.2), an area the size of Texas in the U.S. 

The total area, and sometimes the annual rate, of native vegetation 

clearing is greater in the cerrado than in the Brazilian Amazon rainforest. 

For example, between 1970 and 1975 the average annual rate of land 

clearing in the cerrado region was 40,600 km 2 per year, 1.8 times the estimated 

deforestation rate of the Amazon rainforest during the period 

1978–1988 (see fig. 5.1). Projections for the year 2000 show that the total



Table 5.2 Changes in Land Cover and Land Use 

in the Cerrado, 1970-1996 

1970 1985 1995/1996 a 

Total cleared land (km 2 ) 202,000 508,000 672,000 

Crops (km 2 ) 41,000 95,000 103,000 

Planted pastures (km 2 ) 87,000 309,000 453,000 

Cleared but not in use 74,000 104,000 116,000 b 

Proportion of total cerrado 10.6% 26.7% 33.6% 

area cleared 

a Dates of last national agriculture survey. 

b Estimated from Klink et al. (1995). 

cleared land in the cerrados can reach 800,000 to 880,000 km 2 , roughly 

1.6 times the size of France (Klink et al. 1995). During the 25-year period 

shown in table 5.2, the area under crops had grown 250%, planted pasture 

520%, and land that had been cleared but had not been cultivated, 

150%. This “uncultivated” land represents productive land that had been 

cleared in the past but either had never been used or had been abandoned. 

Several environmental and economical conditions favored these 

transformations. Although the rainfall distribution within the year is 

uneven, the mean total annual rainfall (1500 mm) is considered sufficient 

for crop production. Temperatures are warm year-round, and sunshine 

does not restrict photosynthesis. Level topography and deep well-drained 

soils propitiate mechanization, and the cerrado savannas and woodlands 

are less expensively and more easily cleared for farming or cattle ranching 

than tropical rainforest. 

PROMOTION OF AGRICULTURAL EXPANSION 

The growth of agriculture in the Cerrado Biome is the outcome of a combination 

of factors, including the growth of the demand for agricultural 

products in Brazil and abroad, public investments in infrastructure, technological 

advances in the agronomic sciences, and the implementation of 

policies for regional development, particularly during the rapid growth 

period of 1968–1980. The strong performance of the Brazilian economy, 

associated with a national development policy aimed at integrating the 

“empty’’ spaces of the cerrado and Amazon regions into the capitalist 

economy of the richer southern and southeastern regions of the country, 

created the right atmosphere for investments (Mueller 1990; Klink et al.



Figure 5.1 The total cleared area of (A) natural cerrado vegetation, and (B) 

Amazon rainforest (data for the cerrado from IBGE 1999 and Klink et al. 

1993, 1995; data for the Amazon from INPE 2000). 

1995). The economic stagnation and crises of the 1980s changed these 

prospects, but did not totally eliminate the programs and policies that 

resulted in the clearing of new lands for agricultural purposes. 

The main causes for the expansion of the agricultural frontier in the 

cerrado can be divided into two broad categories: (1) policies aimed at



expanding the agricultural output of the country as a whole; and (2) policies 

specifically targeting the cerrado, mainly developmental programs, 

use of new technologies, and subsidies. 

Subsidized Loans and Inflation 

Credit was one of the main instruments utilized by the Brazilian government 

to promote agricultural development. Probably nowhere are the 

environmental, economic and social contradictions of intervention in the 

economic process as visible as in the credit policy of Brazil in the 

1960–1980 period. Subsidized credit has had a direct impact on the profitability 

expectation of farmers with access to loans, as well as an indirect 

but powerful impact on land prices. The greatest incentive was provided 

by a policy of low-interest farm loans. Because these loans were at fixed 

rates of 13% to 15%, they were further subsidized in that they ignored 

rapidly rising inflation rates (averaging 40% annually during that time). 

Even lower rates were imposed for fertilizers, pesticides and equipment. 

As a result, credit to the rural sector grew between 1969 and 1979 at a 

rate 188% faster than total output (199% for agriculture, 164% for livestock 

production) (Klink et al. 1993, 1995). The growth was particularly 

high during the 1969–1976 period, when agricultural credit rose at a 

yearly 24% compound interest rate (World Bank 1982). After 1977, 

credit flows, especially long-term investment credit, slowed. 

Loans were not evenly distributed among crop types. Over 75% of 

production loans were concentrated in six crops: soybeans, rice, coffee, 

wheat, maize, and sugarcane. Soybeans alone received 20% of the credit 

available for Brazilian farmers (Klink et al. 1993). Since loans were allocated 

based on the size of the planted area, they encouraged extensive and 

inefficient agriculture (Goedert 1990). Credit was concentrated in the 

south and southeast regions of Brazil, which received almost twice as much 

credit per hectare as the cerrado. The cerrado, in turn, received 70% more 

credit than the Amazon and northeast, where the small producers of food 

crops are concentrated. This readily available credit increased the demand 

for land in the cerrado and drove up land prices (Klink et al. 1993). 

POLOCENTRO and PRODECER 

The failure of the settlement initiatives in the Amazon region (Mahar 

1989) and the desire to hoist the cerrado economy (Klink et al. 1993) led 

to the creation of the Program for the Development of the Cerrado 

(POLOCENTRO) in 1975. Its goals were to settle farmers in places with



good farming potential (twelve areas were selected), to improve infrastructure 

(mainly the construction of secondary roads and electricity), and 

to develop agricultural research and technology. The program’s original 

target was to farm 60% of the exploited area and to give preference to 

foodstuffs. Farmers received subsidized loans, and credit lines were at relatively 

low fixed interest rates with no monetary correction. Rising inflation 

and considerable grace periods effectively transformed these loans 

into donations (Klink et al. 1995). 

This program had a major impact on cerrado agriculture. Between 

1975 and 1982, 3,373 agriculture projects were approved, totaling 

U.S.$577 million. Medium and large farmers benefited most (Klink et al. 

1993). Eighty-one percent of the farms were 200 ha or larger in size and 

accounted for 88% of the total funds allocated; farms larger than 1,000 

ha accounted for 39% of all projects and received more than 60% of the 

credit. An estimated 2.4 million ha of native land had been transformed 

between 1975 and 1980 alone. The program’s original target to give preference 

to foodstuffs was never realized. Instead it induced the expansion 

of commercial agriculture in the cerrado region. Most of the land was 

used for cattle ranching, and soybean became the main crop. 

After the inauguration of the “New Republic’‘ in 1985, many developmental 

programs, including the POLOCENTRO, were closed in Brazil. 

However, in the late 1970s a new program, the Brazil-Japan Cooperative 

Program for the Development of the Cerrado (PRODECER), was initiated. 

It selects experienced farmers from the south and southeast of Brazil 

for settlement in the cerrado. It is financed by loans from both the Brazilian 

and the Japanese governments. In contrast with the POLOCENTRO, 

loans are granted at real (not fixed) interest rates. PRODECER is still 

active. At the time of this writing, two projects are being implemented, in 

Balsas (7°30' S, 46°20' W) and Porto Nacional (10°08'S, 48°15' W), and 

each will establish 40,000 hectares of new agricultural land, mainly for 

soybeans. 

New Technologies 

The development of appropriate technologies to enable farmers to deal 

with the nutrient-poor, acidic soils has also helped promote the agricultural 

development in the cerrado. These include technologies for soil 

fertilization, such as application of phosphate fertilizer and lime to correct 

both nutrient deficiency and acidity; Rhizobium-based nitrogen fixation; 

the development of crop varieties; heavy use of herbicides and pesticides; 

and modern machinery (chapter 2). The performance of the



cerrado Agricultural Center of the Brazilian Institute for Agricultural 

Research (EMBRAPA) has been impressive in almost every respect (Paterniani 

and Malavolta 1999). However, the technologies that were developed 

were strongly biased toward medium and large capital-intensive 

farmers, and cash crops, especially soybeans (Klink et al. 1993). 

Because the expansion of agriculture in the cerrado is based on the 

incorporation of technology requiring increased use of machinery in agricultural 

operations, the need for manpower decreases in both time and 

space. As the number of tractors increases, relative employment growth 

decelerates. Between 1970 and 1985, the period of greatest agricultural 

expansion, employment grew 2.7% per year, whereas the farmed area 

expanded 5.4%, planted pastures grew 8.4%, the bovine herd increased 

5.5%, and the stock of tractors, 13.6% (Klink et al. 1993, 1995). 

Minimum Price Policy 

The price support policy in Brazil has been in effect since the 1930s to 

guarantee a minimum price for agricultural products. It assumed a special 

significance for the cerrado in the 1980s. Responding to pressures 

from the World Bank and the International Monetary Fund, farm credit 

lines were restricted, reduced, or eliminated. Consequently, production 

costs increased substantially in the cerrado. The government then started 

to purchase large amounts of cerrado products, particularly soybeans, 

rice, and corn (Mueller and Pufal 1999). This favored farmers from 

remote areas in the cerrado, because they benefited indirectly from the 

nationwide unified fuel price. The net result of the price support policy 

was that it encouraged the expansion of commercial farming in areas that 

could not have supported profitable production without subsidies, and, 

consequently, the deforestation of new land. 

AGRICULTURAL EXPANSION 

Integration of cerrado farming and ranching into the national economy is 

a recent phenomenon. Agricultural activities, however, are not evenly distributed, 

and intraregional differences exist. In the state of Mato Grosso 

do Sul, southern Mato Grosso, central, southwestern and southeastern 

Goiás, the Federal District, the Triângulo Mineiro, and western Minas 

Gerais, modern, consolidated farming activities are well established. The 

basic infrastructure is well developed, along with access to the more 

dynamic markets of the country. The “modern’’ subregion is responsible



for most of the soybean, corn, coffee, and bean production in the cerrado. 

It also produces a large share of the regional rice and cassava, as well as 

most of the bovine herd. In the remaining cerrado areas, agriculture is still 

developing, the road network and commercialization facilities are precarious, 

but further deforestation is expected as agriculture expands. 

The expansion of cash crops, particularly soybean and corn, has been 

considerable. Soybean production, virtually nonexistent in the cerrado in 

the early 1970s, is currently 15 million metric tons (see fig. 5.2A). Production 

advanced slowly until 1989, suffered a fall in 1990 due to the collapse 

of the agricultural policy, and grew rapidly after that. The expansion 

during the 1990s was due to the restructuring of agricultural policies, 

high international prices for soybean and soy products, and productivity 

increases. The latter is evident in the disparity between yields and acreage 

in the 1990s (fig. 5.2A). 

Transport and commercialization difficulties are hampering the expansion 

of soybean in the northern cerrado. This may change if the planned 

export river and railway corridors to the north are built; these will combine 

the Carajá and Norte-Sul railroads, inland waterways and highways, going 

to the port of São Luis in the northern state of Maranhão. 

Corn production is currently expanding in some parts of the cerrado 

region, having increased from 0.5 million metric tons in 1960 to 5.5 million 

metric tons in 1995 (see fig. 5.2B). The “modern’’ subregion is 

responsible for more than 85% of the total increase in planted area of the 

cerrado. Cerrado corn production is part of the south central agri-industrial 

complex, which has experienced dramatic productivity increases 

over the last fifteen years. For example, between 1985 and 1994, production 

almost doubled, whereas the farmed area increased only 20% 

(Klink et al. 1995). 

Rice has been an important crop in the cerrado, its production peaking 

in 1980 (fig. 5.2B). Since 1980, rice production has declined to the 

levels of the 1970s. The plunge in rice production has resulted primarily 

from planting soybean in recently cleared areas, instead of planting rice 

first as was done in the past. 

Cerrado produces 40% of the Brazilian soybeans and 22% of corn. 

Despite the recent declines in production, the share of rice in the national 

market is still 12%. A recent trend is the increase of cotton production in 

the cerrado. As a strategy to reduce risks by diversifying crop production, 

soybean farmers are also investing in cotton, which represents 33% of the 

national production (Mueller and Pufal 1999). 

Coffee production in 1990 was close to 250,000 metric tons. Nevertheless, 

the marked decline in the international price of coffee, the retreat



Figure 5.2 (A) Expansion of soybeans in the cerrado. Area is given in 1,000 

hectares and production in 1,000 metric tons. (B) Growth of corn and rice in 

the cerrado. Production is given in 1,000 metric tons (data from IBGE 1999; 

Klink et al. 1993, 1995; Mueller and Pufal 1999). 

of government support, and the virtual collapse of the International Coffee 

Agreement led to waning interest on the part of the farmers, which in 

turn led to the eradication of some coffee plantations in the Triângulo 

Mineiro (state of Minas Gerais), the main coffee producing zone in the 

region. Bean (Phaseolus) production has had a modest and irregular



expansion in the cerrado. Most of the bean production comes from the 

“modern’’ subregion, where productivity is substantially higher than in 

the remaining cerrado areas because of the use of irrigated cropping systems, 

particularly in the state of Goiás (Klink et al. 1995). 

Livestock 

The cerrado is an important cattle ranching region. Ranching varies from 

relatively modern and efficient farms to extensive operations with rudimentary 

methods and low productivity. Modern techniques have been 

most readily adopted in the areas closest to markets with better access to 

technical assistance, basic and support infrastructure, and relatively 

sophisticated meat packing facilities. 

The bovine herd in the cerrado region increased by 21.4 million head 

between 1970 and 1985: that is, from 16.6 million to almost 38 million 

(see fig. 5.3). The average yearly growth was initially high (3.6%) but has 

decreased since 1995, partly due to the relative saturation in areas with 

better transport, processing, and commercialization infrastructure. Total 

number of cattle in the cerrado is more than 51 million, representing 33% 

of the national herd. The recent frontier expansion areas may have a certain 

growth potential but are constrained by deficient transport networks. 

Ranching in these areas is done mainly on extensive native pastures. 

The increase in the number of bovines is a direct consequence of the 

increase in the area of planted pastures. Planted pastures are by far the 

most important land use in the cerrado, representing 67% of the total 

cleared land (table 5.2), an area the size of Sweden and Denmark combined. 

To establish planted pastures, the savannas are clear-cut and 

burned, and then seeded with grasses of African origin, such as Andropogon 

gayanus, Brachiaria brizantha, B. decumbens, Hyparrhenia rufa, 

and Melinis minutiflora. Legumes, like Centrosema and Stylosanthes, are 

used as a source of protein (Barcellos 1996; Billoz and Palma 1996; Sano 

et al. 1999; see chapter 7). 

Charcoal 

Farming in the cerrado requires the removal of trees and roots, which usually 

are piled up and burned. Today it is becoming common to sell the firewood 

from deforestation for charcoal production to offset the costs of 

land clearing (Klink et al. 1995). This is usually done by itinerant, family-based 

charcoal producers. In the past, the use of cerrado vegetation for 

charcoal production was associated with the installation of large steel



Figure 5.3 Expansion of the number of cattle and planted pastures in the 

cerrado. Cattle is given in 1,000 heads and pasture in 1,000 ha (data from 

IBGE 1999; Klink et al. 1993, 1995). 

factories in the state of Minas Gerais in the 1940s and 1950s. As the natural 

cerrado vegetation around the plants vanished, and the transport 

costs rose in the late 1980s, the steel plants started to reforest extensively 

cleared cerrado land with Eucalyptus trees for charcoal production (Klink 

et al. 1995). 

ENVIRONMENTAL IMPACTS 

The perception of abundant land has driven most of the land use changes 

in the cerrado in the last 40 years. Although the extent of environmental 

modification is less well documented than the economical transformation, 

it is clear that the net impact has been negative. Soil mechanization 

of large tracts of monoculture expose great areas of bare soil, resulting 

in erosion and soil compaction. Soil losses on a 5% slope under bare fallow 

can be as high as 130 metric tons per hectare annually (Goedert 

1990). 

Agricultural expansion has led to an increase in burning, and areas 

still covered with natural vegetation are now burned almost every year



(Klink et al. 1993; chapter 4). Fire controls the proportion of woody and 

herbaceous plants in the cerrado. Because of their negative effect on tree 

and shrub seedlings, fires tend to favor herbaceous plants at the expense 

of woody plants. Protection against fire for sufficiently long periods of 

time favors the appearance of more wooded physiognomies in the cerrado 

(Moreira 2000; chapters 4, 8, 9). 

The cerrado is one of the richest terrestrial ecoregions (“hotspots’’) 

on Earth (Mittermeier et al. 1999). It has a unique fauna and the largest 

diversity of all savanna floras in the world (ca. 10,000 species) (Ratter et 

al. 1997; Conservation International et al. 1999; Mittermeier et al. 1999; 

chapter 18). Large-scale transformation of the cerrado landscape is 

endangering its biodiversity with habitat fragmentation and even animal 

extinction. Three amphibians, 15 reptiles, and 33 bird species are now 

threatened with extinction in the cerrado (Conservation International et 

al. 1999). 

The development of a modern agriculture in the cerrado has not been 

able to maintain and increase the ecosystem yield capacity without degradation. 

For instance, it is estimated that 50% of the planted pastures with 

African grasses in the cerrado (an area of 250,000 km 2 ) is degraded, to 

the detriment of its production capacity (Barcellos 1996). Extensive cultivation 

of African grasses in the cerrado has increased concerns over 

biotic invasions. African grasses proliferate, persist, and spread into a new 

range in which they cause detriment to the environment (Berardi 1994; 

Klink 1994). Given their current scale, African grasses are major agents 

of change (chapter 7). 

One of the most widespread African species is the molassa or fat grass 

(Melinis minutiflora), known for the disruption it brings to biodiversity 

and ecosystem functioning in other parts of the world (Berardi 1994). 

Although it has been superseded by more productive African species, it is 

extensively found in disturbed areas, roadsides, abandoned plantations, 

and nature reserves in the cerrado (Klink et al. 1995). It can attain 

extremely high biomass that, when dried, becomes a highly combustible 

fuel that initiates a grass-fire interaction capable of preventing the 

regrowth of natural vegetation (Berardi 1994). In places where M. minutiflora 

reaches high cover, the diversity of the local flora is considerably 

lower than in native areas (chapter 7). Compared to natural savanna fires, 

fire temperature in M. minutiflora was far higher and had a much longer 

residence time, with flames over 6 meters high (Berardi 1994). 

Large expanses of natural cerrado vegetation have been transformed 

from a mixture of trees and grasses into planted pastures and crops. Many 

cerrado trees and shrubs are deep-rooted and uptake soil water at 8 meters



depth or more (Oliveira 1999; see also chapter 10). Simulations of the 

effects of the conversion of natural cerrado into open grasslands on 

regional climate have shown reduced precipitation by approximately 

10%, an increase in the frequency of dry periods within the wet season, 

changes in albedo, and increased mean surface air temperature by 0.5°C 

(Hoffmann and Jackson 2000). 

Considerable amounts of carbon are stored in roots and soil organic 

matter in the cerrado. Up to 70% of the alive biomass in cerrado vegetation 

is underground (Castro and Kauffmann 1998), and up to 640 tons 

of soil organic carbon has been found to a depth of 620 cm under natural 

vegetation (Abdala et al. 1998). Given the extension of cerrado vegetation 

already transformed into planted pastures and agriculture, it is possible 

that a significant change in both root biomass and the regional carbon 

sink has already occurred. 

STRATEGIES FOR FUTURE CERRADO LAND USE 

The agricultural development in the cerrado has been selective. Subsidies 

have favored commercial crops, and credit concessions have provided a 

strong incentive to open new land. Increases in production are due more 

to the increased area of land under cultivation than to gains in productivity. 

The impact of many of the agricultural development policy initiatives 

pursued since 1960 has also been inequitable. Most of the programs 

have favored wealthy farmers and large landholders. The use of technology-intensive 

agriculture generated only modest gains in farm employment 

and explains the decline of the rural population in the areas of 

dynamic agriculture in the southern cerrado (Cunha et al. 1994). Moreover, 

the environmental impact has been negative. 

The demand for food, fibers, and other agricultural products exerts 

much pressure on the natural resources of the cerrado. Therefore it is necessary 

to identify the best agricultural growth options: whether to encourage 

intensification on areas already cleared, or to continue the expansion 

of the agricultural frontier. Both options have ecological costs. Horizontal 

expansion leads to a larger transformed area, destruction of habitats 

and biodiversity, and natural ecosystems disturbances, while farming 

intensification may increase soil degradation, chemical contamination, 

and water pollution. In the future, both types of agricultural systems shall 

coexist in the cerrado. The prospects for commercial farming are promising: 

the devaluation of Brazil’s currency in January 1999 should help farm 

exporters; the increase of farm credit by the Banco do Brasil should stim-



ulate cash crops; the transport costs should be lowered when the Ferronorte 

railroad is fully running and loading terminals are built; and the 

use of new technologies (e.g., genetically modified crops) is already on the 

horizon. 

The modernization of agriculture should permit more intensive land 

use in the “modern’‘ subregion of the Cerrado Biome, in which case no 

new area needs to be cleared. This tendency has been perceived in some 

parts of Brazil, as well as in other countries. Nevertheless, as isolated areas 

of the cerrado become more accessible by roads and railways, expansion 

of the agriculture frontier should be expected, especially for planted 

pastures. 

In theory cerrado land is appropriate for sustainable agricultural 

activities, but this requires adequate public action. Usually the fragility of 

the natural ecosystems does not rank among farmers’ top priorities, even 

where cash crops or cattle raising is inappropriate. The cerrado has an 

enormous environmental heterogeneity, including ecosystems that are relatively 

stable and resistant to changes, and others that are extremely sensitive 

to anthropogenic modifications. 

Sustainable use of the cerrados will only be achieved if the strategies 

that direct its use are clearly defined. For example, the use of the no-till 

(direct drilling) technique that reduces soil erosion during soybean farming 

(Landers 1996), or the use of organic farming along with pig raising 

(Mueller and Pufal 1999), are methods already in use for large scale farming 

in the cerrado. 

Based on the existing knowledge, many authors have suggested that 

an agri-environmental zoning of cerrado is desirable and possible (Goedert 

1990; Cunha et al. 1994; Klink et al. 1995; Mueller and Pufal 1999). 

For instance, Mueller and Pufal (1999), based on an initial proposition 

made by Cunha et al. (1994), suggest that four major categories can be 

used for this zoning: areas used preferentially for crops, areas for crops 

and cattle raising, areas for crops and afforestation, and areas for conservation, 

particularly those with fragile ecosystems. This last category 

could be as large as a third of the cerrado area, or 630,000 km 2 (Mueller 

and Pufal 1999). 

This chapter has shown that policies were often formulated with little 

attention to their implications for cerrado land use and, as a result, 

have increased rates of land degradation, encouraged inefficient forms of 

development, and caused social conflict. The time is ripe for policies that 

take into consideration the interests of all cerrado land users (including 

small farmers and indigenous people) and the environmental services provided 

by the cerrado (Klink and Moreira 2000). These policies should rely



on mechanisms that couple market forces with the natural economic and 

ecological capacity of cerrado ecosystems. 


We thank Paulo S. Oliveira and an anonymous referee for a thorough 

review that greatly improved the chapter. 

REFERENCES 

Abdala, G., L. Caldas, M. Haridasan, and G. Eiten. 1998. Above and belowground 

organic matter and root: Shoot ratio in a cerrado in central 

Brazil. Braz. J. Ecol. 2:11–23. 

Almeida, F. F. M and M. A. Lima. 1956. The west central plateau and Mato 

Grosso “Pantanal.’’ In Fundação Instituto Brasileiro de Geografia e 

Estatística, ed., 18th International Geographical Congress Excursion 

Guidebook no. 1. Rio de Janeiro: Instituto Brasileiro de Geografia e 

Estatística. 

Barbosa, A. S. and I. V. Nascimento. 1994. Processos culturais associados à 

vegetação. In M. N. Pinto, ed., Cerrado: Caracterização, Ocupação e 

Perspectivas, pp. 155–170. Brasília: Universidade de Brasília. 

Barcellos, A. O. 1996. Sistemas extensivos e semi-intensivos de produção: 

Pecuária bovina de corte nos cerrados. In R. C. Pereira and L. C. B. 

Nasser, eds., Biodiversidade e Produção Sustentável de Alimentos e 

Fibras nos Cerrados, pp. 130–136. VIII Simpósio sobre o cerrado. 

Brasília: Empresa Brasileira de Pesquisa Agropecuária. 

Berardi, A. 1994. “Effects of the African grass Melinis minutiflora on Plant 

Community Composition and Fire Characteristics of a Central Brazilian 

Savanna.’’ Master’s thesis, University College, University of London, 

London, UK. 

Billaz, R. and V. Palma. 1996. La expansion de la agricultura y de la ganaderia 

en las sabanas tropicales de America del Sur. In R. C. Pereira and 

L. C. B. Nasser, eds., Biodiversidade e Produção Sustentável de Alimentos 

e Fibras nos Cerrados (VIII Simpósio sobre o Cerrado), pp. 484–491. 


Castro, E. A. and J. B. Kauffman. 1998. Ecosystem structure in the Brazilian 

cerrado: A vegetation gradient of aboveground biomass, root mass and 

consumption by fire. J. Trop. Ecol. 14:263–284. 

Conservation International, Funatura, Fundação Biodiversitas, and Universidade 

de Brasília. 1999. Ações Prioritárias para a Conservação da Biodiversidade 

do Cerrado e Pantanal. Brasília: www.bdt.org.br/workshop/cerrado/br. 

Cunha, A. S., C. C. Mueller, E. R. A. Alves, and J. E. Silva. 1994. Uma Avali-



ação da Sustentabilidade da Agricultura nos Cerrados. Estudos de Política 

Agrícola no. 11. Brasília: Instituto de Pesquisa Econômica e Aplicada and 

Programa das Nações Unidas para o Desenvolvimento (UNDP). 

Davidson, E. A., D. C. Nepstad, C. A. Klink, and S. E. Trumbore. 1995. Pasture 

soils as carbon sink. Nature 376:472–473. 

The Economist. 1999. Growth in the prairies. April 10th issue, p.34. 

Goedert, W. J. 1990. Estratégias de manejo das savanas. In G. Sarmiento, ed., 

Las Sabanas Americanas: Aspectos de su Biogeografia, Ecologia y Utilizacíon, 

pp. 191–218. Guanare: Acta Científica Venezolana. 

Guidon, N. 1991. Peintures Préhistoriques du Brésil. Paris: ERC. 

Hees, D. R., M. E. P. C. de Sá, and T. C. Aguiar. 1987. A evolução da agricultura 

na região Centro-Oeste na década de 70. Rev. Bras. Geogr. 49:197–257. 

Hoffmann, W. A. and R. B. Jackson. 2000. Vegetation-climate feedbacks in the 

conversion of tropical savanna to grassland. J. Climate 13: 1593–1602. 

IBGE (Instituto Brasileiro de Geografia e Estatística). 1999. Censo Agropecuário 

1995–1996. Brasília: www.ibge.gov.br. 

IBGE (Instituto Brasileiro de Geografia e Estatística). 2001. Censo Demográfico 

2000: Dados Preliminares. Brasília: www.ibge.gov.br. 

INPE (Instituto Nacional de Pesquisas Espaciais). 2000. Monitoring of the 

Brazilian Amazonian Forest by Satellite. São José dos Campos: Ministério 

da Ciência e Tecnologia, www.inpe.mct.gov.br. 

Klink, C. A. 1994. Clipping effects on size and tillering of native and African 

grasses of the Brazilian savannas (the “cerrado”). Oikos 70:365–376. 

Klink, C. A., R. H. Macedo, and C. C. Mueller. 1995. De Grão em Grão o 

Cerrado Perde Espaço. Brasília: World Wildlife Fund–Brazil and Pró-Cer. 

Klink, C. A. and A. G. Moreira. 2000. Valoração do potencial do cerrado em 

estocar carbono atmosférico. In A. G. Moreira and S. Schwartzman, eds., 

As Mudanças Climáticas Globais e os Ecossistemas Brasileiros, pp. 

85–91. Brasília: Editora Foco. 

Klink, C. A., A. G. Moreira, and O. T. Solbrig. 1993. Ecological impacts of 

agricultural development in the Brazilian cerrados. In M. D. Young and 

O. T. Solbrig, eds., The World’s Savannas, pp. 259–282. London: The 

Parthenon Publishing Group. 

Landers, J. N. 1996. O plantio direto na agricultura: O caso do cerrado. In I. 

V. Lopes, G. S. Bastos Filho, D. Bille, and M. Bale, eds., Gestão Ambiental 

no Brasil, pp. 2–33. Rio de Janeiro: Editora da Fundação Getúlio 

Vargas. 

Leeuwenberg, F. and M. Salimon. 1999. Os Xavantes na Balança das Civilizações. 

Brasília: Pratica Gráfica. 

Lucarelli, H. Z., N. R. Innocencio, and O. M. B. L. Fredrich. 1989. Impacto 

da construção de Brasília na organização do espaço. Rev. Bras. Geogr. 

51:99–138. 

Mahar, D.J. 1989. Government Policies and Deforestation in Brazil’s Amazon 

Region. Environment Department Working Paper 7. Washington, 

D.C.: The World Bank.



Maybury-Lewis, D. 1988. The Savage and the Innocent. 2nd edition. Boston: 

Beacon Press. 

Mittermeier, R.A., N. Myers, and C. G. Mittermeier. 1999. Hotspots: Earth’s 

Biologically Richest and Most Endangered Terrestrial Ecoregions. Mexico: 

CEMEX and Conservation International. 

Moreira, A. G. 2000. Effects of fire protection on savanna structure in Central 

Brazil. J. Biogeogr. 27:1021–1029. 

Mueller, C. C. 1990. Políticas governamentais e a expansão recente da 

agropecuária no Centro-Oeste. Planejamento e Políticas Públicas 

3:45–73. 

Mueller, C. C. and D. V. L. Pufal. 1999. Atividades Agropecuárias no Cerrado. 

Brasília: Instituto Sociedade, População e Natureza. 

Oliveira, R. S. 1999. “Padrões Sazonais de Disponibilidade de Agua nos Solos 

de um Cerrado Denso e um Campo Sujo e Evapotranspiração.’’ Master’s 


Paterniani, E. and E. Malavolta. 1999. La conquista del “cerrado’’ en el Brasil: 

Victoria de la investigación científica. Interciencia 24:173–176. 

Ratter, J. A., J. F. Ribeiro, and S. Bridgwater. 1997. The Brazilian cerrado vegetation 

and threats to its biodiversity. Ann. Bot. 80:223–270. 

Roosevelt, A. C., M. L. da Costa, C. L. Machado, M. Michab, N. Mercier, 

H. Valladas, J. Feathers, W. Barnett, M. I. da Silveira, A. Henderson, J. 

Sliva, B. Chernoff, D. S. Reese, J. A. Holman, N. Toth, and K. Schick. 

1996. Paleoindian cave dwellers in the Amazon: The peopling of the 

Americas. Science 272:373–384. 

Sano, E. A., A. O. Barcellos, and H. S. Bezerra. 1999. Área e distribuição espacial 

de pastagens cultivadas no Cerrado brasileiro. Boletim de Pesquisa 

da Embrapa—Cerrados 3:1–21. 

Schmitz, P. I. 1992. A história do velho Brasil. Ciência Hoje, Special Issue Eco- 

Brasil: 95–102. 

World Bank. 1982. A Review of Agricultural Policies in Brazil. Washington, 

DC: World Bank Publications.


6 

Vegetation Physiognomies 

and Woody Flora 

of the Cerrado Biome 

Ary T. Oliveira-Filho and James A. Ratter 

The Cerrado Biome of tropical South America covers 

about 2 million km 2 , an area approximately the same as that of Western 

Europe, representing ca. 22% of the land surface of Brazil, plus small 

areas in eastern Bolivia and northwestern Paraguay (fig. 6.1). It extends 

from the southern borders of the Amazonian forest to outlying areas in 

the southern states of São Paulo and Paraná, occupying more than 2° of 

latitude and an altitudinal range from near sea-level to 1,800 m. The distribution 

of the Cerrado Biome is highly coincident with the plateaux of 

central Brazil, which divide three of the largest South American water 

basins: those of the Amazon, Plate/Paraguay, and São Francisco rivers. 

The cerrados form part of the so-called diagonal of open formations 

(Vanzolini 1963) or corridor of “xeric vegetation’’ (Bucher 1982), which 

includes the much drier Caatinga in northeastern Brazil and the Chaco in 

Paraguay-Bolivia-Argentina. This corridor runs between the two main 

areas of moist forest of tropical South America: the Amazonian forest in 

the northwest, and the Atlantic forest in the east and southeast. 

The Cerrado Biome was named after the vernacular term for its predominant 

vegetation type, a fairly dense woody savanna of shrubs and 

small trees. The term cerrado (Portuguese for “half-closed,’’ “closed,’’ or 

“dense’’) was probably applied to this vegetation originally because of the 

difficulty of traversing it on horseback. 

The typical vegetation landscape of the Cerrado Biome consists of 

savanna of very variable structure, termed cerrado sensu lato, on the 

91


92 the plant community 

Figure 6.1 Geographic distribution of the Cerrado Biome in South America. 

The isolated patches of Amazonian savannas are not indicated because 

they are inserted in a different Biome. Key to state codes: Amazonas (AM), 

Bahia (BA), Ceará (CE), Distrito Federal (DF), Espírito Santo (ES), Goiás 

(GO), Maranhão (MA), Minas Gerais (MG), Mato Grosso (MT), Mato 

Grosso do Sul (MS), Pará (PA), Paraná (PR), Piauí (PI), Rio de Janeiro (RJ), 

Rondônia (RO), São Paulo (SP), Tocantins (TO). After IBGE (1993), and 

Ribeiro and Walter (1998). 

well-drained interfluves, with gallery forests or other moist vegetation following 

the watercourses. In addition, areas of richer soils in the biome are 

clothed in mesophytic forests. In this chapter we consider the main environmental 

factors associated with the Cerrado Biome, its main vegetation 

physiognomies, and the diversity and origin of its woody flora. In general, 

common usage will be followed throughout this chapter, and cerrado 

(sensu lato) will be termed cerrado, cerrados, or cerrado vegetation, while


Vegetation Physiognomies and Woody Flora 93 

the other vegetation types of the Cerrado Biome will be indicated by clear 

distinguishing names. 

ENVIRONMENTAL VARIABLES DETERMINING 

THE DISTRIBUTION OF THE CERRADO BIOME 

The distribution of the Cerrado Biome shown in figure 6.1 is determined 

basically by the predominance of cerrado (sensu lato) in the landscape. 

The factors determining the distribution of cerrado vegetation have long 

been a subject of controversy, but in general the following are considered 

important: seasonal precipitation, soil fertility and drainage, fire regime, 

and the climatic fluctuations of the Quaternary (e.g., Eiten 1972; Furley 

and Ratter 1988; Ratter 1992; Oliveira-Filho and Ratter 1995; Furley 

1999). These are, on the whole, the same factors identified as important 

in maintaining savanna biomes worldwide, although grazing also plays 

an important role in other continents such as Africa (see references in 

Werner 1991; Furley et al. 1992). Although the climate of the Cerrado 

Biome varies considerably, it is mostly typical of the rather moister 

savanna regions of the world. There is a remarkable variation across the 

region in both the average annual temperature, ranging from 18° to 28°C, 

and rainfall, from 800 to 2,000 mm, with a very strong dry season during 

the southern winter (approx. April–September) (Dias 1992). Nevertheless, 

as a number of authors point out, rainfall seasonality cannot 

entirely explain the predominance of cerrado vegetation, as the present 

climatic conditions would favor the establishment of forests in most of 

the Cerrado Biome region (e.g., Rizzini and Pinto 1964; Reis 1971; Klein 

1975; Coutinho 1978; Van der Hammen 1983). 

Soil fertility and moisture are other important factors to be considered 

in the distribution of cerrado vegetation. Most soils of the Cerrado 

Biome are dystrophic, with low pH and availability of calcium and magnesium, 

and high aluminium content (Furley and Ratter 1988; see also 

chapter 2). Moreover, cerrado only grows on well-drained soils, therefore 

concentrating on interfluves and avoiding valley bottoms. In fact, within 

the Cerrado Biome, vegetation physiognomies other than cerrado itself 

are found on the patches of base-rich soils as well as on sites liable to 

waterlogging for considerable periods. 

Although of fundamental importance, seasonal rainfall and low soil 

fertility are apparently insufficient to explain the present distribution of 

cerrado vegetation completely. For instance, vast areas in southeastern 

Brazil with strongly seasonal rainfall and infertile soils (e.g., eastern 

Minas Gerais) have a continuous cover of semideciduous forests and show



no trace of cerrado (Oliveira-Filho and Fontes 2000). This leaves disturbance 

by fire as a possible key determinant of the presence of forest or 

cerrado. Most of the flora of cerrado is of fire-adapted species, involving 

not only fire tolerance but also fire dependency (Coutinho 1990; Braithwaite 

1996; see also chapters 4, 9). Palaeoenvironmental studies have confirmed 

that forests and cerrado vegetation showed successive expansions 

and contractions in the region, following the climatic fluctuations of the 

Quaternary (chapter 3). The last expansion of cerrado occurred during 

the brief dry episode of the Holocene; however, since then, forests have 

not regained their original area, although rainfall apparently returned to 

previous levels (Ledru 1993). This failure of forest to expand into the cerrado 

may be largely determined by human-induced fires, which have been 

important elements in the region at least since the mid-Holocene (Ledru 

et al. 1998). 

Despite past disputes concerning the relative importance of various 

environmental factors in determining the distribution of cerrado vegetation 

(see Eiten 1972; Goodland and Ferri 1979), it is now widely accepted 

that climate, soils, and fire are highly interactive in their effect on vegetation 

within the Cerrado Biome. For example, the seasonal climate favors 

the outbreak of fire during the dry season when cerrado plant cover is 

highly inflammable; recurrent fires tend to prevent the vegetational succession 

into forest and also cause soil impoverishment; and increasingly 

base-poor soils liable to strong water deficit during the dry season restrict 

the establishment of forest species (chapter 4). Therefore one should consider 

these interactions when trying to understand the distribution of both 

the Cerrado Biome and its constituent vegetation types. 

VEGETATION PHYSIOGNOMIES OF THE CERRADO BIOME 

The vegetation of the Cerrado Biome shows a remarkable physiognomic 

variation, and many authors have attempted to produce an efficient classification. 

A revision is found in Ribeiro and Walter (1998), who also proposed 

a comprehensive modern treatment of the subject using a pragmatic 

but more detailed classification than that adopted here. Most problems 

can be explained by the fact that any classification category is actually a 

segment of a multidimensional vegetational continuum (particularly 

within the cerrado sensu lato). Table 6.1 is an attempt to show the major 

vegetational categories and the main determining environmental factors. 

Traditional Brazilian nomenclature for cerrado vegetation is used here 

because it is well accepted, unambiguous, and appropriate.



Table 6.1 Main Vegetation Physiognomies of the Cerrado Biome and 

Their Association with Soil Fertility Levels and Ground Water Regimes 

Soil Fertility in Terms of Base Saturation 

Groundwater 

regime Low Intermediate High 

Strongly drained Cerrado s.l.: campo Mesotrophic facies Mesophytic deciduous 

sites with deep limpo, campo sujo, cerradão or meso- dry forests on interwater 

table and campo cerrado, phytic semideciduous fluves and slopes 

seasonal water cerrado sensu stricto, forests on interfluves 

deficit at topsoil or dystrophic facies 

level cerradão 

High soil moisture Dystrophic facies Mesotrophic facies Valley forests (mesofor 

most of the year cerradão or valley cerradão or valley phytic semideciduous 

within catchment forests (mesophytic forests (mesophytic or deciduous) 

areas evergreen) semideciduous) 

Permanently Riverine forests (evergreen): gallery forests (swampy and wet), alluvial 

waterlogged to forests, or riverside portions of valley forests 

very wet alongside 

rivercourses 

Successive periods Seasonal grasslands: valley-side marshes (veredas), floodplain 

of soil water- grasslands (pantanal), or rocky grasslands (campo rupestre). Both 

logging and strong valley-side marshes and floodplain grasslands may contain scattered 

water deficit earth-mounds (campo de murundus) 

In general, cerrado physiognomies predominate in the landscape 

because well-drained, low-fertility soils are the predominant substratum. 

Cerrado vegetation tends to be replaced by forest physiognomies on sites 

with increased water availability and/or soil fertility, while seasonal grasslands 

appear where periods of strong water deficit follow periods of 

waterlogging. It is worth noting that fire, an important element for Cerrado 

Biome ecosystems, may interfere with the vegetation-environment 

relationships represented in table 6.1, particularly in forest-cerrado transitions. 

A brief description of each main vegetation physiognomy and its 

plant community follows. 

CERRADO SENSU LATO 

Cerrado vegetation is generally characterized by a mixture of plants of 

two fairly distinct layers. The first, hereafter called the woody layer,



includes trees and large shrubs; the other, the ground layer, is composed 

of subshrubs and herbs (chapter 7). In cerrado, it is often difficult to distinguish 

between trees and large shrubs. We define “large shrub’’ here as 

a plant bearing a perennial woody stem (i.e., not a hemixyle) and generally 

attaining a height of at least 1.3 m. The flora of both woody and 

ground layers has typical features of pyrophytic savanna vegetation. The 

trees are of low contorted form with thick, corky, fire-resistant bark. Sclerophylly 

is common: many leaves have thick cuticles, sunken stomata, and 

greatly lignified and sometimes silicified tissues, and are often of considerable 

longevity. Xylopodia (swollen, woody underground structures) are 

well developed in both the woody and ground layers, and the hemixyle 

growth form, where woody shoots of annual duration are developed from 

an underground xylopodium, is particularly common (e.g., Jacaranda 

decurrens, Anacardium humile, Andira humilis). On the other hand, 

annuals are rare: Warming (1892, 1973) calculated that they constituted 

less than 6% of the herbaceous species of the cerrados of Lagoa Santa in 

central Minas Gerais. 

The cerrado sensu lato encompasses a series of vegetation physiognomies 

from open grasslands to dense woodlands, and more or less recognizable 

stages of this continuum are given vernacular names. Dry grassland 

without shrubs or trees is called campo limpo (“clean field’’); grassland with 

a scattering of shrubs and small trees is known as campo sujo (“dirty 

field’’).Where there are scattered trees and shrubs and a large proportion of 

grassland, the vegetation is termed campo cerrado (“closed field’’); the next 

stage when the vegetation is obviously, at least visually, dominated by trees 

and shrubs often 3–8 m tall and giving more than 30% crown cover but 

with still a fair amount of herbaceous vegetation between them is known 

as cerrado (sensu stricto). The last stage is an almost closed woodland with 

crown cover of 50% to 90%, made up of trees, often of 8–12 m or even 

taller, casting a considerable shade so that the ground layer is much reduced; 

this form is called cerradão (Portuguese augmentative of cerrado). It is 

unfortunate that, in common usage, the term cerrado should refer both to 

Brazilian savanna vegetation in its generic sense (cerrado sensu lato), and 

to one of its subvariants (cerrado sensu stricto). Clearly the dividing line 

between the five cerrado physiognomies is somewhat arbitrary, but workers 

in the field usually agree surprisingly well on the classification. Some 

authors exclude campo limpo from cerrado sensu lato as it has no woody 

layer, but we prefer to include it since it is usually composed of the characteristic 

cerrado ground layer and thus forms the most open part of the continuum. 

Examples of these physiognomies are given in figure 6.2. 

Many factors are probably operative in determining which of these 

physiognomies of cerrado vegetation occurs in a given site. Goodland and



Figure 6.2 Cerrado physiognomies. (A) Large expanse of cerrado sensu 

stricto, Gilbués, Piauí; (B) campo limpo, Chapada dos Veadeiros, Goiás; (C) 

campo sujo, Brasília, Federal District; (D) campo cerrado, Alter do Chão, 

Pará; (E) cerrado sensu stricto, Loreto, Maranhão; (F) mesotrophic facies cerradão, 

Doverlândia, Goiás. 

Pollard (1973) correlated increased production of woody elements with 

an increasing soil fertility gradient in the Triângulo Mineiro (western 

Minas Gerais), while Lopes and Cox (1977), who studied over 500 sites 

covering much of the central core cerrado area, arrived at the same conclusion. 

The data of other authors, however, fail to demonstrate this correlation, 

and sometimes show well-developed cerradão on no more fertile 

soils than sparser forms of cerrado nearby (e.g., Ribeiro 1983; Ribeiro 

and Haridasan 1990). The explanation of these contradictory results lies 

at least partly in the occurrence of two floristically different forms of cerradão, 

which have not been distinguished by the majority of authors.



Mesotrophic facies cerradão occurs on soils of intermediate fertility 

in the cerrado landscape, particularly in terms of calcium and magnesium 

levels (see fig. 6.2F). This community is readily recognized by the presence 

of a number of indicator species such as Magonia pubescens, Callisthene 

fasciculata, Dilodendron bipinnatum and Terminalia argentea (Ratter 

1971, 1992; Ratter et al. 1973, 1977; Oliveira-Filho and Martins 

1986, 1991; Furley and Ratter 1988). This facies of cerradão is of very 

widespread occurrence in the Cerrado Biome and is often associated with 

the transition to mesophytic forest, the climax vegetation of the most 

base-rich soils in the Cerrado Biome. In fact, it indicates a soil intermediate 

in fertility between that of the more dystrophic forms of cerrado and 

that of the mesophytic deciduous dry forest (see table 6.1). Many of the 

indicator species also occur in deciduous dry forest and the arboreal 

caatinga vegetation (xeric thorn woodland) of NE Brazil, to which the 

deciduous dry forest is closely related. 

A floristically different type of cerradão, the dystrophic facies cerradão 

(Ratter 1971; Ratter et al. 1973, 1977; Oliveira-Filho and Martins 

1986, 1991; Furley and Ratter 1988), is found on base-poor soils and has 

its own indicator species, such as Hirtella glandulosa, Emmotum nitens, 

Vochysia haenkeana, and Sclerolobium paniculatum. This facies of cerradão 

is also very widespread in the Cerrado Biome and tends to be associated 

with the savanna-forest transition on base-poor and often sandy 

soils (see table 6.1). Therefore, it is commonly found fringing valley 

forests at the feet of sandstone plateaux and in the transition to Amazonian 

forests on sandy soils. 

Fire is undoubtedly an extremely important factor affecting the density 

of the woody layer of cerrado vegetation (chapters 4, 9). Although 

most woody species are strongly fire-adapted, fires at too frequent intervals 

damage them and favor the ground layer, thus producing more open 

physiognomies; conversely protection from fire allows the woody vegetation 

to close, and characteristic cerradão tree species (e.g., Emmotum 

nitens, Protium heptaphyllum, Virola sebifera) to establish themselves. It 

seems probable that, in the past, before the advent of frequent manmade 

fires, the denser arboreal physiognomies of cerrado occupied a much 

larger area than they do today. Some authors, such as Warming (1892, 

1973) and Coutinho (1978, 1990), believe that the climax vegetation of 

most of the Cerrado Biome is actually cerradão and that the other more 

open physiognomies exist as successional phases determined mainly by 

fire regime, all tending to evolve to cerradão in the absence of fire. Succession 

from cerrado to cerradão in the Distrito Federal and Mato Grosso 

is discussed in Ratter (1980, 1986) and Ratter et al. (1973, 1978a), respectively. 

In both cases there are strong indications that the succession pro-



ceeds to forest, probably reflecting the re-expansion of forests after the 

brief Holocenic dry episode. 

Soil moisture is another factor which may affect the physiognomic 

continuum of cerrado, as this vegetation, mostly restricted to soils which 

are well drained throughout the year, commonly shows pronounced physiognomic 

and floristic changes as it approaches seasonally waterlogged 

grasslands. Where interfluvial cerrados are bordered by veredas (valleyside 

marshy grasslands, see below) it is common to observe a decline in 

mean tree height and an increasing density of woody plants toward the 

cerrado margin, where a distinct community of cerrado tree species more 

resistant to soil saturation occurs (Ratter et al. 1973; Oliveira-Filho et al. 

1989). A similar community appears on raised islands of ground which 

appear in both veredas and seasonally flooded alluvial grasslands (Diniz 

et al. 1986; Oliveira-Filho 1992a). The distribution of these islands frequently 

produces a landscape known as campo de murundus, consisting 

of an expanse of open grasslands dotted with a regular pattern of raised 

earthmounds bearing cerrado trees and shrubs, and often termitaria. 

Landscapes of this type are particularly common in such seasonally inundated 

floodplains of the Mato Grosso Pantanal (Ratter et al. 1988; Dubs 

1992), Varjão do Araguaia, and Ilha do Bananal (Ratter 1987). Tree 

species which tolerate soil saturation and are commonly found in these 

marginal cerrado communities are Curatella americana, Byrsonima crassifolia, 

B. orbignyana, Dipteryx alata, Tabebuia aurea, and Andira 

cuiabensis (Furley and Ratter 1988; Oliveira-Filho 1992b). 

Goodland (1971) focused attention on the role of aluminium in the 

cerrado, and other authors have since concentrated on the subject. Levels 

of this element are so high in the dystrophic cerrado soils (chapter 2) as 

to be extremely toxic to most cultivated plants, but most native species 

are aluminium-tolerant, as would be expected. The tolerant species 

include a number of diverse unrelated families that accumulate aluminium 

in their tissues, particularly in leaves, but also in roots (Haridasan 1982), 

such as the Vochysiaceae, various Rubiaceae, some Myrtaceae, Miconia 

spp. (Melastomataceae), Symplocos (Symplocaceae), Strychnos pseudoquina 

(Loganiaceae), Myrsine spp. (Myrsinaceae), and Vellozia (Velloziaceae). 

Some families, including the Vochysiaceae, are obligate aluminium 

accumulators and cannot grow in its absence. 

MESOPHYTIC SEASONAL FORESTS 

The occurrence of mesophytic seasonal forests—comprising both deciduous 

and semideciduous forests—within the Cerrado Biome is very



extensive and generally underestimated (Oliveira-Filho and Ratter 1995). 

Where more fertile soils occur in the region, the climax vegetation is definitely 

mesophytic forest (see table 6.1). These forests are found, for 

example, on base-rich alluvial deposits in the Mato Grosso Pantanal 

(Prance and Schaller 1982; Ratter et al. 1988; Dubs 1992), on calcareous 

outcrops (Ratter et al. 1973, 1977, 1978b, 1988; Prado et al. 1992), on 

soils originated from basalt (Oliveira-Filho et al. 1998), and in valleys 

where the topography has cut into more mineral-rich underlying rocks 

(e.g., silts and mudstones). One of the largest extensions of these more fertile 

areas covered by mesophytic forests, the “Mato Grosso de Goiás,’’ is 

estimated to have had an area of 40,000 km 2 before agriculture destroyed 

it almost entirely (chapter 5). During a road journey from the Distrito Federal 

to Estreito in Maranhão, crossing ca. 1,400 km of the states of Goiás 

and Tocantins, one of us (J.R.) estimated that probably nearly 50% of the 

vegetation traversed was degraded mesophytic forest or the closely related 

mesotrophic facies cerradão. 

The soils of mesophytic forests are particularly good for agriculture; 

consequently, the vegetation has been devastated to such an extent that in 

many areas the past role of this forest as an important or even dominant 

land cover has been obscured. In our experience it is rare to encounter cerrado 

regions where at least some of these forests do not occur: even the 

most dystrophic plateaux have small fertile forested valleys. These socalled 

valley forests (see fig. 6.3A) are favored not only by higher soil fertility, 

but also by higher water availability for most of the year; normally 

the higher the soil fertility, the higher the forest deciduousness (see table 

6.1). In fact, there seems to be no clear-cut ecological and floristic differences 

between deciduous and semideciduous forests. The level of deciduousness 

probably depends on the conjunction of soil moisture and chemical 

properties. Often there are quite local differences in deciduousness in 

a single valley forest: for instance, at Vale dos Sonhos, Mato Grosso, the 

same forest community is deciduous on the well-drained valley sides but 

semideciduous in the moister valley bottom (Ratter et al. 1978b). In the 

Chapada dos Guimarães, Mato Grosso, the composition and deciduousness 

of the valley forest changes as the underlying bedrock changes from 

sandstone to slate, thereby increasing soil fertility (Pinto and Oliveira- 

Filho 1999). In an area surveyed at Três Marias, Minas Gerais, the forest 

changes from evergreen to semideciduous and deciduous before meeting 

the cerrado in a forest strip no more than 150 m wide along the Rio São 

Francisco (Carvalho et al. 1999). 

Deciduous dry forests are particularly common on the base-rich soils 

of the peripheral areas that connect the Cerrado to the Caatinga Biome in 

the northeast, and to the Chaco Biome in the western boundaries of the



Figure 6.3 Forest and grassland physiognomies of the Cerrado Biome. (A) 

Expanse of mesophytic semideciduous (valley) forest, Serra da Petrovina, 

Mato Grosso; (B) interior of mesophytic deciduous dry forest during the rainy 

season, Torixoreu, Mato Grosso (note abundant Maranthaceae in ground 

layer); (C) deciduous dry forest during the dry season, Sagarana, northern 

Minas Gerais; (D) the same site during the rainy season; (E) swampy gallery 

forest flanked by veredas, Chapada dos Veadeiros, Goiás; (F) vereda with 

buriti palmery, Nova Xavantina, Mato Grosso. 

Mato Grosso Pantanal (Ratter et al. 1988; Prado et al. 1992; see figs. 

6.3B–D). On the other hand, large extensions of semideciduous forests 

predominate in the complex transition between the Cerrado and the 

Atlantic Rainforest Biomes in southeastern Brazil (Oliveira-Filho and 

Fontes 2000). Deciduous dry forests in Central Brazil are characterized



by a species-poor woody plant community dominated by a few indicator 

species such as Myracrodruon urundeuva, Anadenanthera colubrina, 

Aspidosperma subincanum, Tabebuia impetiginosa, Dipteryx alata, and 

Dilodendron bipinnatum (Ratter 1992). Semideciduous forests, however, 

tend to be considerably richer in species and actually represent a floristic 

crossroad. The flora is intermediate between those of deciduous dry 

forests and either rainforests (both Amazonian and Atlantic), on a geographic 

scale, or evergreen riverine forests, on a local scale. They also 

share many species with the two types of cerradão, thereby representing 

a connection between the cerrado and rainforest floras (Oliveira-Filho 

and Fontes 2000). 

A particular form of deciduous dry forest is the so-called mata calcárea 

(calcareous forest) which appears on calcareous outcrops throughout 

the Cerrado Biome. These have already been regarded by Prado and 

Gibbs (1993) as relics of a once even more extensive deciduous dry forest 

that during the glacial maxima would have connected the caatingas, in 

northeastern Brazil to the semideciduous forests in southeastern Brazil 

and southern Paraguay, and to the piedmont forests in central-western 

Argentina (see chapter 3 for palynological evidence). It is reasonable to 

think that an intense process of soil leaching and acidification, possibly 

helped by fire, following the return of more humid climates to the cerrado 

region, would have favored the establishment of cerrado vegetation in 

most places and the isolation of decidous dry forests on the present-day 

islands of mesotrophic and calcareous soils (Ratter et al. 1988). 

RIVERINE FORESTS 

Riverine forests are ubiquitous throughout the Cerrado Biome; nearly all 

water bodies of the region are fringed by forests. This forest net is determined 

by the year-round high soil moisture, which, despite the long dry 

season of the region, provides a suitable habitat for a large number of typical 

moist forest species. The nomenclature of riverine forests in Central 

Brazil is complex, as many names are given to the various forms throughout 

the region (Mantovani 1989; Ribeiro and Walter 1998). Most of the 

striking variation of riverine forests both in physiognomy and floristic 

composition results from variation in topography and drainage characteristics, 

together with soil properties (Felfili et al. 1994; Silva Júnior et 

al. 1996; Haridasan et al. 1997; Silva Júnior 1997). Narrow forest strips, 

found along streams and flanked by grasslands or cerrados, are often 

called gallery forests because tree crowns form a “gallery’’ over the watercourse 

(fig. 6.3E). These may be swampy galleries, where slow water flow



increases soil anoxia, or wet (nonswampy) galleries, where flow is faster 

and the soil better drained (Oliveira-Filho et al. 1990; Felfili and Silva 

Júnior 1992; Walter and Ribeiro 1997). When they border wider rivers, 

riverine forests are often called matas ciliares (literally, eyelash forests) 

because they fringe both margins like eyelashes (Ribeiro and Walter 

1998). Where these forests are on alluvial beds under the strong influence 

of the river flooding regime, they may alternatively be called alluvial 

forests (Oliveira-Filho and Ratter 1995). Such forests are often characterized 

by the presence of a raised levee running along their riverside margin. 

On some steep valleys, where the riverine forests are flanked by wider 

areas of mesophytic forest instead of cerrado or seasonal grasslands, they 

are part of valley forests (see previous section). 

These many forms of riverine forest are not always well defined in the 

field, as they may replace each other either very gradually over large areas 

or through short local transitions. Furthermore, the separation of mesophytic 

seasonal forests and riverine forests sometimes breaks down, as 

mixed associations often occur in areas of valley forest. In fact, many 

forests in the Cerrado Biome are formed by narrow strips of evergreen 

riverine forest stretching alongside the river courses, sided by more or less 

wide tracts of mesophytic seasonal forests on the adjacent slopes. 

A number of species are good indicators of the groundwater regime 

(Ratter 1986; Oliveira-Filho et al. 1990; Walter and Ribeiro 1997; Schiavini 

1997; Felfili 1998). For example, Xylopia emarginata, Talauma 

ovata, Calophyllum brasiliense, Hedyosmum brasiliense, and Richeria 

grandis are typical species of swampy conditions; Protium spruceanum, 

Endlicheria paniculata, Pseudolmedia laevigata, and Hieronyma alchorneoides 

are characteristic of wet but better drained soils; while Inga vera, 

Salix humboldtiana, and Ficus obtusiuscula are common in seasonally 

flooded forests. The abundant light at the sharp transition to cerrado or 

grassland favors the occurrence of typical forest edge species such as Piptocarpha 

macropoda, Lamanonia ternata, Vochysia tucanorum, and Callisthene 

major. The shady interior favors species such as Cheiloclinium 

cognatum and Siparuna guinensis. However, many of riverine forest 

species are habitat-generalists (e.g., Schefflera morototoni, Casearia 

sylvestris, Protium heptaphyllum, Tapirira guianensis, T. obtusa, Virola 

sebifera, Copaifera langsdorffii, and Hymenaea courbaril), many of 

which are abundant in the interface with the semideciduous forest and/or 

the cerrado vegetation. 

The key factor for the occurrence of riverine forest within the Cerrado 

Biome is high soil water availability throughout the year, making up 

for the overall water deficit during the dry season. This has led a number 

of authors to suggest that the central Brazilian riverine forests represent



floristic intrusions of the Amazonian and/or Atlantic forests into the cerrado 

domain (Cabrera and Willink 1973; Rizzini 1979; Pires 1984). In 

fact, a considerable number of species shared by the two great South 

American forest provinces do cross the Cerrado Biome via the riverine 

forests (e.g., Ecclinusa ramiflora, Protium spruceanum, Cheiloclinium 

cognatum, and Margaritaria nobilis). Others extend their range into the 

Cerrado Biome along the riverine forests but do not complete the crossing 

(e.g., from the Amazonian forest, Tapura amazonica, Elaeoluma 

glabrescens, Oenocarpus distichus, and even species of rubber-tree 

[Hevea spp.], and from the Atlantic side, Euterpe edulis, Hedyosmum 

brasiliense, Geonoma schottiana, and Vitex polygama). However, there 

are also a few species that are exclusive to these forests, such as Unonopsis 

lindmannii, Vochysia pyramidalis, and Hirtella hoehnei. 

SEASONAL GRASSLANDS 

The alternation of periods of water excess and deficit normally favors the 

occurrence of seasonal grasslands within the Cerrado Biome. There are 

three main vegetation physiognomies of the type in the region. Veredas 

are valley-side marshes where the water table reaches or almost reaches 

the surface during the rainy season; they are commonly found in the middle 

of topographic sequences, between gallery forests and cerrado. Veredas 

are very widespread in the Cerrado Biome, particularly near headwaters, 

and may include palm groves of Mauritia flexuosa (buriti-palm) (see fig. 

6.3E–F). Floodplain grasslands are found on areas of even topography 

liable to more or less long periods of inundation; they are usually restricted 

to the vicinity of large rivers, such as the Paraguay (Mato Grosso Pantanal) 

and Araguaia (Varjão and Bananal island). Rocky grasslands (campo 

rupestre, campo de altitude) are mostly restricted in the Cerrado Biome to 

the tops of plateaux and mountain ridges, where the soils are shallow or 

confined to cracks between rocks. As they have very limited water storage 

capacity, these soils are often soaked during the rainy season but extremely 

dry during the dry season. As these physiognomies are poor or totally lacking 

in woody vegetation, they are better treated in chapter 7. 

THE BIODIVERSITY OF THE CERRADO BIOME 

The combination of the great age of the Cerrado Biome and the relatively 

recent (Quaternary) dynamic changes in vegetation distribution patterns 

has probably led to its rich overall biodiversity, estimated at 160,000



species of plants, animals, and fungi by Dias (1992). The figure for vascular 

plants is still very approximate, but Mendonça et al. (1998) list 

6,429 native species from all communities of the biome. Future investigations 

will certainly add many species to the list. For instance, recent surveys 

over a large part of the cerrado area show many more woody species 

in the cerrado sensu lato. The eventual total may indeed reach the 10,500 

estimate given by Dias (1992). In a recent publication, Myers et al. (2000) 

recognize the cerrado among 25 global biodiversity “hotspots’’ and estimate 

that it contains 4,400 endemic higher plant species, representing no 

less than 1.5% of the world’s total vascular plant species. 

An important aspect of biodiversity of the Cerrado Biome, and one 

of profound ecological importance, is the loss of large mammalian fauna, 

as Ratter et al. (1997) explain: 

unlike the African savannas, it [the cerrado] has lost the fauna of large 

mammals with which it must have co-evolved throughout the Tertiary. 

The large herbivores (grazers and browsers) must have been eliminated 

as a result of competition with North American fauna which 

migrated across the Panama Land-Bridge in the Great American Interchange 

3 million years ago in the late Pliocene, or later in Man’s Pleistocene 

and Holocene Overkill. The only remnants of the ancient 

neotropical mammalian fauna now occurring in the Cerrado Biome 

are some Edentates (such as the tamanduá anteaters and armadillos), 

marsupials (such as opossums), platyrrhine monkeys (such as marmosets, 

howlers and capuchins), and various rodents (such as agoutis, 

pacas, capybaras, and many mouse-sized species). Many larger-fruited 

plants species probably lost their natural mode of dispersal as a result 

of the extinction of their native mammalian vectors (see Janzen and 

Martin 1982). The reintroduction of grazers in the form of cattle and 

horses into the natural cerrado vegetation in the last few hundred years 

probably partially restored the balance of the vegetation to the situation 

prior to the Great American Interchange. 

The levels of information on the diversity of the various communities 

of the Cerrado Biome are very unequal and are considered separately below. 

CERRADO SENSU LATO 

The characteristic arboreal flora of the savanna elements of the Cerrado 

Biome is relatively well known. A useful base list was provided by Rizzini 

(1963) and added to by Heringer et al. (1977). In all, these authors record



774 woody species belonging to 261 genera, of which 336 species (43%) 

are regarded as endemic to cerrado sensu lato. Since 1977 much research 

has been carried out on the floristics and phytosociology of the cerrados, 

and the number of species recorded has increased. A recent compilation 

by Castro et al. (1999) gives 973 species and 337 genera identified “with 

confidence’’ and in addition mentions a large number of records of undetermined, 

or partially determined, taxonomic entities. These authors suggest 

that the total arboreal and large shrub flora of the cerrado sensu lato 

may be 1,000–2,000 species. The latter figure, however, must be approached 

with a great degree of caution, as stressed by Castro et al. (1999) 

in an extremely succint discussion. They consider that a reasonably secure 

minimum estimate for the arboreal-shrubby cerrado flora is around 1,000 

species, 370 genera, and 90 families. However, they point out that “three 

main objections might be raised: (1) the list includes a large number of 

species that certainly would not be regarded as typical cerrado species (e.g. 

Talauma ovata); (2) a number of species that are not typically woody in 

most sites are also included (e.g., Oxalis); (3) some unrecorded rarer species 

are likely to be ‘hidden,’ having been misidentified as common cerrado 

species’’ (Castro et al. 1999). We consider (1) and (2) to be particularly 

potent factors in inflating estimates of cerrado woody species, although 

fully accepting the arguments of Castro et al. on the difficulties of separating 

(a) “characteristic’’ and “accessory’’ species, and (b) “ground’’ from 

“arboreal or large shrub’’ species when the same taxon may show contrasting 

growth forms in different localities. Cerrado research is in a very 

dynamic phase, and more accurate estimates will be available in the next 

few years. In the meantime it is interesting to note that the present data 

base of the ongoing Anglo-Brazilian collaborative Conservation and Management 

of the Biodiversity of the Cerrado Biome (BBC) project records 

approximately 800 species for 300 surveys throughout the cerrado region 

(Ratter et al. 2000). 

The most important families in terms of species numbers, using the 

fairly conservative figures of Heringer et al. (1977), are Leguminosae (153 

spp., all three subfamilies), Malpighiaceae (46 spp.), Myrtaceae (43 spp.), 

Melastomataceae (32 spp.), and Rubiaceae (30 spp.). However, in many 

areas the vegetation is dominated by Vochysiaceae (with 23 arboreal 

species in the cerrado) because of the abundance of the three species of 

pau-terra (Qualea grandiflora, Q. parviflora and Q. multiflora). The 

largest genera are Byrsonima (Malpighiaceae, 22 spp.), Myrcia (Myrtaceae, 

18 spp.), Kielmeyera (Guttiferae, 16 spp.), Miconia (Melastomataceae, 

15 spp.) and Annona (Annonaceae, 11 spp.). 

Heringer et al. (1977) analyzed the geographic affinity of the 261 gen-



era they listed and found that 205 had species in common with the Brazilian 

Atlantic Forest, 200 with the Amazonian forest, 30 with the mesophytic 

forests, and 51 with the cerrado ground layer, while seven (three 

of which are monotypic) did not occur in any other vegetation type. 

Recent work by Ratter and Dargie (1992), Ratter et al. (1996), Castro 

(1994), and Castro et al. (1999) has been directed toward discovering 

patterns of geographic distribution of cerrado vegetation by comparison 

of large numbers of floristic surveys using multivariate techniques. The 

studies have covered almost the entire cerrado area and have also included 

some isolated Amazonian savannas. In all, Ratter et al. (1996) compared 

98 sites, while Castro studied 78 areas and 145 species lists. The results 

of the two groups seem to be very much in accord. Ratter et al. (1996) 

demonstrated a strong geographic pattern in the distribution of the flora, 

which allowed the provisional recognition of southern (São Paulo and 

south Minas Gerais), southeastern (largely Minas Gerais), central (Federal 

District, Goiás and parts of Minas Gerais), central-western (largely 

Mato Grosso, Goiás and Mato Grosso do Sul), and northern regions 

(principally Maranhão, Tocantins and Pará), as well as a disjunct group 

of Amazonian savannas (see fig. 6.4). This work is continuing as a part 

of the Biodiversity of the Cerrado Biome (BBC) project, and results based 

on the comparison of more than 300 sites will soon be available. 

Diversity of trees and large shrubs occurring at a single site (alpha 

diversity) may reach 150 species per hectare, but is generally much lower 

than this, while at the other extreme it can be less than 10 species in isolated 

Amazonian savannas. Diversity tends to be lower on the more 

mesotrophic sites where dominance of characteristic “indicator’’ species 

such as Callisthene fasciculata, Magonia pubescens, Terminalia argentea, 

and Luehea paniculata occurs. The comparison of 98 sites by Ratter et al. 

(1996) revealed a remarkable intersite heterogeneity (beta diversity). In 

total 534 tree and large shrub species were recorded at these sites. Of 

these, 158 (30%) occurred at a single site only; no species occurred at all 

sites; and only 28 (5%) were present at 50% or more of the sites. The 

most widespread species was Qualea grandiflora, which occurred at 82% 

of sites. High levels of instersite heterogeneity have also been demonstrated 

in surveys of the same land unit, the Chapada Pratinha, in the Federal 

District, Goiás, and Minas Gerais (Felfili and Silva Júnior 1993; 

Filgueiras et al. 1998). The extreme floristic heterogeneity (beta diversity) 

of cerrado vegetation has important consequences for conservation planning, 

since many protected areas will need to be established in order to 

represent biodiversity adequately. The sites we have recorded with the 

highest species numbers (“biodiversity hotspots’’) are in Mato Grosso,



Figure 6.4 Provisional geographic pattern of cerrado vegetation based on 

floristic analyses of woody flora. After Ratter et al. (1996). 

Goiás (Araguaia region), Tocantins, and the Federal District. It is interesting 

that some of these “hotspots’’ lie in the periphery of the Cerrado 

Biome close to the transition into the Hylaean forest (e.g., on the Rio 

Xingu drainage in Mato Grosso, where the cerrado community comprised 

107 woody species). However, despite their outlying location, their diversity 

is composed of species typical of the cerrado, not of neighboring vegetation 

types. 

The number of species growing in a particular small area of cerrado 

can be surprising. For instance, at Fazenda Água Limpa, the ecological 

reserve of the University of Brasília in the Federal District, one can find 

six species of Miconia and Byrsonima, and five species of Erythroxylum 

and Kielmeyera (Ratter 1986). This is far less than the 45 Pouteria, 40



Ocotea, 31 Protium, 30 Inga, or 19 Eschweilera species found in the 

10,000 hectares of Amazonian rainforest of the Reserva Ducke near Manaus 

(Ribeiro et al. 1999), but the number of individuals of each species 

found in these cerrado mixed populations is probably significantly 

greater. In addition, the diversity of growth form of the cerrado congeners 

is much greater than in the forest. 

The diversity of plants of the ground layer (the so-called vegetação 

rasteira, consisting of herbs, subshrubs, and smaller shrubs) is much richer 

than for trees and large shrubs, and species numbers are so high that 

detailed floristic lists are only available for comparatively few localities 

(chapter 7). Rizzini (1963) gives the figure of more than 500 genera for 

smaller plants against less than 200 for trees and large shrubs; more 

detailed information can be extracted from a number of works. In an 

exhaustive survey of the IBGE Ecological Reserve in the Federal District 

conducted over many years, Pereira et al. (1993) record 636 species in the 

vegetação rasteira against 84 arboreal species (a ratio of 7.6:1), while Ratter 

(1986) lists 400 ground species against 110 trees (3.6:1) in the nearby 

Fazenda Água Limpa. The difference between these two ratios is probably 

partly explained by the length of time spent in observations at the two 

localities: at Fazenda Água Limpa the bulk of the work was done in one 

year’s intensive study, while at the IBGE Reserve detailed observation was 

extended over 15 years, thus allowing many rare smaller species to be 

recorded. Figures for São Paulo state are somewhat lower, with ratios of 

3:1 and 2:1 recorded by Mantovani and Martins (1993). Ratter and 

Ribeiro (1996) suggested that extrapolation of the Federal District figures 

taking 1,000 as the number of tree/large shrub species would give an estimate 

of over 5,000 ground species, while Castro et al. (1999) estimate 

2,919–6,836, working on the same basis from their data (see chapter 7). 

MESOPHYTIC FORESTS 

As already mentioned, mesophytic deciduous and semideciduous forests 

occur throughout the cerrado landscape where richer (mesotrophic to 

eutrophic) soils occur (table 6.1). Their flora belongs to a floristic province 

which was probably continuous, or almost continuous, during drier and 

cooler periods in the Pleistocene and is now represented by three main 

nuclei: the arboreal caatingas of NE Brazil; the “Misiones’’ forests of 

Corumbá–Puerto Suarez, extending into Paraguay, Argentinian Misiones, 

and Brazilian Santa Catarina; and the “Piedmont’’ forests of Bolivia and 

northen Argentina (Ratter et al. 1988; Prado 1991; Prado and Gibbs



1993; Pennington et al. 2000). The islands of mesophytic forests crossing 

the region of the Cerrado Biome now form a very discontinous bridge 

between the caatinga vegetation and the “Misiones forests.’’ 

The diversity of arboreal species of deciduous dry forests is much 

lower than that of the cerrado or the riverine forests of the Cerrado Biome. 

Semideciduous forests also tend to be considerably richer than deciduous 

dry forests, as they share many species with riverine forests. The total 

arboreal floristic list is approximately 100 species for deciduous dry 

forests occurring in the Planalto and central-western areas of the biome, 

but normally the number of species found in any one locality is very much 

less than this. Most of the species require higher levels of soil calcium, but 

a few of the taller trees such as the leguminous Apuleia leiocarpa, 

Copaifera langsdorffii, and Hymenaea courbaril var. stilbocarpa, and the 

anacardiaceous Tapirira guianensis, are more tolerant of dystrophic conditions 

and thus can be found in other forest communities (dystrophic galleries, 

Amazonian forest, etc.). About 20% of the species occurring in 

central Brazilian deciduous dry forests are also found in mesotrophic cerradão 

(e.g., Astronium fraxinifolium, Dilodendron bipinnatum, Dipteryx 

alata, and Platypodium elegans). The floristic relationship between deciduous 

dry forest and mesotrophic cerradão has been noted by various 

authors (Rizzini and Heringer 1962; Magalhães 1964; Ratter et al. 1977; 

Furley and Ratter 1988) and is to be expected considering the pedological 

similarities of the two communities. Deciduous dry forests usually 

show a high degree of dominance of a few species, which in our experience 

is a characteristic of communities inhabiting richer soils. Extreme 

examples of this are shown in surveys of deciduous dry forests in the 

Triângulo Mineiro. For example, Araújo et al. (1997) found 65 tree 

species in a forest near Uberlândia, Minas Gerais, where Anadenanthera 

colubrina var. cebil (under the synonym A. macrocarpa) comprised 

60.51% of the relative dominance and had an IVI of 73.41. In Santa 

Vitória, Oliveira-Filho et al. (1998) registered 60 species, with the top five, 

including A. colubrina, accounting for 61% of all individuals and 70% 

of the relative dominance. 

There is a considerable variation in the floristics and other characteristics 

of mesophytic forests across the vast area of the Cerrado Biome. 

Thus forests such as the remnants of the “Mato Grosso de Goiás’’ on deep 

soils in the core area of the biome are much more mesic than those on calcareous 

outcrops and/or near the drier areas lying close to the Caatinga 

Biome margin. As would be expected, the latter show a greater abundance 

of more extreme caatinga species (associated with more xeric conditions) 

such as tall Cereus jamacaru cacti growing among the trees, Commiphora



leptophloeos, Spondias tuberosa, Zizyphus joazeiro, Schinopsis brasiliensis, 

Cavanillesia arborea, etc. Conversely, in forests closer to the “Misiones’’ 

floristic nucleus, particularly those south of the Mato Grosso Pantanal, 

species characteristic of that element, such as Calycophyllum multiflorum, 

Pterogyne nitens, Chrysophyllum marginatum, Terminalia triflora, Prosopis 

spp., and, once again, Commiphora leptophloeos are found. 

RIVERINE FORESTS 

The riverine forests following the drainage throughout the vast Cerrado 

Biome cover probably less than 10% of its total area but harbor an enormous 

floristic and faunal diversity. A number of recent surveys have 

shown a much greater diversity of the tree/large shrub species in the 

gallery forests than in the cerrado vegetation itself; for instance, Ramos 

(1995) found 260 species in the galleries of the Brasília National Park but 

only 109 species in the cerrado, while Pereira et al. (1993) list 183 species 

from the galleries and 84 from the cerrado of the IBGE Ecological 

Reserve, also in the Federal District. In a wider area on the Planalto Central 

(Federal District, Goiás, and Minas Gerais), Silva Júnior, Nogueira, 

and Felfili (1998) recorded 446 woody species in 22 gallery forests. Figures 

for total floristic diversity of riverine forests throughout the complete 

Cerrado Biome are not yet available. The database of Oliveira-Filho and 

Ratter (1994) contains 627 species of trees for 17 riverine forests of the 

Cerrado Biome, while a total of 771 species of trees and shrubs of riverine 

forests can be extracted from the list prepared by Mendonça et al. 

(1998) for all communities of the Cerrado Biome. However, as there are 

still relatively few surveys of these forests, it is certain that they contain 

many more species than suggested by those figures. 

The reason for so much diversity can be ascribed to two main factors: 

(1) the environmental heterogeneity occurring both within and between 

riverine forests and (2) the diverse floristic elements from which the communities 

of riverine forests are derived in different parts of the region. 

A number of authors have carried out detailed studies of particular 

galleries and demonstrated how the habitat, generally defined according 

to topography and drainage, can be broken into environmental subdivisions 

and characterized by floristic differences (Felfili et al. 1994; Felfili 

1998). Oliveira-Filho et al. (1990) studied the headwaters of the Córrego 

da Paciência, near Cuiabá, Mato Grosso, and recognized four distinct 

communities: dry cerradão (dystrophic), wet cerradão (mesotrophic), wet 

forest (semideciduous), and swampy forest (evergreen). An extremely



detailed study was carried out in the Reserva do Roncador, Federal District 

(Silva Júnior 1995; Silva Júnior et al. 1996) to investigate the association 

of particular floristic communities with differing environmental 

conditions. Dry upslope, intermediate, and wet downslope communities 

were defined, with all species showing strong habitat preferences and only 

one species, Tapirira guianensis (a geographically widespread generalist) 

occurring in all three communities. Similarly, Schiavini (1997) and Van 

den Berg and Oliveira-Filho (1999) have demonstrated the distinct association 

of woody species with three bands of differing humidity and exposure 

to light (edge-effect) in two gallery forests near Uberlândia and 

Itutinga, respectively, both situated in Minas Gerais. 

A recent study by Silva Júnior, Felfili, Nogueira, and Rezende (1998) 

demonstrates the heterogeneity of gallery forests in the Federal District. 

In all, 15 forests containing a total of 446 arboreal species were compared. 

These were reduced for analysis to the 226 species with more than five 

individuals, of which only two species (Copaifera langsdorfii and Tapirira 

guianensis) were present at all sites. The percentage breakdown was 

27.4% of species occurring only at a single site, 38.9% at two to four 

sites, 17.2% at five to eight, 10.6 at nine to 12, and 1.8% at 13–15. The 

results showed the presence of a few species with a wide distribution and 

many with a very restricted occurrence. The forests with the lowest diversity 

were, following the normal pattern, the mesophytic examples associated 

with better soils. Sørensen similarity indices were calculated using 

the full species lists, including rare species. They were as high as 73.8% 

for two nearby sites on the IBGE Ecological Reserve and as low as 11.0% 

for two more distant sites. The majority, however, lie between 30.1% and 

47.0%, indicating fairly low levels of similarity between the galleries of 

the Federal District. All the galleries studied were in the drainage of the 

Plate/Paraguay system. Had the observations included those galleries also 

occurring in the Federal District but on the Araguaia-Tocantins and São 

Francisco drainages, the results would probably have shown greater 

diversity and many low-similarity indices. 

A study of the origins of Central Brazilian forests by the analysis of 

plant species distribution patterns (Oliveira-Filho and Ratter 1995) 

involved the comparison of 106 forest surveys containing a total of 3,118 

species. The result of multivariate analysis revealed that the riverine 

forests of the north and west of the Cerrado Biome have a strong relationship 

with the Amazonian rainforests, while those of the center and 

south show stronger affinity with the montane semideciduous forests of 

southeastern Brazil. Introduction of species from adjacent forest nuclei 

into the riverine forests crossing the cerrado region has probably been



very active during postglacial forest expansion, although a great number 

of the characteristic gallery forest species are widespread generalists. An 

important aspect of gallery forests is that they have interfaces with many 

other types of vegetation, including rainforests, mesophytic forests and 

the cerrado itself. They are thus subjected to very different floristic influences 

resulting in great heterogeneity. 

CONCLUDING REMARKS ON CONSERVATION 

The Cerrado Biome is one of the richest savanna biomes in the world and 

harbors an immense floral and faunal diversity. Much of this is a consequence 

of its great antiquity, possibly going back to a prototypic cerrado 

in the Cretaceous, followed by long evolution over the Tertiary and a 

dynamic phase leading to much speciation over the glacial and interglacial 

periods of the Quaternary. Endemicity is very high, estimated at as high 

as 4,400 species of higher plants by Myers et al. (2000), who rank the 

Cerrado among 25 global biodiversity “hotspots’’ of absolute importance 

for conservation. Already about 40% of the original Cerrado Biome area 

has been converted to “anthropic landscape,’’ principally as “improved’’ 

pastures or intense arable cultivation, and the need for implementation of 

conservation plans is urgent (Klink et al. 1995; Ratter et al. 1997; Cavalcanti 

1999; see chapters 5, 18). The contents of this book demonstrate the 

multifaceted importance of the biome, and we hope that we are now seeing 

an eleventh-hour awakening to the needs of conservation of cerrado 

and dry forest. 


This chapter was prepared under the sponsorship of the Royal Society of 

London (International Exchange Award to A. T. Oliveira-Filho) and the 

Royal Botanic Garden Edinburgh. We acknowledge their support with gratitude. 

We also thank Samuel Bridgewater, Toby Pennington, and Luciana 

Botezelli for their invaluable help during the preparation of the manuscript. 

REFERENCES 

Araújo, G. M., L. A. Rodrigues, and L. Ivizi. 1997. Estrutura fitossociológica 

e fenologia de espécies lenhosas de mata decídua em Uberlândia, MG.



In L. L. Leite and C. H. Saito, eds., Contribuição ao Conhecimento 

Ecológico do Cerrado: Trabalhos Selecionados do 3° Congresso de 

Ecologia do Brasil (Brasília, 6–11/10/96), pp. 22–28. Brasília: Editora da 

Universidade de Brasília. 

Braithwaite, R. W. 1996. Biodiversity and fire in the savanna landscape. In 

O. T. Solbrig, E. Medina, and J. F. Silva, eds., Biodiversity and Savanna 

Ecosystem Processes, pp. 121–142. Berlin: Springer-Verlag. 

Bucher, E. H. 1982. Chaco and Caatinga: South American arid savannas, 

woodlands and thickets. In B. J. Huntley and B. H. Walker, eds., Ecology 

of Tropical Savannas, pp. 48–79. Berlin: Springer-Verlag. 

Cabrera, A. L. and A. Willink. 1973. Biogeografia de America Latina. Washington: 

Secretaria General de la Organización de los Estados Americanos. 

Carvalho, D. A., A. T. Oliveira-Filho, E. A. Vilela, N. Curi, E. Van den Berg, 

and M. A. L. Fontes. 1999. Estudos Florísticos e Fitossociológicos em 

Remanescentes de Florestas Ripárias do Alto São Francisco e Bacia do 

Rio Doce–MG. Boletim técnico, CDD 33.7, Belo Horizonte: Companhia 

Energética de Minas Gerais. 

Castro, A. A. J. F. 1994. “Comparação Florístico-Geográfica (Brasil) e 

Fitossociológica (Piauí–São Paulo) de Amostras de Cerrado.’’ Ph.D. thesis, 

Universidade Estadual de Campinas, Campinas, Brazil. 

Castro, A. A. J. F., F. R. Martins, J. Y. Tamashiro, and G. J. Shepherd. 1999. 

How rich is the flora of Brazilian cerrados? Ann. Miss. Bot. Gard. 

86:192–224. 

Cavalcanti, R. B. (scientific coordinator). 1999. Ações Prioritárias para Conservação 

da Biodiversidade do Cerrado e Pantanal. Brasília: Ministério 

do Meio Ambiente, Funatura, Conservation International, Fundação 

Biodiversitas, Universidade de Brasília. 

Coutinho, L. M. 1978. O conceito de Cerrado. Rev. Bras. Bot. 1:17–23. 

Coutinho, L. M. 1990. Fire in the ecology of the Brazilian cerrado. In 

J. G. Goldammer, ed., Fire in Tropical Biota, pp. 82–105. Berlin: 

Springer-Verlag. 



Recursos Naturais Renováveis, pp. 11–25. Brasília: Fundação Pró- 

Natureza. 

Diniz, M. A. N., P. A. Furley, C. Johnson, and M. Haridasan. 1986. The 

murundus of central Brazil. J. Trop. Ecol. 2:17–35. 

Dubs, B. 1992. Observations on the differentiation of woodland and wet 

savanna habitats in the Pantanal of Mato Grosso, Brazil. In P. A. Furley, 

J. Proctor, and J. A. Ratter, eds., Nature and Dynamics of Forest-Savanna 

Boundaries, pp. 431–449. London: Chapman and Hall. 


Felfili, J. M. 1998. Determinação de padrões de distribuição de espécies em 

uma mata de galeria no Brasil Central com a utilização de técnicas de 

análise multivariada. Bol. Herb. E. P. Heringer 2:35–48.



Felfili, J. M., T. S. Filgueiras, M. Haridasan, R. Mendonça and A. V. Rezende. 

1994. Projeto biogeografia do bioma cerrado: Vegetação e solos. Cad. 

Geoc. IBGE 12:75–166. 

Felfili, J. M. and M. C. Silva Júnior. 1992. Floristic composition, phytosociology 

and comparison of cerrado and gallery forests at Fazenda Água 

Limpa, Federal District, Brazil. In P. A. Furley, J. Proctor, and J. A. Ratter, 

eds., Nature and Dynamics of Forest-Savanna Boundaries, pp. 

393–416. London: Chapman and Hall. 

Felfili, J. M. and M. C. Silva Júnior. 1993. A comparative study of cerrado 

(sensu stricto) vegetation in Central Brazil. J. Trop. Ecol. 9:277–289. 

Filgueiras, T. S., J. M. Felfili, M. C. Silva Júnior, and P. E. Nogueira. 1998. 

Floristic and structural comparison of cerrado (sensu stricto) vegetation 

in Central Brazil. In F. Dallmeier and J. A. Comisky, eds., Forest Biodiversity 

in North, Central, and South America, and the Caribbean, pp. 

633–648. Man and the Biosphere Series, vol. 21. Paris: UNESCO and 

Parthenon Publishing Group. 

Furley, P. A. 1999. The nature and diversity of neotropical savanna vegetation 

with particular reference to the Brazilian cerrados. Glob. Ecol. Biogeogr. 

8:223–241. 

Furley, P. A., J. Proctor, and J. A. Ratter. 1992. Nature and Dynamics of Forest-Savanna 

Boundaries. London: Chapman and Hall. 

Furley, P. A. and Ratter, J. A. 1988. Soil resources and plant communities of the 

Central Brazilian cerrado and their development. J. Biogeogr. 15:97–108. 

Goodland, R. A. 1971. A physiognomic analysis of the “cerrado’’ vegetation 

of Central Brazil. J. Ecol. 59:411–419. 

Goodland, R. A. and M. G. Ferri. 1979. Ecologia do cerrado. Belo Horizonte 

and São Paulo: Livraria Itatiaia Editora and Editora da Universidade de 

São Paulo. 

Goodland, R. A. and R. Pollard. 1973. The Brazilian cerrado vegetation: A 

fertility gradient. J. Ecol. 6:219–224. 

Haridasan, M. 1982. Aluminium accumulation by some cerrado native 

species in Central Brazil. Plant and Soil 65:265–273. 

Haridasan, M., M. C. Silva Júnior, J. M. Felfili, A. V. Rezende, and P. E. N. 

Silva. 1997. Gradient analysis of soils properties and phytosociological 

parameters of some gallery forests on the Chapada dos Veadeiros in the 

cerrado region of Central Brazil. In Proceedings of the International 

Symposium on Assessment and Monitoring of Forests in Tropical Dry 

Regions with Special Reference to Gallery Forests, pp. 259–275. Brasília: 

Editora da Universidade de Brasília. 

Heringer, E. P., G. M. Barroso, J. A. Rizzo, and C. T. Rizzini. 1977. A flora 

do cerrado. In Anais do IV Simpósio sobre o Cerrado, pp. 211–232. Belo 

Horizonte: Livraria Itatiaia Editora and São Paulo: Editora da Universidade 

de São Paulo. 

IBGE. 1993. Mapa de vegetação do Brasil. Rio de Janeiro: Fundação Instituto 

Brasileiro de Geografia e Estatística.



Janzen, D. H. and P. Martin. 1982. Neotropical anachronisms: What the 

gomphotheres ate. Science 215:19–27 

Klein, R. M. 1975. Southern Brazilian phytogeographic features and the 

probable influence of upper Quaternary climatic changes in the floristic 

distribution. Bol. Paran. Geoc. 33:67–88. 

Klink, C.A., Macedo, R., and Mueller, C. 1995. Bit by bit the Cerrado loses 

space. In Alho, C. J. R. and E. de S. Martins, eds., p. 66. Brasilia: World 

Wildlife Fund and Pró-Cer. 

Ledru, M. P. 1993. Late Quaternary environmental and climatic changes in 


Ledru, M. P., M. L. Salgado-Labouriau, and M. L. Lorscheitter. 1998. Vegetation 

dynamics in southern and central Brazil during the last 10,000 yr 

BP. Rev. Palaeobot. Palynol. 99:131–142. 

Lopes, A. S. and E. R. Cox. 1977. Cerrado vegetation in Brazil: An edaphic 

gradient. Agronom. J. 69:828–831. 

Magalhães, G. M. 1964. Fitogeografia do Estado de Minas Gerais. In Anon. 

ed., Recuperação do Cerrado, pp. 69–82. Rio de Janeiro: Ministério da 

Agricultura, Edições SIA, Estados Brasileiros n° 21. 

Mantovani, W. 1989. Conceituação e fatores condicionantes. In Anais do I 

Simpósio sobre Mata Ciliar, pp. 2–10. Campinas: Fundação Cargill. 

Mantovani, W. and F. R. Martins. 1993. Florística do cerrado na Reserva 

Biológica de Moji Guaçu, SP. Acta Bot. Bras. 7:33–60. 

Mendonça, R. C., J. M. Felfili, B. M. T. Walter, M. C. Silva Júnior., A. V. 

Rezende, T. S. Filgueiras, and P. E. Nogueira. 1998. Flora vascular do 

cerrado. In S. M. Sano and S. P. Almeida, eds., Cerrado: Ambiente e 

Flora, pp. 288–556. Planaltina, DF, Brazil: Empresa Brasileira de 

Pesquisa Agropecuária. 

Myers, N., R. A. Mittermeir, C. G. Mittermeir, G. A. B. Fonseca, and J. Kent. 

2000. Biodiversity hotspots for conservation priorities. Nature 403: 

853–858. 

Oliveira-Filho, A. T. 1992a. Floodplain “murundus’’ of Central Brazil: Evidence 

for the termite-origin hypothesis. J. Trop. Ecol. 8:1–19. 

Oliveira-Filho, A. T. 1992b. The vegetation of Brazilian “murundus’’: The 

island-effect on the plant community. J. Trop. Ecol. 8:465–486. 

Oliveira-Filho, A. T., N. Curi, E. A. Vilela, and D. A. Carvalho. 1998. Effects 

of canopy gaps, topography and soils on the distribution of woody species 

in a central Brazilian deciduous dry forest. Biotropica 30:362–375. 

Oliveira-Filho, A. T. and M. A. L. Fontes. 2000. Patterns of floristic differentiation 

among Atlantic forests in south-eastern Brazil, and the influence 

of climate. Biotropica 32:793–810. 

Oliveira-Filho, A. T. and F. R Martins. 1986. Distribuição, caracterização e 

composição florística das formações vegetais da região da Salgadeira, na 

Chapada dos Guimarães (MT). Rev. Brasil. Bot. 9:207–223. 

Oliveira-Filho, A. T. and F. R. Martins. 1991. A comparative study of five cerrado 

areas in southern Mato Grosso, Brazil. Edinb. J. Bot. 48:307–332.



Oliveira-Filho, A. T. and J. A. Ratter. 1994. Database: Woody Flora of 106 

Forest Areas of Eastern Tropical South America. Edinburgh: Royal 

Botanic Garden Edinburgh. 

Oliveira-Filho, A. T. and J. A. Ratter. 1995. A study of the origin of central 

Brazilian forests by the analysis of plant species distribution patterns. 

Edinb. J. Bot. 52:141–194. 

Oliveira-Filho, A. T., J. A. Ratter, and G. J. Shepherd. 1990. Floristic composition 

and community structure of a central Brazilian gallery forest. 

Flora 184:103–117. 

Oliveira-Filho, A. T., G. J. Shepherd, F. R. Martins, and W. H. Stubblebine. 

1989. Environmental factors affecting physiognomical and floristic variations 

in a cerrado of central Brazil. J. Trop. Ecol. 5:413–431. 

Pennington, R. T., D. E. Prado, and C. A. Pendry. 2000. Neotropical seasonally 

dry forests and Quaternary vegetation changes. J. Biogeogr. 27 (in 

press). 

Pereira, B. A. S., M. A. Silva, and R. C. Mendonça. 1993. Reserva Ecológica 

do IBGE, Brasília–DF: Lista das plantas vasculares. Rio de Janeiro: Fundação 

Instituto Brasileiro de Geografia e Estatística. 

Pinto, J. R. R. and A. T. Oliveira-Filho. 1999. Perfil florístico e estrutura da 

comunidade arbóreo-arbustiva de uma floresta de vale no Parque 

Nacional da Chapada dos Guimarães. Rev. Bras. Bot. 22:53–67. 

Pires, J. M. 1984. The Amazonian forest. In H. Sioli, ed., The Amazon: Limnology 

and Landscape Ecology of a Mighty Tropical River and its Basin, 

pp. 581–602. Dordrecht: Junk Pub. 

Prado, D. E. 1991. “A Critical Evaluation of the Floristic Links Between 

Chaco and Caatingas Vegetation in South America.’’ Ph.D. thesis. University 

of Saint Andrews, Saint Andrews, Scotland. 

Prado, D. E. and P. E. Gibbs. 1993. Patterns of species distribution in the dry 

seasonal forests of South America. Ann. Miss. Bot. Gard. 80:902–927. 

Prado, D. E., P. E. Gibbs, A. Pott, and V. J. Pott. 1992. The Chaco-Pantanal 

transition in southern Mato Grosso, Brazil. In P. A. Furley, J. Proctor, and 

J. A. Ratter, eds., Nature and Dynamics of Forest-Savanna Boundaries, 

pp. 431–470. London: Chapman and Hall. 

Prance, G. T. and G. B. Schaller. 1982. Preliminary study of some vegetation 

types of the Pantanal, Mato Grosso, Brazil. Brittonia 34:228–251. 

Ramos, P. C. M. 1995. “Vegetation Communities and Soils of Brasília National 

Park.” Ph.D. thesis, University of Edinburgh Edinburgh, Scotland. 

Ratter, J. A. 1971. Some notes on two types of cerradão occurring in northeastern 

Mato Grosso. In M. G. Ferri, ed., III Simpósio sobre o cerrado, 

pp. 100–102. São Paulo: Editora da Universidade de São Paulo. 

Ratter, J. A. 1980. Notes on the Vegetation of Fazenda Água Limpa (Brasília, 

DF, Brazil). Edinburgh: Royal Botanic Garden Edinburgh. 

Ratter, J. A. 1986. Notas sobre a vegetação da Fazenda Água Limpa (Brasília, 

DF, Brasil). Textos Universitários no. 3. Brasília: Editora da Universidade 

de Brasília.



Ratter, J. A. 1987. Notes on the vegetation of the Parque Nacional do 

Araguaia (Brazil). Notes Roy. Bot. Gard. Edinb. 44:311–342. 

Ratter, J. A. 1992. Transitions between cerrado and forest vegetation in 

Brazil. In P. A. Furley, J. Proctor, and J. A. Ratter, eds, Nature and 

Dynamics of Forest-Savanna Boundaries, pp. 417–429. London: Chapman 

and Hall. 

Ratter, J. A., G. P. Askew, R. F. Montgomery, and D. R. Gifford, 1977. Observações 

adicionais sobre o Cerradão de solo mesotrófico no Brasil Central. 

In M. G. Ferri, ed., IV Simpósio sobre o Cerrado, pp. 303–316. São 

Paulo: Editora da Universidade de São Paulo. 

Ratter, J. A., G. P. Askew, R. F. Montgomery, and D. R. Gifford. 1978a. 

Observations on the vegetation of northeastern Mato Grosso: II. Forests 

and soils of the Rio Suiá-Missu area. Proc. Roy. Soc. London B. 

203:191–208. 

Ratter, J. A., G. P. Askew, R. F. Montgomery, and D. R. Gifford. 1978b. 

Observations on forests of some mesotrophic soils in central Brazil. Rev. 

Bras. Bot. 1:47–58. 

Ratter, J. A., S. Bridgewater, R. Atkinson, and J. F. Ribeiro. 1996. Analysis of 

the floristic composition of the Brazilian Cerrado vegetation II: Comparison 

of the woody vegetation of 98 areas. Edinb. J. Bot. 53:153–180. 

Ratter, J. A. and T. C. D. Dargie. 1992. An analysis of the floristic composition 

of 26 cerrado areas in Brazil. Edinb. J. Bot. 49:235–250. 

Ratter, J. A., A. Pott, V. J. Pott, C. N. Cunha, and M. Haridasan. 1988. Observations 

on woody vegetation types in the pantanal and at Corumbá, 

Brazil. Notes Roy. Bot. Gard. Edinb. 45:503–525. 

Ratter, J. A. and J. F. Ribeiro. 1996. Biodiversity of the flora of cerrado. In R. 

C. Pereira and L. C. B. Nasser, eds., Anais do VIII Simpósio sobre o Cerrado 

and Proceedings of the I International Symposium on Tropical 

Savannas, pp. 3–6. Planaltina: Empresa Brasileira de Pesquisa 

Agropecuaria. 

Ratter, J. A., J. F. Ribeiro, and S. Bridgewater. 1997. The Brazilian cerrado 

vegetation and threats to its biodiversity. Ann. Bot. 80:223–230. 

Ratter, J. A., J. F. Ribeiro, and S. Bridgewater. 2000. Distribuição das espécies 

lenhosas da fitofisionomia cerrado sentido restrito nos estados compreendidos 

no bioma cerrado. Bol. Herb. E. P. Heringer 5 (in press). 


on the vegetation of northeastern Mato Grosso: I. The woody vegetation 

types of the Xavantina-Cachimbo Expedition Area. Phil. Trans. 

Roy. Soc. London B 226:449–492. 

Reis, A. C. S. 1971. Climatologia dos cerrados. In M. G. Ferri, ed., III Simpósio 

sobre o Cerrado, pp. 15–26. São Paulo: Editora da Universidade 


Ribeiro, J. E. L. S., M. J. G. Hopkins, A. Vicentini, C. A. Sothers, M. A. S. 

Costa, J. M. Brito, M. A. D. Souza, L. H. P. Martins, L. G. Lohman, P. 

A. C. L. Assunção, E. C. Pereira, C. F. Silva, M. R. Mesquita, and L. C.



Procópio. 1999. Flora da Reserva Ducke: Guia de identificação das plantas 

vasculares de uma floresta de terra-firme na Amazônia central. Manaus: 

INPA-DFID. 

Ribeiro, J. F. 1983. “Comparação da concentração de nutrientes na vegetação 

arbórea e nos solos de um cerrado e um cerradão no Distrito Federal.” 


Ribeiro, J. F. and M. Haridasan. 1990. Comparação fitossociológica de um 

cerrado denso e um cerradão em solos distróficos no Distrito Federal. In 

Anais do XXXV Congresso Nacional de Botânica, pp. 342–353. 

Brasília: Sociedade Botânica do Brasil. 

Ribeiro, J. F. and B. M. T. Walter. 1998. Fitofisionomias do Bioma Cerrado. 

In S. M. Sano and S. P. Almeida, eds., Cerrado: Ambiente e Flora, pp. 

87–166. Planaltina: Empresa Brasileira de Pesquisa Agropecuária. 

Rizzini, C. T. 1963. A flora do cerrado: Análise florística das savanas centrais. 

In M. G. Ferri, ed., Simpósio sobre o Cerrado, pp. 127–177. São Paulo: 


Rizzini, C. T. 1979. Tratado de Fitogeografia do Brasil: Vol. 2, Aspectos Sociológicos 

e Florísticos. São Paulo: Hucitec-Editora da Universidade de São 

Paulo. 

Rizzini, C. T. and E. P. Heringer. 1962. Preliminares Acêrca das Formações 

Vegetais e do Reflorestamento no Brasil Central. Rio de Janeiro: Edições 

SIA, Ministério da Agricultura. 

Rizzini, C. T. and M. M. Pinto. 1964. Areas climático-vegetacionais do Brasil 

segundo os métodos de Thornthwaite e Mohr. Rev. Bras. Geogr. 26: 

523–547. 

Schiavini, I. 1997. Environmental characterization and groups of species in 

gallery forests. In Proceedings of the International Symposium on Assessment 

and Monitoring of Forests in Tropical Dry Regions with Special 

Reference to Gallery Forests, pp. 107–113. Brasília: Editora da Universidade 

de Brasília. 

Silva Júnior, M. C. 1995. “Tree Communities of Gallery Forests of the IBGE 

Ecological Reserve, Distrito Federal, Brazil.” Ph.D. thesis, University of 

Edinburgh, Edinburgh, Scotland. 

Silva Júnior, M. C. 1997. Relationships between the tree communities of the 

Pitoco, Monjolo and Taquara gallery forests and environmental factors. 

In Proceedings of the International Symposium on Assessment and Monitoring 

of Forests in Tropical Dry Regions with Special Reference to 

Gallery Forests, pp. 287–298. Brasília: Editora da Universidade de 

Brasília. 

Silva Júnior, M. C., J. M. Felfili, P. E. Nogueira, and A. V. Rezende. 1998. 

Análise florística das matas de galeria no Distrito Federal. In J. F. Ribeiro, 

ed., Cerrado: Matas de Galeria, pp. 51–84. Planaltina: Empresa 

Brasileira de Pesquisa Agropecuaria. 

Silva Júnior, M. C., P. A. Furley, and J. A. Ratter. 1996. Variation in the tree 

communities and soils with slope in gallery forest, Federal District,



Brazil. In M. G. Anderson and S. M. Brooks, eds., Advances in Hillslope 

Processes, vol. 1, pp. 451–469. London: John Wiley and Sons. 

Silva Júnior, M. C., P. E. Nogueira, and J. M. Felfili. 1998. Flora lenhosa das 

matas de galeria no Brasil Central. Bol. Herb. E. P. Heringer 2:57–76. 

Van den Berg, E. and A. T. Oliveira-Filho. 1999. Spatial partitioning among 

tree species within an area of tropical montane gallery forest in southeastern 

Brazil. Flora 194:249–266. 

Van der Hammen, T. 1983. The palaeoecology and palaeogeography of 

savannas. In F. Bourliere, ed., Tropical Savannas, pp. 19–35. Amsterdam: 

Elsevier. 

Vanzolini, P. E. 1963. Problemas faunísticos do cerrado. In M. G. Ferri, ed., 

Simpósio Sobre o Cerrado, pp. 307–320. São Paulo: Editora da Universidade 


Walter, B. M. T. and J. F. Ribeiro. 1997. Spatial floristic patterns in gallery 

forests in the cerrado region. In Proceedings of the International Symposium 

on Assessment and Monitoring of Forests in Tropical Dry 

Regions with Special Reference to Gallery Forests, pp. 339–349. Brasília: 

Editora da Universidade de Brasília. 

Warming, E. 1892. Lagoa Santa: Et Bidrag til den Biologiske Plantegeographi. 


Warming, E. 1973. Lagoa Santa: Contribuição para a Geographia Phytobiologica 

(reprint of the 1908 Portuguese translation of the work). In M. G. 

Ferri, ed., Lagoa Santa e a Vegetação de Cerrados Brasileiros, pp. 1–284. 

Belo Horizonte: Livraria Itatiaia Editora and Editora da Universidade de 

São Paulo. 

Werner, P. A. 1991. Savanna Ecology and Management, Australian Perspectives 

and Intercontinental Comparisons. London: Blackwell.


7 

Herbaceous Plant Communities 

Tarciso S. Filgueiras 

The herbaceous plant communities, consisting of plant 

life forms not considered trees, can be found as the ground layer of forest 

habitats such as gallery and semideciduous forests and woodland cerradão. 

But they reach their highest diversity in open habitats such as 

campo limpo, campo sujo, campo rupestre, cerrado sensu stricto, and 

campo de murundus (see chapter 6). In such habitats a surprising number 

of life forms, taxonomic diversity, and adaptations can be found. These 

features make these communities very attractive both to biologists and to 

the general public. A campo limpo or an area of campo rupestre in full 

bloom during December and January is arguably one of the prettiest natural 

sights in central Brazil. These rich, challenging communities are 

described here. 

FLORISTICS AND PHYTOSOCIOLOGY 

Herbaceous communities are dominated by chamaephytes, hemicryptophytes, 

geophytes, therophytes, lianas, and epiphytes (Ellenberg and 

Mueller- Dombois 1967). These diverse life forms are collectively designated 

by an array of names such as the herbaceous layer (Medina 1987), 

ground layer (Eiten 1990, 1991), nontrees (Gentry and Dodson 1987), 

field layer (Munhoz and Proença 1998), herbaceous category (Filgueiras 

et al. 1998), nonarboreal category (Felfili et al. 1998), and nonwoody vegetation 

(Castro et al. 1999). These plants are less conspicuous in forests 

but are the most important elements in the open habitats, where they frequently 

cover the entire ground and dominate the scene in number of both 

species and individuals. 

121



The number of species in the herbaceous communities is high, and 

current data indicate at least 4,700 species (Mendonça et al. 1998) for the 

cerrado region. The number of species in the nonarboreal versus arboreal 

category is estimated to vary from 3:1 (Felfili et al 1994) to 4.5:1 (Mendonça 

et al. 1998). Thus, for each tree species in the cerrados there are 

between 3 and 4.5 nonarboreal species. The corollary to this statement is 

that any botanical survey that does not take into account the herbaceous 

members of the flora is grossly underestimating the total species diversity 

of the area. Therefore, the available data unequivocally disqualify earlier 

statements that the herbaceous flora in the Cerrado region is poorer than 

the woody component. Estimates around 2.5:1 by Castro et al. (1999) 

should also be disregarded because these authors overlooked recent data 

such as those of Mendonça et al. (1998), Felfili et al. (1998) and Filgueiras 

et al. (1998). 

The taxonomic composition of the plants in the nonarboreal category 

is varied. The dominant families are Leguminosae (ca. 780 spp.), Compositae 

(ca. 560 spp.), Gramineae (ca. 500 spp.), and Orchidaceae (ca. 

495 spp.). The genera with the highest number of species are Chamaecrista 

(ca. 120 spp.), Paspalum (ca. 117 spp.), Mimosa (ca. 113 spp.) Vernonia 

(ca. 100 spp.), Habenaria (ca. 70 spp.), and Panicum (ca. 63 spp.). 

On the other hand, a large number of genera are represented by a single 

species, such as Langsdorffia, Ottonia, Paragonia, Sanderella, Soaresia, 

and Tatianyx (Mendonça et al. 1998). 

The diversity and abundance of species vary greatly in the different 

vegetation types (Ratter 1987; Mantovani 1990; Mantovani and Martins 

1993) to the extent that the cerrado has aptly been described as a mosaic 

of resources under any resolution scale (Alho 1982). For example, while 

surveying the herbaceous communities of Chapada Pratinha (states of 

Goiás and Minas Gerais; 15°–20° S and 46°–49° W), Felfili et al. (1994) 

found that the gallery forests showed the lowest number of species (47) 

against 210 in the cerrado sensu stricto. The cerradão showed an intermediate 

number (63 spp.). In the various herbaceous communities at this 

locality, five families (Gramineae, Leguminosae, Euphorbiaceae, Compositae, 

and Rubiaceae) out of a total of 64 comprised 53% of all species 

reported. The remaining 59 families were represented by one to seven 

species. The grasses alone represented 25% of all species, and the legumes 

13%. About 51% of all individuals surveyed were grasses. 

In forest formations the more frequent genera and species are as follows: 

Coccocypselum and Psychotria, the species Olyra ciliatifolia, Oplismenus 

hirtellus, Serjania lethalis, and the ground orchid Craniches 

candida. Cerrado sensu stricto and campo sujo (chapter 6) are dominated


Herbaceous Plant Communities 123 

by species of the genera Axonopus, Chamaecrista, Croton, Hyptis, 

Mimosa, and Oxalis; and the species Echinolaena inflexa, Cissampelos 

ovalifolia, Trachypogon spicatus, Galactia glaucescens, and Andira 

humilis (Felfili et al. 1994; Mendonça et al.1998). 

Campo limpo is dominated by the Gramineae, and common grasses 

include Echinolaena inflexa, Leptocoryphium lanatum, Trachypogon 

spicatus, Paspalum spp., Axonopus spp., Mesosetum loliiforme, Schizachyrium 

tenerum, Tristachya leiostachya, and Aristida spp. Other commonly 

encountered taxa are Pradosia brevipes ( = Chrysophyllum 

soboliferum), Parinari obtusifolia, Smilax spp., Banisteriopsis campestris, 

Campomanesia spp., Cambessedesia espora, Myrcia linearifolia, Spiranthera 

odoratissima, Senna rugosa, Centrosema bracteosum, Anemopaegma 

spp., Byrsonima spp., Calea spp., Vernonia spp., and Mimosa 

spp. (Mendonça et al. 1998). 

Campos rupestres are predominantly found in the states of Minas 

Gerais, Bahia, and Goiás (chapter 6). The rich flora of these habitats is 

estimated at around 4,000 species (Giulietti et al. 1997). Serra do Cipó, 

a significant part of the Espinhaço Range (10°–20° S and 40°–44° W) in 

Minas Gerais, has been carefully surveyed by Giulietti et al. (1987). These 

authors have documented 1,590 species, the great majority in the nonarboreal 

category. The best-represented families are Compositae (169 spp.), 

Gramineae (130 spp.), Leguminosae (107 spp.), Melastomataceae (90 

spp.), Eriocaulaceae (84 spp.), and Orchidaceae (80 spp.). Characteristic 

genera are Paepalanthus, Leiothrix, Syngonanthus, Vellozia, Barbacenia, 

Xyris, Marcetia, Lychnophora, Declieuxia, Cambessedesia, Mimosa, and 

Microlicia. 

The flora of the campo rupestres is largely endemic, especially in 

groups such as Velloziaceae (ca. 70%), Eriocaulaceae (ca. 68%), and 

Xyridaceae (ca. 30%). Some species are narrow endemics, known only 

from a small area. Examples are found in the genera Barbacenia, Paepalanthus, 

Syngonanthus, and Xyris. Other families with endemic species 

are Iridaceae (Pseudotrimezia), Cactaceae (Cipocereus, Uebelmannia), 

Sterculiaceae (Raylea), and Compositae (Bishopella, Morithamnus). 

Aulonemia effusa, a shrubby bambusoid grass, is known only from the 

rocky outcrops of the Serra do Espinhaço, Minas Gerais. 

The Chapada dos Veadeiros (13°46' S and 47°30' W) is the bestknown 

campo rupestre in the state of Goiás. Several vegetation types 

occur there, and herbaceous communities abound everywhere. In a recent 

account of the local flora (both woody and herbaceous), Munhoz and 

Proença (1998) listed 1,310 species distributed in 498 genera and 120 

families. The five richest families were Leguminosae (144 spp.), Com-



positae (125 spp.), Gramineae (115 spp.), Melastomataceae (54 spp.), and 

Orchidaceae (47 spp.). There, too, the flora is largely herbaceous, and there 

are many endemics. The flora of this Chapada is only partially known, and 

further species of trees and ground vegetation are still being discovered. An 

unexpected biogeographic connection between the Chapada dos Veadeiros, 

Africa, and Australia recently came to light with the description of a new 

grass species of the genus Triraphis. The genus was previously known only 

from Africa (6 spp.) and Australia (1 sp.). The discovery of T. devia in a 

campo rupestre of the Chapada dos Veadeiros thus presents a puzzling biogeographical 

enigma (Filgueiras and Zuloaga 1999). 

Derived from ultramafic rocks, serpentine soils are characterized by 

high concentrations of magnesium, iron, nickel, chromium, and cobalt, 

and abnormally low levels of some essential plant nutrients such as phosphorus, 

potassium, and calcium (Brooks 1987). Such soils were found by 

Brooks et al. (1990) in several municipalities in Goiás, along with a serpentine-adapted 

flora almost exclusively dominated by shrubs, subshrubs, 

and herbs. In Niquelândia (ca. 14°18' S and 48°23' W) several 

serpentine endemic grass taxa have been described (Filgueiras et al. 1993; 

Filgueiras 1995). Because of this high level of endemism, Niquelândia has 

been included as one of 87 areas selected for special conservation efforts 

in the cerrado region (chapter 18; locality 33 in fig. 18.1). 

The wet habitats (such as the veredas, campo de murundus; see 

chapter 6) tell a different story. In the “Pantanal’’ of Mato Grosso, a huge 

seasonally flooded area with a large diversity of landscapes (Allem and 

Valls 1987; A. Pott 1988; V. J. Pott 1998), the dominant genera are 

Nymphaea, Nymphoides, Mayaca, Lycopodium, Pontederia, Eichhornia, 

Pistia, Ludwigia, Utricularia, and graminoids, such as grasses, sedges, and 

Eriocaulaceae. 

Outside the Pantanal, wet areas in general are dominated by the 

grasses Paspalum hydrophylum, P. pontonale, Leersia hexandra, Sorghastrum 

setosum, Reimarochloa acuta, and Axonopus purpusii; but 

Habenaria, a genus of ground orchids typical of wet habitats, is represented 

by at least 70 species (Mendonça 1998). In the Parque Nacional 

Grande Sertão Veredas (15°00' S and 45°46' W) the aquatic flora is also 

quite rich (Filgueiras, unpublished data), represented by Lycopodium, 

Lycopodiella, Ludwigia, Eleocharis, Bulbostylis, Paspalum morichalense, 

Panicum pernambucense, Cyperus giganteus, Eriocaulon, Philodice, and 

Syngonanthus. The aquatic flora of the savannas of Roraima (5°16' N and 

1°27' W) is dominated by Alismataceae, Eriocaulaceae, Iridaceae, Mayacaceae, 

Menyanthaceae, Nymphaeaceae, Pontederiaceae, and Orchidaceae 

(Miranda 1998).



It is important to stress that various sites differ greatly in species composition. 

In the Chapada Pratinha (15°–20° S and 46°–49° W), for instance, 

the five sites surveyed had 52 to 121 species (Filgueiras et al.1998), 

but only eight species were common to all sites. The total density varied 

from 595 to 3,278 plants per hectare. Echinolaena inflexa, a perennial, 

rhizomatous, C3 grass, represented 36% of all plants found in that Chapada. 

In Roraima, a predominantly “savanna’’ state, 66% of all species 

surveyed by Miranda (1998) were herbs, including 58 grasses and 40 

sedges. 

Great taxonomic diversity is commonly found in the same area, 

throughout the cerrados. For instance, Goldsmith (1974) found that 

species composition of a vereda changed gradually as one goes downslope. 

Filgueiras (unpublished data) also found in a cerrado sensu stricto 

at Gama/Cabeça de Veado (ca. 15°52' S and 47°58' W) 8 to 28 species 

per m 2 (mean 11.5 species per m 2 ). This range is relatively high, considering 

that the most distant plots were not more than 1,000 meters apart. 

The consequences of a mosaic habitat to conservation will be addressed 

later in this chapter. 

The general trend for herbaceous communities is to encounter high 

species densities in the cerrados. Monodominance is rare. A few documented 

exceptions are (1) at the Parque Nacional das Emas (18°10' S and 

52°55' W), where around 70% of the area (ca. 131,800 ha) is dominated 

by Tristachya leiostachya, a caespitose, perennial grass; and (2) huge 

tracts of land are totally covered by Actinocladum verticillatum 

(“taquari’’), resulting in an almost impenetrable thicket (Verdesio and 

Garra 1987). This thin-culmed bamboo reaches 80–200 cm in height and 

about 1 cm in diameter. The immense area where it grows is locally called 

grameal. The name denotes a special kind of vegetation and can be loosely 

translated as “vast quantity of grass.’’ 

Essentially herbaceous stemmed vines and lianas are fairly common. 

The families with the largest number of species with these life forms are 

Aristolochiaceae (Aristolochia), Asclepiadaceae (Blepharodon, Ditassa 

and Oxypetalum), Bignoniaceae (Arrabidaea, Cuspidaria, and Paragonia 

pyramidata), Convolvulaceae (Ipomoea), Cucurbitaceae (Cayaponia, 

Melancium), and Sapindaceae (Serjania). Mikania, a genus of about 18 

vine species, is a member of the Compositae. The parasite Cassytha filiformis 

(Lauraceae) is ubiquitous. 

Palms are not very diverse in the Cerrado, but their ecological, economic, 

and cultural importance greatly surpasses their low taxonomic 

diversity. They represent valuable resources for wildlife, domesticated animals, 

and many rural populations. A few acaulescent and shrubby genera



and species occur: Allagoptera spp., Astrocaryum campestre, Attalea spp., 

Butia archeri, Syagrus graminea, and S. petraea (Mendonça et al. 1998). 

There are about 10 bambusoid grasses with a shrubby or herbaceous habit 

in the Cerrado (e.g., Actinocladum verticillatum, Apoclada cannavieira, A. 

arenicola, Aulonemia aristulata, Olyra ciliatifolia, O. taquara, Pharus lappulaceus, 

and Raddiella esenbeckii) (Judziewicz et al. 1998). The latter, a 

diminutive plant up to 30 cm tall, resembles a fern and is found on river 

banks in forests and occasionally in cerradão (chapter 6). The bambusoid 

species of open habitats are fairly common throughout the cerrado and 

under certain ecological conditions form vast colonies. 

A small number of species of cacti occur. A few examples of lowgrowing 

taxa are Cipocereus, Epiphyllum phyllanthus, Melocactus paucipinus, 

Pilosocereus pachycladus, Rhipsalis spp., and Uebelmannia 

(Taylor and Zappi 1995). Other genera and species of cacti have been 

described (Diers and Esteves 1989; Braun 1990). The epiphytic flora is 

not particularly diverse in the cerrado, as compared with other regions 

such as Amazonia (Daly and Prance 1989) or the Atlantic forest (Lima 

and Guedes-Bruni 1994). The families best represented are Orchidaceae 

(Bulbophyllum, Cattleya, and Oncidium), Bromeliaceae (Aechmaea, Bilbergia, 

and Tillandsia), Araceae (Philodendron), Piperaceae (Peperomia), 

and Cactaceae (Rhipsalis spp.). Epiphytic ferns are also common (Adiantum, 

Anemia, and Elaphoglossum) (Mendonça et al. 1998). 

Hemiparasitic taxa are restricted to the Loranthaceae (Phthirusa, 

Psittacanthus, and Struthanthus). The most abundant species is Phthirusa 

ovata, whose fruits (red and sticky when ripe) are easily dispersed by 

birds. Species of Psittacanthus and Struthanthus are also rather common 

in cerradão and cerrado sensu stricto. When in full bloom, their showy 

yellowish or reddish flowers stand out in the cerrado vegetation. 

The commonest true parasites and saprophytes are Apteria aphylla, 

Cuscuta spp., Helosia brasiliensis, Langsdorffia hypogea, Cassytha spp., 

Pilostyles spp., and Voyria flavescens. Insectivorous droseras are found in 

swamps, damp places, veredas, and wet campos. Although they are quite 

small (usually only 3 to 10 cm tall), they are easily recognizable by their 

reddish color and conspicuous glandular hairs. A single shrubby gymnosperm 

(Zamia bronguiartii) is found in cerrado sensu stricto and on 

limestone outcrops in the state of Mato Grosso (Guarim Neto 1987). 

Invasive species are found practically in all anthropic habitats, be it 

forest, cerrado sensu stricto, campo sujo, vereda, agricultural field, pasture, 

or land around human dwellings (chapter 5). Mendonça et al. (1998) 

listed 456 invasive species in the cerrado. Invasive species can pose a serious 

threat to the cerrado flora by successfully competing with the natives



or even eliminating them altogether (Filgueiras 1990). Loss of biodiversity 

is therefore one of the most serious consequences of the introduction 

of weeds in cerrado ecosystems. Some introduced African grasses of special 

concern include Andropogon gayanus, Brachiaria spp, Hyparrhenia 

rufa, Melinis minutiflora, and Panicum maximum. It is postulated that 

plants in the nonarboreal category are probably more sensitive to humanrelated 

disturbances than trees in the cerrados (Filgueiras et al. 1998). 

Therefore, plants in that category may be used as a parameter to assess 

the level of human intervention in cerrado ecosystems. 

SPECIAL ADAPTATIONS AND COMPARISONS 

WITH OTHER ECOSYSTEMS 

Many cerrado species display an unusual behavior: their aerial stems die 

and revert to the underground once a year, during the dry season, even in 

the absence of fire. When conditions improve, they sprout again from special 

subterranean structures, the xylopodia. These plants are called recurrent 

shrubs and semi-shrubs by Eiten (1984), and hemixyles or geoxyles 

by others (Ellenberg and Mueller-Dombois 1967). An even more peculiar 

trait of several species in this category is the existence of extensive subterranean 

woody systems. Apparently Reinhardt (in Warming 1973:58) 

was the first author to note this peculiarity. While traveling in central 

Brazil he noted that “les grands arbres souterrains à tige verticale cachée 

dans le sol, sont une des peculiarités les plus curieuses de la flore de ces 

régions.’’ 

Warming (1973) himself was quite impressed by this singularity of 

some cerrado species at Lagoa Santa, Minas Gerais. He described and 

illustrated a plant of Andira laurifolia ( = A. humilis) with a well-developed 

subterranean system. He emphasized that an area of about 10 m in 

diameter was occupied by a single individual plant. The investigation of 

the subterranean growth forms of several cerrado species has been subsequently 

pursued by researchers such as Rawitscher and Rachid (1946), 

Rizzini and Heringer (1966), and Paviani (1987). 

Some of these subterranean systems are clearly means of vegetative 

reproduction. The plant sprouts in different directions, but all the shoots 

are interconnected below ground, forming an extensive, complex system. 

An example is Parinari obtusifolia, in which the aerial stems give the 

appearance of individual plants, but excavation reveals that all the stems 

are shoots from the same system of subterranean axes. If considered in its 

entirety, this classical example of a cerrado subshrub could more accurately



be classified as a kind of “subterranean tree.’’ A similar pattern can be 

seen in Anemopaegma arvense, where the aerial stems can be as far as 120 

cm apart. 

This phenomenon seems to be quite widespread and should be investigated 

in depth because it has serious implications for the definition of 

the individual plant in certain cerrado vegetation types, clearly a critical 

definition in phytosociological studies. In some cases, what has been 

regarded as several plants is in reality a single individual whose aerial 

stems are sometimes more than one meter apart but all of whose underground 

parts are connected. Colonies of Annona pygmaea, Andira 

humilis, Pradosia brevipes, Parinari obtusifolia, and Smilax goyazana, 

among others, can reach several meters in diameter. Each of these colonies 

may very well derive from a single seed. The same can be said of some 

grasses with extensive stoloniferous growth habit, such as Axonopus purpusii, 

Paspalum morichalense, and Reimarochloa acuta. 

Another intriguing aspect of the cerrado herbaceous flora is the low 

occurrence of annual species, representing less than 5%. Thus, perenniality 

is clearly at an advantage. The success of the perennial versus the 

annual habit in the cerrado may be due to the limited amount of “topsoil’’ 

water during some months of the year (May–September), such that 

plants need to possess extensive root systems or other adaptations to withstand 

the dry periods (chapter 10). However, in Gilbués, Piauí (ca. 9°34' 

S and 44°55' W), a marginal cerrado area (Castro et al. 1999) subjected 

to serious desertification processes, Filgueiras (1991) found 30 times more 

annuals than perennials. Under prolonged dry periods, many perennials 

cannot survive except as seeds. Apparently this is precisely what happens 

in some areas of the adjacent arid caatinga where, among the herbaceous 

plants, annuality is the rule. 

Data on the biomass of plants in the herbaceous layer of cerrados is 

meager. However, it is well established that they provide the bulk of the 

fuel for the seasonal fires (Coutinho 1990). In the Distrito Federal, central 

Brazil, Kauffman et al. (1994) found that the biomass of fuel loads 

varied greatly from campo limpo (7,128 kg/ha) to cerrado sensu stricto 

(10,031 kg/ha) (chapter 4). Grasses comprised 91% to 94% of the total 

aboveground biomass in campo limpo and campo sujo, whereas in the 

campo cerrado and cerrado sensu stricto communities the graminoids represented 

only 27%. The remainder of fuel load biomass was composed of 

deadwood (18%), dicot leaf litter (36%), and dicot and shrub leaves 

(18%). In Roraima, Miranda and Absy (1997) found that grasses and 

sedges were the most important families in terms of total biomass. In the 

10 areas sampled, the grasses comprised 43% of the biomass; the sedges,



35%. Silva and Klink (1996) found significant variation in biomass at the 

species level, in a study that compared the biomass of two native grasses, 

Schizachyrium tenerum (C4) and Echinolaena inflexa (C3). After 24 

weeks of growth, S. tenerum was taller, had more tillers and leaves, and 

allocated more biomass to roots and shoots than E. inflexa. 

ECOLOGICAL, ECONOMIC, 

AND ANTHROPOLOGIC IMPORTANCE 

It is not surprising that the highly diverse flora of the cerrado is useful to 

humans and fauna in a variety of ways. The usefulness of these plants falls 

into several categories, such as food, forage, medicine, ornament, and 

genetic resources. Because they cover vast tracts of land, these plant communities, 

especially the veredas and the aquatic and semi-aquatic communities, 

play a key role in watershed protection. Invariably, wherever 

there is a stream or river source, the presence of herbaceous communities 

covering the soil minimizes erosion. On the other hand, their absence indicates 

that erosion processes will begin and much soil will be lost in a short 

time. To maintain the landscape intact, it is essential to allow a rich, 

diverse herbaceous community to establish and thrive locally; the important 

role of nonarboreal plants in natural habitats becomes apparent 

when they are removed and the soil is exposed. The removal of the herbaceous 

layer in the cerrado means the disruption of countless ecological 

processes. Besides the obvious demise of the plant life involved, the local 

fauna that depends directly or indirectly on the plants will be greatly 

reduced or, more likely, altogether eliminated. Animals lose their food, 

vital space, breeding places, and escape routes. Erosion processes may 

begin quickly. The consequences of erosion and the silting of river channels 

are well documented, especially in agricultural systems (Goedert et 

al. 1982; Dedecek et al. 1986). 

An important but little discussed aspect of the ecological function of 

the nonarboreal flora is the effective protection offered to some native 

bird and small mammal populations (chapters 13, 14). These animals 

feed, breed, and raise their young in the habitats dominated by these 

plants. Some small mammals, especially rodents, have very limited home 

ranges (Alho 1982), and their spatial distribution is closely connected 

with the distribution of natural resources. Some are generalists, but a few 

are specialists. For example, the rodent Bolomys ( = Zygodontomys) lasiurus 

feeds mainly on grass “seeds’’ and is found in a large number of habitats, 

whereas Oxymycterus robertii occurs mostly in open grasslands



dominated by Tristachya leiostachya. Many birds, mammals, and especially 

insects are obligatory pollinators and seed dispersal agents of a 

number of nonarboreal cerrado species. Entire guilds of insects depend 

exclusively on these plants during their life cycles (chapter 17). The role 

that all these animal species play in the maintenance of cerrado biodiversity 

is crucial albeit not readily apparent. 

A great extent of cerrado land is used for grazing (chapter 5). About 

40% of the Brazilian cattle industry depends on this native grazing land 

(Filgueiras and Wechsler 1992). A. Pott (1988) presented a list of 145 

grasses, 70 legumes, and 60 forb and browse species that feed four million 

cattle in the pantanal of Mato Grosso, western Brazil. The more 

important taxa in the pantanal are Axonopus purpusii, Mesosetum 

chaseae, Hemarthria altissima, Leersia hexandra, Paratheria prostrata, 

Paspalidium paludivagum, Paspalum plicatulum, and species of the genera 

Reimarochloa, Aeschynomene, Discolobium, Galactia, Rhynchosia, 

Teramnus, and Vigna. In the Distrito Federal alone, central Brazil, 

Filgueiras (1992) listed 134 native forage grass species, 13 of which are 

of very high quality. 

Legumes are extremely important in the cerrados. Kirkbride (1984) 

listed 548 legume species in the region, distributed in 50 genera. Among 

the genera with fodder species are Stylosanthes, Zornia, Desmodium, 

Aeschynomene, Eriosema, and Arachis. Because of their high forage 

value, Aeschynomene (ca. 52 spp.; Fernandes 1996) and Stylosanthes (ca. 

11 spp.; Ferreira and Costa 1979) deserve special attention. Besides 

increasing the nutritional value of fodder, legumes also increase natural 

soil fertility by adding nitrogen to the system through the fixation of 

atmospheric nitrogen by associated Rhizobium. An evaluation of the 

genetic resources of legumes and grasses native to Brazil is presented by 

Valls and Coradin (1986). 

Although grasses and legumes make up the bulk of the grazed plants 

of the cerrado, they are by no means the only fodder resources there. 

Macedo et al. (1978) found 83 cerrado species, belonging to 33 families, 

eaten by cattle in the cerrado. The pattern of consumption of leaves, flowers, 

and seeds by cattle and other domestic animals depends on the plant 

species, habitat, and season. Similarly, Pereira (1984) cites 35 species in 

the cerrado of the Distrito Federal that are frequently eaten by cattle, 26 

of which are neither grasses nor legumes. 

There is a wealth of medicinal plants amongst the nonarboreal 

species: roots, bark, stems, leaves, fruits, and seeds of over one hundred 

species are used by local populations and are also regularly commercialized 

in local markets and even exported to other areas outside the cerrado



region. In many large Brazilian cities it is possible to buy medicinal plants 

to “cure’’ almost any kind of malady. The so-called raizeiros (literally, root 

dealers, or root healers) display their plants in fairs and open markets. 

The plants are sold in natura or as garrafadas (i.e., plant parts bottled in 

alcohol; Guarim Neto 1987), wine, or pinga (i.e., white rum, also known 

as cachaça). The plants supplying this market are harvested directly from 

the wild. It is noteworthy that there is always a fresh supply of plants to 

sell. This means that the harvesting is continuous throughout the year and 

through the years. The impact of continuous harvesting of wild plants 

from natural populations has not yet been evaluated. 

The effectiveness of popular herbal treatments is understandably subject 

to dispute, but a few species have been scientifically tested with 

extremely good results (e.g., Psychotria ipecacuanha and Renealmia exaltata). 

A growing number of Brazilian botanists are studying the native 

medicinal flora (Siqueira 1988; Guarim Neto 1987; Barros 1982). Graduate 

programs that include research on medicinal plants have been established 

in several Brazilian universities. A small number of selected species 

has become the prime target for detailed investigation (e.g., Heteropterys 

aphrodisiaca, Zamia brogniartii, Gomphrena officinalis, Pfaffia jubata 

and Macrosiphonia velame). Other medicinal plants commonly used are 

Centrosema bracteosum, Lychnophora ericoides, Palicourea coriacea, P. 

marcgravii, P. rigida, Anemopaegma arvense, and Galactia glaucescens. 

Aromatic plants are also found. The roots of Croton adenodontus 

(“arcassu’’) are used to flavor milk (the taste and odor are described as 

better than chocolate!), and the fleshy roots of Escobedia grandiflora are 

used as a local substitute for expensive saffron (Pereira 1992). 

Toxic plants are found in several taxonomic groups (Ferreira 1971), 

but they are particularly abundant in the 35 species of Psychotria. Palicourea 

marcgravii (“erva-de-rato’’) is a major cattle poisoner in certain 

areas. It is generally held that frequently even toxic plants can be of medicinal 

importance, depending on the specific dosage of the active substance 

(Hoehne 1939). 

Exploration of the phytochemistry and pharmacology of cerrado 

herbaceous plants can be rewarding (Gomes et al. 1981, Gottlieb et al. 

1996). The potential of extracting lauric acid from Cuphea is being 

explored (Arndt 1985; Hirsinger 1985), and there are 43 native species of 

Cuphea in the cerrado (Mendonça et al.1998). The possibility of extracting 

essential oils from the 68 native species of Hyptis or even from the 

lemon grasses (Elionurus spp.) should also be explored. The cultivation 

(domestication) of the more promising species is a possible solution for 

the exploitation of natural stocks. The biosynthesis of new products and



drugs is an area where close cooperation between governmental agencies 

and the private initiative is likely to make a difference. The search for new 

drugs and new food items has been strongly defended by prominent biologists 

such Raven (1990), Heywood (1992), and Wilson (1992). 

Sustainability in the exploitation of medicinal plants must be a priority. 

The most-used species are suffering great reduction on their natural 

stocks due to decades of relentless exploitation. Such is the case of 

Lychnophora ericoides and of Pseudobrickellia pinifolia (both called 

“arnica’’), whose stems and leaves are collected by the tons and sold everywhere 

in central Brazil as a kind of natural antibiotic. Xylopodia and 

roots of Anemopaegma arvense (“verga-tesa’’) and of Centrosema bractesoum 

(“rabo-de-tatu’’) are also harvested and sold everywhere as an 

aphrodisiac and against stomach maladies, respectively. The worst case is 

that of Psychotria ipecacuanha (“poaia,’’ “ipeca,’’ or “ipecacuanha’’). At 

Serra do Tapirapuã, state of Mato Grosso, there used to be a unique vegetation 

type (known as “poaia forest’’) where the undergrowth was dominated 

by this species (Ferreira 1999). After many decades of exploitation 

of the roots of “poaia,’’ the species is now scarce in the area and is doomed 

to local extinction. One might solve this problem through cultivation of 

promising wild species using simple but sound agronomic techniques such 

as those described by Mattos (1996). 

Honey-producing herbaceous plants are also abundant and exploited 

commercially. Pereira (1990) found 220 species (of which 50 are Leguminosae 

and 40 Compositae) regularly visited by honeybees in the cerrado 

of the Distrito Federal in central Brazil. 

The flourishing business of dried wildflower arrangements is a source 

of income for many families. It is the so-called sempre-viva industry 

(Giulietti et al. 1997). In Goiás and Distrito Federal the trade is called 

“Flores do Planalto’’ (Ferreira 1974; Filgueiras 1997). In the state of 

Minas Gerais this type of business has been quite active for more than 30 

years. The plants are dried, tied in bundles, and sold as long-lasting bouquets. 

Eriocaulaceae, Gramineae and Compositae supply the bulk of the 

plants collected for this purpose. Many families in the states of Minas 

Gerais and Goiás set up “firms’’ and live exclusively on the “sempre-viva’’ 

trade. During past decades this was a thriving business. Since then, the 

rate of exportation has declined considerably. 

A large number of native cerrado herbaceous species have obvious 

potential as ornamentals. Orchids, bromeliads, and cacti need not be discussed 

here, because there is already an active market that thrives on the 

legal and illegal commercialization of these plants, a large percentage of 

which are harvested in the wild. But there is a great potential to establish 

legal enterprises to exploit the trade (through cultivation) of ornamental



cerrado herbaceous plants. All the known species of Paepalanthus are very 

ornamental, but P. speciosus and P. hillairei are outstanding. The same can 

be said of many other species, including all the previously mentioned 

species of bamboos and palms. The genera Hippeastrum, Allamanda, 

Mandevilla, Anthurium, Asterostygma, Philodendron, Spathiphyllum, 

Begonia, Lobelia, Cochlospermum, Calea, Lagenocarpus, Calliandra, 

Chamaecrista, Collaea, Mimosa, Hibiscus, Pavonia, Maranta, Cambessedesia, 

Lavoisiera, and Microlicia are examples from an extensive list 

of plants whose ornamental value merits close investigation. The perfume 

industry might have potential in exploring the unique fragrance of certain 

cerrado flowers, such as Spiranthera odoratissima, Protium ovatum, and 

Mimosa spp. 

The conservation of crop genetic resources is vital for humankind. It 

is well established that genetic uniformity can make crops vulnerable to 

epidemics of pests and diseases (Myers 1983). The nonarboreal flora of 

the cerrados encompasses a large number of species related to crops that 

represent invaluable resources in crop improvement programs. Despite 

their obvious potential as sources of wild genes, they have been largely 

neglected by agronomists and plant breeders. The following list will provide 

a small sample of the potential in this field: in the cerrados there exist 

at least 42 species of Manihot, 37 Dioscorea, 33 Arachis, 30 Ipomoea, 28 

Solanum, 26 Psidium, 25 Piper, 22 Passiflora, 12 Annona, 11 Cissus 

(related to Vitis), 8 Vanilla, 7 Vigna, 5 Hibiscus, 4 Anacardium, 3 Oryza, 

2 Ananas, and 1 Phaseolus (P. uleanus) (Mendonça et al. 1998). 

Hoyt (1988) argues that most protected areas are established to preserve 

a unique landscape or a rare animal species, but seldom to preserve 

a wild relative of an important crop. It is time for Brazil to protect the 

native cerrado genetic resources effectively, especially those related to 

crops and medicinal gene pools. A significant step in this direction has 

been the founding of the Center for Genetic Resource and Biotechnology 

(CENARGEN) of the Brazilian Research Enterprise (EMBRAPA). 

CENARGEN maintains a germplasm bank and an active cerrado plant 

collecting program (Schultze-Kraft 1980) considered as a model for developing 

countries. Of special interest are the collections of Arachis spp. and 

forage grass species (especially Paspalum spp.). 

RESEARCH NEEDS AND CONCLUSIONS 

A remarkable paper by Constanza et al. (1997) demonstrates, in economic 

terms, the value of the services that natural ecosystems provide to humanity. 

The various ecosystems are viewed as natural capital. The idea that



natural ecosystems are worth money is not new; nonetheless it is astonishing 

to see that the actual estimated mean value of the ecological services 

rendered by the world’s ecosystems and the natural capital that 

generates them amounts to U.S.$33 trillion. 

In a recent workshop held in Brasília, and sponsored by the Ministry 

of the Environment and three nongovernmental organizations to address 

this issue, 87 areas were selected for special conservation efforts (see chapter 

18). The document resulting from this workshop further states that 

the conservation of cerrado biodiversity can be achieved by establishing 

conservation units, doing inventories, supporting herbaria, and monitoring 

natural populations. The nonarboreal flora of the cerrado is so rich 

and vital to many human endeavors that its study should be greatly 

encouraged. Nonetheless, most species in this category are of notorious 

taxonomic difficulty (such as orchids, grasses, sedges, Compositae, legumes, 

etc.). A way to overcome this difficulty would be to publish inexpensive 

field guides, such as those published for the herbs (Correa et al. 

1998a) and for ferns and epiphytes of Panama (Correa et al. 1998b). In 

these publications the specimens representing each species are simply photocopied 

and bound. This simple initiative would most certainly widen 

interest in the study of these valuable components of the cerrado flora. 

Wilson (1992) offered two important suggestions regarding the future 

preservation and conservation of biodiversity in general: a survey of the 

world’s total flora, and the search for new biological products such as foods 

and drugs. A cerrado checklist such as that of Mendonça et al. (1998) is 

extremely useful and a promising beginning, but a “Cerrado Flora’’ is now 

desirable. If a cerrado flora is to be produced, inventories have to be made, 

herbaria have to be equipped and supported, and human resources must be 

trained and subsequently hired to work on their specialties. 

Special attention should be given to the education of a new generation 

of plant taxonomists. They should be motivated and financially supported. 

Once properly trained, they should document the immense 

floristic wealth of the Cerrado Biome while opening new avenues for the 

rational uses of cerrado natural resources. The data presented here indicate 

that efforts and investments toward this most timely goal will be 

highly rewarding. 


I gratefully thank Drs. J. H. Kirkbride, Jr. and Jeanine M. Felfili for helpful 

suggestions on the manuscript. Two anonymous reviewers greatly



improved the text. Dr. Taciana Cavalcanti provided valuable literature 

assistance, and Maria S.S. Amorim checked the bibliography for errors 

and inconsistencies. I also thank the Conselho Nacional de Desenvolvimento 

Científico e Tecnológico (CNPq) for a productivity research grant 

(Proc. No 301190/86-0). 

REFERENCES 

Alho, C. J. R. 1982. Brazilian rodents: Their habitats and habits. The 

Pymatuning Symposia in Ecology. Special Publication Series. University 

of Pittsburgh, 6:143–166. 

Allem, A . C. and J. F. M. Valls. 1987. Recursos Forrageiros Nativos do Pantanal 

Mato-grossense. Brasília: Empresa Brasileira de Pesquisa Agropecuária. 

Arndt, S. 1985. Cuphea: Diverse fatty acid composition may yield oleochemical 

feedstock. J. Am. Oil Chem. Soc. 62:6–12. 

Barros, M. A. G. 1982. Flora medicinal do Distrito Federal. Brasil Florestal 

12:35–45. 

Braun, P. J. 1990. Melocactus estevesii P.J.Braun: Eine neue Kakteenart aus 

Nord Brasilien. Kakteen und andere Sukkulenten 41:6–10. 

Brooks, R. R. 1987. Serpentine and its Vegetation. Portland: Dioscorides 

Press. 

Brooks, R. R., R. D. Reeves, A. J. Baker, J. A. Rizzo, and H. D. Ferreira. 1990. 

The Brazilian serpentine expedition (Braspex), 1988. Nat. Geogr. Res. 

6:205–219. 


How rich is the flora of Brazilian cerrados? Ann. Miss. Bot. Gard. 

86:192–224. 

Constanza, R., R. d`Arge, R. de Groot et al. 1997. The value of the world’s 

ecosystem services and the natural capital. Nature 387:1–13. 

Correa, M. D. A. 1998a. Guía Preliminar de Campo: Vol. 3, Flora del Parque 

Nacional Altos de Campana, Pananá: Hierbas. Chicago: Universidad 

de Panamá and Smithsonian Tropical Research Institute. 

Correa, M. D. A. 1998b. Guía preliminar de campo: Vol. 4, Flora del Parque 

Nacional Altos de Campana, Pananá: Bejucos y Epífitas. Chicago: Universidad 

de Panamá and Smithsonian Tropical Research Institute. 

Coutinho, L. M. 1990. Fire in the ecology of the Brazilian cerrado. In J. G. 

Goldammer, ed., Fire in Tropical Biota, pp. 82–105. Berlin: Springer-Verlag. 

Daly, D. C. and G. T. Prance. 1989. Brazilian Amazon. In D. G. Campbell 

and H. D. Hammond, eds., Floristic Inventory of Tropical Countries: 

The Status of Plant Systematics, Collections, and Vegetation, plus Recommendations 

for the Future, pp. 401–426. New York: The New York 

Botanical Garden.



Dedecek, R. A., D. V. S. Resck, and E. Freitas Jr. 1986. Perdas de solo, água 

e nutrientes por erosão em latossolo vermelho-escuro nos Cerrados em 

diferentes cultivos sob chuva natural. Rev. Bras. Ciênc. Solo 10:265–272. 

Diers, L. and E. Esteves. 1989. Arrajadoa beatae Braun and Esteves, eine neue 

Art (Cactaceae) aus Minas Gerais/Brasilien. Kakteen und andere Sukkulenten 

40:250–256. 

Eiten, G. 1984. Vegetation of Brasília. Phytocoenologia 12:271–292. 

Eiten, G. 1990. A vegetação do Cerrado. In M. N. Pinto, ed., Cerrado: Caracterização, 

ocupação e perspectivas, pp.17–66. Brasília: Editora Universidade 

de Brasília. 

Eiten, G. 1991. What’s a herb? Veröff. des Geobotanishchen Institutes der 

Eidg. Techn. Hochschule Stiftung Rübel 106:288–304. 

Ellenberg, H. and D. Mueller-Dombois. 1967. A key to Raunkiaer plant life 

forms with revised subdivisions. Berichte des Geobotanishchen Institutes 

der Eidg. Techn. Hochschule Stiftung Rübel 37:56–73. 

Felfili, J. M., T. S. Filgueiras, M. Haridasan, M. C. da Silva Jr., Mendonça, R. 

C., and A. V. Rezende. 1994. Projeto biogeografia do bioma Cerrado: 

Vegetação e solos. Cad. Geoc. IBGE 12:75–166. 

Felfili, J. M., M. C. da Silva Jr., T. S. Filgueiras and P. E. Nogueira. 1998. 

Comparison of Cerrado (s.s.) vegetation in central Brazil. Ciênc. Cult. 

50:237–243. 

Fernandes, A.. 1996. O Táxon Aeschynomene no Brasil. Fortaleza: Edições 

Editora da Universidade Federal do Ceará. 

Ferreira, M. B. 1971. As plantas tóxicas no Distrito Federal. Cerrado 4:26–30. 

Ferreira, M. B. 1974. “Flores do Planalto,’’ divisas para Brasília. Cerrado 6:4–7. 

Ferreira, M. B. and N. M. S. Costa. 1979. O Gênero Stylosanthes no Brasil. 

Belo Horizonte: Empresa de Pesquisa Agropecuária de Minas Gerais. 

Ferreira, M. S. F. D. 1999. Mato Grosso: Um Pouco de seus Ecossistemas. 

Cuiabá: Instituto de Biociências, Universidade Federal de Mato Grosso. 

Filgueiras, T. S. 1990. Africanas no Brasil: Gramíneas introduzidas da África. 

Cad. Geoc. 5:57–63. 

Filgueiras, T. S. 1991. Desertificação em Gilblués, Piauí: Uma análise agrostológica. 

Cad. Geoc. 7:23–27. 

Filgueiras, T. S. 1992. Gramíneas forrageiras nativas no Distrito Federal. 

Pesquisa Agropecuária Brasileira 27:1103–1111. 

Filgueiras, T. S. 1997. Distrito Federal: Brazil. In S. D. Davis, V.H. Heywood, 

O. Herrera-MacBryde, J. Villa-Lobos. and A . C . Hamilton, eds., Centres 

of Plant Diversity, pp. 405–410. Oxford: World Widlife Fund and 

International Union for Conservation of Nature and Natural Resource. 

Filgueiras, T. S. 1995. Paspalum niquelandiae (Poaceae: Paniceae), a new 

species from the serpentine outcrops of central Brazil. Novon 5:30–33. 

Filgueiras, T. S., G. Davidse, and F. O. Zuloaga. 1993. Ophiochloa: A new 

endemic serpentine grass genus (Poaceae:Paniceae) from the Brazilian 

Cerrado vegetation. Novon 3:310–317. 

Filgueiras, T. S., J. M. Felfili, M. C. da Silva Jr., and P. E. Nogueira. 1998.



Floristic and structural comparison of cerrado (sensu stricto) vegetation in 

central Brazil. In F. Dallmeier and J. A. Comiskey, eds., Forest Biodiversity 

in North, Central and South America, and the Caribbean: Research and 

Monitoring, pp. 633–647. Paris: The Parthenon Publishing Group. 

Filgueiras, T. S. and F. S. Wechsler. 1992. Pastagens nativas. In B. F. S. Dias, 

ed., Alternativas de Desenvolvimento dos Cerrados: Manejo dos Recursos 

Naturais Renováveis, pp. 47–49. Brasília: Fundação Pró-Natureza. 

Filgueiras, T. S. and F. O. Zuloaga. 1999. A new Triraphis (Poaceae: Eragrostideae) 

from Brazil: The first record of a native species in the New 

World. Novon 9:36–41. 

Gentry, A. and Dodson, C. 1987. Contribution of nontrees to the species richness 

of a tropical rain forest. Biotropica 19:149–156. 

Giulietti, A. M., N. L. Menezes, J. R. Pirani, M. Meguro, and M. G. L. Wanderley. 

1987. Flora da Serra do Cipó, Minas Gerais: Caracterização e lista 

das espécies. Boletim de Botânica 9:1–151. 

Giulietti, A. M., J. R. Pirani, R. M. Harley. 1997. Espinhaço range region: 

Eastern Brazil. In S. D. Davis, V.H. Heywood, O. Herrera-MacBryde, J. 

Villa-Lobos, and A . C . Hamilton, eds., Centres of Plant Diversity, pp. 

397–404. Oxford: World Widlife Fund and International Union for Conservation 

of Nature and Natural Resource. 

Goedert, W. J., E. Lobato, and M. Resende. 1982. Management of tropical 

soils and World food prospects. Proceedings of the 12th International 

Congress of Soil Science, pp. 317–325. New Delhi, India. 

Goldsmith, F. B. 1974. Multivariate analysis of of tropical grassland communities 

in Mato Grosso, Brazil. J. Biogeogr. 1:111–122. 

Gomes, C. M. R., O. R. Gottlieb, G.-B. Marini-Bettòlo, F. Delle Monache, 

and R. Polhill. 1981. Systematic significance of flavonoids in Derris and 

Lonchocarpus Bioch. Syst. Ecol. 9:129–147. 

Gottlieb, O. R., M. A. C. Kaplan, and M. R. de M. R. Borin. 1996. Biodiversidade: 

Um Enfoque Químico-Biológico. Rio de Janeiro: Editora Universidade 

Federal do Rio de Janeiro. 

Guarim Neto, G. 1987. Plantas Utilizadas na Medicina Popular do Estado de 

Mato Grosso. Brasília: Conselho Nacional de Desenvolvimento Científico 

e Tecnológico. 

Heywood, V. H. 1992. Conservation of germplasm of wild species. In O. T. 

Sandlund, K. Hindar and A. H. D., eds., Conservation of Biodiversity for 

Sustainable Development, pp. 189–203. Oslo: Scandinavian University 

Press. 

Hirsinger, H. 1985. Agronomic potential and seed composition of Cuphea, an 

annual crop for lauric and capric seed oils. J. Am. Oil Chem. Soc. 62:76–80. 

Hoehne, F. C. 1939. Plantas e Susbtâncias Tóxicas e Medicinais. Rio de 

Janeiro: Graphicars. 

Hoyt, E. 1988. Conserving the wild relatives of crops. Gland, Switzerland: 

World Widlife Fund and International Union for Conservation of Nature 

and Natural Resource.



Judziewicz, E. J., L. G. Clark, X. Londoño, and M. J. Stern. 1998. American 

Bamboos. Washington, DC: Smithsonian Institution Press. 

Kauffman, J. B., D. L. Cummings, and D. E. Ward. 1994. Relationships of 

fire, biomass and nutrient dynamics along a vegetation gradient in the 

Brazilian cerrado. J. Ecol. 82:519–531. 

Kirkbride, J. H. 1984. Legumes of the Cerrado. Pesquisa Agropecuária 

Brasileira 19:23–46. 

Lima, M. P. de and R. R. Guedes-Bruni.1994. Reserva Ecológica de Macaé 

de Cima, Nova Friburgo, RJ: Aspectos Florísticos das Espécies Vasculares, 

vol. 1. Rio de Janeiro: Jardim Botânico do Rio de Janeiro. 

Macedo, G. A. R., M. B. Ferreira, and C. J. Escuder. 1978. Dieta de Novilhos 

em Pastagem Nativa de Cerrado. Belo Horizonte: Empresa de Pesquisa 

Agropecuária de Minas Gerais. 

Mantovani, W. 1990. Variação da flora arbustivo-herbácea de diversas 

fisionomias do cerrado de Itirapina, estado de São Paulo. Anais do 36º 

congresso nacional de botânica, pp. 125–135. Brasília: Sociedade 

Botânica do Brasil and IBAMA. 

Mantovani, W. and F. R. Martins. 1993. Florística do cerrado na Reserva 

biológica de Moji Guaçu, SP. Acta Bot. Brasil. 7:33–60. 

Mattos, J. K. de A. 1996. Plantas Medicinais: Aspectos Agronômicos. 

Brasília: Edição do Autor. 

Medina, E. 1987. Requirements, conservation and cycles of nutrients in the 

herbaceous layer. In B. H. Walter, ed., Determinants of Tropical Savannas, 

pp. 39–65. Paris: IUBS Special Issue no. 3. 

Mendonça, R. C., Felfili, J. M.., B. M. T. Walter, M. C. da Silva Jr., 

A . V. Rezende, T. S. Filgueiras and P. E. Nogueira. 1998. Flora vascular 

do Cerrado. In Sano, S. M. and S. P. de Almeira, eds., Cerrado: Ambiente 

e Flora, pp. 289–556. Planaltina: Empresa Brasileira de Pesquisa 

Agropecuária. 

Miranda, I. S. 1998. “Flora, fisionomia e estrutura das savanas de Roraima, 

Brasil.” Ph.D. thesis, Instituto Nacional de Pesquisas da Amazônia, Manaus, 

Brasil. 

Miranda, I. S. and M. L. Absy. 1997. A flora fanerogâmica das savanas de 

Roraima. In R. I.. Barbosa, E. J. G. Ferreira and E. G. Castellón, eds., 

Homem, Ambiente e Ecologia no Estado de Roraima, pp. 445–462. 

Manaus: Instituto Nacional de Pesquisas da Amazônia. 

Munhoz, C. B. and C. E. B. Proença. 1998. Composição florística do município 

de Alto Paraíso na Chapada dos Veadeiros. Bol. Herb. Ezechias 

Paulo Heringer 3:102–150. 

Myers, N. 1983. A Wealth of Wild Species. Colorado: Westview Press. 

Paviani, T. I. 1987. Anatomia do desenvolvimento do xilopódio de Brasilia 

sickii G. M.Barroso. Estágio inicial. Ciênc. Cult. 39:399–405. 

Pereira, B. A. S. 1984. Plantas nativas do cerrado pastadas por bovinos na 

região geoeconômica do Distrito Federal. Rev. Bras. Geogr. 46:381–388.



Pereira, B. A. S. 1990. Espécies apícolas da flora da Área de Proteção Ambiental 

(APA) da bacia do Rio São Bartolomeu, Distrito Federal: Estudo 

preliminar. Cad. Geoc. 5:7–19. 

Pereira, B. A. S. 1992. A flora nativa. In B. F. S. Dias, ed., Alternativas de 

Desenvolvimento dos Cerrados: Manejo dos Recursos Naturais Renováveis, 

pp. 53–57. Brasília: Fundação Pró-Natureza. 

Pott, A. 1988. Pastagens no Pantanal. Corumbá: Empresa Brasileira de 


Pott, V. J. 1998. A família Nymphaeaceae no Pantanal, Mato Grosso e Mato 

Grosso do Sul. Acta Bot. Brasil. 12:183–194. 

Ratter, J. A. 1987. Notes on the vegetation of the Parque Nacional do 

Araguaia (Brazil). Notes Roy. Bot. Gard. Edinb. 44:311–342. 



Raven, P. H. 1990. The politics of preserving biodiversity. BioScience 40:769. 

Rawitscher, F. and M. Rachid. 1946. Troncos subterrâneos de plantas 

brasileiras. An. Acad. Bras. Ciênc. 18:262–280. 

Rizzini, C. T. and E. P. Heringer. 1966. Underground organs of trees and 

shrubs from some Brazilian savannas. An. Acad. Bras. Ciênc. 38:85–112. 

Schultze-Kraft, R. 1980. Recolección de plantas nativas con potencial forrajero. 

In Empresa Brasileira de Pesquisa Agropecuária, ed., Plantas Forrageiras, 

pp. 61–72. Brasília: Empresa Brasileira de Pesquisa Agropecuária. 

Silva, D. A. da and C. A. Klink. 1996. Crescimento e alocação de biomassa 

de duas gramíneas nativas do cerrado. In L. L. Leite and C. H. Saito, eds., 

Contribuição ao Conhecimento Ecológico do Cerrado, pp. 59–63. 

Brasília: Universidade de Brasília. 

Siqueira, J. C. 1988. Plantas Medicinais: Identificação e Uso das Espécies dos 

Cerrados. São Paulo: Edições Loyola. 

Taylor, N. P. and D. C. Zappi. 1995. Cactaceae. In B. L. Stannard, ed., Flora 

of the Pico das Almas, Chapada Diamantina, Bahia, Brazil, pp.157–164. 

Kew: Royal Botanic Gardens. 

Valls, J. F. M. and L. Coradin. 1986. Recuros genéticos de plantas forrageiras 

nativas no Brasil. Anais do 3º Simpósio Sobre Produção Animal, pp. 

19–34. Campo Grande: Fundação Cargill. 

Verdesio, J. J. and F. D. Garra. 1987. Caracterização Física dos Cerrados do 

Oeste da Bahia. Brasília: Instituto Interamericano de Cooperação para a 

Agricultura. 

Warming, E. 1973. Lagoa Santa: Contribuição para a Geographia Phytobiologica 

(reprint of the 1908 Portuguese translation of the work). In M. G. 

Ferri, ed., Lagoa Santa e a Vegetação de Cerrados Brasileiros, pp. 1–284. 

Belo Horizonte: Livraria Itatiaia Editora and Editora da Universidade de 

São Paulo. 

Wilson, E. O. 1992. The Diversity of Life. Cambridge, MA: Belknap Press, 

Harvard.


8 

Patterns and Dynamics 

of Plant Populations 

Raimundo P. B. Henriques and John D. Hay 

The understanding of the dynamics of plant communities 

is based on measurements of rates of mortality and recruitment. These 

population parameters are essential to the study and prediction of 

responses of vegetation to global changes (Phillips and Gentry 1994; 

Pimm and Sugden 1994) and short-term climatic change (Condit et al. 

1992), as well as patterns in species richness (Phillips et al. 1994). The 

issue takes on special significance because of its implications for community 

conservation and management (Primack and Hall 1992). 

Long-term monitoring of plant populations has been undertaken in 

various tropical forests around the world (Swaine, Lieberman, and Putz 

1987; Hartshorn 1990; Phillips and Gentry 1994; Phillips et al. 1994). 

For savannas, the dynamics of plant communities protected from fire have 

been monitored, and data presented (Brookman-Amissah et al. 1980; San 

José and Fariñas 1983; Devineau et al. 1984; Bowman and Panton 1995). 

In contrast, although the Cerrado Biome is the second largest in South 

America, there are very few long-term studies using permanent plots to 

observe recruitment, growth, and mortality of woody plants in this vegetation 

type. Silberbauer-Gottsberger and Eiten (1987) recorded three 

years of cerrado change in ten quadrats of 100 m 2 each. 

Currently the Cerrado Biome is suffering strong anthropogenic 

change associated with deforestation, high fire incidence, and invasion by 

alien species (chapter 5). Even protected reserves in central Brazil have 

only a few dozen hectares that have been free from disturbance of this 

kind for more than 20 years. Studies of parameters such as annual mortality 

and recruitment rates in areas protected from human impact are 

needed to determine their natural dynamics. 

140


Patterns and Dynamics of Plant Populations 141 

THE CERRADO BIOME IN THE CONTEXT 

OF SAVANNA DYNAMICS 

The pattern of dynamic processes that emerges in savannas appears to 

depend on four site factors: water, nutrients, herbivory, and fire. Using 

water and nutrients and, to some extent, herbivory, savannas were classified 

in four types (Frost et al. 1986): (1) low water availability and high 

nutrient supply; (2) high water availability with extremely nutrient poor 

soil; (3) limited in both water and nutrients; and (4) no limitations of 

water and nutrients. Most cerrado vegetation can be included in type (2), 

however some are of type (3). 

Fire is a widespread disturbance factor in savannas, particularly in the 

cerrados (chapter 4). The response of vegetation to fire depends largely 

on season, fire temperature, fuel biomass, and subsequent events such as 

rainfall, drought, and herbivory. Several studies indicate that fire reduces 

the recruitment of seedlings, saplings, and small trees, leading to an interruption 

or stabilization of vegetation development (Walker 1981; Frost 

and Robertson 1986; Frost et al. 1987). Hoffmann (1996) showed that, 

even two years after burning, seedling establishment in cerrado sensu 

stricto was lower in a burned area than in a control area (see chapter 9). 

The influence of fire suppression on cerrado vegetation, and subsequent 

effects on invasion, recruitment, and structural stability, are fundamental 

in determining whether cerrado is a self-maintaining community. 

In the 19th century the naturalist Lund (1843) proposed that the campos 

and cerrado sensu stricto were created by the frequent burning of an original 

forest that he recognized as the cerradão (see chapter 6). This idea 

was reinforced by the floristic analysis of the woody flora of the cerrados 

(Rizzini 1963). Based on several lines of evidence, Rizzini (1979) agreed 

with Lund and stated that in areas without edaphic limitations the cerrado 

sensu stricto arose from frequent burning of cerradão. Several other 

authors recognize the cerradão as the original (or climax) vegetation of 

the Cerrado Biome (Ab’Saber and Junior 1951; Aubréville 1959; Schnell 

1961; Eiten 1972). 

A similar pattern of origin for the African savannas was proposed by 

Aubréville (1949) who worked in Africa prior to studying the Brazilian 

cerrados. African savannas, once established, are often maintained by frequent 

fires, both natural and man-made. Several studies showing the origin 

of savannas by burning of forests in Africa and India are reviewed by 

Backéus (1992). 

The complete protection of savanna vegetation from fire results in an 

increase in tree density and species richness, particularly of fire-sensitive



species, evolving toward a woodland community (Brookman-Amissah et 

al. 1980, Lacey et al. 1982, San José and Fariñas 1983, Swaine et al. 1992, 

Devineau et al. 1984, Bowman and Panton 1995). 

In fire-protected cerrado areas, observations indicate that the vegetation 

changes, both in species composition and tree density, toward a 

closed woodland (Coutinho 1982, 1990; Moreira 1992). Simulations 

using data for five cerrado plant species (Hoffmann 1999) suggest that if 

current fire frequencies were lowered, woody plant cover and tree density 

would increase (chapters 4, 9). Unfortunately there are few existing experiments 

of fire suppression in savannas, and to our knowledge none for the 

cerrado, which would permit a rigorous evaluation of this hypothesis. 

The nature of savanna structure and dynamics reflects the set of 

species attributes (Noble and Slayter 1980). Many studies have reported 

the widespread occurrence of vegetative reproduction in savanna woody 

species (Medina 1982; Menaut 1983). In the Cerrado Biome many woody 

species are known to reproduce vegetatively via root suckers and rhizomes 

(Rawitscher 1948), and some studies attribute great importance to vegetative 

growth (Rizzini 1971). However, these early studies are limited by 

a lack of quantitative data with which to assess the prevalence of vegetative 

reproduction at the community level. Henriques (1993) showed that 

vegetative reproduction in cerrado sensu stricto occurs in approximately 

60% of the woody species. The role of vegetative growth as a stabilizing 

element has been observed in frequently burned savannas (Lacey et al. 

1982). Hoffmann (1998) showed that fire also tends to increase the 

importance of vegetative versus sexual reproduction, and that, under current 

fire regimes, fire may be causing a shift in species composition, favoring 

fire resistant species. 

Questions 

This chapter deals with succession and stand stability in a 0.5 ha plot of 

cerrado sensu stricto that has been protected from fire since 1973. First 

we address the following questions about structure and composition: (1) 

What are the floristic characteristics of a cerrado stand? (2) What is the 

size class and abundance distribution of species? Second, we summarize 

questions concerning floristic changes, mortality, recruitment, growth, 

and population dynamics: (1) Is species richness balanced by immigration 

and extinction? (2) Are stand structural components (number of individuals, 

number of stems and basal area) increasing with time? (3) Does mortality 

match recruitment for various species? (4) What is the pattern of 

growth rate? (5) What are the rates of change of populations? Finally, we



address questions about cerrado stability and succession: (1) Is the cerrado 

(sensu stricto) in an unstable equilibrium state? (2) Is the cerrado a 

self-maintained community? 

THE CERRADO SENSU STRICTO 

AS A NEOTROPICAL SAVANNA CASE STUDY 

Study Area 

The sample plot was located in the Ecological Reserve of the IBGE (Instituto 

Brasileiro de Geografia e Estatística). The reserve (15°57'S, 47°53'W) 

is located 16 km SW of the center of Brasília and occupies 1,350 ha of 

protected area. Altitude ranges between 1,045 m and 1,125 m, with an 

average slope of 6%. Soils are mostly Oxisols (U.S. Soil Taxonomy 

System). Chemically, the soils are acid (pH ≈ 5.4), with low available 

P levels (90% falling from October 

to April. The dry season occurs from May to September, when 

monthly precipitation can drop to 0 mm. The temperature regime is typical 

of a continental subtropical climate, with an annual mean of 21.9°C 

over the same 17-year period. 

Methods 

The plot was established within a cerrado area that has been protected 

from fire since 1973. The stand was sampled with a transect of 20 m × 

250 m (0.5 ha), divided into 10 × 10 m subplots. The first census of the 

plot was conducted in January 1989. Within each subplot all plants ≥ 15 

cm circumference (4.8 cm diameter) at ground level were measured and 

identified by species. Plants with multiple stems resulting from vegetative 

growth were included in the census if the sum of their circumferences was 

≥ 15 cm. Each plant was labeled with numbered aluminum tag and 

mapped. Since many cerrado plants exhibit vegetative reproduction, the 

identification of separate individuals was made by excavations to 10 cm 

depth, between all conspecific stems separated by less than 1 m. Excavations 

were made at the mid-point of the shortest distance between the center 

of neighboring stems, and the exposed soil profile was examined for



connections. After inspection the soil was replaced into the hole. Care was 

taken not to harm the subterranean organs of the plants, and no mortality 

was observed. This excavation procedure facilitated a separate calculation 

of the number of individuals and the number of stems. These values 

should be viewed with caution, however, since they do not include individuals 

with connections below 10 cm depth or whose connections had 

become separated. 

The species were classified into the following growth forms: tree (≥10 

cm diameter, and ≥3 m height); thick shrub (≥10 cm diameter, and < 3 m 

height); thin shrub (



species), followed by thick shrubs (7 species), thin shrubs (6 species), 

palms (4 species), and one treelike rosette, Vellozia squamata (Velloziaceae). 

The number of stems drops rapidly with increasing diameter class, 

and the number of trees with a diameter greater than 10 cm is remarkably 

low when compared with tropical forest (Manokaran and Kochummen 

1987; Swaine, Hall, and Alexander 1987; Primack and Hall 1992). The 

inverted-J shape of the frequency distribution is typical of natural forest 

regenerating from seeds, with high numbers in the smaller size classes and 

a logarithmic decline with increasing size. 

CERRADO DYNAMICS 

Floristic Changes 

The number of families, genera, and species in the plot increased from 

1989 to 1991. This corresponded to two new families, five new genera, 

and eight new species. No species disappeared during this period. The 

immigration rate of new species into the plot was 6.6 spp. ha −1 yr −1 . In 

another cerrado area the calculated immigration rate during a 3-year 

period was 3.3 spp. ha −1 yr −1 , with no losses of species recorded (Silberbauer-Gottsberger 

and Eiten 1987). These values are greater than the 

immigration rate of species for the Llaños savanna protected from fire in 

central plains of Venezuela (0.23 spp. ha −1 yr −1 ), calculated from 16 years 

of records (San José and Farina 1983), but less than the range of 

15.3–19.3 spp. ha −1 yr −1 for the Lamto savanna in Africa, also protected 

16 years from fire (Devineau et al. 1984). Immigration rate curves for 

savannas protected from fire show a faster increase initially followed by 

a diminution (Braithwaite 1996; Devineau et al. 1984). The unbalanced 

and high immigration rates in our study area and the Lamto savanna indicate 

that these areas are in the early stages of succession. 

Stand Dynamics 

There was a large structural change in the cerrado plot over time (see table 

8.1). In 1991 the density, number of stems, and basal area rose to 1,253 

individuals (annual net increase of +11.6%), 1,534 stems (+13.5%), and 

73,068 cm 2 (+17.1%), respectively. This high net increase corresponded 

to a low turnover of plants in the plot, since in all growth forms the high 

recruitment was not balanced by mortality. There were 304 new individuals 

recruited (+31.0%) and only 31 deaths (3.2%) over the period. There


Figure 8.1 Species abundance and diameter distribution for all woody 

plants ≥ 4.8 cm in diameter in a sample plot of cerrado sensu stricto in the 

Ecological Reserve of IBGE, central Brazil. Top: Species abundance curve; 

species ranked in descending density. Bottom: Diameter class distribution for 

all woody plants ≥ 4.8 cm in diameter in a 0.5-ha plot in 1989. Inset: the same 

data plotted on log scale.



was a total of 420 new stems (36.3%), and the increase in basal area due 

to recruitment and growth was 22,456 cm 2 (43.3%). Losses due to mortality 

were 44 stems and 1,194 cm 2 of basal area. 

The changes in growth form distribution showed that thick shrubs 

had the largest annual net increase in density (+18.9%); palms the largest 

increase in number of stems (+29.3%) and basal area (+63.6%). Since 

palm species do not have true secondary growth, the high net increase in 

basal area was due to the high vegetative reproduction of stipes. 

The distribution of individuals among growth forms was not significantly 

different between 1989 and 1991 (χ 2 = 5.98, P > 0.05), but the proportion 

of number of stems and basal area differed significantly (χ 2 = 

15.59, P < 0.005; χ 2 = 6,850.91, P < .0001, respectively). The significantly 

higher number of stipes and basal area of palms and the treelike rosette 

plant in 1991 accounted for this difference. 

The proportion of stems originating from vegetative reproduction 

was 18.2% in 1989, and the value increased to 22.4% in 1991 (χ 2 = 28.50, 

P < .005). This increase was due both to vegetative reproduction of existing 

individuals and to new recruits. The proportion of vegetatively reproduced 

stems in other cerrado areas was 9.9% and 11.5% for southeastern 

and central Brazil, respectively (Silberbauer-Gottsberger and Eiten 1987, 

Henriques 1993). 

Mortality, Recruitment, and Growth 

Deaths were recorded in 12 of the 48 species present in 1989. The number 

of individuals that died over the study period was 31, corresponding to an 

annual mortality rate of 1.34% yr −1 . The losses by deaths of number of 

stems and basal area were 1.61% yr −1 and 0.97% yr −1 , respectively (table 

8.1). The mortality rate calculated for another cerrado area by Silberbauer- 

Gottsberger and Eiten (1987) was 5.50% yr −1 . These values are in the middle 

to maximum range of the annual mortality rate recorded in studies 

of tropical forests (Swaine, Lieberman, and Putz 1987; Hartshorn 1990; 

Phillips and Gentry 1994; Phillips et al. 1994). Sato and Miranda (1996) 

showed that after fire the mortality rate of cerrado plants > 5 cm in diameter 

increases strongly to values ranging between 6.4% to 13.0% yr −1 . 

The annual mortality rate for the various size classes were: 4.9–7.5 

cm (1.61% yr −1 ); 7.6–10.2 cm (1.19% yr −1 ); >10.3 cm (0.79% yr −1 ). 

These results suggest that the proportion of plants dying diminished with 

increasing size class, but the sample size is small for a definite conclusion. 

The mortality of individuals was highest for thick shrubs and the treelike 

rosette plant (table 8.1) and lowest for trees. This pattern of mortality



rate was observed for both number of stems and basal area. The mortality 

rate differed markedly among species; the highest rate was shown by 

Byrsonima coccolobifolia (24.9% yr −1 ). A group of three species, 

comprising Kielmeyera coriacea (Guttiferae), Connarus suberosus (Connaraceae), 

and Mimosa claussenii (Leguminosae), showed an intermediate 

mortality rate, between 1.8% and 7.9% per year. The lowest mortality 

rate (0.5% to 0.9% per year) was shown by a group of five species, including 

Eremanthus goyazensis (Compositae), Qualea parviflora (Vochysiaceae), 

and Palicourea rigida (Rubiaceae). The only other data available 

for a cerrado tree species, Vochysia thrysoidea, showed an average mortality 

of 1.9% per year, but with temporal and spatial variation (Hay and 

Barreto 1988). 

Of the 48 species marked in 1989, 30 (62.5%) recruited new individuals, 

and seven (14.6%) produced new stems by vegetative reproduction. 

By growth form, the proportion of species with recruitment 

increased in the following order: thin shrubs (50.0%), trees (53.3%), 

thick shrubs (71.4%), and palms (100%). More than half of the recruits 

(226, 58.3%) occurred in the 4.8–7.2-cm-diameter class. Another, larger 

number occurred with less than a 4.8-cm diameter (150, 38.8%), because 

we included as recruits connected stems with less than a 4.8-cm diameter 

(see Methods). A few stems (7, 2.9%) were recruited into a larger-diameter 

class (7.2–9.5 cm). 

The annual recruitment rates calculated with the logarithmic model 

for number of individuals (11.57% per year), number of stems (13.33% 

per year), and basal area (15.30% per year) are much higher than for any 

known mature tropical forest (Phillips and Gentry 1994; Phillips et al. 

1994). By growth forms the recruitment rates increased in the following 

order: treelike rosette plant, thin shrubs, palms, trees, and thick shrubs. 

This order was consistent for number of individuals, number of stems, 

and basal area. 

The most abundant species recruited more than the rare species. All 

species with more than 42 individuals present in 1989 (nine species) had 

new recruits, while for those with less than 42 individuals, 46.2% (18 

species) failed to recruit. We calculated the per capita recruitment rates for 

each species as the ratio of new recruits to population size in 1989. Species 

with less than 42 individuals in 1989 averaged only 53% per capita recruitment, 

while species with more than 42 individuals averaged 76%. 

The mean diameter growth rate, pooled for all individuals and 

species, but excluding palms and the treelike rosette, was quite low at 1.59 

mm per year (SD = 2.43, N = 791). Five percent of all individuals showed



apparent negative growth rate in diameter, probably due to natural bark 

losses or measurement errors. Altogether 41% did not show growth; 51% 

had a growth rate between 1–5 mm per year; and 3% grew between 5 and 

33 mm. Silberbauer-Gottsberger and Eiten (1987) found a mean growth 

rate of 2.7 mm per year for another cerrado. In their plots 24% of the 

individuals did not increase in circumference, 52% of individuals 

increased by 1–3 mm, and 24% above 4 mm per year. 

These results for cerrado plants are close to the minimum growth 

rates reported for tropical forests (Lang and Knight 1983, Primack et al. 

1985, Manokaran and Kochummen 1987, Korning and Balslev 1994) 

and gallery forests in central Brazil (Felfili 1995). The low growth rates 

observed are consistent with the poor nutrient content of cerrado soils 

(chapter 2) and a long dry season (4–5 months). 

Mean diameter increments, pooled for all individuals and species, 

increased gradually with diameter. We found a significant linear relationship 

between individual diameter (x) and mean growth rate (y): y = −0.97 

+ 0.30x (r 2 = 0.65, P < .0005). Similar relationships were observed in tropical 

forest communities (Swaine et al. 1987; Condit et al. 1992; Felfili 

1995). We hypothesize that plants that achieve larger size grow faster due 

to deep roots, which provide greater access to water. 

Species Population Dynamics 

For the whole cerrado community recruitment and mortality were unequal; 

this was also observed for individuals at the population level (see 

fig. 8.2). Only two species, Salacia crassifolia and Eremanthus glomerulatus, 

showed a net change close to the expected line of zero net change, 

as would be expected for balanced populations. The other 10 species 

recruited more individuals than were lost by deaths. Eremanthus goyazensis 

showed the highest population net increase with 33 individuals, 

followed by Roupala montana (21 individuals) and Kielmeyera coriacea 

(15 individuals). Particularly interesting was the persistence of rare 

species, suggesting that the increase in abundant species does not result in 

the loss of rare species in this community. 

The species studied showed a strong net increase in density (see fig. 

8.3). Most species (54%) increased in density by ≤ 2 individuals, 33% 

increased by 3–15 individuals, and 13% by 16–34 individuals. No species 

declined during the study period. In terms of growth form, the mean net 

increase in population density was highest in trees (24.6 ± 6.1 individuals), 

as compared with thick (11.7 ± 4.2) and thin shrubs (6.9 ± 2.3).



Figure 8.2 Recruitment and mortality for 12 plant species in a cerrado 

sensu stricto at the Ecological Reserve of IBGE, central Brazil, between 1989 

and 1991. The diagonal line represents equal mortality and recruitment. The 

species are: Bco, Byrsonima coccolobifolia; Bcr, Byrsonima crassa; Csu, Connarus 

suberosus; Egl, Eremanthus glomerulatus; Ego, Eremanthus goyazensis; 

Kco, Kielmeyera coriacea; Mcl, Mimosa claussenii; Pri, Palicourea rigida; 

Qpa, Qualea parviflora; Rmo, Roupala montana; Scr, Salacia crassifolia; and 

Vsq, Vellozia squamata. 

STABILITY AND DEVELOPMENT OF CERRADO 

Stability and Turnover 

Stability could indicate the state of vegetation changes in the absolute 

number of individuals and basal area over time (Korning and Baslev 

1994). Using the terminology of Hallé et al. (1978), after a disturbance, 

the successional sequence of changing composition and structure begins 

with a growing phase with a net increase in the number of individuals and


Figure 8.3 Distribution of net increase in density for plant species and stability 

and turnover for cerrado sensu stricto compared with other communities. 

Top: Distribution of net increase in number of individuals for 48 species. 

Bottom: Stability and turnover for cerrado (black dot), Lamto savanna (black 

triangle), and 13 neotropical forests (white dots). Stability is measured as the 

numerical difference between stand half-life and doubling time. Turnover is 

measured as the average of doubling time and stand half-life. Source: Phillips 

and Gentry 1994 (neotropical forests); Devineau et al 1984 (Lamto savanna).



basal area. This phase is followed by a homeostatic phase with accumulation 

of basal area due to growth, but in which mortality and recruitment 

are balanced. Another aspect of vegetation dynamics is turnover, which 

indicates the changes of composition or structure with time (i.e., the time 

necessary to replace lost stems and basal area). Turnover is measured by 

averaging rates of recruitment and mortality (Phillips and Gentry 1994: 

Phillips et al. 1994), calculating the stand half-life, which is the number of 

years necessary for the initial population to lose 50% (Lieberman et al. 

1985; Lieberman and Lieberman 1987; Hartshorn 1990). In this study we 

measured both stability and turnover using the Korning and Baslev (1994) 

approach. Stability was estimated as the numerical difference between 

stand half-life and doubling time. Doubling time is the time needed to double 

the initial population at the present recruitment rate, using a log model. 

Turnover was estimated as the average of doubling time and stand half-life. 

Since low numerical values indicate high stability and turnover, we subtracted 

the numerical values from 100 to facilitate graphical interpretation. 

A comparison of our data with those from neotropical forests 

(Phillips and Gentry 1994) and a savanna plot protected from fire in the 

Ivory Coast (Devineau et al. 1984) showed that the cerrado and savanna 

plots are moderately dynamic but in an unstable growing phase, with a 

net increase in the number of individuals and basal area (fig. 8.3). 

Other results point out the instability of cerrado: (1) the diameter distribution 

(fig. 8.1) suggests a community with an abrupt decline in size 

above 10 cm diameter and a decreasing mortality rate with increasing size 

classes; (2) the species immigration rate was high, and no extinction was 

recorded for individuals ≥ 4.8 cm diameter; (3) the balance between 

recruitment and mortality over the study period showed a high positive 

net change (table 8.1). 

The Cerrado as a Nonequilibrium Community 

Here we present one of the first studies of cerrado dynamics. Data on 

floristic changes, recruitment, mortality, and growth for 48 plant species 

were recorded for a 2.4-year period. We are still cautious with these 

results, because both the period of study and the plot size are in the minimum 

range for similar studies in tropical forests (Swaine, Lieberman and 

Putz 1987; Hartshorn 1990; Phillips and Gentry 1994; Phillips et al. 

1994). Compared with tropical forests, studies of cerrado vegetation 

dynamics are still in their infancy. Nevertheless, the data are sufficient to 

evaluate the state of stand dynamic parameters by growth forms as well 

as to suggest differences between species. In the light of these results we



Table 8.1 Changes in Density, Number of Stems, and Basal Area (cm 2 ) by 

Growth Form Over Census Interval of 1989-1991 in 0.5 ha of Cerrado Sensu 

Stricto, in the Ecological Reserve of IBGE, Brasília, Central Brazil 

Annual 

net Annual Annual 

Deaths Recruits change mortality recruitment 

1989 1989–91 1989–91 Growth 1991 (%) %/yr %/yr 

Density 

Trees 598 13 156 741 59.6 (+9.9) 0.92 9.84 

Thick shrub 202 17 109 294 38.3 (+18.9) 3.66 19.30 

Thin shrub 29 0 6 35 2.5 (+8.6) 0.00 7.84 

Palm 72 0 17 89 7.1 (+9.8) 0.00 8.83 

Treelike rosette 79 1 16 94 6.3 (+7.9) 0.53 7.77 

plant 

Total 980 31 304 1253 113.8 (+11.6) 1.34 11.57 

Stems 

Trees 671 15 167 6 829 65.8 (+9.8) 0.94 9.75 

Thick shrub 234 22 113 3 328 39.2 (+16.7) 4.11 18.18 

Thin shrub 30 0 8 0 38 3.3 (+11.1) 0.00 9.84 

Palm 94 0 62 2 158 26.6 (+28.4) 0.00 21.64 

Treelike rosette 129 7 37 22 181 21.7 (+16.8) 2.32 16.43 

plant 

Total 1158 44 387 33 1534 156.7 (+13.5) 1.61 13.33 

Basal area 

Trees 34,470 651 3570 6396 43,785 3881 (+11.3) 0.79 10.76 

Thick shrub 6292 523 2409 288 8466 906 (+14.4) 3.62 15.98 

Thin shrub 682 0 120 122 924 101 (+14.8) 0.00 12.65 

Palm 6060 0 604 8649 15313 3855 (+63.6) 0.00 38.62 

Treelike 4302 20 458 -160 4580 116 ( +2.7) 0.19 2.80 

rosette 

plant 

Total 51,806 1194 7161 15,295 73,068 8859 (+17.1) 0.97 15.30 

Note: Annual net change was computed as follows: [(density, stems or basal area in 1991 – deaths 

+ recruits + growth) – (density, stems or basal area in 1989)]/2.4 years. Annual mortality was 

estimated as log e survivorship vs. time (Swaine and Lieberman 1987). Annual recruitment was 

estimated using a logarithmic model based on final recruitment and annual mortality (Phillips et 

al. 1994). See chapter 6 for descriptions of cerrado physiognomies. 

evaluate the idea concerning cerrado evolution under fire suppression and 

reappraise the hypothesis of Lund-Rizzini about the origin of the cerrado 

physiognomy. 

This study found strong evidence that the cerrado sensu stricto (chapter 

6) has a clear successional nature, as suggested in earlier works 

(Coutinho 1982, 1990), corroborating the expected increase in cover and



density if fire frequency is diminished (Hoffmann 1999; chapter 9). Our 

results showed that the cerrado, when protected from fire, increases in 

richness with a net increase in the number of individuals, number of stems, 

and basal area. The high immigration rate, recruitment greater than mortality, 

and high positive net growth indicate that the cerrado is in a very 

early stage of succession, after 16 years of fire suppression. 

With fire suppression we expected an increase in fire sensitive species. 

Particularly noticeable in this respect is the immigration of the thick 

shrub Myrsine guianensis and the high population growth of the tree 

Roupala montana, both indicated by Hoffmann (1996, 1998) as showing 

positive population growth rates for intervals between fire of 5 and 9 

years, respectively. 

In an unburned cerrado there was a negative relationship between 

plant height and population growth rate (Hoffmann 1996). The high net 

increase in number of individuals and number of stems of trees and thick 

shrubs in the present study confirm this finding. The same trend was 

observed in the mean net population density, with values decreasing in the 

following order: trees, thick shrubs, and thin shrubs. 

The great similarity in the results of the present study with those 

observed in other savannas deserves future investigation, since there are 

so few comparative studies. Nevertheless, all the present evidence strongly 

suggests that the savannas have a successional nature and are not equilibrium 

communities, and that the present structure and physiognomy are 

maintained by fire frequency below the minimum interval to permit establishment 

of tree species populations. 

Based on the present study, we can only make an educated guess 

about the origin of the cerrado physiognomy. The extent of impact of fire 

in the cerrados is yet to be determined. Although there is some evidence 

that fire has long been present in the Cerrado Biome, the past fire frequency 

was well below that of the present (Vincentini 1999). The high 

mortality caused by fire in cerrado plants (Sato and Miranda 1996) and 

the high sensitivity of cerradão species to fire (Moreira 1992; chapter 6), 

as well as the clearly early successional state of the area studied, is consistent 

with the Lund-Rizzini hypothesis. The absence of immigration of 

cerradão species into the study plot should not be used to challenge the 

Lund-Rizzini hypothesis, because the cerradão areas that could serve as 

seed sources are small in total area (< 1%) and distant from the study site. 

Nevertheless there is strong evidence showing that cerradão species are 

capable of establishment in cerrado sensu stricto physiognomy, and that 

their survival is higher in areas which have been protected from fire for 

20 years compared to more recently burned areas (Hoffmann 1996).



Long-term studies are needed to corroborate this result and evaluate 

whether the savanna physiognomy results from repeated burning of the 

original forest cover. 

REFERENCES 

Ab’Saber, A.N. and M.C. Junior 1951. Contribuição ao estudo do Sudoeste 

Goiano. Bol. Geográfico 9:123–138. 

Aubréville, A. 1949. Ancienneté de la destruction de la couverture forestière 

primitive de l’Afrique tropicale. Bull. Agric. Congo Belg. 40:1347–1352. 

Aubréville, A. 1959. As florestas do Brasil. An. Brasil. Econ. Flor. 11:201–243. 

Backéus, I. 1992. Distribution and vegetation dynamics of humid savannas 

in Africa and Asia. J. Veg. Sci. 3:345–356. 

Bowman, D. M. J. and W. J. Panton. 1995. Munmalary revisited: Response 

of a north Australian Eucalyptus tetrodonta savanna protected from fire 

for 20 yr. Aust. J. Ecol. 20:526–531. 

Braithwaite, R.W. 1996. Biodiversity and fire in the savanna landscape. In O. 

T. Solbrig, E. Medina, and J. F. Silva, eds., Biodiversity and Savanna 

Ecosystem Processes, pp. 121–142. Berlin: Springer-Verlag. 

Brookman-Amissah, J., J. N. Hall, M. D. Swaine, and J. Y. Attakorah. 1980. 

A reassessment of a fire protection experiment in north-eastern Ghana 

savanna J. Appl. Ecol. 17:85–99. 

Condit, R., S. P. Hubbell, and R. B. Foster. 1992. Short-term dynamics of a 

neotropical forest. BioScience 42:822–828. 

Coutinho, L. M. 1982. Ecological effect of fire in Brazilian cerrado. In B. J. 

Huntley and B. H. Walker, eds., Ecology of Tropical Savannas, pp. 


Coutinho, L. M. 1990. Fire in the ecology of Brazilian cerrado. In J. G. 

Goldammer, ed., Fire in the Tropical Biota: Ecosystem Processes and 

Global Changes, pp. 82–105. Berlin: Springer-Verlag. 

Devineau, J. L., C. Lecordier, and R. Vuattoux. 1984. Evolution de la diversité 

esécifique du peuplement ligneux dans une succession préforestière 

de colonisation d’une savane protégée des feux (Lamto, Côte d’Ivoire). 

Candollea 39:103–134. 


Eiten, G. and R. H. R. Sambuichi. 1996. Effect of long-term periodic fire on 

plant diversity in a cerrado region. In R. C. Pereira and L. C. B. Nasser, 

eds., Biodiversity and Sustainable Production of Food and Fibers in the 

Tropical Savannas, pp. 46–55. Planaltina: Empresa Brasileira de Pesquisa 

Agropecuaria. 

Felfili, J. M. 1995. Growth, recruitment and mortality in the Gama gallery 

forest in Central Brazil over a six-year period (1985–1991). J. Trop. Ecol. 

11:67–83.



Frost, P., E. Medina, J. C. Menaut, O. Solbrig, M. Swift, and B. Walker. 1986. 

Responses of Savannas to Stress and Disturbance. IUBS Special Issue no. 

10:1–82. 

Frost, P. G. H. and F. Robertson. 1987. The ecological effects of fire in savannas. 

In B. Walker, ed., Determinants of Tropical Savannas, pp. 93–140. 

Paris: IUBS Special Issue no. 3. 

Hallé, F., R. A. A. Oldeman, and P. B. Tomlinson. 1978. Tropical Trees and 

Forests. Berlin: Springer-Verlag. 

Hartshorn, G. 1990. An overview of neotropical forest dynamics. In A. H. 

Gentry, ed., Four Neotropical Rainforests, pp. 585–199. New Haven: 

Yale University Press. 

Hay, J. D. and E. M. J. Barreto. 1988. Natural mortality of Vochysia thrysoidea 

in an unburnt cerrado ecosystem near Brasília. Biotropica 20:274–279. 

Henriques, R. P. B. 1993. “Organização e Estrutura das Comunidades Vegetais 

de Cerrado em um Gradiente Topográfico no Brasil Central.” Ph.D. 

thesis, Universidade Estadual de Campinas, Campinas, Brazil. 

Hoffmann, W. A. 1996. The effects of fire and cover on seedling establishment 

in a neotropical savanna. J. Ecol. 84:383–393. 

Hoffmann, W. A. 1998. Post-burn reproduction of woody plants in a neotropical 


J. Appl. Ecol. 35:422–433. 

Hoffmann, W. A. 1999. Fire and population dynamics of woody plants in a 

neotropical savanna: Matrix model projections. Ecology 80:1354–1369. 

Huntley, B. J. 1982. Southern African savannas. In B. J. Huntley and B. H. 

Walker, eds., Ecology of Tropical Savannas, pp. 101–119. Berlin: 


Korning, J. and H. Baslev. 1994. Growth and mortality of trees in Amazonian 

tropical rain forest in Ecuador. J. Veg. Sci. 4:77–86. 

Lacey, C. J., J. Walker, and I. R. Noble. 1982. Fire in Australian tropical 

savannas. In B. J. Huntley and B. H. Walker, eds., Ecology of Tropical 

Savannas, pp. 246–272. Berlin: Springer-Verlag. 

Lang, G. E. and D. H. Knight. 1983. Tree growth, mortality, recruitment, and 

canopy gap formation during a 10-year period in a tropical moist forest. 


Lieberman, D. and M. Lieberman. 1987. Forest tree growth and dynamics at 

La Selva, Costa Rica (1969–1982). J. Trop. Ecol. 3:347–358. 

Lieberman, D., M. Lieberman, R. Peralta, and G. Hartshorn. 1985. Mortality 

patterns and stand turnover rates in a wet tropical forest in Costa 

Rica. J. Ecol. 73:915–924. 

Lund, P. W. 1843. Blik paa Brasilien dryeverden. Selsk Skrifter 11:1–82. 

Manokaran, N. and K. M. Kochummen. 1987. Recruitment, growth and 

mortality of tree species in a lowland dipterocarp forest in Peninsular 

Malaysian. J. Trop. Ecol. 3:315–330. 

Medina, E. 1982. Physiological ecology of neotropical savanna plants. In B.



J. Huntley and B. H. Walker, eds., Ecology of Tropical Savannas, pp. 


Menaut, J. C. 1983. The vegetation of African savannas. In F. Bourlière, ed., 

Tropical Savannas, pp. 109–150. Amsterdam: Elsevier Scientific Publishing 

Company. 

Moreira, A. G. 1992. “Fire Protection and Vegetation Dynamics in the Brazilian 

Cerrado.’’ Ph.D. thesis, Harvard University, Cambridge, MA, U.S.A. 

Noble, I. R. and R. O. Slayter. 1980. The use of vital attributes to predicts 

successional changes in plant communities subject to recurrent disturbance. 

Vegetatio 43:5–21. 

Pereira, B. A. da S., M. A. da Silva, and R. C. de Mendonça. 1993. Reserva 

ecológica do IBGE, Brasília, (DF): Lista das plantas vasculares. Rio de 

Janeiro: Instituto Brasileiro de Geografia e Estatística. 

Phillips, O. L. and A. H. Gentry. 1994. Increasing turnover through time in 

tropical forests. Science 263:954–958. 

Phillips, O. L., A. H. Gentry., S. A. Sawyer, and R. Vásquez. 1994. Dynamics 

and species richness of tropical rain forest. Proc. Natl. Acad. Sci. USA 

9:2805–2809. 

Pimm, L. S. and A. M. Sugden. 1994. Tropical diversity and global change. 

Science 263:933–934. 

Primack, R. B., P. S. Ashton., P. Chai, and H. S. Lee. 1985. Growth rates and 

population structure of Moraceae trees in Sarawak, East Malaysia. Ecology 

66:577–588. 

Primack, R. B. and P. Hall. 1992. Biodiversity and forest change in Malaysian 

Borneo. BioScience 42:829–837. 

Rawitscher, F. K. 1948. The water economy of the vegetation of campos cerrados 

in southern Brazil. J. Ecol. 36:237–268. 

Rizzini, C. T. 1963. A flora do cerrado. In M. G. Ferri, ed., Simpósio sobre o 

Cerrado, pp. 126–177. São Paulo: Editora da Universidade de São Paulo. 

Rizzini, C. T. 1971. Aspéctos ecológicos da regeneração em algumas plantas 

do cerrado. In M. G. Ferri, ed., III Simpósio sobre o Cerrado, pp. 61–64. 

São Paulo: Editora Edgard Blücher. 

Rizzini, C. T. 1979. Tratado de Fitogeografia do Brasil: Aspectos Sociológicos 

e Florísticos. São Paulo: Hucitec Ltda. 

San José, J. J. and M. R. Fariñas. 1983. Changes in tree density and species 

composition in a protected Trachypogon savanna, Venezuela. Ecology 

64:447–453. 

Sato, M. N. and H. S. Miranda. 1996. Mortalidade de plantas lenhosas do 

cerrado após duas queimadas prescritas. In R. C. Pereira and L. C. B. 

Nasser, eds., Biodiversidade e Produção Sustentável de Alimentos e 

Fibras nos Cerrados, pp. 204–207. Planaltina: Empresa Brasileira de 


Schnell, R. 1961. Le problème des homologies phytogéographiques entre 

l’Afrique et l’Amérique tropicales. Mem. Mus. Hist. Nat., Bot. 11:137–241.



Silberbauer-Gottsberger, I. and G. Eiten. 1987. A hectare of cerrado: I. General 

aspects of the trees and thick-stemmed shrubs. Phyton 27:55–91. 

Silberbauer-Gottsberger, I. and G. Gottsberger. 1984. Cerrado-Cerradão: A 

comparison with respect to number of species and growth forms. Phytocoenologia 

12:293–303. 

Swaine, M. D., J. B. Hall, and I. J. Alexander. 1987. Tree population dynamics 

at Kade, Ghana (1968–1982). J. Trop. Ecol. 3:331–346. 

Swaine, M. D., W. D. Hawthorne, and T. K. Orgle. 1992. The effects of fire 

exclusion on savanna vegetation at Kpong, Ghana. Biotropica 24: 

106–172. 

Swaine, M. D. and D. Lieberman. 1987. Note on the calculation of mortality 

rates. J. Trop. Ecol. 3:ii–iii. 

Swaine, M. D., D. Lieberman, and F. E. Putz. 1987. The dynamics of tree populations 

in tropical forest: A review. J. Trop. Ecol. 3:359–363. 

Vincentini, K. R. F. 1999. “História do Fogo no Cerrado: Uma Análise Palinológica.’’ 

Ph.D. thesis, Universidade de Brasília, Brasília, Brazil. 

Walker, B. H. 1981. Is succession a viable concept in African savanna ecosystems? 

In D. C. West, H. H. Schugart, and D. B. Botkin, eds., Forest Succession: 

Concepts and Application, pp. 431–447. New York: Springer- 

Verlag.


9 

The Role of Fire in Population 

Dynamics of Woody Plants 

William A. Hoffmann and Adriana G. Moreira 

Understanding the factors responsible for the great variation 

in woody plant density has been a challenge for ecologists in the 

cerrado and other tropical savannas. It is becoming evident that no single 

factor determines tree density in the cerrado; rather, nutrient availability, 

water stress, and fire interact to determine woody plant cover. Of these 

three factors, the role of fire is perhaps most important to understand, 

since it alone is largely under human control and is probably the factor 

most variable at the interannual to interdecadal scale. Therefore it is likely 

responsible for most temporal changes in woody plant density within the 

time frame of human observation (see chapter 4). 

Evidence of fires in the cerrado has been recorded for before 27,000 

years before present (Vicentini 1993) and was probably present long 

before then. Thus, woody plants have been exposed to its selective pressure 

for a considerable part of their evolutionary history. Although the 

flora appears very well adapted to normal levels of fire, human activity 

has almost certainly increased fire frequency above the natural rate (chapter 

5). So while the cerrado plants are in general tolerant of fire, in many 

cases they are now subjected to frequencies in excess of the environment 

in which they evolved. 

The dynamics of plant populations is determined by a suite of vital 

rates, including survival, growth, sexual reproduction, vegetative reproduction, 

and seedling establishment (chapter 8). The impact of fire is 

sufficiently severe to affect all of these vital rates, having implications 

for population dynamics and thus community dynamics. In this chapter, 

we review the known effects of fire on each of these vital rates. Then we 

159



demonstrate the consequences of these changes for population dynamics 

and finally community dynamics. 

EFFECT OF FIRE ON LIFE HISTORIES 

OF WOODY PLANTS 

Plant Survival 

In many fire-prone ecosystems, two principal life history strategies of 

woody plants can be identified (Keeley and Zedler 1978; van Wilgen and 

Forsyth 1992; Bond and van Wilgen 1996). The first strategy is that of 

obligate seeders, species that do not survive fire as adults but have high 

seedling establishment following burns. The second strategy is that of 

resprouters, which survive fire by resprouting from stems or roots. In the 

cerrado, it is this second strategy which predominates, with all or most 

woody plant species capable of resprouting after fire. Observed community-wide 

adult mortality rates after fire have ranged from 5% to 19% in 

the cerrado. The high degree of variability in community-wide estimates 

may result from differences in fire intensity (Sato 1996), timing, or species 

composition. Among species, there is considerable variation in adult mortality, 

with observed mortality rates ranging from 1% for Piptocarpha 

rotundifolia (Hoffmann 1996) to 42%, as in the case of Eremanthus 

suberosum (Sato 1996). As a whole, woody plant mortality rates appear 

similar to other moist tropical savannas (Rutherford 1981; Lonsdale and 

Braithwaite 1991) but are much lower than Brazilian tropical forests (see 

fig. 9.1A) which are not adapted to such high fire frequencies. A number 

of traits contribute to the high capacity of cerrado species to survive fire, 

including thick bark (Silva and Miranda 1996), large investment in carbohydrate 

and nutrient reserves (Miyanishi and Kellman 1986; Hoffmann 

et al. 2000), and the capacity to resprout from dormant or 

adventitious buds. 

The difference in survival between cerrado and forest species is also 

observed at the seedling stage. For three forest species studied, no 

seedlings survived fire when less than 1 year old, whereas 12 of 13 cerrado 

species were able to survive fire (see fig. 9.1B). Among these 13 

species, survival was strongly correlated to seed mass. At a very young 

age, seedlings develop a deep taproot, which serves not only for carbohydrate 

and nutrient storage but also to reach moist soil layers during the 

dry season (Moreira 1992; Oliveira and Silva 1993). Larger seeds should 

provide the seedling with extra resources for the quick development of 

root reserves sufficient for postfire resprouting (Hoffmann 2000).


Figure 9.1 Mortality of burned cerrado and forest species. (A) Communitywide 

estimates of adult mortality in burned stands (mean ± SE). Sample size 

(n) is the number of burns. Data are from Uhl and Buschbacher (1985); Kauffman 

(1991); Sato (1996); Sato and Miranda (1996); Silva, Sato, and Miranda 

(1996); Holdsworth and Uhl (1997); Silva (1999); Cochrane and Schulze 

(1999); and Peres (1999). (B) The effect of fire on survival of first-year 

seedlings. Data are from Franco et al. (1996) and Hoffmann (2000). There is 

a significant correlation between seed mass and survival for savanna species 

(r 2 = 0.59, P < .0001).



Growth 

Although fire may not kill an individual, it generally destroys part of the 

plant. Many larger individuals may lose only leaves and thin branches due 

to scorching, but smaller individuals may lose most or all of their aerial 

biomass. In either case, the loss of aerial tissue represents negative growth 

(i.e., a reduction in the size of an individual). It is well established that size 

largely determines a plant’s rates of growth, survival, and reproduction 

(Harper 1977), so a reduction in individual size can have negative impacts 

on future population growth. 

The primary trait preventing topkill is the thick bark present in many 

species of trees and shrubs. The thermal insulation provided by bark protects 

cambium layers from the high fire temperature (Silva and Miranda 

1996; Gignoux et al. 1997). While height is also effective in protecting 

sensitive tissues from fire, the low stature of cerrado species relative to 

closely related forest species precludes the possibility that cerrado species 

have evolved height as an adaptive response to fire (Rizzini 1971). 

After the initial reduction in plant size, regrowth of woody plants 

tends to be vigorous. It is possible that for some species the rate of regrowth 

is high enough that the plants eventually surpass the size they would have 

attained had they not been burnt, particularly for subshrubs. However, for 

larger growth forms, fire normally causes a net reduction in size. As a 

result, it becomes nearly impossible for small individuals to recruit into 

large size categories under frequent burning (Gignoux et al. 1997). During 

two fire cycles of 2 years, Sato (1996) found that in an area initially encompassing 

1,212 individuals with stem diameter greater than 5 cm, only 37 

new individuals had recruited into this minimum stem diameter. This contrasts 

with 277 individuals that died within this same interval. Without a 

prolonged fire-free interval, few individuals are likely to reach a size at 

which the aerial stem can resist fire. This should be particularly detrimental 

for tree species that reach sexual maturity at a large size. 

Sexual Reproduction 

High fire frequency poses a serious constraint to sexual reproduction. If 

a species has no adaptation to protect seeds from fire, successful reproduction 

requires that individuals flower and produce seeds and that the 

resulting seedlings reach a fire-tolerant size in the short period of time 

between burns. Therefore there should be a selective pressure to accelerate 

this sequence of events. It was previously shown that many seedlings 

develop fire tolerance when less than a year old. On the other hand, there 

is a much greater interspecific variation in capacity for ensuring seedling


The Role of Fire in Population Dynamics of Woody Plants 163 

establishment after burning, whether via fire-protected seeds or rapid 

postfire flowering. 

Seed maturation in the cerrado is concentrated at the end of the dry 

season and beginning of the wet season (Oliveira 1998). Therefore burns 

in the dry season destroy many flower buds, flowers, developing fruit, and 

mature seeds, greatly reducing seed availability (Hoffmann 1998). While 

some cerrado species produce fruits that protect seeds from burning 

(Coutinho 1977; Landim and Hay 1996), many do not. And for many 

species that produce woody fruits that could potentially protect seeds from 

fire, many seem ineffective because of the phenology of fruit maturation 

and seed release. For some species, there appears to be a short window of 

time during which the seeds are mature but have not yet been dispersed 

(W. A. Hoffmann, personal observation). For many such species, the timing 

of fire probably determines the success of this strategy; this relation has 

been observed in the savannas of Northern Australia (Setterfield 1997) but 

not in the cerrado to date. Similarly, there is a need to understand the contribution 

of seed banks in postfire regeneration in the cerrado. 

For the majority of species, it appears that postfire seedling establishment 

is dependent upon seeds produced after burning. Among subshrubs 

and herbs, fire frequently stimulates flowering after fire (Coutinho 1990), 

but in larger species the opposite is often true. In a study of six species of 

trees and shrubs, five, Periandra mediterranea, Rourea induta, Miconia 

albicans, Myrsine guianensis, and Roupala montana, exhibited reduced 

seed production after fire. For the first two species, the reduced postburn 

production was entirely due to the reduction in mean plant size, rather than 

a reduction in size-specific seed production. For the remaining three 

species, reduced plant size was compounded by reduced size-specific reproduction. 

Not all trees exhibit a reduction in postburn seed output, as 

demonstrated by Piptocarpha rotundifolia, which produced nearly twice 

the number of seeds in the first year after fire (Hoffmann 1998). 

Other trees and shrubs reproduce successfully after fire (e.g., Silva, 

Hays, and Morais 1996; Landim and Hay 1996; Cavalheiro and Miranda 

1999), but, in these cases, it is uncertain whether fire has a net positive or 

negative effect on the number of seeds produced. Since the effect of fire 

on seed production is often quantitative rather than qualitative, and spatial 

and temporal variation is high, detailed study is needed to ascertain 

the response of cerrado species to fire. 

Seedling Establishment 

In many environments, fire is known to stimulate seedling establishment 

by removing competing adults and inhibitory litter. Hoffmann (1996)



found the opposite to be true in the cerrado. For seven of twelve species, 

establishment success of experimentally placed seeds was found to be significantly 

lower in recently burned sites than in sites burned 1 or more 

years previously, whereas none of the species exhibited enhanced seedling 

establishment. A possible explanation is that fire reduces litter and canopy 

cover, which facilitates seedling establishment of several species, probably 

by ameliorating water stress (Hoffmann 1996). 

The poor seedling establishment following fire might explain why 

many species have not evolved protective fruits that ensure seed availability 

after fire. If conditions for seedling establishment are poor after 

burning, there should be little selective pressure to develop such fruits. 

Vegetative Reproduction 

Rather than produce large cohorts of seedlings following burning, some 

cerrado species have evolved the strategy of producing large cohorts of 

suckers (Hoffmann 1998). Following fire, Rourea induta, Roupala montana, 

and Myrsine guianensis produced large numbers of root suckers, a 

response documented in species of other regions (Lacey 1974; Farrell and 

Ashton 1978; Lamont 1988; Lacey and Johnston 1990; Kammesheidt 

1999). It is necessary to emphasize that we are defining root suckers as 

new stems originating from root buds at some distance from the parent 

individual. When referring to vegetative reproduction, we do not include 

resprouting from the root crown of the original individual. 

Many other cerrado species reproduce vegetatively (Ferri 1962; 

Rizzini and Heringer 1962; Raw and Hay 1985), but it remains to be confirmed 

that fire increases vegetative reproduction in these species. The 

advantages of vegetative reproduction appear to be strong in the cerrado. 

In general, vegetatively produced offspring tend to be larger than 

seedlings, making them less prone to stress and disturbance (Abrahamson 

1980; Peterson and Jones 1997). This was confirmed in the cerrado for 

the three species mentioned above. Suckers were much larger than 

seedlings, and for two of the species, exhibited a greater capacity to survive 

fire (Hoffmann 1998). 

Suckers produced after fire probably benefit from the high light and 

nutrient availability at this time. However, it is not reasonable to argue 

that the timing of vegetative reproduction has evolved in response to 

these selective advantages. There is strong evidence that the timing of 

sucker production has originated as a physiological constraint on the 

process of root bud formation. The formation of root buds can be stimulated 

by the reduction in auxin content resulting from the loss of aerial



biomass (Peterson 1975). The influence of auxins on root bud formation 

has been found in many taxa (Peterson 1975), suggesting that root 

sucker formation is inextricably linked to loss of aerial biomass. 

Although this may constrain the timing of vegetative reproduction, natural 

selection has probably strongly affected not only the quantity of 

suckers produced, but also the capacity of a species to produce root suckers 

in the first place. 

EFFECT OF FIRE ON POPULATIONS 

As shown above, fire simultaneously affects seed production, seedling 

establishment, vegetative reproduction, survival, and growth. The 

changes in these vital rates have direct and interactive effects on the population 

growth rate, so understanding how fire affects only a part of the 

life cycle yields incomplete information on the response of a species to fire. 

However, with quantitative information about the effect of fire on the 

entire plant life history, it becomes possible to synthesize the information 

into a complete picture using matrix population models. Details on the 

use of such models for simulating the effects of fire frequency are provided 

by Silva et al. (1991) and Hoffmann (1999). 

Hoffmann (1999) used information on the growth, survival, seed production, 

seedling establishment, and vegetative reproduction in burned 

and unburned plots to construct matrix population models for a subshrub 

(Periandra mediterranea), two shrubs (Miconia albicans and 

Rourea induta), and two trees (Myrsine guianensis and Roupala montana). 

Using these models, four of the five species are predicted to decline 

under frequent burning but increase under low fire frequencies (see 

fig. 9.2). The fifth species, R. induta, showed little response to burning 

(fig. 9.2). 

The effect of fire on a population depends on its size distribution. 

Under a constant fire frequency, a population will tend toward a particular 

stable size distribution. The population growth rates in figure 9.2 represent 

growth rates after the stable size distribution has been attained. 

However, for a different size distribution, fire can have a very different 

effect. Take, for example, the case of Roupala montana, as shown in figure 

9.3. If a population has not been burned for many years, large adults 

are common, and the population is predicted to increase at a rate of 5% 

per year. If this population is then subjected to triennial burning, large 

individuals produce numerous vegetative offspring, causing a large 

increase in population density for the first two fire cycles. However, after


Figure 9.2 The effect of fire frequency on population growth rate (lambda) 

of five woody plant species, as predicted by matrix population models. Lambda 

> 1 indicates population growth and lambda < 1 indicates population decline. 

Growth forms are P. mediterranea, subshrub; R. induta and M. albicans, shrub; 

M. guianensis and R. montana, tree. Adapted from Hoffmann (1999).



repeated burns, the population becomes dominated by small individuals, 

which do not produce sufficient vegetative suckers to balance mortality, 

so the population experiences decline. Eventually, as the population 

approaches a new size distribution, the population is expected to decline 

at a rate of 5% per year (fig. 9.3). This illustrates the importance of longterm 

studies to shed light on the consequences of fire on population and 

community dynamics, since short-term studies may yield very different 

results. 

With matrix population models, it is possible to quantify the contributions 

of sexual and vegetative reproduction to population growth of 

Figure 9.3 The effect of a change in fire frequency for a size-structured population. 

This graph shows relative population size of Roupala montana as 

simulated by a matrix population model. At the beginning of this simulation, 

the population is assumed to have been unburnt for many years. The first five 

years show the population increase under unburned conditions. After the fifth 

year, the population is subjected to triennial burning, with burns being indicated 

by arrows. Initially there is an increase in population density, but eventually 

population decreases as it becomes dominated by small individuals.



these species using elasticity analysis (de Kroon et al. 1986). This analysis 

revealed that, for the five study species, the contribution of sexual 

reproduction to population growth was low under frequent burning and 

increased under less frequent burning (see fig. 9.4). In contrast, for the 

three suckering species, vegetative reproduction made a large contribution 

to population growth under frequent burning, whereas it decreased 

or remained relatively constant under less frequent burning (fig. 9.4). As 

a result of these contrasting responses to fire frequency, the relative importance 

of vegetative to sexual reproduction increases under increasing fire 

frequency (Hoffmann 1999). 

For the three species capable of vegetative reproduction, sexual reproduction 

contributed little to population growth (fig. 9.4), a trend widely 

observed in clonal plants in other habitats (Abrahamson 1980; Eriksson 

1992). 

CHANGE IN COMMUNITY STRUCTURE 

AND PHYSIOGNOMY 

As described earlier, woody plants differ widely in their tolerance to fire 

and in their capacity to recover afterwards. Fire has considerable potential 

to influence the structure and composition of the vegetation, particularly 

those considered fire-type vegetation where fires are likely to be 

recurrent, such as seasonal tropical savannas (Myers 1936; Soares 1990). 

Savannas are modified by natural fires (Coutinho 1990; Sarmiento 1984), 

and many tropical savannas are maintained today by frequent anthropogenic 

fires (Walker 1981). In the cerrado, fire plays a fundamental role 

for the floristics and physiognomy of savanna stands (Moreira 1996; 

Meirelles et al. 1997; see also chapter 6). 

To understand the role of fire in maintaining vegetation structure in 

the cerrado, it is useful to examine the consequences of excluding fire from 

the system. Fire protection in moist savannas induces gradual changes in 

the density of tree species, leading to denser savannas (Menaut 1977; 

Brookman-Amissah et al. 1980; Frost and Robertson 1987; San José and 

Fariñas 1991; Swaine et al. 1992). The same processes occur in the cerrado, 

where a gradual and progressive increase of woody vegetation after 

fire exclusion has been reported (see chapter 8). Coutinho (1990) reports 

that a campo sujo became a taller, dense cerradão after 43 years of fire 

and cattle exclusion (chapter 6). 

Fire protection increases the frequency of woody plants in both open 

and closed cerrado physiognomies, particularly in the more open ones,


Contribution to population growth 

Fire Return Interval (yr) 

Figure 9.4 The effect of fire frequency on the importance of vegetative and 

sexual reproduction (data from Hoffmann 1999). The contribution of the two 

forms of reproduction were calculated from elasticity analysis of the matrix 

models (de Kroon et al. 1986).



indicating that protection against burning permits woody vegetation 

regeneration (Moreira 1996). Conversely, reintroduction of fire into previously 

protected sites causes a reduction in tree and shrub density (chapter 

4). Fire reduces the woody plant cover as a whole regardless of 

physiognomy, but denser formations experience the greatest declines 

(Moreira 2000). 

The reduction in woody plant density in burned areas is compounded 

by a reduction in mean plant size. In a study comparing the effect of long 

term fire protection on five cerrado physiognomies, ranging from open 

campo sujo to cerradão forest, Moreira (1992, 2000) showed that the 

number of plants of the smallest size class was higher in burned than in 

protected sites, while fire protection led to a shift towards taller plants 

(see fig. 9.5). Contrary to claims that fire affects only individuals below 

the fire line, usually around 1 to 1.5 m (Brookman-Amissah et al. 1980; 

San José and Fariñas 1983; Frost and Robertson 1987; Coutinho 1990), 

fire actually affects woody plants in all size classes, which can lead to striking 

differences on woody cover (Moreira 1992, 1996). 

The negative effect of fire on woody plant density does not occur uniformly 

across all species. Rather, fire influences species composition, in 

general driving a shift towards smaller growth forms. It has been widely 

noted that savanna fire favors herbaceous plants at the expense of woody 

plants; however, even among woody plants, fire causes a shift toward 

smaller growth forms (fig. 9.5). The abundance of trees and large shrubs 

tended to be reduced by fire, whereas subshrubs were favored. It appears 

that this relationship between growth form and fire tolerance reflects the 

effect of topkill (Hoffmann 1999). For the same rate of regrowth, larger 

growth forms require more time to regain their preburn size following 

topkill. Similarly, topkill makes it unlikely that immature individuals of 

trees and large shrubs reach a mature size under frequent burning, thus 

curtailing population growth. 

There are many notable exceptions to this relationship between 

growth form and sensitivity to fire, such as the tree Bowdichia virgilioides, 

which is considered a fire-tolerant species (San José and Fariñas 1983) and 

was more common in frequently burned sites than in adjacent protected 

sites (Moreira 2000). The existence of fire-sensitive shrubs has not been 

previously reported in the cerrado literature, and the shrub flora is always 

considered to be “typically pyrophytic” (Coutinho 1990). The genus 

Miconia, however, has two fire-sensitive shrub species, M. albicans and 

M. pohliana. In fact, M. albicans was the most abundant species in a cerrado 

sensu stricto protected for more than 20 years against fire, but was



Figure 9.5 Response of three growth forms to frequent fire. The figure shows 

the proportion of species that were more common in a frequently burned site 

than in an adjacent unburned site. Smaller growth forms were more likely to 

be favored by fire (data from Moreira 1992). 

completely absent in the adjacent unprotected side (Moreira 1992, 1996). 

Because this species produces very few seeds in the first two years after 

burning, and its seedlings are extremely sensitive to fire for at least the 

first two years of life (Hoffmann 1998), establishment is virtually eliminated 

under frequent burning. 

Fire has a particularly strong effect on species composition of more 

closed physiognomies such as cerradão (chapter 6). Moreira (1996) found 

that five of the ten most abundant species in a fire-protected cerradão 

were totally absent in sampled areas of adjacent unprotected cerradão. 

Cerradão generally includes fire-sensitive species typical of forest, as 

shown earlier. Emmotum nitens, Ocotea spixiana and Alibertia edulis are 

forest species typical of cerradão that rarely, if ever, are found in more 

open cerrado (Furley and Ratter 1988). The establishment of E. nitens in 

open cerrado is probably constrained by the absence of a resprouting ability 

in seedlings of this species. Emmotum nitens and A. edulis have highly 

branched roots without any of the enlarged taproot that characterizes



many cerrado woody plants (Labouriau et al. 1964; Rizzini 1965; Moreira 

1987, 1992; Oliveira and Silva 1993). 

Other fire-sensitive species typical of cerradão can expand into other 

physiognomies in the absence of fire. The fire-sensitive cerradão species 

Blepharocalix salicifolius and Sclerolobium paniculatum were present in 

fire-protected campo sujo and campo cerrado. In general fire protection 

leads to an increase of tree species in these shrub-dominated physiognomies, 

particularly in campo sujo, corroborating the idea that fire protection 

can lead to the substitution of open campo sujo by a treedominated 

denser cerrado physiognomy (Moreira 1992; chapter 6). 

In general burning decreases species richness of woody plants in cerrado 

(Eiten and Sambuichi 1996; Moreira 1996; Sato 1996; Silva 1999). 

This reduction represents a reduction not only in the number of species 

per area, as would be expected from an overall reduction in the density 

of individuals, but also in the total stock of species in an area (Eiten and 

Sambuichi 1996). 

CONCLUSIONS 

Fire has a large impact on the population ecology of woody plants in the 

cerrado. Fire tends to increase plant mortality, reduce plant size, and 

increase vegetative reproduction. Sexual reproduction is also affected, but 

the response is highly species specific. The net effect of these changes is a 

reduction in woody plant cover under the high fire frequecies currently 

observed. This change is manifested as a reduction in mean plant size 

within populations, a reduction in the density of individuals, and a shift 

in species composition towards smaller growth forms. Composed of firesensitive 

species, forest formations within the cerrado region are most susceptible 

to fire. Because of the large impact of fire on the cerrado, the 

current high frequency of anthropogenic burning is capable of effecting 

widespread change in the cerrado ecosystem. 

REFERENCES 

Abrahamson, W. G. 1980. Demography and Vegetative Reproduction. In O. 

T. Solbrig, ed., Demography and Evolution in Plant Populations, pp. 

89–106. Berkeley, CA: University of California Press. 

Bond, W. J. and B. W. van Wilgen. 1996. Fire and Plants. London: Chapman 

and Hall.



Brookman-Amissah, J., Hall J. N., Swaine M. D., and J. Y. Attakorah. 1980. 

A re-assessment of a fire protection experiment in a north-eastern Ghana 

savanna. J. Appl. Ecol. 17:85–99. 

Cavalheiro, M. C. and H. S. Miranda. 1999. Estudo da produção de flores 

em áreas de campo sujo submitidos a diferentes regimes de queima, 

Brasília, DF. Rev. Bras. Fisiol. Vegetal, Sup. 11:82. 

Cochrane, M. A. and M. D. Schulze. 1999. Fire as a recurrent event in tropical 

forests of the eastern Amazon: Effect on forest structure, biomass, 

and species composition. Biotropica 31:2–16. 

Coutinho, L. M. 1977. Aspectos ecológicos do fogo no cerrado: II. As 

queimadas e a dispersão de sementes de algumas espécies anemocóricas 

do estrato herbáceo-subarbustivo. Bol. Bot. USP 5:57–64. 

Coutinho, L. M. 1990. Fire ecology of the Brazilian cerrado. In J. G. 

Goldammer, ed., Fire in the Tropical Biota, pp. 82–105. Berlin: Springer- 

Verlag. 

De Kroon, H., A. Plaisier, J. van Groenendael, and H. Caswell. 1986. Elasticity: 

The relative contribution of demographic parameters to population 

growth rate. Ecology 67:1427–1431. 

Eiten, G. and R. H. R. Sambuichi. 1996. Effect of long-term periodic fire on 

plant diversity in a cerrado area. In R. C. Pereira and L. C. B. Nasser, 

eds., Biodiversidade e Produção Sustentável de Alimentos e Fibras nos 

Cerrados. Proceedings of the VIII Simpósio Sobre o Cerrado, pp. 46–55. 


Eriksson, O. 1992. Evolution of seed dispersal and recruitment in clonal 

plants. Oikos 63:439–448. 

Farrel, T. P. and D. H. Ashton. 1978. Population studies on Acacia melanoxylon 

R.Br.: I. Variantion in seed and vegetative characteristics. Aust. J. Bot. 

26:365–379. 

Ferri, M. G. 1962. Histórico dos trabalhos botânicos sobre o cerrado. In M. 

G. Ferri, ed., Simpósio Sobre o Cerrado, pp. 7–35. São Paulo: Editora 

Edgard Blucher. 

Franco, A. C., M. P. Souza, and G. B. Nardoto. 1996. Estabelecimento e 

crescimento de Dalbergia miscolobium Benth. em áreas de campo sujo e 

cerrado no DF. In H. S. Miranda, C. H. Saito, and B. F. de S. Dias, eds., 


Brasília: ECL/Universidade de Brasília. 

Frost, P. G. H. and F. Robertson. 1987. The ecological effects of fire in savannas. 

In B. J. Walker, ed., Determinants of Tropical Savannas, pp. 93–140. 

Paris: IUBS Special Issue no. 3. 

Furley, P. A. and J. A. Ratter. 1988. Soil resources and plant communities of 

Central Brazilian cerrado and their development. J. Biogeogr. 15:97–108. 

Gignoux, J., J. Clobert, and J.-C. Menaut. 1997. Alternative fire resistence 

strategies in savanna trees. Oecologia 110:576–583. 

Harper, J. L. 1977. Population Biology of Plants. London: Academic Press.



Hoffmann, W. A. 1996. The effects of cover and fire on seedling establishment 

in a neotropical savanna. J. Ecol. 84:383–393 

Hoffmann, W. A. 1998. Post-burn reproduction of woody plants in a neotropical 


J. Appl. Ecol. 35:422–433. 

Hoffmann, W. A. 1999. Fire and population dynamics of woody plants in a 

neotropical savanna: Matrix model predictions. Ecology 80:1354–1369. 

Hoffmann, W. A. 2000. Post-establishment seedling success of savanna and 

forest species in the Brazilian Cerrado. Biotropica 32:62–69. 

Hoffmann, W. A., F. A. Bazzaz, N. J. Chatterton, P. Harrison, and R. B. Jackson. 

2000. Elevated CO 2 enhances resprouting of a tropical savanna tree. 

Oecologia 123:312–317. 

Holdsworth, A. R. and C. Uhl. 1997. Fire in Amazonian selectively logged 

rain forest and the potential for fire reduction. Ecol. Appl. 7:713–725. 

Kammesheidt, L. 1999. Forest recovery by root suckers and above-ground 

sprouts after slash-and-burn agriculture, fire, and logging in Paraguay 

and Venezuela. J. Trop. Ecol. 15:143–158. 

Kauffman, J. B. 1991. Survival by sprouting following fire in tropical forests 

of the Eastern Amazon. Biotropica 23:219–224. 

Keeley, J. E. and P. H. Zedler. 1978. Reproduction of chaparral shrubs after 

fire: A comparison of sprouting and seeding strategies. Am. Nat. 

99:142–161. 

Labouriau, L. G., I. F. M. Válio, and E. P. Heringer. 1964. Sôbre o Sistema 

Reprodutivo de Plantas dos Cerrados. An. Acad. Bras. Ciên. 

36:449–464. 

Lacey, C. J. 1974. Rhizomes in tropical Eucalypts and their role in recovery 

from fire damage. Aust. J. Bot. 22:29–38. 

Lacey, C. J. and Johnston, R. D. 1990. Woody clumps and Clumpwoods. 

Aust. J. Bot. 38:299–334. 

Lamont, B. B. 1988. Sexual versus vegetative reproduction in Banksia elegans. 

Bot. Gaz. 149:370–375. 

Landim, M. F. and Hay, J. D. 1996. Impacto do fogo sobre alguns aspectos 

da biologia reprodutiva de Kielmeyera coriacea Mart. Rev. Bras. Biol. 

56:127–134. 

Lonsdale, W. M. and R. W. Braithwaite. 1991. Assessing the effects of fire on 

vegetation in tropical savannas. Aust. J. Ecol. 16:363–374. 

Meirelles, M. L., C. A. Klink, and J. C. S. Silva. 1997. Un modelo de estados 

y transiciones para el Cerrado brasileño. Ecotrópicos 10:45–50. 

Menaut, J. C. 1977. Evolution of plots protected from fire since 13 years in 

a Guinea savanna of Ivory Coast. Actas del IV Simposio Internacional 

de Ecologia Tropical (Panama) 2:541–558. 

Miyanishi K. and M. Kellman. 1986. The role of root nutrient reserves in 

regrowth of two savanna shrubs. Can. J. Bot. 64:1244–1248. 

Moreira, A. G. 1987. “Aspectos Demograficos de Emmotum nitens (Benth.)



Miers em um Cerradão Distrófico do Distrito Federal.’’ Master’s thesis, 



Cerrado.’’ Ph.D. thesis, Harvard University, Cambridge, MA, U.S.A. 

Moreira, A. G. 1996. Proteção contra o fogo e seu efeito na distribuição e 

composição de espécies de cinco fisionomias de cerrado. In H. S. 

Miranda, C. H. Saito, and B. F. de S. Dias, eds., Impactos de Queimadas 

em Areas de Cerrado e Restinga, pp. 112–121. Brasília: ECL/Universidade 

de Brasília. 

Moreira, A. G. 2000. Effects of fire protection on savanna structure in Central 

Brazil. J. Biogeogr. 27:1021–1029. 

Myers, J. G. 1936. Savanna and forest vegetation of the interior Guiana 

Plateau. J. Ecol. 24:162–184. 

Oliveira, P. E. 1998. Fenologia e biologia reprodutiva das espécies de cerrado. 

In S. M. Sana and S. P. de Almeida, eds., Cerrado: Ambiente e Flora, pp. 


Oliveira, P. E. and J. C. S. Silva. 1993. Biological reproduction of two species 

of Kielmeyera (Guttiferae) in the cerrados of central Brazil. J. Trop. Ecol. 

9:67–80. 

Peres, C. A. 1999. Ground fires as agents of mortality is a central Amazonian 

Forest. J. Trop. Ecol. 15:535–541. 

Peterson, C. J. and Jones, R. H. 1997. Clonality in woody plants: A review 

and comparison with clonal herbs. In H. de Kroon and J. van Groenendael, 

eds., The Ecology and Evolution of Clonal Plants, pp. 263–289. 

Leiden, Netherlands: Backhuys Publishers. 

Peterson, R. L. 1975. The initiation and development of root buds. In G. Torrey 

and D. T. Clarkson, eds., The Development and Function of Roots, 

pp. 125–161. San Diego: Academic Press. 

Raw, A. and J. Hay. 1985. Fire and other factors affecting a population of 

Simarouba amara in cerradão near Brasília. Rev. Bras. Bot. 8:101–107. 

Rizzini, C. T. 1965. Experimental studies on seedling development of cerrado 

woody plants. Ann. Miss. Bot. Gard. 52:410–426. 

Rizzini, C. T. 1971. A flora do cerrado. In M. G. Ferri, ed., Simpósio Sôbre 

o Cerrado, pp. 105–154. São Paulo: Editora Edgard Blücher. 

Rizzini, C. T. and Heringer, E. P. 1962. Studies on the underground organs of 

trees and shrubs from some southern Brazilian savannas. An. Acad. Bras. 

Ciênc. 34:235–247. 

Rutherford, M. C. 1981. Survival, regeneration and leaf biomass changes in 

woody plants following spring burns in Burkea africana–Ochna pulchra 

savanna. Bothalia 13:531–552. 

San José, J. J. and M. R. Fariñas. 1983. Changes in tree density and species 

composition in a protected Trachypogon savana, Venezuela. Ecology 

64:447–453. 

San José, J. J. and M. R. Fariñas. 1991. Changes in tree density and species



composition in a protected Trachypogon savanna protected for 25 years. 

Acta Oecol. 12:237–247. 



Sato, M. N. 1996. “Taxa de Mortalidade da Vegetação Lenhosa do Cerrado 

Submetida a Diferentes Regimes de Queima.’’ Master’s thesis, Universidade 


Sato, M. N. and H. S. Miranda. 1996. Mortalidade de plantas lenhosas do 

cerrado sensu stricto submetidas a diferentes regimes de queima. In H. S. 

Miranda, C. H. Saito, and B. F. S. Dias, eds., Impactos de Queimadas em 

Áreas de Cerrado e Restinga, pp.102–111. Brasilia: Ecologia, Universidade 

de Brasília. 

Setterfield, S. A. 1997. The impact of experimental fire regimes on seed production 

in two tropical eucalypt species in northern Australia. Aust. J. 

Ecol. 22:279–287. 

Silva, D. M. S., J. D. Hay, and H. C. Morais. 1996. Sucesso reprodutivo de 

Byrsonima crassa (Malpighiaceae) após uma queimada em um cerrado 

de Brasilia–DF. In H. S. Miranda, C. H. Saito, and B. F. S. Dias, eds., 


Brasília: Ecologia, Universidade de Brasília. 

Silva, E. P.da R. 1999. “Efeito do Regime de Queima na Taxa de Mortalidade 

e Estrutura da Vegetação.” Master’s thesis, Universidade de Brasília, 


Silva, E. P. da R. and H. S. Miranda 1996. Temperatura do câmbio de espécies 

lenhosas do cerrado durante queimadas prescritas. In R. C. Pereira 

and L. C. B. Nasser, eds., Biodiversidade e Produção Sustentável de Alimentos 

e Fibras nos Cerrados. Proceedings of the VIII Simpósio Sobre o 

Cerrado, pp. 253–257. Brasília: Empresa Brasileira de Pesquisa 


Silva, G. T., M. N. Sato, and H. S. Miranda. 1996. Mortalidade de plantas 

lenhosas em um campo sujo de cerrado submitidos a queimadas prescritas. 

In H. S. Miranda, C. H. Saito, and B. F. de S. Dias, eds., Impactos 

de Queimadas em Áreas de Cerrado e Restinga, pp. 93–101. Brasília: 

ECL/Universidade de Brasília. 

Silva, J. F., J. Raventos, H. Caswell, and M. C. Trevisan. 1991. Population 

responses to fire in a tropical savanna grass: A matrix model approach. 

J. Ecol. 79:345–356. 

Soares, R. V. 1990. Fire in some tropical and subtropical South American vegetation 

types: An overview. In J. G. Goldammer, ed., Fire in the Tropical 

Biota, pp. 63–81. Berlin: Springer-Verlag. 

Swaine, M. D., W. D. Hawthorne, and T. K. Orgle. 1992. The effects of fire 

exclusion on savanna vegetation at Kpong, Ghana. Biotropica 

24:166–172. 

Uhl, C. and R. Buschbacher. 1985. A disturbing synergism between cattle



ranching burning practices and selective tree harvesting in the eastern 

Amazon. Biotropica 17:265–268. 

Van Wilgen, B. W. and G. G. Forsyth. 1992. Regeneration strategies in Fynbos 

plants and their influence on the stability of community boundaries 

after fire. In B. W. van Wilgen, D. M. Richardson, F. J. Kruger, and H. J. 

van Hensbergen, eds. Fire in South African Mountain Fynbos,. pp. 


Vicentini, K. R. C. F. 1993. “Análise Palinológica de uma Vereda em Cromínia–GO.’’ 


Walker, B. H. 1981. Is succession a viable concept in African savanna ecosystems? 

In D. C. West, H. H. Shugart, and D. B. Botkin, eds., Forest Succession: 

Concepts and Applications, pp. 431–447. New York: Springer- 

Verlag.


10 

Ecophysiology of Woody Plants 

Augusto C. Franco 

The central plains of Brazil are occupied by a complex of 

plant physiognomies such as cerradão, cerrado sensu stricto, and campo 

sujo (chapter 6). The great spatial variation in woody plant density across 

the cerrado landscape results in a complex pattern of resource availability, 

which changes both horizontally across the landscape and vertically within 

each vegetation type. This is of particular importance for seedlings that colonize 

the grass matrix typical of a campo sujo vegetation or a closed 

canopy woodland such as a cerradão. Like any other neotropical savanna, 

the cerrados are characterized by a strongly seasonal climate with distinctive 

wet and dry seasons (see fig. 10.1). Soils are deep and well drained, 

acidic, extremely low in available nutrients and with high Al content 

(Goodland and Ferri 1979; Haridasan 1982; Sarmiento 1984; chapter 2). 

Recurrent fires in the dry season place additional stress on the survival of 

woody plant seedlings (fire effects are discussed in chapters 4, 9). 

Models explaining the structure and function of savanna ecosystems 

Figure 10.1 (opposite page) Integrated monthly rainfall, monthly variation in 

the daily number of hours of sunshine, minimum and maximum air temperatures 

and relative humidity, and changes in soil water potential for a campo 

sujo (solid squares) and a cerradão (open circles). Weather data come from the 

climate station of Reserva Ecológica do IBGE, Brasília (15°56' S, 47°53' W). 

Soil water potential is the mean + SE for 5 soil psychrometers (models PCT- 

55-15-SF or PST-55-15-SF, Wescor Inc., Logan, Utah, U.S.) at 5, 30 and 85 cm 

depth, placed in a campo sujo and a cerradão of Fazenda Água Limpa (15°56' 

S, 47°55' W), near the Reserva Ecológica do IBGE. The campo sujo site had 

533 individuals per ha with stem diameter greater than 5 cm at 30 cm from 

the ground. The cerradão site had 2,800 individuals of the same size class per 

ha. See chapter 6 for descriptions of cerrado physiognomies. 

178




typically involve water and nutrients as limiting resources and a two-layered 

(grasses versus woody species) soil-water system (Walker and Noy- 

Meir 1982; Knoop and Walker 1985). According to this model, the 

shallow roots of grasses make them superior competitors for water in the 

upper part of the soil profile, whereas deeply rooted woody plants have 

exclusive access to a deeper, more predictable water source. This two-layered 

model implies that tree recruitment is dependent on the capacity of 

seedlings to withstand grass root competition during early growth stages, 

until their roots reach deeper and more reliable soil water sources to sustain 

them during dry periods (Medina and Silva 1990). 

While the tropical savannas of northern South America are characterized 

by relatively low woody plant diversity, cerrado communities contain 

a remarkably complex community structure rich in endemic woody 

species. More than 500 species of trees and large shrubs are present in the 

cerrado region, and individual sites may contain up to 70 or more woody 

species (Felfili et al. 1998). As discussed in chapter 6, the establishment of 

modern floristic composition is primarily the result of historic and biogeographic 

events. However, resource availability certainly plays a prominent 

role in maintaining such high species diversity. Tilman (1982) and 

Cody (1986) have independently proposed models to explain plant species 

diversity in resource-limited environments. Their models, based on selection 

for increased resource partitioning, suggest that seasonality in cerrado 

rainfall regime and low soil nutrient availability are critical factors allowing 

the coexistence of many woody species in relatively small areas. The 

theoretical model of Tilman (1982) is particularly significant in this respect 

since it explicitly considers the temporal and spatial dynamics of resource 

availability in selection for divergent strategies of resource capture. 

In this chapter I will show that in savannas of high woody species 

diversity such as the cerrado, plants exhibit an array of physiological and 

morphological mechanisms to cope with rainfall seasonality. Because these 

plants are growing on extremely infertile soils, I will also discuss the effects 

of nutrient deficiency on plant growth and biomass allocation. Finally, I 

will focus on establishment and growth of seedlings of woody perennials, 

with an emphasis on the effects of spatial variation in light, an important 

environmental factor that is generally neglected in savanna studies. 

RAINFALL SEASONALITY AND PATTERNS 

OF WATER USE 

The cerrados are characterized by strong rainfall seasonality (fig. 10.1). 

The winters are dry and cool. Frost events are uncommon and occur only


Ecophysiology of Woody Plants 181 

at the southern limit of the cerrado region. Low relative humidities result 

in a high evaporative demand during the dry period that generally extends 

from May to September. During the dry season, the soil begins to dry out 

from the surface downwards, and soil water potentials (Ψ s ) in the upper 

soil layers can reach values below −3.0 MPa. Upper soil layers dry much 

faster in an open campo sujo formation than in a closed cerradão formation. 

Deeper soil layers exhibit a much higher degree of water constancy, 

but they are drier in the cerradão (fig. 10.1). Indeed, soil gravimetric water 

content at a campo sujo averaged 4% higher than at a nearby cerradão 

site for the 1–2-m depth range, and 3% higher for the 2–3-m depth range, 

toward the end of the dry period (Jackson et al. 1999). Thus, increases in 

woody cover and in tree diversity result in a larger exploitation of soil 

water resources. However, individuals of three evergreen species showed 

similar water status and daily water loss in these two contrasting vegetation 

types (Bucci et al. 2002). 

A schematic representation of the most relevant factors affecting the 

water balance of cerrado woody plants is presented in figure 10.2. Imbalance 

between water supply and demand results in changes in plant water 

status, generally assessed by measurements of leaf water potential (Ψ l ). 

Because nocturnal stomatal closure curtails transpiration and potentially 

allows for plant rehydration, a water balance between the plant and the 

soil should be reached by the end of the night. Thus, predawn Ψ l can be 

used as a measure of the water potential of the soil adjacent to the root 

system and therefore of the maximum water status the plant can achieve. 

Predawn Ψ l remains between −0.1 and −0.3 MPa in the wet season, and 

−0.3 to −0.8 MPa in the dry season (Mattos et al. 1997; Mattos 1998; 

Franco 1998; Meinzer et al. 1999; Bucci et al. 2002). Minimum Ψ l is 

reached between midday and early afternoon, when the evaporative 

demand of the atmosphere is higher. Values are in the range of −1 to −3 

MPa in the wet season and −2 to −4 MPa in the dry season (Perez and 

Moraes 1991; Franco 1998; Meinzer et al. 1999; Bucci et al. 2002). Thus, 

Ψ l of adult shrubs and trees decreases in the dry season but is still higher 

than Ψ s of upper soil layers. Excavation experiments reported that about 

70% of the root biomass was contained in the first 100 cm depth (Abdala 

et al. 1998). However, a small proportion of root biomass was found to 

depths of 6 to 8 meters. Thus, these few deeper roots would extract 

enough water to ensure that rainfall seasonality exerts only a small effect 

on Ψ l . 

Not all cerrado woody species are deep-rooted. Earlier excavation 

experiments have already observed a wide range of rooting habits, from 

shallow-rooted to deep-rooted species (Rawitscher 1948). Measurements 

of hydrogen isotope differences between plant tissue water and soil water



Figure 10.2 A diagrammatic representation of the most relevant factors 

affecting the water balance of cerrado woody plants. The solid arrows indicate 

processes that are well established in the literature or that have been 

reported for cerrado plants. The broken arrows indicate processes that are 

either controversial (most of them are probably controlled by endogenous 

growth regulators, including receptor pigments) or poorly known in the cerrado. 

For the sake of simplicity, the processes affecting soil water availability 

(e.g., soil water evaporation, precipitation, and soil characteristics) are not 

detailed. 

samples collected at different depths have indicated a complex pattern of 

water exploitation of the soil profile (Jackson et al. 1999). Deciduous 

species were extracting water from deep soil layers in the dry season, 

whereas evergreeen species showed a broader range of water extraction 

patterns, from shallow-rooted to deep-rooted. These studies suggest that 

root patterns in cerrado ecosystems are much more complex than predicted 

by the classical two-layered soil-water model for savannas. 

Loss of water vapor by transpiration is driven by the leaf-to-air vapor 

pressure difference. It occurs mainly through the stomatal pore because 

of the presence of a highly waterproof cutinized epidermis. The adequate 

availability of soil water for deep-rooted woody plants would imply that



they would not need to rely on regulation of water losses even during the 

peak of the dry season. This was in agreement with early studies, relying 

on rapid weighing of detached leaves, that concluded that many cerrado 

woody species transpired freely throughout the year; some restricted their 

transpiration only at the end of the dry season; and only a few restricted 

their transpiration from the beginning of the dry season (e.g., Ferri 1944; 

Rawitscher 1948). This weak stomatal control of transpiration was also 

reported in four characteristic woody trees of the Venezuelan savannas 

(Sarmiento et al. 1985) also common in the Brazilian cerrados. Because 

of the diurnal and seasonal changes in vapor pressure deficit, conclusions 

based on measurements of transpiration rates have to be examined with 

caution. Indeed, studies using gas exchange techniques consistently 

reported lower maximum photosynthetic rates and stomatal conductances 

(mainly a function of the degree of stomatal opening), as well as a 

moderate midday depression of photosynthesis and stomatal conductances 

in most species during the dry season (Johnson et al. 1983; Moraes 

et al. 1989; Perez and Moraes 1991; Moraes and Prado 1998). In most 

species lower stomatal conductances in the dry season resulted in an 

increase in water use efficiency (the amount of CO 2 assimilated by photosynthesis/water 

lost by transpiration), which was determined by comparing 

the carbon isotope ratios of leaves collected in the wet and dry 

season (Mattos et al. 1997). Medina and Francisco (1994) reported similar 

results for two common woody species of the Venezuelan savannas. 

Control of water loss and reductions of photosynthetic rates as a result of 

partial stomatal closure both in the wet and dry season were also reported 

(Franco 1998; Moraes and Prado 1998; Maia 1999). 

Several studies of regulation of water use at the plant level with sap 

flow sensors have shown that, despite the potential access to soil water at 

a depth where availability was relatively constant throughout the year, 

plants exhibited reduced transpiration due to partial stomatal closure during 

both the dry and wet seasons (Meinzer et al. 1999; Bucci et al. 2000; 

Naves-Barbiero et al. 2000). For most species, sap flow increased sharply 

in the morning, briefly attained a maximum value by about 09:30 to 

12:00 h, then decreased sharply, despite steadily increasing solar radiation 

and atmospheric evaporative demand. This decrease was particularly 

strong in the dry season, when high values of vapor pressure deficit prevail 

during most of the daylight hours. In some cases, transpiration rates 

briefly recovered in the late afternoon (Naves-Barbiero et al. 2000). Thus 

the reductions in stomatal opening are the result of hydraulic constraints, 

or a direct response to changes in leaf-to-air vapor pressure deficit. 

At the ecosystem level, measurements of water fluxes using eddy



covariance methods showed that evapotranspiration increased linearly 

with solar irradiation in the wet season, but not in the dry season, when 

this linear increase was maintained only up to solar irradiation values of 

200 Wm −2 (about one-fourth the energy of full sunlight at this time of 

year). This was followed by a considerable decline in the slope of the 

curve, despite steadly increasing solar irradiation (Miranda et al. 1997). 

These authors also reported that ecosystem surface conductances (an estimate 

of stomatal opening at the canopy level) were much lower in the dry 

season and fell gradually as the day progressed both in the wet and dry 

seasons, probably in response to increases in the evaporative demand of 

the atmosphere. 

In conclusion, stomatal conductance plays a major role in the control 

of water flow, minimizing the effects of increases in the evaporative 

demand of the atmosphere (increase in vapor pressure deficit) as a force 

driving transpiration (fig. 10.2). The values of stomatal conductance 

which restrict transpiration to levels compatible with the supply of water 

from the xylem will vary, mainly depending on the magnitude of the vapor 

pressure deficit and total leaf area, which changes seasonally, but also 

reflecting some limitation due to soil water availability and stem water 

storage. Thus, this rapid decrease in flow rate after an early peak may represent 

the limits of an internal reservoir that is recharged at night, or may 

be the result of limited capacity of the root system to absorb water in sufficient 

quantities to sustain high rates of transpiration over a longer 

period. Little is known about the internal structure of the vascular systems 

of these species and their effects on the transpiration rates. Morphological 

studies coupled with physiological measurements of hydraulic 

conductivity, root distribution, and water extraction patterns at different 

soil layers are necessary to provide a more complete picture of water relations 

of cerrado woody plants. Moreover, it is yet to be determined 

whether leaf abscission and changes in stomatal opening of cerrado plants 

are biochemically controlled in response to environmental signals, or 

whether they are a response to physical factors related to changes in cell 

turgor driven by water availability. 

SEASONAL WATER AVAILABILITY, PHOTOSYNTHESIS, 

LEAF PHENOLOGY, AND GROWTH 

The drop in predawn Ψ l by the end of the dry season indicates that full 

nocturnal recharge of water reservoirs is not attained and that soil water 

availability is becoming a critical factor. Plant water reservoirs and noc-



turnal recharge are still sufficient to accommodate high rates of plant 

water loss in the dry season, but shoot growth depends on the relief of 

water stress. Table 10.1 presents leaf characteristics of 13 woody species 

commonly found throughout the cerrado region. In general, evergreen 

species tend to rely on shallower water sources and to flush during the 

wet season. Vochysia elliptica, an evergreen with a deep root system, 

flushes at the end of the dry season, a pattern similar to deep-rooted deciduous 

trees. However, the evergreen Roupala montana and the deciduous 

Pterodon pubescens apparently have a shallow-rooted system, but both 

show a leaf flush pattern typical of a deep-rooted tree: that is, they flush 

at the end of the dry season. Partial (R. montana) or total (P. pubescens ) 

Table 10.1 Patterns of Leaf Flush, Maximum CO 2 Assimilation 

Rates (A max ; µmol CO 2 m -2 s –1 ) and Depth of the Root System 

of 13 Cerrado Woody Species 

Species Family Leaf flush A max Root depth 

Evergreen 

Didymopanax Araliaceae Throughout the 18 a Shallow 

macrocarpum year 

Miconia ferruginata Melastomataceae Wet season 14 f Shallow 

Rapanea guianensis Myrsinaceae Wet season 12 e Not available 

Roupala montana Proteaceae End of the dry season 14 b Shallow 

Sclerolobium Caesalpinoidae Wet season 20 f Shallow 

paniculatum 

Vochysia elliptica Vochysiaceae End of the dry season 14 f Deep 

Briefly deciduous 

Blepharocalyx Myrtaceae End of the dry season 12 f Not available 

salicifolia 

Dalbergia Caesalpinoidae End of the dry season 15 c Deep 

miscolobium 

Pterodon pubescens Faboideae End of the dry season 11 c Shallow 

Qualea grandiflora Vochysiaceae End of the dry season 16 f Deep 

Deciduous 

Caryocar brasiliense Caryocaraceae End of the dry season 11 e Not available 

Kielmeyera coriacea Guttiferae End of the dry season 12 d Deep 

Qualea parviflora Vochysiaceae End of the dry season 12 c Intermediate 

Sources: a = Franco 1983; b = Franco 1998; c = Kozovits 1997; d = Cardinot 1998; e = Maia 

1999; f = previously unpublished data. 

Note: A max was measured during morning hours under field conditions in the wet season. 

Depth of the root system was based on comparisons of stable hydrogen isotope composition 

of stem xylem water, and soil water that was collected at different depths (Jackson et 

al. 1999). Patterns of leaf flush were compiled from Franco (1998); Jackson et al. (1999); 

Maia (1999).



leaf fall precedes new leaf production in both species. Stem diameter of 

several deciduous and evergreen cerrado species increases continuously 

throughout the entire rainy season, but this stem expansion ceases at the 

onset of the dry period (Alvin and Silva 1980). 

Deciduous and evergreen trees of the tropical dry forest of Guanacaste, 

Costa Rica, are also able to flush or flower during drought, following stem 

rehydration (Borchert 1994a,b). Rehydration and bud break during the 

dry season occur only in trees with access to soil water or to stored water 

in their trunks, whereas desiccated trees at dry sites rehydrate and flush 

only after the first heavy rains of the rainy season (Borchert 1994c). In a 

study of leaf phenology patterns of 49 woody species (tall shrubs and 

trees) in a tropical eucalypt savanna of North Australia, Williams et al. 

(1997) reported that leaf fall in all species coincided with the attainment 

of seasonal minima in predawn Ψ l , which were about −1.5 to −2.0 MPa 

in the evergreen and semideciduous (canopy fell below 50% of full canopy 

in the dry season) species, and about −0.5 to −1.0 MPa in the fully deciduous 

species. Leaf flushing occurred primarily in the late dry season. 

However, two evergreen species flushed throughout the dry season with 

a major peak in the late dry season, while leaf flushing in two fully deciduous 

species occurred only at the onset of rainy season. Soil moisture at 

1m depth did not fall below the permanent wilting point (10% v/v); hence, 

reserves of soil water were sufficient to support whole-plant rehydration 

that preceded leaf flushing in the absence of rain. Production of new 

foliage more or less ceased soon after the beginning of the wet season, and 

most species were dormant by mid-late wet season. Therefore, we can 

expect that stem rehydration that precedes leaf flushing in cerrado woody 

species is the result of adjustments in the tree-water supply and demand, 

driven by a complex interplay of partial stomatal closure and reductions 

in leaf area that affects the root:shoot ratio as well as the depth of the root 

system and the size of the internal water reservoirs (fig. 10.2). Integrative 

studies of plant water status, cambium activity and leaf phenology of a 

range of cerrado woody species are critically needed. 

The degree of soil water partitioning and variation in the timing of 

leaf production and loss among cerrado woody species suggests that 

resource partitioning may play an important role in maintenance of the 

high diversity of woody species in the cerrado. However, resource partitioning 

may imply a series of tradeoffs. For example, although a deep 

rooting pattern may allow for new leaf production during the driest 

months of the year, it may also impose hydraulic limitations on the 

amount of water that the plants can extract and transpire daily (Meinzer 

et al. 1999). Stomatal control of transpiration could be especially critical



for evergreen species that maintain a considerable amount of photosynthetically 

active leaves throughout the whole dry season. Moreover, leaf 

flushing in the dry season may have important implications in competition 

for water and nutrients with perennial grasses, which grow only after 

the first rains. 

NUTRIENT DEFICIENCY, LEAF SCLEROMORPHISM, 

AND PHOTOSYNTHETIC CAPACITY 

A well-developed cuticle, epidermal cells with thick cell walls, presence of 

hypodermis, and abundance of structural elements characterize leaves of 

most cerrado woody plants. Arens (1958a,b) proposed that the leaf scleromorphism 

in cerrado woody plants was a consequence of nutrient deficiency, 

especially nitrogen. According to his hypothesis, high light levels 

and lack of water stress would result in an abundance of assimilated CO 2 , 

consumed in the production of structural elements in the leaves because 

of low availability of nutrients for growth. Cerrado soils are generally 

very deep Oxisols with a high percentage of clay; well drained; strongly 

acidic; high in Al saturation; and extremely low in available nutrients 

(Goodland and Ferri 1979; Sarmiento 1984; chapter 2). P levels are generally 

below 1 ppm. In a study of 40 woody cerrado species, Miranda et 

al. (1997) reported that N levels (% of dry matter) in mature leaves were 

between 0.7% and 1.0% for 33% of the species, 1.1% and 1.5% for 50% 

of the species, and 1.6 and 2.2% for 10% of the species. Only 3 species 

had N levels above 2.2%. In a study of 8 common cerrado species, 

Medeiros and Haridasan (1985) reported that leaf nutrient levels varied 

between 0.05% and 0.07% for P, 0.14% and 0.80% for Ca, 0.28 and 

0.87% for K, and 0.07% and 0.28% for Mg. These leaf nutrient levels 

are within the range of values reported for sclerophyllous leaves with similar 

specific leaf areas (leaf area: leaf dry mass) on extremely oligotrophic 

soils of the upper Rio Negro, Venezuela (Medina et al. 1990; Reich et al. 

1995). Therefore, low nutrient levels could be a major constraint for plant 

growth in cerrado ecosystems. 

One of the major physiological processes that affects growth is photosynthesis. 

Low nutrient levels could result in low photosynthetic rates 

and, as a consequence, low growth rates. On a leaf area basis, maximum 

photosynthetic rates of cerrado woody plants are moderate, generally in 

the range of 6 to 20 µmol CO 2 m −2 s −1 (Prado and Moraes 1997; Moraes 

and Prado 1998; table 10.1). These values are similar to those of tropical 

canopy trees (Hogan et al. 1995; Zotz and Winter 1996) and savanna



trees of Venezuela (Sarmiento et al. 1985; Medina and Francisco 1994) 

and Australia (Eamus et al. 1999). On a dry weight basis, values are also 

moderate, ranging from 39 to 146 µmol kg −1 s −1 (Prado and Moraes 

1997). Factors such as a higher diversion of plant resources to non-photosynthetic 

tissues, larger leaf construction and maintenance costs, or the 

degree and duration of crown deciduousness could potentially offset the 

benefits of maintaining a photosynthetically active crown in the dry season 

and may explain similar maximum photosynthetic rates in evergreen 

and deciduous species (table 10.1). Indeed, the combined effects of herbivory, 

partial leaf loss, and reductions in photosynthetic rates greatly 

reduced the estimated daily carbon gain of the evergreen Roupala montana 

by end of the dry season (Franco 1998). 

Conditions where photosynthetic capacity is high but N availability 

is low favor carbon storage and biomass partitioning to roots rather than 

leaves (Fichtner et al. 1995). At the plant level, root:shoot ratios of cerrado 

woody species are fairly large, and underground structures for storage 

or vegetative propagation are common (Rizzini and Heringer 1961, 

1962). Both evergreen and deciduous species have scleromorphic leaves 

and a large carbon investment in root biomass. At the ecosystem level, a 

cerrado sensu stricto vegetation (chapter 6) has a root:shoot ratio of about 

1, significantly greater than ratios for tropical forests (ca. 0.1 to 0.5), but 

within the range of values measured for other savanna ecosystems (0.5 to 

2.1; Abdala et al. 1998). 

Changes in nutrient availability affect growth and biomass partitioning 

of cerrado woody plants. Growth of seedlings of Miconia albicans and 

Copaifera langsdorffii was enhanced in gallery forest soils richer in nutrients 

as compared with growth on dystrophic cerrado soils (Haridasan 

1988; Machado 1990). High nutrient availability had a positive effect on 

plant biomass and a negative effect on root:shoot ratio and nonstructural 

carbohydrate concentrations in seedlings of Dalbergia miscolobium (Sassaki 

and Felippe 1998) and K. coriacea (Hoffmann et al. 2000). Melo 

(1999) reported that N fertilization decreased the leaf mass:leaf area ratio 

of Eugenia dysenterica and Sclerolobium paniculatum seedlings but did 

not affect the leaf mass:leaf area ratio of Dypterix alata and Hancornia 

speciosa. N fertilization did not enhance growth of these four species, 

instead decreasing biomass allocation to roots, while P fertilization had a 

positive effect on growth of all four species. 

Mycorrhizal fungi generally play a critical role in soils with low availability 

in phosphorus. The information about mycorrhizal fungi in cerrado 

soils is scant. In a survey of mycorrhizal colonization in cerrado species, 

Thomazini (1974) reported that all species were infected with mycorrhiza.



In a greenhouse experiment with seedlings of four cerrado woody species, 

Reis (1999) concluded that mycorrhiza effectively colonized and increased 

growth of these species, but moderate additions of P enhanced the response 

of these plants to inoculation, increasing their growth. 

The presence of nodules could overcome N deficiency in legumes, 

provided P deficiency was not a constraint for nodule activity. Legumes 

native to low-phosphorus soils often fix N well on them (Barnet and Catt 

1991). The presence of nodules was observed in many cerrado legume 

trees (Faria et al. 1987, 1994). Total leaf N in cerrado legumes is higher 

than in other tree and shrub species (Miranda et al. 1997; Kozovits 1997). 

Based on δ 15 N measurements, Sprent et al. (1996) presented some evidence 

that N fixation by nodules is a significant N source for small nodulated 

legume shrubs and herbs of the cerrado, but no such studies were 

performed with cerrado legume trees. Tripartite symbiosis (Rhizobiummycorrhizal 

fungi-legume) were reported in hemicryptophyte legumes of 

Trachypogon savannas in Venezuela (Medina and Bilbao 1991). However, 

no evidence was found that this symbiosis was effective in reducing 

P deficiency in these small legumes under natural conditions. There is a 

need for studies evaluating the contribution of mycorrhiza-Rhizobium 

associations as N and P sources and the cost of such associations for cerrado 

legume trees and shrubs. 

Many species of the cerrado vegetation accumulate aluminum in large 

quantities in their leaves, but the accumulation of Al does not interfere in 

the absorption of other cations like K, Ca, and Mg (Haridasan 1982; 

Medeiros and Haridasan 1985). Aluminum accumulation is particularly 

common in cerrado species of the families Vochysiaceae, Melastomataceae, 

and Rubiaceae. High concentrations of Al were found in the leaf 

phloem of Al-accumulating species and in the walls and contents of the 

collenchyma of the midrib, epidermal cells, guard cells of the stomata, and 

spongy parenchyma (Haridasan et al. 1986, 1987). However, the physiological 

significance of Al accumulation for cerrado plants is still unknown. 

Miconia albicans, an Al-accumulating shrub of the cerrado region, failed 

to grow in calcareous soils and produced chlorotic leaves, but showed 

complete recovery when parts of their root systems were grown in an 

AlCl 3 solution containing 10 mg Al/L or transplanted into an acid latosol 

(Haridasan 1988). Similar results were also found for seedlings of 

Vochysia thyrsoidea (Machado 1985). 

In conclusion, the scleromorphic, nutrient-poor leaves of cerrado 

trees and shrubs do maintain relatively high photosynthetic rates. It 

appears that most of the assimilated carbon is not used for growth but 

stored in underground structures or diverged for leaf structural compo-



nents, although experiments designed specifically to test this hypothesis 

are lacking. It is, however, somewhat surprising that leaf longevity is not 

an overwhelming feature in such nutrient-poor soils. Even in most species 

that are considered evergreen, leaf lifespan is less than a year. 

ESTABLISHMENT AND GROWTH 

OF SEEDLINGS OF WOODY PERENNIALS 

Tree seedling establishment in neotropical savannas is heavily constrained 

by grass root competition, drought, and fire (Medina and Silva 1990). 

Survival of tree seedlings during a given rainy season depends on the water 

availability in the topsoil, where most of the roots of the herbaceous layer 

are found. Probability of seedling establishment depends on their capability 

to reach moist soil layers beyond the grass root zone, and on the 

buildup of underground energy reserves, which allow regrowth of aerial 

biomass after fire or drought. In this model, water is the basic constraint 

on seedling establishment and growth. 

Grass root competition for soil water should not be a critical factor 

in the wet season, because the topsoil layers remain wet (high Ψs ) most of 

the time (fig. 10.1). However, unpredictable dry spells in the wet season 

may limit the survival of newly germinated seedlings (Hoffmann 1996). 

Other factors, such as herbivory and pathogen attack, have to be considered 

and may play a major role, at least for some species (Nardoto et al. 

1998; Braz et al. 2000; chapter 16). On the other hand, seasonal drought 

was not an important mortality factor for seedlings of three common cerrado 

trees (Nardoto et al. 1998; Braz et al. 2000; Kanegae et al. 2000). 

Seasonal drought may have a major impact not only on seedling survival, 

but also on seedling growth and carbon metabolism. Information 

on physioecological characteristics of cerrado tree seedlings such as photosynthetic 

responses to water stress and carbon budgets is scant. Net 

CO2 assimilation rates (A ) of seedlings of cerrado woody plants reach 

CO2 

the compensation point (ACO2 = 0) at Ψl of −2.4 to −3.9 MPa (Prado et 

al. 1994; Sassaki et al. 1997; Moraes and Prado 1998). Soil water potential 

of upper soil layers reaches values within this range during the dry 

season (fig. 10.1). Cerrado woody species allocate a larger proportion of 

the their biomass to roots than to shoots during the initial growth period 

(Arasaki and Felippe 1990; Sassaki and Felippe 1992; Paulilo et al. 

1993). However, roots of seedlings that germinated in the rainy season 

would still be exposed to these dry soil layers during the subsequent 

drought period, and perhaps in the next drought as well (Rizzini 1965; 

Moreira 1992).



Although generally not considered a limiting factor in savanna environments, 

canopy shading can restrict seedling growth in the initial phases 

of plant development. Leaves of cerrado woody species typically reach 

90% of the maximum photosynthetic values at photosynthetic photon flux 

densities (PPFD; the flux of photons between 400 and 700 nm wavelength 

per unit area) of 600 to 1,200 µmol m −2 s −1 , which is about 30% to 60% 

of full sunlight (Prado and Moraes 1997). PPFD compensation point 

ranges from 10 to 50 µmol m −2 s −1 at leaf temperatures in the range of 25° 

to 30°C. Open cerrado vegetation types such as campo sujo are covered 

with a grass layer, typically 40 to 50 cm tall (chapter 6). For instance, PPFD 

measurements suggested that 5-cm-tall Kielmeyera coriacea and Dalbergia 

miscolobium would not receive enough light to reach even 50% of their 

photosynthetic capacity during the daylight period in a campo sujo site 

(Nardoto et al. 1998; Braz et al. 2000). The effects of canopy shading on 

CO 2 assimilation can become critical for seedling growth and survival in 

closed canopy vegetation such as cerradão physiognomies. Because of 

shading, species characteristic of open habitats may not be able to grow in 

closed canopy sites, whereas photoinhibition can be an important stress 

factor for young plants in fully sun-exposed habitats (Mattos 1998). Thus, 

reported differences in the range of several species along a gradient from 

campo sujo to cerradão (Goodland 1971; Goodland and Ferri 1979; chapter 

6) may reflect species differences in shade tolerance. 

Cerrado trees grow slowly in natural conditions (table 10.2; Rizzini 

1965). This is probably the result of inherently low growth rates, larger 

Table 10.2 Size of Cerrado Woody Plants in a Campo Sujo 

Formation Near Brasília, Central Brazil 

Species Plant size (cm) Plant age (years) 

Bowdichia virgilioides a 8.3 (0.5; n = 20) 2 

Dalbergia miscolobium b 23 (2.7; n = 10) 7 

Kielmeyera coriacea c 8.3 (1.8; n = 7) 5 

Qualea grandiflora a 5.3 (0.3; n = 38) 1 

19.0 (1.3; n = 4) 5 

a Established from seeds that were planted into the field site. 

b Nine-month-old seedlings were transplanted into the field site. This site burned once. At 

the time of the fire event, plants had an approximate age of 20 months. 

c Two-month-old seedlings were transplanted into the field site. 

Note: The study site (15 o 56' S, 47 o 55'W) is at the center of the cerrado region. Plant size 

is the combined length of the main stem and branches, if present. See chapter 6 for description 

of cerrado physiognomies.



carbon allocation to roots linked to low availability of nutrients, and light 

limitation by canopy shading. Thus, seedlings of woody plants develop a 

tree canopy layer in the grass matrix through a slow process. Tree canopy 

recovery after disturbance in cerrado vegetation is mainly the result of 

resprouting of existing trees and shrubs. The effects of grass root competition 

for nutrients need to be evaluated, and the light regimes along the 

gradient from campo sujo to cerradão need a better characterization. 

Research is also needed to characterize shade and high light tolerance and 

the contribution of light acclimatization to increased carbon gain for 

plants growing in different cerrado physiognomies. 


This research was supported by the Conselho Nacional de Desenvolvimento 

Científico e Tecnológico (CNPq), the Inter-American Institute for 

Global Change Research, and the Programa de Apoio a Núcleos de 

Excelência-PRONEX. I thank Raimundo P. B. Henriques and William 

Hoffmann for their helpful comments. 

REFERENCES 

Abdala, G. C., L. S. Caldas, M. Haridasan, and G. Eiten. 1998. Above and 

belowground organic matter and root: Shoot ratio in a cerrado in central 

Brazil. Braz. J. Ecol. 2:11–23. 

Alvin, P. de T. and J. E. da Silva. 1980. Comparação entre os cerrados e a 

região amazônica em termos agroecológicos. In D. Marchetti and A. D. 

Machado, eds., Simpósio sobre o Cerrado: Uso e Manejo, pp. 143–160. 

Brasília: Editora Editerra. 

Arasaki, F. R. and G. M. Felippe. 1990. Crescimento inicial de Kielmeyera 

coriacea. Cienc. Cult. 42:715–720. 

Arens, K. 1958a. Considerações sôbre as causas do xeromorfismo foliar. B. 

Fac. Fil. Ci.. Letr. Univ. São Paulo, Botânica 15:23–56. 

Arens, K. 1958b. O cerrado como vegetação oligotrófica. B. Fac. Fil. Ci.. Letr. 

Univ. São Paulo, Botânica 15:59–77. 

Barnet, Y. M. and P. C. Catt. 1991. Distribution and characteristics of root 

nodule bacteria isolated from Australian Acacia spp. Plant and Soil 

135:109–120. 

Borchert, R. 1994a. Soil and stem water storage determine phenology and distribution 

of tropical dry forest trees. Ecology 75:1437–1449. 

Borchert, R. 1994b. Electric resistance as a measure of tree water status during 

a seasonal drought in a tropical dry forest in Costa Rica. Tree Physiol. 

14:299–312.



Borchert, R. 1994c. Water status and development of tropical trees during 

seasonal drought. Trees 8:115–125. 

Braz, V. S., M. F. Kanegae, and A. C. Franco. 2000. Estabelecimento e desenvolvimento 

de Dalbergia miscolobium Benth. em duas fitofisionomias 

típicas dos cerrados do Brasil Central. Acta Bot. Bras. 14:27–35. 

Bucci, S., S. Naves, G. Cardinot, G. Nardoto, M. Bustamante, F. R. Meinzer, 

G. Goldstein and, A. C. Franco. 2002. Convergence in regulation of 

water use among Brazilian savanna woody species: The role of plant 

hydraulic architecture. Trees (in press). 

Cardinot, G. K. 1998. “Efeitos de Diferentes Regimes de Queimas nos 

Padrões de Rebrotamento de Kielmeyera Coriacea Mart. e Roupala 

Montana Aubl., Duas Espécies Típicas do Cerrado.” Master’s thesis, 


Cody, M. L. 1986. Structural niches in plant communities. In J. Diamond and 

T. Case, eds., Community Ecology, pp. 381–405. New York: Harper and 

Row. 

Eamus, D., B. Myers, G. Duff, and D. Williams. 1999. Seasonal changes in 

photosynthesis of eight savanna species. Tree Physiol. 10:665–671. 

Faria, S. M. de, H. C. de Lima, A. M. Carvalho, V. F. Conçalves, and J. I. 

Sprent. 1994. Occurrence of nodulation in legume species from Bahia, 

Minas Gerais and Espírito Santo states of Brazil. In J. I. Sprent and D. 

McKey, eds., Advances in Legume Systematics 5: The nitrogen factor, pp. 

17–23. Kew: Royal Botanic Gardens, Kew. 

Faria, S. M. de, H. C. de Lima, A. A. Franco, E. S. F. Mucci, and J. I. Sprent. 

1987. Nodulation of legume trees from south-east Brazil. Plant and Soil 

99:347–356 

Felfili, J. M., M. C. Silva Junior, T. S. Filgueiras, and P. E. Nogueira. 1998. A 

comparative study of cerrado (sensu stricto) vegetation in Central Brazil. 

Cienc. Cult. 50:237–243. 

Ferri, M. G. 1944. Transpiração de plantas permanentes dos “Cerrados.” 

B. Fac. Fil. Ci.. Letr. Univ. São Paulo, Botânica 4:161–224. 

Fichtner, K., G. W. Koch, and H. A. Mooney. 1995. Photosynthesis, storage 

and allocation. In Schulze, E.-D. and M. M. Caldwell, eds., Ecophysiology 

of Photosynthesis, pp. 133–146. Berlin: Springer-Verlag. 

Franco, A. C. 1983. “Fotossíntese e Resistência Foliar em Didymopanax 

macrocarpum.” Master’s thesis, Universidade de Brasília, Brasília, Brazil. 

Franco, A. C. 1998. Seasonal patterns of gas exchange, water relations and 

growth of Roupala montana, an evergreen savanna species. Plant Ecol. 

136:69–76. 

Goodland, R. 1971. A physiognomic analysis of the cerrado vegetation of 

Central Brazil. J. Ecol. 59:411–419. 

Goodland, R. and M. G. Ferri. 1979. Ecologia do Cerrado. São Paulo: Editora 

da Universidade de São Paulo. 

Haridasan, M. 1982. Aluminium accumulation by some cerrado native species 

of central Brazil. Plant and Soil 65:265–273.



Haridasan, M. 1988. Performance of Miconia albicans (SW.) Triana, an aluminium-accumulating 

species, in acidic and calcareous soils. Commu. In 

Soil Sci. Plant Anal. 19:1091–1103. 

Haridasan, M., P. G. Hill, and D. Russel. 1987. Semiquantitative estimates of 

Al and other cations in the leaf tissues of some Al-accumulating species 

using probe microanalysis. Plant and Soil 104:99–102. 

Haridasan, M., T. I. Paviani, and I. Schiavini. 1986. Localization of aluminium 

in the leaves of some aluminium-accumulating species. Plant and 

Soil 94:435–437. 

Hoffmann, W. A. 1996. The effects of cover and fire on seedling establishment 

in a neotropical savanna. J. Ecol. 84:383–393. 

Hoffmann, W. A., F. A. Bazzaz, N. J. Chatterton, P. A. Harrison, and R. B. 

Jackson. 2000. Elevated CO 2 enhances resprouting of a tropical savanna 

tree. Oecologia 123:312–317. 

Hogan, K. P., A. P. Smith, and M. Samaniego. 1995. Gas exchange in six tropical 

semi-deciduous forest canopy tree species during the wet and dry season. 


Jackson, P. C., F. C. Meinzer, M. Bustamante, G. Goldstein, A. Franco, P. W. 

Rundel, L. Caldas, E. Igler, and F. Causin. 1999. Partitioning of soil water 

among tree species in a Brazilian Cerrado ecosystem. Tree Physiol. 

19:717–724. 

Johnson, C., A. C. Franco and L. S. Caldas. 1983. Fotossíntese e resistência 

foliar em espécies nativas do cerrado: Metodologia e resultados preliminares. 

Rev. Bras. Bot. 6:91–97. 

Kanegae, M. F., V. da S. Braz, and A. C. Franco. 2000. Efeitos da disponibilidade 

sazonal de água e luz na sobrevivência de Bowdichia virgilioides 

em duas fitofisionomias típicas dos cerrados do Brasil Central. Rev. Bras. 

Bot. 23:457–466. 

Knoop, W. T. and B. H. Walker. 1985. Interactions of woody and herbaceous 

vegetation in a Southern African savanna. J. Ecol. 73:235–254. 

Kozovits, A. R. 1997. “Assimilação de Nitrogênio em Espécies Lenhosas do 

Cerrado.” Master’s thesis, Universidade de Brasília, Brasília, Brazil. 

Machado, J. W. B. 1985. “Acumulação de Alumínio em Vochysia thyrsoidea 

Pohl.’’ Master’s thesis, Universidade de Brasília, Brasília, Brazil. 

Machado, J. W. B. 1990. “Relação Origem/solo e tolerância à saturação 

hídrica de Copaifera langsdorffii Desf.’’ Ph.D. thesis, Universidade de 

Campinas, Campinas, Brazil. 

Maia, J. M. F. 1999. “Variações Sazonais das Relações Fotossintéticas, Hídricas 

e Crescimento de Caryocar brasiliense e Rapanea guianensis em um 

Cerrado sensu stricto.’’ Master’s thesis, Universidade de Brasília, Brasília, 

Brazil. 

Mattos, E. A. 1998. Perspectives in comparative ecophysiology of some 

Brazilian vegetation types: leaf CO 2 and H 2 O exchange, chlorophyll a 

fluorescence and carbon isotope discrimination. In Scarano, F. R. and A. 

C. Franco, eds., Ecophysiological Strategies of Xerophytic and Amphibi-



ous Plants in the Neotropics. Series Oecologia Brasiliensis, vol. 4, pp. 

1–22. Rio de Janeiro: Universidade Federal do Rio de Janeiro. 

Mattos, E. A. de, F. Reinert, and J. A. P. V. de Moraes. 1997. Comparison of 

carbon isotope discrimination and CO 2 and H 2 O exchange between the 

dry and the wet season in leaves of several cerrado woody species. Rev. 


Medeiros, R. A. de and M. Haridasan. 1985. Seasonal variations in the foliar 

concentrations of nutrients in some aluminium accumulating and nonaccumulating 

species of the cerrado region of central Brazil. Plant and 

Soil 88:433–436. 

Medina, E. and B. Bilbao. 1991. Significance of nutrient relations and symbiosis for the 

competitive interaction between grasses and legumes in tropical savannas. In Esser, 

G. and G. Overdieck, eds., Modern Ecology: Basic and Applied Aspects, pp. 

295–319. Amsterdam: Elsevier Science Publications. 

Medina, E. and M. Francisco. 1994. Photosynthesis and water relations of 

savanna tree species differing in leaf phenology. Tree Physiol. 14: 

1367–1381. 

Medina, E., V. Garcia, and E. Cuevas. 1990. Sclerophylly and oligotrophic 

environments: relationships between leaf structure, mineral nutrient content, 

and drought resistance in tropical rain forests on the upper Río 

Negro region. Biotropica 22:51–64. 

Medina, E. and J. F. Silva. 1990. Savannas of northern South America: A 

steady state regulated by water-fire interactions on a background of low 

nutrient availability. J. Biogeogr. 17:403–413. 

Meinzer, F. C., G. Goldstein, A. C. Franco, M. Bustamante, E. Igler, P. Jackson, 

L. Caldas, and P. W. Rundel. 1999. Atmospheric and hydraulic limitations 

on transpiration in Brazilian cerrado woody species. Funct. Ecol. 

13:273–282. 

Melo, J. T. de. 1999. “Resposta de Mudas de Algumas Espécies Arbóreas 

Nativas do Cerrado aos Nutrientes em Latossolo Vermelho Escuro.” 

Ph.D. thesis, Universidade de Brasília, Brasília, Brazil. 

Miranda, A.C., H. S. Miranda, J. Lloyd, J. Grace, R. J. Francey, J. A. Mcintyre, 

P. Meir, P. Riggan, R. Lockwood, and J. Brass. 1997. Fluxes of carbon, 

water and energy over Brazilian cerrado: An analysis using eddy 

covariance and stable isotopes. Plant Cell Environ. 20:315–328. 

Moraes, J. A. P. V. de, S. C. J. G. de A. Perez, and L. F. Carvalho Jr. 1989. 

Curso diário do potencial de água e resistência estomática em plantas de 

cerradão. Ann. Miss. Bot. Gard. 27:13–23. 

Moraes, J. A. P. V. and C. H. B. A. Prado. 1998. Photosynthesis and water 

relations in cerrado vegetation. In Scarano, F. R. and A. C. Franco, eds., 

Ecophysiological Strategies of Xerophytic and Amphibious Plants in the 

Neotropics. Series Oecologia Brasiliensis, vol. 4, pp. 45–63. Rio de 

Janeiro: Universidade Federal do Rio de Janeiro. 


Cerrado.’’ Ph.D. thesis, Harvard University, Cambridge, MA, U.S.A.



Nardoto, G. B., M. P. Souza, and A. C. Franco. 1998. Estabelecimento e 

padrões sazonais de produtividade de Kielmeyera coriacea (Spr) Mart. 

nos cerrados do Planalto Central: Efeitos do estresse hídrico e sombreamento. 

Rev. Bras. Bot. 21:313–319. 

Naves-Barbiero, C. C., A. C. Franco, S. J. Bucci, and G. Goldstein. 2000. Fluxo 

de seiva e condutância estomática de duas espécies lenhosas sempre-verdes 

no campo sujo e cerradão. Rev. Bras. Fisiol. Vegetal 12: 119–134. 

Paulilo, M. T. S., G. M. Felippe, and J. E. Dale. 1993. Crescimento e desenvolvimento 

inicial de Qualea grandiflora. Rev. Bras. Bot. 16:37–46. 

Perez, S. C. J. G. de A. and J. A. P. V. de Moraes. 1991. Determinações de 

potencial hídrico, condutância estomática e potencial osmótico em espécies 

dos estratos arbóreo, arbustivo e herbáceo de um cerradão. Rev. 


Prado, C. H. B. A. and J. A. P. V. de Moraes. 1997. Photosynthetic capacity 

and specific leaf mass in twenty woody species of Cerrado vegetation 

under field conditions. Photosynthetica 33:103–112. 

Prado, C. H. B. A., J. A. P. V. de Moraes. and E. A. de Mattos. 1994. Gas 

exchange and leaf water status in potted plants of Copaifera langsdorffii: 

1. Responses to water stress. Photosynthetica 30:207–213. 

Rawitscher, F. 1948. The water economy of the vegetation of the “campos 

cerrados” in southern Brazil. J. Ecol. 36:237–268. 

Reich, P. B., D. S. Ellsworth, and C. Uhl. 1995. Leaf carbon and nutrient 

assimilation in species of differing successional status in an oligotrophic 

Amazonian forest. Funct. Ecol. 9:65–76. 

Reis, M. J. de O. dos. 1999. “Eficiência Micorrízica em Plantas Nativas do 

Cerrado.” Ph.D. thesis, Universidade de Brasília, Brasília, Brazil. 



Rizzini, C. T. and E. P. Heringer. 1961. Underground organs of plants from 

southern Brazilian savannas, with special reference to the xylopodium. 

Phyton 17:105–124. 

Rizzini, C. T. and E. P. Heringer. 1962. Studies on the underground organs of 

trees and shrubs from some Southern Brazilian savannas. An. Acad. Bras. 

Ciênc. 34:235–251. 



Sarmiento, G., G. Goldstein, and F. Meinzer. 1985. Adaptive strategies of 

woody species in neotropical savannas. Biol. Rev. 60:315–355. 

Sassaki, R. M. and G. M. Felippe. 1992. Remoção dos cotilédones e desenvolvimento 

inicial de Dalbergia miscolobium. Rev. Bras. Bot. 15:5–16. 

Sassaki, R. M. and G. M. Felippe 1998. Response of Dalbergia miscolobium 

Benth. seedlings, a cerrado tree species, to mineral nutrient supply. Rev. 

Bras. Bot. 21:55–72. 

Sassaki, R. M., Machado, E. C., Lagôa, A. M. M. A., and G. M. Felippe. 

1997. Effect of water deficiency on photosynthesis of Dalbergia mis-



colobium Benth., a cerrado tree species. Rev. Bras. Fisiol. Vegetal 9: 

83–87. 

Sprent, J. I., I. E. Geoghegan, and P. W. Whitty. 1996. Natural abundance of 

15 N and 13 C in nodulated legumes and other plants in the cerrado and 

neighbouring regions of Brazil. Oecologia 105:440–446. 

Thomazini, L. I. 1974. Mycorrhiza in plants of the “Cerrado”. Plant and Soil 

41:707–711. 

Tilman, D. 1982. Resource Competition and Community Structure. Princeton, 

NJ: Princeton University Press. 

Walker, B. H. and I. Noy-Meir. 1982. Aspects of the stability and resilience 

of savanna ecosystems. In B. J. Huntley and B. H. Walker (eds), Ecology 

of Tropical Savannas, pp. 577–590. Berlin: Springer-Verlag. 

Williams, R. J., B. A. Myers, W. J. Muller, G. A. Duff, and D. Emaus. 1997. 

Leaf phenology of woody species in a north Australian tropical savanna. 


Zotz, G. and K. Winter. 1996. Diel patterns of CO 2 exchange in rainforest 

canopy plants. In Tropical Forest Plant Ecophysiology, S.E. Mulkey, R. 

L. Chazdon and A. P. Smith (eds), pp. 89–113. New York: Chapman and 

Hall.



11 

Lepidoptera in the Cerrado 

Landscape and the Conservation 

of Vegetation, Soil, and 

Topographical Mosaics 

Keith S. Brown Jr. and David R. Gifford 1 

Many chapters in this book emphasize the complexity, 

antiquity, and singularity of the biological systems of the Central Brazil 

Plateau. The widespread misconception that these mixed-savanna systems 

are species-poor has been definitively set aside by these chapters, as well 

as by those in a recent book on gallery forests in the region (Rodrigues 

and Leitão-Filho 2000). The only poverty now apparent is that of our data 

and sampling of the many profoundly different biological systems that 

occur in bewilderingly complex mosaics throughout the Cerrado Biome, 

often determined by varying soil characteristics (chapter 2) and water 

availability (chapter 6, see also Oliveira-Filho and Ratter 1995, 2000; 

Castro et al. 1999; Oliveira-Filho and Fontes 2000). 

Many authors have also noted the difficulty in biological inventory, 

monitoring, and conservation priority-setting in the region, as a result of 

the great variation of environments in time and space and the resulting 

unstable ecological mosaics (chapter 18). Such heterogeneous landscapes 

defy analysis or classification by large mobile mammals like us. Smaller 

animals and most plants are very sensitive to immediate environmental 

1 Dr. David Gifford was working in the Ecology Program of the University of 

Brasília when he passed away in June 1981, having left parts of this paper as an 

unpublished manuscript (1979). 

201


202 the animal community 

factors, however, and thus can be used effectively as indicators of system 

structure, richness, and history (see Brown 1991, 1997, 2000; Brown and 

Freitas 2000). How can we “get a handle’’ on the diversity and importance 

of a given site in order to describe and characterize its ecology and 

compare it with others for management and conservation of its biological 

diversity? 

This chapter will examine various groups of specialized plant-feeding 

Lepidoptera (moths and butterflies) of the Planalto region in central 

Brazil, sampled throughout this area dominated by various types of cerrado 

vegetation (chapter 6) and at its peripheries (see fig. 11.1), to evaluate 

their usefulness as indicators both of the history and biogeographical 

subdivision of the region and of the variable community structure and 

species richness at the landscape and local levels. The objectives will be to 

answer the following three questions: 

a. Do Lepidoptera show clear endemism in the cerrado region, with 

biogeographical divisions or transitions within the region or at its 

peripheries, similar to those seen in woody plants, eventually referable 

to broad historical factors acting on the landscape? 

b. Which local environmental factors have the greatest effects on the 

structure and richness of the Lepidopteran community in a given 

site? 

c. Are the broad or local patterns revealed in Lepidoptera coherent 

enough to qualify as good indicators for landscape evaluation, conservation, 

and management in the region? 

DATABASE: COLLECTION AND ANALYSIS 

The data (regional and site lists of species) for answering these questions 

were compiled from many sources, both published (Brown and Mielke 

1967a, 1967b, 1968; Brown 1987a; Mielke and Casagrande 1998; 

Camargo and Becker 1999) and unpublished (Brown and Mielke 1972; 

Gifford 1979; Motta 2002; Callaghan and Brown in preparation). 

Acceptably complete lists (at least 50% of the expected community in 

smaller groups, or 24 species in larger groups, with two exceptions) were 

obtained by repeated sampling in 23 to 33 sites (see appendix table) of 

species in four readily encountered and recognized groups of Lepidoptera: 

three of Nymphalid butterflies (bait-attracted groups except Satyrinae— 

126 taxa, Heliconiini—25 taxa, and Ithomiinae—43 taxa) and saturniid 

moths (169 species; see also Camargo and Becker 1999). Less complete 

data, useful for regional lists, were obtained simultaneously for sphingid


Lepidoptera in the Cerrado and Conservation 203 

Figure 11.1 Biogeographical division of the Brazilian cerrado region (see 

fig. 6.4). Sampling sites 1–45 for Lepidoptera are in order of increasing latitude, 

with 11 Federal District sites in the enlarged rectangle at top (see appendix 

table). S = Rio Suiá-Missu, G = northern sector of the “Mato Grosso de 

Goiás.’’ The “Araguaia subspecies-endemic center’’ (AR, encircled by a heavy 

line) follows Brown (1979, 1982a, 1982b, 1987b, 1987c). Black areas within 

AR are regions of rich mesotrophic and eutrophic soils. Stars indicate four 

outlying limits of the distribution of a suite of cerrado-endemic stenotopic 

species of Hypoleria and Pseudoscada (sedentary transparent Ithomiinae confined 

to superhumid habitats) to the south, southeast, and northwest of the 

region. 

moths and many other butterfly groups. Butterflies, most common, 

diverse, and easily sampled from February to July, were censused with 

transect walks and bait-traps (Brown 1972; Brown and Freitas 2000). A 

permanent notebook was used to record lists of all species observed in a 

given day and place, as well as observation conditions including observers, 

effort, maps, and biological data. Moths were attracted to and 

identified on a light-colored solid surface or screen reflecting near-UV



light (“black’’ 15-W fluorescent tubes or 250-W mixed mercury-vapor 

lamps); they were best inventoried at the beginning to middle of the rainy 

season (September to January). 

These data were used for quantitative analyses and comparison with 

geographical and ecological factors, whose purpose was to reveal patterns 

of species occurrence and distribution in relation to environment, landscape, 

and the fauna of adjacent regions (see table 11.1, fig. 11.1). Since 

estimates of the abundance of species were not fully standardized among 

the years and sites, the records were kept in binary form (recorded/ 

unrecorded in each site). 

Lepidoptera in the cerrado region, like most humidity-sensitive 

insects, tend to be concentrated in gallery forests and other dense vegetation 

near water (see Brown 2000). There are also many endemic species 

of more open vegetation, however. The UV light source is invariably 

placed in an open area for nocturnal census and attracts adults from all 

habitats within many hundreds of meters. Flowers and larval host plants 

on borders and in open vegetation, helpful in the attraction and maintenance 

even of shade-loving butterflies, were also regularly monitored in 

many sites, along with transects in larger savanna areas. 

The high levels of natural disturbance in the cerrado landscape frustrated 

attempts at objective evaluation of the degree of short- and longterm 

anthropic disturbance in many sites, especially those censused over 

many years during various landscape reorganization episodes. 

The site lists were analyzed by PC-ORD, STATISTICA (StatSoft 1995), 

FITOPAC (Shepherd 1995), and CANOCO (Ter Braak 1987–1992), in 

search of patterns of biogeographical distribution, and for discovery of 

principal local environmental factors acting on the Lepidoptera communities 

in each site and in the region (fig. 11.1; appendix table). 

BIOGEOGRAPHY OF CERRADO LEPIDOPTERA, 

ENDEMISM, AND HISTORY 

Distribution analysis of a variety of cerrado Lepidoptera showed that the 

proportions of endemic, widespread, and peripheral-affinity species vary 

widely among groups (table 11.1). Of the 802 taxa analyzed, over 33% 

(including 70% of the wide-flying Sphingidae) are widespread in South 

America, while only 19% are endemic to the Cerrado region. Species 

showing primary affinities with the dry areas to the northeast (caatinga) 

and southwest (chaco) are very few (only 2.6%, two-thirds of these Saturniidae), 

though a number of species native to these regions have been 

found in some collections on their borders (see table 3 of Camargo and



Table 11.1 Biogeographical Affinities of Cerrado Species 

in Various Lepidopteran Groups 

REGIONS: SE/S NW/N SW/S NE/E Endemic to All Areas 

Wide- Cerrado 

spread Atlantic Amazon Chaco Caatinga (%) Total 

Saturniidae 

Arsenurinae 5 [1] 7 [6] 2 [3] — 4 (14) [10] 18 

Ceratocampinae 18 [1] 11 [2] 6 [3] 1 [1] 1 8 (15) [7] 45 

Hemileucinae 12 [3] 20 [9] 17 [1] 3 1 15 (19) [13] 68 

Saturniinae 5 3 — — — — 8 

Saturniidae (total) 40 [5] 41 [17] 25 [7] 4 [1] 2 27 (16) [30] 139 

Sphingidae (total) 67 11 2 2 — 9 (10) 91 

Papilionidae (total) 4 [9] 10 [9] 2 [2] [1] 10 (21) [21] 26 

Pieridae (total) 17 10 [2] 1 — — 3 (9) [2] 31 

Riodininae (total) 40 39 51 — — 42 (24) 172 

Myrmecophilous 14 9 8 — — 21 (40) 52 

genera 

Nonmyrmeco- 26 30 43 — — 21 (18) 120 

philous genera 

Nymphalidae 

Libytheana-Dana.- 4 [5] 13 [3] — — 15 (38) [8] 32 

Ithomiinae 

Morph.-Brassol.- 40 21 [1] 12 — — 23 (36) [1] 96 

Satyrinae 

Apat.-Colob.- 10 11 [3] 3 — — 2 (7) [3] 26 

Cyrest.-Limen. 

Charaxinae- 34 [1] 17 5 — [1] 14 (31) [2] 70 

Biblidinae 

Nymphalinae- 10 [1] 21 [9] 3 [1] — 7 (29) [11] 41 

Heliconiinae 

Nymphalidae (total) 98 [7] 83 [16] 23 [1] [1] 61 (21) [25] 265 

Grand total: 266 [21] 194 [44] 104 [10] 6 [3] 2 152 (19) [78] 724 

Percent of total: 33.2% 26.8% 18.5% 2% 0.6% 19% 100% 

Sources: Brown and Mielke 1972 (all groups); Gifford 1979 (Heliconiini and Ithomiinae); Tyler et al. 

1994 (Papilionidae); Camargo and Becker 1999 (Saturniidae); Callaghan and Brown in preparation 

(Riodininae). 

Note: Bracketed numbers include marginal species invading the Cerrado from the indicated adjacent 

biomes. 

Becker 1999). The primary link, as with plants (chapter 6), is with the 

Atlantic forests to the southeast (26.8%, seen in all groups) (table 11.1), 

followed by the Amazonian forests to the northwest (18.5% of taxa, 

almost a third peripheral, found only in central Mato Grosso sites 3 and 

5 and the region marked S in fig. 11.1). 

Progressively higher levels of endemism in the cerrado region are



shown by well-marked species and geographical subspecies of Papilionidae 

(21% endemic, or 47% for the Troidini alone), Riodininae (24%, or 40% 

for myrmecophilous species), Satyrinae (27%), Biblidinae ( = Eurytelinae 

auctt.) (27%), and Ithomiinae (43%, not considering the four widespread 

Danainae and one Libytheine) (table 11.1). When the mostly grass-feeding 

satyrs and hesperiine skippers (not analyzed here) are better studied, 

they should show still higher endemism, considering the great wealth of 

potential host species in the cerrado (chapter 7). The large proportion of 

endemic myrmecophilous Riodininae (40%), with many still awaiting 

description, may be related to the specific and intense interactions between 

ants and plants in the cerrado vegetation (chapter 15); endemism is much 

lower in species with non-myrmecophilous juveniles (18%). 

Does this endemism have primarily historical or ecological roots? As 

with most basic ecological questions, the most likely answer is “yes’’— 

that is, both are important and necessary. Geomorphologic analysis 

(chapter 2) and paleopollen records (Ledru 1993; chapter 3) show that 

the region has had a long history of landscape changes (especially in its 

contacts with the Amazon Basin to the north), but a surprising apparent 

stability of interlinked forest/savanna matrices during the climatic fluctuations 

of the late Pleistocene. The very complex geology and sharp relief 

of the central plateau (chapter 2) probably always included gallery and 

headwater swamp forests, providing refuge and connectivity (especially 

southward with the Atlantic forests) to diversified forest biotas even during 

the least favorable climatic periods. Orographically and edaphically 

determined deciduous and semideciduous forests (including mesotrophic 

headwater woods and spring-fed copses at the base of escarpments) are 

predictably green at least at certain seasons of each year (Ratter et al. 

1973, 1978) and were probably also widely distributed at all times, preserving 

endemic taxa adapted to this mosaic landscape. 

An “Araguaia endemic center’’ for subspecies of Heliconiini and 

Ithomiinae, recognized by Brown (1979, 1982a, 1982b, 1987b, 1987c; 

see also Gifford 1979), is outlined in figure 11.1 (AR, one-third of maximum 

isocline for corrected endemism based on eight taxa, reinforced with 

the 18 taxa included today, table 11.1). This center covers a large region 

characterized by cerrados mixed with forests in a complex dystrophic/ 

eutrophic soil matrix (Ratter et al., 1973, 1978). Gifford’s observations 

in the upper Xingu (Suiá-Mussu, Serra do Roncador; marked S in fig. 

11.1), along the Araguaia (site 1), and in the northern part of the “Mato 

Grosso de Goiás’’ (12, 17, 22, G in fig. 11.1) show many species in these 

two groups as recent marginal invaders from the Amazonian forests, mixing 

with the usual endemic fauna found in greater abundance farther



southeast, often on rich soils (black areas in fig. 11.1). The relatively stable 

topographical, soil, and vegetation mosaics in the Brazilian Planalto 

could have conserved the adapted and differentiated “Araguaia’’— 

endemic Lepidoptera and plants through all the ecological cataclysms that 

so greatly affected the lowland sedimentary regions and their biotas in the 

Neotropics. As a typical example, five well-differentiated, stenotopic 

transparent Ithomiinae species in the genera Hypoleria and Pseudoscada 

are ubiquitous in interconnected humid gallery forests (Brown 2000) with 

deep shade throughout the cerrado landscape all the way to its various 

peripheries (see fig. 11.1), along with many other similarly restricted and 

endemic species of insects, plants, birds, and reptiles (chapters 6, 7, 12–14; 

Rodrigues and Leitão-Filho 2000). All of these animals are excellent indicators 

of complex forest/cerrado mosaic landscapes in the region today, 

and suggest their relative continuity in the past, not as climatically induced 

and edaphically defined “forest islands’’ or refuges (as proposed for the 

more level Amazon Basin, Brown 1979), but as interlinked and geomorphologically 

stabilized vegetation mosaics. These mosaics could conserve 

and select both forest and savanna species over time (Gifford 1979). 

LANDSCAPE HETEROGENEITY 

IN THE CERRADO REGION 

The sites (fig. 11.1) were compared for similarity of their faunal lists for 

each of the four censused Lepidoptera groups (appendix table). The 

results (see fig. 11.2) support a landscape- and vegetation-based grouping 

in the saturniid moths (fig. 11.2A) but a geographic base for community 

composition in the butterflies (fig. 11.2B-D). Thus, sites in a given biogeographical 

subregion (fig. 11.1; Ratter et al. 1997), even those with different 

vegetation types, tend to cluster together in the three butterfly 

analyses (fig. 11.2B-D); note the proximity of nearby site numbers in the 

dendrograms—the average difference in numbers between pairs and triads 

for all three groups is 6.6. In contrast, adjacent sites are scattered over 

the dendrogram in the Saturniidae analysis (fig. 11.2A); note the lack of 

clusters of proximal site numbers (average difference in pairs and triads 

13.3), with, however, a homogeneity of vegetation in two of the three 

large clusters. This indicates a strong association of the Planalto saturniid 

moths with specific landscapes and vegetation (Ce, FCe, SDF) rather than 

with geographical subregions; the light-sampling procedure would bring 

them in from their typical habitats within the broad local vegetation 

mosaic (see also below, and Camargo 1999; Camargo and Becker 1999).


Figure 11.2 Dendrograms showing similarity among Lepidoptera communities 

in the cerrado region and its peripheries. Linkage (percentage of maximum) 

is by Ward’s minimum-variance clustering, favoring formation of pairs 

and triads; similarity is as 1-Pearson’s r. The four groups have different sets 

of sites. Note the strict vegetation grouping in Saturniidae moths (A), but 

more biogeographical grouping (proximate site numbers) in butterflies (B-D). 

See appendix table for vegetation codes, site characteristics, and numbers of 

species in each group recorded in each site.



The butterflies, more restricted to microhabitats and food plants, are 

recorded along the specific transects chosen for census. They will thus 

reflect a biogeographical affinity smaller in scale than the saturniid moths’ 

affinity with general vegetation or landscape (fig. 11.1). 

Within each subregion (fig. 11.1), the presence of a species in a site is 

most likely to be determined by various factors in the local environment, 

often a mosaic of cerrado and forest physiognomies including galleries along 

permanent watercourses. This is evident in Principal Components and 

CANOCO analyses of the sites and their butterfly faunas (see tables 11.2 

and 11.3 and fig. 11.3). The PCA (with latitude and longitude removed) 

grouped the various sites along major axes reflecting not only geography 

through broad climatic patterns (fig. 11.3A), but also a wide variety of factors 

of topography, soil, and vegetation, especially their complex mosaics 

(table 11.2), thereby revealing each site’s strong relationship to landscape 

heterogeneity. Even with the inclusion of many factors (especially edaphic) 

Table 11.2 Statistics of the PCA Analysis for 14 

Environmental Factors in 45 Sites in the Brazil Planalto 

Environmental Factors Axis 1 Axis 2 Axis 3 

% of variation explained 25.0% 14.7% 13.8% 

Mean altitude –.360 .270 –.399 

Topographical Surface (1–5) .368 –.119 .173 

General topography .130 .333 –.154 

Permanent water availability –.155 .028 .327 

Vegetation category .368 –.125 –.252 

Vegetation mosaic .169 .089 .240 

Bamboos .338 .081 .049 

Soil category .213 –.222 –.493 

Soil mosaic .144 .512 –.142 

Soil bases (fertility) .404 .075 –.231 

Annual rainfall –.242 .268 .029 

Length of the dry season (days) .189 .430 –.076 

Temperature (yearly average) .305 .099 .452 

Temperature variation –.004 –.434 –.172 

“Flavors” of each axis Soil bases, Soil mosaic, Soil type, 

vegetation, climate (dry, temperature, 

altitude temp.variat.) altitude 

Note: Data taken from the appendix table, not including latitude or longitude. The most 

important factors are in bold. Axis 4 (vegetation mosaic, water) explained a further 10.6%; 

Axis 5 (topography, bamboos), 8.3% of the variation. See figure 11.3A for site ordination 

along the first two axes, with maximized scatter of points representing different types of 

environment. Note significant inclusion of variation in soil, vegetation, and topographical 

factors in the first three axes. This would suggest a large variation of microhabitats within 

and between cerrado sites.



Table 11.3 Statistics of the Principal Vectors in the CCA Analysis 

Factors F P % 

(1) Saturniidae (153 taxa, 28 sites) (Fig. 3B) 

Mean altitude 1.01 0.001 19 

Temperature variation 1.72 0.001 16 

Annual rainfall 1.49 0.004 14 

Vegetation category 1.38 0.022 12 

(Soil mosaic —-not significant) 1.24 0.097 11 

Total explained (four significant 61 

vectors) 

(2) Heliconiini (25 taxa, 23 sites) 

Temperature (annual mean) 4.07 0.001 26 

Bamboos 2.59 0.004 16 

Vegetation mosaic 2.39 0.006 13 


Soil mosaic 1.81 0.044 9 

Total explained (five significant 73 

vectors) 

(3) Bait-attracted Nymphalidae (126 taxa, 24 sites) (Fig. 3C) 



Length of the dry season 1.63 0.051 10 



Total explained (five significant 57 

vectors) 

(4) Ithomiinae (43 taxa, 33 sites) (Fig. 3D) 




Length of the dry season 1.84 0.014 7 

Vegetation category 1.84 0.015 7 



General topography 1.54 0.056 5 

Total explained (eight significant 65 

vectors) 

Note: The data represent 14 environmental factors (latitude/longitude eliminated) and four 

Lepidoptera communities in 45 sites in the Brazilian cerrado region. 

in the first three PCA axes, together they explained only about half the 

variations among the sites (table 11.2). 

When the species of the four Lepidoptera groups were directly compared 

with various local geoecological variables (table 11.3, fig. 11.3) by 

Canonical Community Ordination, the most influential ecological factors


Figure 11.3 Site ordination by Principal Components Analysis (A) of environmental 

variables and by Canonical Community Ordination of three Lepidoptera 

groups (B-D, see table 11.3). The most significant relations (other 

than latitude and longitude, not ecological) are expressed as vectors in B (Saturniidae), 

C (bait-attracted Nymphalidae), and D (Ithomiinae; Heliconiini are 

very similar) (see tables 11.2, 11.3). The appendix table gives the 45 site 

names, environmental characteristics, and species richness of each group. In 

15 sites, only one of the four groups was sampled, typically (10) Saturniidae 

(Camargo and Becker 1999). The second axis in B and the first in D have been 

inverted, to place the temperature variation vector always to the upper right 

and site 5 in the upper left quadrant. The variable positions of some vectors 

in B-D is due to the smaller number of significant factors in site ordination 

for each lepidopteran group, in accord with their different environmental 

responses.



on the community in each site (also after removing latitude and longitude) 

invariably included climate (a regional phenomenon), altitude (a 

restricted topographical factor, highly correlated with lower temperature), 

and vegetation (especially its fine mosaic, an intensely local factor; 

soil mosaic was also very important, see table 11.3). All these factors 

helped to explain the community structure and composition (table 11.3), 

while reflecting the geographical subregions and contributing to the principal 

axes defining the environment in each site (table 11.2, fig. 11.3A). 

Their combination can be expressed as a single composite term, environmental 

heterogeneity, that effectively determines Lepidoptera community 

composition and richness throughout the region. This is hardly a surprising 

result, given the predominance of landscape mosaics, predictability of 

a marked dry season, continued presence of ever-humid swamp and 

gallery forests, and strong ecological specializations of the animals in the 

region. This is true even at a fine scale in the nuclear cerrado region, as 

seen in the elongation of the cluster of positions of neighboring sites in 

the Federal District and the Mato Grosso de Goiás (fig. 11.1; open or 

black triangles in fig. 11.3) along the vegetation mosaic and altitude vectors 

in the canonical analyses, and in the separation of sites outside this 

“central’’ region from this cluster. Much less scattering is seen along the 

climatic vectors (fig. 11.3). 

LEPIDOPTERA AND THE 

“CONSERVATION LANDSCAPE’’ 

The close relationship of the structure of the Lepidoptera communities 

with the complex landscape mosaics in the Cerrado region (fig. 11.3, 

tables 11.2, 11.3), and their typical association with ecotones and gallery 

forests (Pinheiro and Ortiz 1992; Brown 2000), give a clear direction for 

effective conservation of the widest range of landscapes and genetic variation 

in the region. This can start with rigorous protection of the landscape 

factors that have always been designated as reserves by Brazilian 

law: water-springs, marshes, riparian forests, and areas of steep or complex 

topography. In the cerrado, these will include most of the vegetation 

types and the species endemic to the region. A plan for a “conservation 

landscape’’ occupied by humans and their economic activities should also 

take other humid areas into account (such as depressions on high 

plateaus, with many endemic monocots and their herbivores), as well as 

any topographic factors (such as breaks between the geomorphic surfaces; 

see chapter 2) that create complex mosaics. Fragile but fertile Surface-3



mesotrophic soils and associated semideciduous forest mosaics are likely 

to be important also (Ratter et al. 1973, 1978, 1997). Indeed, a Cerrado 

landscape without mesotrophic, headwater, and gallery forests would be 

much poorer in insect species, like most of the hydrologic savannas in the 

Amazon or Llanos regions (Ratter et al. 1997). 

Lepidoptera are easily monitored in the Cerrado landscape. In any season, 

butterflies are attracted in the morning to flowers and fruits on borders, 

and later in the day to resources or baits within the forest (Brown 

1972; Brown and Freitas 2000). Moths can be called from afar by near- 

UV light in the open areas and identified when they sit down on the illuminated 

light-hued sheet, net, or wall (as in the microwave tower 

blockhouses, in Santa Maria, Anápolis, Cilu, and Ponte Funda among others), 

especially on foggy or moonless nights in the spring and summer (see 

Camargo 1999). The richness, composition, and mosaic structure of the 

local biota can thus be continuously censused and monitored as landscape 

conversion or effective use increases. Each site will have its characteristic 

community (note the scatter of points in figure 11.3), closely tied with local 

edaphic and vegetation mosaics and water availability (fig. 11.3, tables 

11.2, 11.3; Brown 2000) and different from those in other sites (see also 

Camargo 1999; Camargo and Becker 1999). When typical species in a 

local community disappear, it may be suspected that the use of the landscape 

is no longer sustainable. If they continue to be present, their habitats 

will probably remain adequately conserved, unless systemic poisons are 

introduced into the landscape (even excess fertilizer can greatly change the 

vegetation along watercourses, and pesticides affect all animals). 

In this way, the “bewildering complexity’’ of soil and vegetation 

mosaics in the Planalto can continue to be a stimulating source of 

resources, both economic and intellectual, for humans during and beyond 

the present occupation of the region by agroindustry. 

PRIORITIES FOR CONTINUING RESEARCH 

The cerrado region has been extensively transformed by large-scale cattle 

ranching and industrial plantations of soybeans, maize, and other crops 

in the past 30 years (chapter 5). Most of the data on Lepidoptera used 

here were gathered before. Recent visits suggest that the flora and fauna 

continue to persist, at least in steeper areas and in reserves of various sorts. 

It is still necessary to evaluate the effects of anthropic disturbance in areas 

adjacent to agriculture, especially those with open native savanna vegetation; 

the diversity can be increased (due to edge effects) but is often



decreased due to agrochemicals and leveling, greatly affecting small 

marshes and other open microhabitats. 

The data presented here need to be updated both in preserved areas 

and in those adjacent to human occupation, whose effects need to be recognized 

and separated from those of fires, excess seasonality (as in 

1998–2000, with greatly reduced rainfall), population fluctuations of 

host plants, parasites, and predators, and the long-term dynamics of natural 

savannas. New censuses of Lepidoptera and other insects, more complete 

and effective than past data, could help in the formulation of 

effective management protocols for the singular landscapes characteristic 

of the central Brazil plateau, a rich biological resource whose description 

and recognition have lagged behind those for forests and more homogeneous 

vegetation in other parts of the Neotropics. The baselines established 

for Lepidoptera previous to extensive human occupation should be 

useful in the design and monitoring of sustainable programs for human 

use of the region and its resources. 


We are grateful to J. Ratter and A. Oliveira-Filho for orientation on the 

analysis of cerrado vegetation, and to A. V. L. Freitas for substantial contributions 

in data analysis. DRG received support from the Royal Society 

and the Royal Geographical Society in the Xavantina-Cachimbo expedition 

(1967–1968), the Royal Society and the CNPq (1971), the Brazilian 

Academy of Sciences (1972, 1978–1979), the Universities of Edinburgh 

and Brasília, FINEP (field study, 1976–1979), and the IBDF (Xingu, 1978; 

Bananal, 1979). Drs. G. P. Askew and R. F. Montgomery also participated 

in these field projects. Support from the BIOTA/FAPESP program contributed 

to the data analysis in 2000; data collection by KSB in the 1960s 

and 1970s was supported by many of the above agencies and a fellowship 

from the CNPq (“Pesquisador-Conferencista’’). O. H. H. Mielke, H. and 

K. Ebert, S. Nicolay, and N. Tangerini made significant contributions to 

the database, and R. Marquis and P. S. Oliveira gave many suggestions 

for the text and figures. 

REFERENCES 

Brown Jr., K. S. 1972. Maximizing daily butterfly counts. J. Lep. Soc. 

26:183–196.



Brown Jr., K. S. 1979. Ecologia Geográfica e Evolução nas Florestas Neotropicais. 

Campinas: Universidade Estadual de Campinas, Brazil. 

Brown Jr., K. S. 1982a. Historical and ecological factors in the biogeography 

of aposematic Neotropical Lepidoptera. Amer. Zool. 22:453–471. 

Brown Jr., K. S. 1982b. Paleoecology and regional patterns of evolution in 

Neotropical butterflies. In G.T. Prance, ed., Biological Diversification in 

the Tropics, pp. 255–308. New York: Columbia University Press. 

Brown Jr., K. S. 1987a. Zoogeografia da região do Pantanal Matogrossense. 

In Anais do 1º Simpósio sobre Recursos Naturais e Sócio-Econômicos 

do Pantanal (Corumbá, Mato Grosso), pp. 137–178. Brasília: Empresa 

Brasileira de Pesquisa Agropecuária. 

Brown Jr., K. S. 1987b. Biogeography and evolution of Neotropical butterflies. 

In T. C. Whitmore and G. T. Prance, eds., Biogeography and Quaternary 

History in Tropical America, pp. 66–104. Oxford, England: 

Clarendon Press. 

Brown Jr., K. S. 1987c. Conclusions, synthesis, and alternative hypotheses. In 

T. C. Whitmore and G. T. Prance, eds., Biogeography and Quaternary 

History in Tropical America, pp. 175–196. Oxford, England: Clarendon 

Press. 

Brown Jr., K. S. 1991. Conservation of Neotropical environments: Insects as 

indicators. In N. M. Collins and J. A. Thomas, eds., The Conservation 

of Insects and their Habitats, pp. 349–404. London: Academic Press. 

Brown Jr., K. S. 1997. Diversity, disturbance, and sustainable use of Neotropical 

forests: Insects as indicators for conservation monitoring. J. Insect 

Cons. 1:25–42. 

Brown Jr., K. S. 2000. Insetos indicadores da história, composição, diversidade, 

e integridade de matas ciliares tropicais. In R. R. Rodrigues and 

H. F. Leitão Filho, eds., Matas Ciliares: Conservação e Recuperação, pp. 

223–232. São Paulo: Editora da Universidade de São Paulo and Fundação 

de Amparo à Pesquisa do Estado de São Paulo. 

Brown Jr., K. S. and A. V. L. Freitas. 2000. Atlantic Forest Butterflies: Indicators 

for landscape conservation. Biotropica 32:934–956. 

Brown Jr., K. S. and O. H. H. Mielke. 1967a. Lepidoptera of the Central 

Brazil Plateau:. I. Preliminary list of Rhopalocera: Introduction, 

Nymphalidae, Libytheidae. J. Lep. Soc. 21:77–106. 

Brown Jr., K. S. and O. H. H. Mielke. 1967b. Lepidoptera of the Central Brazil 

Plateau: I. Preliminary list of Rhopalocera (continued): Lycaenidae, Pieridae, 

Papilionidae, Hesperiidae. J. Lep. Soc. 21:145–168. 

Brown Jr., K. S. and O. H. H. Mielke. 1968. Lepidoptera of the Central Brazil 

Plateau: III. Partial list for the Belo Horizonte area, showing the character 

of the southeastern “blend zone.’’ J. Lep. Soc. 22:147–157. 

Brown Jr., K. S. and O. H. H. Mielke. 1972. Lepidoptera of the Central Brazil 

Plateau: VI. Preliminary list of Saturniidae and Sphingidae; VII. Supplementary 

list of Rhopalocera. Unpublished manuscripts.



Brown Jr., K. S. and J. Vasconcellos-Neto. 1976. Predation on aposematic 

ithomiine butterflies by tanagers (Pipraeidea melanonota). Biotropica 

8:136–141. 

Callaghan, C. J., and K. S. Brown Jr. In prep. An annotated list of the Riodinine 

butterflies of the Brazilian Planalto. 

Camargo, A. J. A. 1999. Estudo comparativo sobre a composição e a diversidade 

de lepidópteros noturnos em cinco áreas da Região dos Cerrados. 

Rev. Bras. Zool. 16:368–380. 

Camargo, A. J. A. and V. O. Becker. 1999. Saturniidae (Lepidoptera) from the 

Brazilian Cerrado: Composition and biogeographic relationships. 



How rich is the flora of the Brazilian Cerrados? Ann. Miss. Bot. Gard. 

86:192–224. 

DNMet. 1992. Normais Climatográficos, 1961–1990. Brasília: Ministério da 

Agricultura, Serviço Nacional de Irrigação. 

Gifford, D. R. 1979. Edaphic factors in dispersion of Heliconian and Ithomiine 

butterflies in the Araguaia region, Central Brazil. Unpublished manuscript. 

Ledru, M.-P. 1993. Late Quaternary environmental and climatic changes in 


Mielke, O. H. H., and M. M. Casagrande. 1998. Papilionoidea e Hesperioidea 

(Lepidoptera) do Parque Estadual do Morro do Diabo, Teodoro 

Sampaio, São Paulo, Brasil. Rev. Bras. Zool. 14:967–1001. 

Motta, P. C. 2002. Butterfly diversity in Uberlândia, central Brazil: Species list 

and biological comments. Rev. Bras. Biol. 62:151–163. 

Oliveira-Filho, A. T. and M. A. L. Fontes. 2000. Patterns of floristic differentiation 

among Atlantic Forests in south-eastern Brazil, and the influence 

of climate. Biotropica 32:793–810. 

Oliveira-Filho, A. T. and J. A. Ratter. 1995. A study of the origin of central 

Brazilian forests by the analysis of plant species distribution patterns. 

Edinb. J. Bot. 52:141–194. 

Oliveira-Filho, A. T. and J. A. Ratter. 2000. Padrões florísticos das matas 

ciliares na região do Cerrado e a evolução das paisagens do Brasil 

Central durante o Quaternário tardio. In R. R. Rodrigues and H. F. 

Leitão Filho, eds., Matas Ciliares: Conservação e Recuperação, pp. 

73–89. São Paulo: Editora da Universidade de São Paulo and Fundação 

de Amparo à Pesquisa do Estado de São Paulo. 

Pinheiro, C. E. G. and J. V. C. Ortiz. 1992. Communities of fruit-feeding butterflies 

along a vegetation gradient in central Brazil. J. Biogeogr. 

19:505–511. 

RADAMBRASIL, Projeto (Ministério de Minas e Energia). 1978–1983. Levantamento 

de Recursos Naturais (folhas Porto Velho, Goiás, Cuiabá, 

Corumbá, Campo Grande, Brasília, Goiânia, and Rio de Janeiro). Volumes 

16, 25–29, 31–32. Salvador and Rio de Janeiro: Departamento



Nacional de Produção Mineral and Instituto Brasileiro de Geografia e 

Estatística. 

Ratter, J. A., G. P. Askew, R. F. Montgomery, and D. R. Gifford. 1978. Observations 

on forests of some mesotrophic soils in central Brasil. Rev. Bras. 

Bot. 1:47–58. 




on the vegetation of northeastern Mato Grosso: I. The woody vegetation 

types of the Xavantina-Cachimbo Expedition area. Phil. Trans. 

Royal. Soc. Lond. 266B:449–492. 

Rodrigues, R. R., and H. F. Leitão-Filho, eds. 2000. Matas Ciliares: Conservação 

e Recuperação. São Paulo: Editora da Universidade de São Paulo 

and Fundação de Amparo à Pesquisa do Estado de São Paulo. 

Shepherd, G. J. 1995. FITOPAC: Manual de Usuários. Campinas: Departamento 

de Botânica, Universidade Estadual de Campinas. 

StatSoft, Inc. 1995. Statistica for Windows. Tulsa, Oklahoma. 

Ter Braak, C. J. F. 1987–1992. CANOCO: A FORTRAN program for 

Canonical Community Ordination. Ithaca: Microcomputer Power. 

Tyler, H. A., K. S. Brown Jr., and K. H. Wilson. 1994. Swallowtail Butterflies 

of the Americas: A Study in Biological Dynamics, Ecological Diversity, 

Biosystematics, and Conservation. Gainesville: Scientific Publishers.

Appendix Environmental and Biological Characteristics of 45 Sampling Sites in the Brazilian 

Cerrado. Conventions for values of variables below 

No. Site Vegetation 

and State Type Lat Lon Alt Sur Top Wat Veg Vmo Bam Soil Smo Sba Plu Dry Tem Tva Heli Ith Bait Satur 

01 Ilha do SDF 12.0 50.0 190 5 1 4 3 4 3 2 2 2 17 110 25.6 3.0 10 14 x x 

Bananal, TO 

02 Barreiras, BA CaCe 12.1 45.0 440 4 3 4 5 4 3 4 4 4 10 150 24.3 3.5 x x x 25 

03 Vilhena, RO FCe 12.6 60.1 615 4 2 2 5 3 3 2 2 3 20 120 23.0 3.0 x x x 29 

04 Chapada dos Ce 14.0 47.7 1100 3 4 4 3 3 2 3 5 3 15 130 21.1 4.0 x 9 31 17 

Veadeiros, GO 

05 *Chapada dos FCe 15.5 55.7 700 3 3 3 4 5 4 3 5 4 17 120 23.0 4.0 15 24 82 34 

Guimarâaes, MT 

06 Alto Rio FCe 15.5 47.7 650 4 4 4 5 3 3 3 3 5 16 120 22.0 5.0 9 14 33 37 

Maranhão, DF 

07 Itiquira + Ce 15.5 47.4 1000 4 5 4 3 4 2 3 3 2 16 130 21.5 5.0 5 10 x 64 

Formosa, DF/GO 

08 FERCAL/Chap. Ce 15.6 47.9 1200 2 4 2 3 2 1 3 3 2 16 120 21.2 5.0 7 8 29 31 

Contagem, DF 

09 *Planaltina FCe 15.6 47.7 960 3 2 4 2 2 1 5 4 3 16 120 21.2 5.0 x x x 98 

(EMBRAPA), DF 

10 Sobradinho, Ce 15.7 47.8 1050 2 2 2 3 3 1 4 2 2 16 120 21.2 5.0 10 12 38 39 

DF 

11 *Brasília Ce 15.8 47.9 1100 3 2 5 3 3 1 4 2 3 16 130 21.2 5.0 x x x 67 

Microwave 

Tower, DF 

12 Goiás Velho, SDF 15.9 50.1 490 5 3 3 5 2 3 4 4 5 18 140 24.7 4.0 8 20 x x 

GO 

13 *Jardim Ce 15.9 48.0 1000 3 1 4 6 3 1 4 2 3 16 120 21.2 5.0 9 17 24 25 

Zoobotânico, 

DF 


Appendix (continued) 



14 Est. Exp. FCe 15.9 47.8 1050 3 2 4 4 2 2 4 2 2 16 120 21.2 5.0 11 21 44 43 

Cabeça do 

Veado, DF 

15 Brasília Country FCe 16.0 48.0 1200 2 2 3 4 2 3 4 2 2 16 120 21.2 5.0 10 11 37 51 

Club, DF 

16 Parque do FCe 16.0 48.1 900 4 4 2 5 4 4 4 3 3 16 120 21.5 5.0 11 17 64 72 

Gama, DF 

17 Pirenópolis, GO SDF 16.0 49.1 900 4 4 3 5 3 2 4 4 5 18 120 22.5 3.6 5 17 26 x 

18 Santa Maria Ce 16.1 48.0 1240 1 1 4 3 3 1 4 3 3 16 20 21.2 5.0 x x x 27 

MW Tower, DF 

19 Anápolis, incl. Ce 16.3 49.0 1020 3 2 3 5 2 3 4 2 4 16 120 21.4 4.0 x 11 32 38 

MW Tower, GO 

20 Unaí de Minas, CaCe 16.4 46.9 580 4 2 3 5 3 2 4 3 3 13 140 22.0 4.0 x x x 28 

MG 

21 Cilú Microwave Ce 16.4 48.2 900 4 3 2 4 2 2 5 2 4 15 130 21.2 5.0 x x x 48 

Tower, GO 

22 Iporá to SDF 16.4 51.3 550 5 3 2 5 3 3 5 4 4 15 140 23.0 4.0 7 11 x x 

Piranhas, GO 

23 Leopoldo Ce 16.6 48.8 1030 4 2 2 5 3 3 5 5 4 16 130 21.2 4.0 x 11 26 x 

Bulhões, GO 

24 *Goiânia + SDF 16.7 49.3 750 5 3 4 5 2 3 4 4 4 16 150 23.2 4.0 5 17 47 26 

Campinas, GO 

25 Vianópolis+ Ce 16.7 48.5 1000 4 2 2 5 2 3 4 4 4 16 130 21.2 5.0 x 12 x 88 

Ponte Funda 

MWT 

26 Paracatu-K231 FCe 17.2 46.9 920 4 4 2 4 3 5 3 3 3 15 130 22.6 5.0 6 x 36 44 

Bras.-Belo,MG 

27 Mineiros-K163 FCe 17.4 52.5 830 4 2 3 5 3 2 4 4 4 15 120 22.5 4.0 x x 26 x 

Jataí-A.Arag. 





28 Mineiros-K123 FCe 17.4 52.5 830 4 2 3 4 2 2 4 3 4 14 120 22.5 4.0 4 10 x x 

Jataí-A.Arag. 

29 Piracanjuba, GO SDF 17.4 49.2 750 4 4 2 5 2 5 5 3 6 15 150 22.0 4.0 x x 28 x 

30 Morrinhos, GO FCe 17.7 49.1 770 4 3 2 5 3 4 5 3 4 15 150 22.0 4.0 x 14 27 24 

31 Três Marias, Ce 18.2 45.2 540 5 2 5 4 2 3 3 2 2 12 120 22.0 3.0 x x x 19 

MG 

32 Felixlândia- Fce 18.7 44.9 630 4 2 4 5 3 2 4 3 3 12 150 21.0 4.0 x 8 x x 

K222 Belo-Bras 

33 Córrego Boa FCe 18.9 49.0 780 3 2 3 4 2 3 4 3 3 16 130 21.4 5.1 x 15 x 25 

Vista,M.Alegre 

34 Cabeceira FCe 19.1 44.6 800 3 2 3 5 3 3 3 3 4 12 150 21.0 5.0 6 9 29 x 

Córrego Leitão, 

MG 

35 *Uberlândia, FCe 19.2 48.4 800 3 3 4 5 3 3 4 3 4 15 130 21.4 5.1 8 18 47 x 

MG 

36 *Paraopeba, SDF 19.3 44.4 730 5 2 2 6 3 3 5 3 4 12 150 20.9 5.5 6 16 35 42 

MG 

37 *Belo Horizonte, SDF 19.9 43.9 1000 2 5 4 6 3 3 5 5 3 15 150 21.1 5.1 12 21 61 40 

MG 

38 *Salobra, MS ChCe 20.3 56.7 180 4 3 3 5 5 3 5 3 6 10 150 25.0 6.1 x x x 27 

39 *Mirassol, SP SDF 20.8 49.6 470 4 3 2 6 2 3 5 2 4 13 120 24.0 5.0 x 19 31 x 

40 Rio Brilhante, ChCe 21.8 54.5 400 4 3 3 5 5 3 6 2 4 13 90 21.9 8.2 x x x 26 

MS 

41 Itirapina, SP Ce 22.2 48.0 600 3 2 3 3 2 3 3 2 2 15 50 22.3 6.4 x 17 x x 

42 *Faz.Camp- Ce 22.3 47.2 530 3 3 4 3 3 2 3 2 2 15 70 20.5 6.0 6 22 x x 

ininha, Mogi- 

Guaáu 





43 Parque Est. DF 22.6 52.3 350 4 2 4 5 3 4 4 3 3 13 150 24.0 8.0 10 20 64 x 

Morro do Diabo, 

SP 

44 *Horto Florestal SDF 22.8 47.3 600 5 2 2 6 2 1 5 2 4 14 60 20.7 5.7 x 22 x x 

de Sumaré,SP 

45 *Santa Genebra, SDF 22.9 47.1 600 5 2 3 6 2 4 5 2 4 14 60 20.7 5.7 13 28 83 x 

Campinas, SP 

* = Sites with considerable anthropic disturbance at some time or sectors. Site numbers, in order of latitude and then longitude, are those used in figures 11.1 and 

11.2. 

Vegetation types: Ca = Caatinga, Ce = Cerrado, Ch = Chaco, DF = Deciduous or dry forest, FCe = Forest mixed with Cerrado, SDF = Semi-deciduous forest. 

Latitude and Longitude (0.1°), Altitude (m), Pluviosity = Annual rainfall (dm), average length of the Dry season (in days), Temperature (annual mean), and Temperature 

Variation (between means for warmest and coldest months, usually September and June in the Cerrado region) (in °C) are quantitative. Many of the climatic 

data came from Oliveira-Filho and Fontes (2000), DNMet (1992), or RADAMBRASIL (1978—83); most soil and vegetation data are from the latter. 

Explanation of codes for categorical variables in matrix (ordered to show increasing humidity, richness, or complexity; see also Brown and Freitas 2000): Surface 

(Geomorphic, see chapter 2) is coded as 1 = highest (oldest) Surface 1, 2 = transition 1—2, 3 = Surface 2, 4 = transition 2—3, 5 = Surface 3 (youngest). 

Topography is given as relief type, 1 = level, 2 = depression or gently rolling, 3 = rolling, 4 = strongly rolling, and 5 = steep, mountainous or escarpment. 

Permanent Water is coded as 1 = dry landscape, 2 = small rivulets, 3 = broader streams, 4 = river and larger swamps, 5 = large open water bodies in site. 

Vegetation categories correspond to inceasing humid arboreal physiognomies in the region, and are averaged for a site; base numbers are 1 = campo (open grassland), 

2 = campo cerrado (sparse small trees), 3 = cerrado or caatinga (many trees, many grasses), 4 = cerradão or agreste (dominant higher woody layer, sparse grasses), 5 

= dry or deciduous forest, or a mixture of cerrado and gallery forest, 6 = semi-deciduous, headwater, or very broad gallery forest. 

Vegetation mosaic is the number of different vegetation types (as categorized above) prominent and interdigitating in the site. See also chapter 6. 

Bamboo abundance in the site is coded as 1 = essentially absent to 6 = present in many large patches, up to 30% of the vegetation stems at ground level. 

Soils are averaged among the following categories: 1 = rock, hardpan, laterite; 2 = sandy or concretionary, 3 = cambisols, plinthic soils, 4 = moderate-texture latosols, 

5 = moderate-textured podzolized soils, and 6 = humic or very argyllic soils. 

Soil mosaic is the number of different soil types (as categorized above) in the site, with 1 = over 80% of a single class of soils, 2 = 50—80% of a single class, 



3 = three types, 4 = four types, 5 = five or more soil types present in a fine mosaic. 

Soil bases (fertility) is as 1 = hardpan or coarse sand, 2 = alic, 3 = alic + dystrophic, 4 = dystrophic, 5 = dystrophic + eutrophic, and 6 = eutrophic soils. 

Heli = Number of distinct Heliconiini taxa recorded in the site, Ith = Ithomiinae, Bait = Bait-attracted Nymphalidae except Satyrinae, Satur = Saturniidae. 

Additional data for Lepidoptera lists of various sites were obtained from N. Tangerini (material from the Microwave stations), Brown and Vasconcellos-Neto 

(1975, Sumaré), Brown (1987a, central Mato Grosso), Mielke and Casagrande (1998, Morro do Diabo), P. C. Motta (2002, Uberlândia), E.G. 

Munroe, C. E. G. Pinheiro and H. C. Morais (unpublished lists, Cabeça do Veado and other Federal District sites), and Fernando Corrêa Campos (Belo 

Horizonte). 



12 

The Character and Dynamics 

of the Cerrado Herpetofauna 

Guarino R. Colli, Rogério P. Bastos, 

and Alexandre F. B. Araujo 

The earliest work on the herpetofauna of the cerrado is 

a list of 54 reptiles and amphibians from Lagoa Santa, state of Minas 

Gerais, prepared by Warming (1892). More than 50 years later, Vanzolini 

(1948) presented an annotated list of 22 snake and 11 lizard species from 

Pirassununga, state of São Paulo. Moreover, Vanzolini (1974, 1976, 

1988) examined the distribution patterns of cerrado and caatinga lizards, 

concluding that no characteristic lizard fauna is harbored by either biome, 

both biomes belonging to a corridor of open vegetation ranging from 

northwestern Argentina to northeastern Brazil (see chapter 6). According 

to Vanzolini (1976), the lack of differentiation of the herpetofauna along 

his “great diagonal belt of open formations’’ results from present-day 

ample, interdigitating contacts between the caatinga and the cerrado and 

past climatic cycles that promoted the expansion and retraction of both 

biomes, leading to extensive faunal mixing. This idea received further 

support from Duellman (1979), Webb (1978), and Silva and Sites (1995). 

More recently, in a preliminary survey conducted near Alto Araguaia, 

Mato Grosso, Vitt (1991) concluded that the cerrado lizard fauna is 

depauperate when compared to its caatinga and Amazonia counterparts, 

most noticeably with respect to the Gekkonidae, and that ecological factors 

such as low habitat structural diversity and the occurrence of fires 

may account for the low lizard diversity. 

In sum, the aforementioned works indicate that the cerrado herpetofauna 

(1) has low levels of species diversity and endemism; (2) lacks a 

character; and (3) is more similar to the caatinga than to other South 

223



American biomes. We advance, however, that these claims, largely the 

outcome of inadequate sampling and/or analyses, do not adequately 

describe the nature of the cerrado herpetofauna. For instance, Heyer 

(1988) analyzed the adequacy of the distributional database of 10 speciesgroups 

of frogs east of the Andes and concluded that frog distributional 

data were largely inadequate for biogeographical analysis, that the cerrado 

and caatinga were the poorest-sampled biomes, and that the question 

whether the cerrado and caatinga share a common frog fauna was 

unanswerable with the data available at the time. 

In a similar vein, of the 101 cerrado lizard localities we obtained from 

three Brazilian institutions, 97% contain fewer than 12 species (see fig. 

12.1). Records from the three most extensively sampled localities 

(Brasília, Distrito Federal; Minaçu, state of Goiás; and Chapada dos 

Guimarães, state of Mato Grosso) suggest that the local lizard diversity 

in the cerrado is around 25 species. Therefore, most collecting localities 

have been inadequately sampled in the cerrado, and our estimates indicate 

that, contrary to Vitt’s (1991) impression, the local diversity of lizards 

in the cerrado (around 25 species) is greater than that of the caatinga (18 

Figure 12.1 Frequency distribution of the cerrado localities where lizards 

have been collected according to the number of lizard species. Data from the 

Museu Paraense Emílio Goeldi (MPEG), Museu Nacional do Rio de Janeiro 

(MNRJ), and Coleção Herpetológica da Universidade de Brasília (CHUNB).


The Character and Dynamics of the Cerrado Herpetofauna 225 

species; Vitt 1995). However, it is possible that the three best-studied 

localities are unusual and that, in fact, the other 98 studied localities, 

reported to have 12 or fewer species, do have low lizard diversity. This 

latter view implies that species diversity in the cerrados is highly variable 

on a regional scale. Further studies are necessary to clarify this issue and 

identify possible determinants of lizard diversity in the cerrado region. 

Several species of reptiles and amphibians of the cerrado have been 

described recently, and several undescribed species still await adequate 

studies. At present, 10 species of turtles, 5 crocodilians, 15 amphisbaenians, 

47 lizards, 107 snakes, and 113 amphibians are known to occur in 

the Cerrado Biome (appendix). Among these species, some are typical of 

the Atlantic forest or Amazonia biomes but often occur deep in the cerrados, 

along gallery forests. As an example, Anilius scytale has been 

recorded in gallery forests of the Tocantins river drainage, at Minaçu, 

state of Goiás, in the core region of the cerrados. Other species, such as 

Leptodactylus petersii (Heyer 1994), have a more marginal distribution 

in the biome. The species list we present roughly corresponds to 20% of 

the Brazilian species of amphibians and 50% of the reptiles. The estimates 

for lizards and snakes are almost twice those previously recorded by Vanzolini 

(1988) and Silva and Sites (1995). Moreover, contrary to early 

beliefs, the cerrados harbor a large number of endemics: 8 species of 

amphisbaenians (50% of the total amphisbaenian species), 12 species of 

lizards (26%), 11 species of snakes (10%), and 32 (28%) species of 

amphibians (appendix). This level of endemicity differs sharply from the 

3.8% recorded for cerrado birds (Silva 1995) and, at least for amphisbaenians, 

is comparable to that registered for the cerrado flora, which is 

approximately 50% (Heringer et al. 1977; chapter 6). The cerrado herpetofauna 

also includes three endangered species of anurans, four turtles, 

five crocodilians, five lizards, and six snakes (see appendix). With the 

exception of Caiman niger (sensu Poe 1996), listed in appendix I of 

CITES, the remaining endangered species are listed in appendix II. 

It is commonly assumed that savanna biomes, due to their lower habitat 

heterogeneity, harbor an impoverished herpetofauna relative to forest 

biomes (e.g., Duellman 1993; Lamotte 1983). Further, it seems to be taken 

for granted that species richness in South America is centered in Amazonia. 

Taking lizards as an example, there are about 100 species in Brazilian 

Amazonia (Ávila-Pires 1995), whereas our list for the cerrados 

contains about half that number. Nevertheless, Brazilian Amazonia covers 

an area of approximately 4 million km 2 , roughly twice the area covered 

by the cerrados. Therefore, the difference in species richness between 

the two biomes may stem from an area effect, rather than from a habitat



heterogeneity effect. In terms of local species richness, moreover, there is 

no difference in the mean number of lizard species between cerrado and 

neotropical forest sites (tables 12.1, 12.2). This is a remarkable and 

apparently counterintuitive result. If local species assemblages are but a 

random sample of the regional pool of species, it follows that neotropical 

forest sites should be richer than cerrado sites (Schluter and Ricklefs 

1993). We advance that the similar richness in local lizard assemblages 

between cerrado and neotropical forest sites results from two factors. 

First, the pronounced horizontal variability in the cerrado landscape balances 

the predominantly vertical variability typical of forested biomes 

when it comes to allowing the coexistence of species. The horizontal variability 

of the landscape may be further enhanced by fire history, as documented 

in other open vegetation regions (e.g., Pianka 1989). Second, 

there is a greater regional differentiation of the herpetofauna in neotropical 

forests relative to the cerrados: the average species turnover (beta 

diversity) among the five forest sites listed in Duellman (1990) is significantly 

higher than that among the five cerrado sites in table 12.1 (forest: 

xm = 0.78 ± 0.19; cerrado: xm = 0.27 ± 0.09; Mann-Whitney U = 0.00, p < 

.001). Thus, even though neotropical forests as a whole harbor a richer 

herpetofauna, the local lizard diversity is comparable to that recorded in 

the cerrado. 

The much lower species turnover in the cerrados, relative to Amazonia, 

suggests that the high regional heterogeneity of the cerrado has little 

effect upon the composition of the herpetofauna. However, some areas, 

such as high altitude habitats, harbor an unusually high number of 

endemics. For instance, the campos rupestres (rocky grasslands; see chapter 

6) along the Espinhaço Range include large numbers of endemic 

amphibians and reptiles (Heyer 1999; Rodrigues 1988). There may be 

other areas of high endemism in the biome, such as the Jalapão sand dunes 

in the state of Tocantins, but this region as well as large tracts of the cerrados 

have never been adequately sampled. 

NATURAL HISTORY 

Most of the available information on the natural history of the cerrado 

herpetofauna is restricted to lizards and anurans. Salamanders are absent 

from the cerrados, and practically no information is available for caecilians, 

amphisbaenians, crocodiles, and testudines. Our coverage, therefore, 

is concentrated on lizards and anurans, reflecting this knowledge 

bias. A striking feature of the Cerrado Biome is its marked seasonality,



Table 12.1 Composition of Lizard Assemblages from Cerrado Sites 

Site Ecology 

Micro- 

Taxon CG BG MI PI BR Diel Habitat habitat 

Hoplocercidae 

Hoplocercus X X X — X D CE G 

spinosus 

Iguanidae 

Iguana iguana X X X — — D F T 

Polychrotidae 

Anolis meridionalis X X X X X D CE G,T 

Anolis nitens X X X X X D F,CE G,T 

Enyalius bilineatus — — — — X D F G,T 

Polychrus X X X X X D CE T 

acutirostris 

Tropiduridae 

Stenocercus — X — — — D CE,F G,T 

caducus 

Tropidurus X — — — — D CE,F T 

guarani 

Tropidurus — — — X X D CE G 

itambere 

Tropidurus X X X — — D CE G 

montanus 

Tropidurus X X X X X D CE G 

oreadicus 

Tropidurus X X X — X D F G,T 

torquatus 

Gekkonidae 

Coleodactylus — — X — — D F,CE L 

brachystoma 

Coleodactylus X — X — — D F L 

meridionalis 

Gymnodactylus X X X X — N(?) CE G 

geckoides 

Hemidactylus X — — — X N CE G,T 

mabouia 

Phyllopezus X — X — — N(?) CE G 

pollicaris 

Scincidae 

Mabuya — — — — X D CE,F G,L 

dorsivittata 

Mabuya frenata X X X X X D CA,CE G,L 

Mabuya guaporicola — — — — X D CE,F G,L 

Mabuya X X X X X D CE G,L 

nigropunctata



Table 12.1 (continued) 

Site Ecology 

Micro- 

Taxon CG BG MI PI BR Diel Habitat habitat 

Gymnophthalmidae 

Bachia bresslaui X — — — X D CE F 

Cercosaura — X X — X D E,CA,F L 

ocellata 

Colobosaura — X X — X D CE,F L 

modesta 

Micrablepharus — X X X X D CE L 

atticolus 

Micrablepharus X — X X X D CE L 

maximiliani 

Pantodactylus X X X X X D CE L 

schreibersii 

Teiidae 

Ameiva ameiva X X X X X D CE,F G 

Cnemidophorus X X X X X D CE G 

ocellifer 

Kentropyx X X X — X D CE,CA G 

paulensis 

Tupinambis duseni — — — — X D CE,CA G 

Tupinambis X X X X X D CE,F G 

merianae 

Tupinambis X X X — — D F,CE G 

quadrilineatus 

Anguidae 

Ophiodes striatus X X X — X D CE,F,CA F 

Total number of 24 22 25 14 25 

species 

Sources: Araujo and records from the Coleção Herpetológica da Universidade de Brasília (CHUNB). 

Note: CG = Chapada dos Guimarães, Mato Grosso; BG = Barra do Garças, Mato Grosso; MI = 

Minaçu, Goiás; PI = Pirenópolis, Goiás; BR = Brasília, Distrito Federal. Diel: D = diurnal, N = nocturnal. 

Habitat: CA = campo cerrado, campo limpo, or campo sujo, CE = cerrado sensu stricto, F = forest. 

Microhabitat: F = fossorial, G = ground, L = litter, T = tree. In columns under sites, X = present, 

— = presumably absent. See chapter 6 for descriptions of cerrado physiognomies. 

with well-defined dry and rainy seasons. Moreover, the cerrado is a highly 

heterogeneous landscape, with a physiognomic mosaic that ranges from 

treeless grass fields to relatively dense gallery forests, with a predominance 

of open vegetation physiognomies (chapter 6). These two aspects, marked 

seasonality and high spatial heterogeneity, have profound effects on the 

ecology of the herpetofauna in general, since these animals are ectothermic, 

and amphibians in particular, because of their highly permeable skin.

Table 12.2 Ecology and Families of Lizards from Cerrado and Neotropical Forest Sites 

Cerrado Neotropical rainforest 

CG BG MI PI BR mean ± sd LS BCI MN MA SC mean ± sd 

Diel activity 

Diurnal 87.5 95.5 92.0 92.9 96.0 92.8 ± 3.4 92.0 92.3 95.8 93.7 96.7 94.1 ± 2.1 

Nocturnal 12.5 4.5 8.0 7.1 4.0 7.2 ± 3.4 8.0 7.7 4.2 6.3 3.3 5.9 ± 2.1 

Habitat 

Fossorial 8.3 4.5 4.0 - 8.0 5.0 ± 3.4 - 3.8 8.3 6.3 3.3 4.3 ± 3.2 

Ground a 66.7 63.6 56.0 64.3 64.0 62.9 ± 4.1 40.0 15.4 25.0 43.8 26.7 30.1 ± 11.7 

Litter 20.8 27.3 36.0 35.7 36.0 31.2 ± 6.9 44.0 30.8 33.3 12.5 36.7 31.5 ± 11.7 

Trees a,b 29.2 27.3 20.0 21.4 24.0 24.4 ± 3.9 52.0 57.7 45.8 56.3 43.3 51.0 ± 6.3 

Family 

Hoplocercidae 4.2 4.5 4.0 - 4.0 3.3 ± 1.9 - - - 6.3 6.7 2.6 ± 3.6 

Corytophanidae - - - - - - 12.0 7.7 - - - 3.9 ± 5.6 

Iguanidae 4.2 4.5 4.0 - - 2.5 ± 2.3 4.0 3.8 4.2 - - 2.4 ± 2.2 

Polychrotidae c 12.5 13.6 12.0 21.4 16.0 15.1 ± 3.8 36.0 46.2 16.7 18.8 23.3 28.2 ± 12.5 

Tropiduridae 16.7 18.2 12.0 14.3 12.0 14.6 ± 2.8 - - 16.7 18.8 6.7 8.4 ± 9.0 

Gekkonidae 16.7 4.5 16.0 7.1 4.0 9.7 ± 6.2 16.0 15.4 12.5 18.8 13.3 15.2 ± 2.5 

Scincidae c 8.3 9.1 8.0 14.3 16.0 11.1 ± 3.7 8.0 3.8 4.2 6.3 3.3 5.12 ± 2.0 

Xantusiidae - - - - - - 4.0 3.8 - - - 1.6 ± 2.1 

Gymnophthalmidae 12.5 18.2 20.0 21.4 24.0 19.2 ± 4.3 - 7.7 33.3 6.3 33.3 16.1 ± 16.0 

Teiidae c 20.8 22.7 20.0 21.4 20.0 21.0 ± 1.1 8.0 11.5 12.5 18.8 13.3 12.8 ± 3.9 

Anguidae 4.2 4.5 4.0 0.00 4.0 3.3 ± 1.9 12.0 - - 6.3 - 3.7 ± 5.4 

Number of species 24 22 25 14 25 22.0 ± 4.6 25 26 24 16 30 24.2 ± 5.1 

Sources: Cerrado sites are from Araujo (1992) and records from the Coleção Herpetológica da Universidade de Brasília (CHUNB); neotropical rainforest sites are from 

Duellman (1990). 

Note: CG = Chapada dos Guimarães, Mato Grosso; BG = Barra do Garças, Mato Grosso; MI = Minaçu, Goiás; PI = Pirenópolis, Goiás; BR = Brasília, Distrito Federal. 

Numbers indicate percentage of lizard species occurring in each category. Some species are present in more than one category. 

a Means of arcsin transformed values are statistically different between cerrado and neotropical rainforest (Ground: F = 31.4, p < .001; Trees: F = 64.7, p < .001). 

b Species in “bushes” category of Duellman (1990) were placed in the “trees” category. 

c Means of arcsin transformed values are statistically different between cerrado and neotropical rainforest (Polychrotidae: F = 5.5, p < .05; Scincidae: F = 11.8, p < .01; 

Teiidae: F = 17.7, p < .01). 




Below, we review different aspects of the ecology of the cerrado herpetofauna, 

such as habitat and microhabitat preferences, diel activities, and 

reproductive cycles. 

Habitat, Microhabitat, and Diel Activity 

Considering the pronounced seasonality and heterogeneity of the cerrado 

landscape (chapter 11), the restriction of moist habitats to the periphery 

of river drainages, and their great physiological requirements, it is 

expected that cerrado anurans should adopt one or more of the following 

strategies: (a) to be nocturnal; (b) to live close to or in moist places, 

like swamps; (c) to show activity only during the rainy season; or d) to 

live permanently buried in the soil (Lamotte 1983). Conversely, because 

of their highly impermeable skin, strictly pulmonary respiration, and 

amniotic eggs, reptiles can explore a broader range of environments. 

The anuran assemblage at the Estação Ecológica de Águas Emendadas, 

Brasília (Distrito Federal), contains 26 species distributed in four 

habitats: campo-úmido/vereda (22 species), vereda (16), cerrado/cerradão 

(12), and gallery forest (10) (Brandão and Araujo 1998; chapter 6). Most 

species use two or more habitats and the campo-úmido/vereda, a seasonally 

flooded field with buriti-palms (Mauritia flexuosa) harboring the 

greatest number of species, including adults and tadpoles. Only two species 

are restricted to gallery forest: Hyla biobeba and Aplastodiscus perviridis. 

Motta (1999) compared the leaf-litter anuran assemblages of cerrados and 

gallery forest, by using pitfall traps and drift fences, at the Estação Florestal 

de Experimentação, Silvânia, state of Goiás. Barycholos savagei and 

Physalaemus cuvieri were the most abundant species in both habitats. 

Despite the greater abundance and diversity of anurans in the gallery forest, 

there was a high similarity between the two assemblages. 

A majority of the lizard species distributed within the cerrado region 

occurs primarily in “cerrado-type’’ physiognomies (see chapter 6), whereas 

Iguana iguana, Enyalius bilineatus, Anolis chrysolepis brasiliensis, Tropidurus 

torquatus, Coleodactylus brachystoma, and Tupinambis quadrilineatus 

are mainly restricted to forested habitats (table 12.1). However, 

apparently there is no difference in the mean richness of lizards between 

cerrado and forested habitats (Araujo 1992). Some species are strongly 

associated with specific habitat characteristics. For example, Phyllopezus 

pollicaris and Gymnodactylus geckoides are found primarily in rock outcrops. 

On the other hand, Bachia bresslaui and Cnemidophorus ocellifer 

are strongly associated with sandy soils (Magnusson et al. 1986; Colli et 

al. 1998). Still other species, such as Ameiva ameiva (Colli 1991; Vitt and



Colli 1994; Sartorius et al. 1999) and Tropidurus torquatus (Wiederhecker 

et al. 2002), seem to benefit from human disturbance, as they are 

very abundant in anthropic areas. 

Among reptiles, only crocodilians, most turtles, some snakes such as 

Eunectes murinus and Helicops modestus, and the lizard Iguana iguana 

are strongly associated with water in the cerrados. Several species of cerrado 

reptiles are fossorial, including all amphisbaenids, all species of the 

gymnophthalmid genus Bachia, the anguid Ophiodes striatus, and several 

snake genera including Liotyphlops, Leptotyphlops, Typhlops, Anilius, 

Phalotris, Apostolepis, Atractus, Tantilla, and Micrurus. A number of 

species also uses cavities in trees, termite and ant nests, and armadillo burrows 

as shelter. Cavities in trees are often used by the arboreal Tropidurus 

guarani (Colli et al. 1992). Several snakes such as Bothrops neuwiedi, 

Tantilla melanocephala, and Oxyrhopus trigeminus, and the lizards Anolis 

meridionalis and Gymnodactylus geckoides are commonly found 

under termite nests. Apparently, the amphisbaenid Amphisbaena alba 

(Colli and Zamboni 1999) and the gymnophthalmid Micrablepharus atticolus 

(Vitt and Caldwell 1993) are associated with nests of leaf-cutter 

ants. The ability to use these microhabitats seems especially critical during 

the periodic fires. 

Another behavioral adaptation of the herpetofauna to resist desiccation 

is nocturnal activity. Diurnal species of anurans, relatively common in 

Amazonic and Atlantic Forests due to high humidity, are relatively rare in 

the cerrados (Heyer et al. 1990; Hödl 1990). Some species, however, can 

call during the day after intense rains (Brandão and Araujo 1998). Vanzolini 

(1948) noticed a preponderance of nocturnal and fossorial species in a cerrado 

assemblage of squamates, suggesting that these traits would be an 

adaptation to high temperatures and visibility to predators at the soil surface, 

coupled with thermal stability, good aeration, and an abundance of 

prey beneath the surface. We evaluated Vanzolini’s statement by classifying 

five lizard assemblages from the cerrado according to diel and microhabitat 

categories, following Duellman (1990). The number of nocturnal species 

in each site varied from one to three (table 12.1), and the percentage of nocturnal 

species averaged 7.2% (table 12.2). Curiously, we observed no significant 

difference in the average of nocturnal species between the cerrado 

and neotropical forest (table 12.2). The number of fossorial species ranged 

from zero to two species (table 12.1), averaging 5% of the species in the 

cerrado (table 12.2). Again, we observed no difference between the mean 

percentage of fossorial species between the two biomes (table 12.2). This 

pattern seemingly derives from a comparable representation of nocturnal 

(Gekkonidae) and fossorial (Anguidae, Gymnophthalmidae) lineages in



both biomes. Thus, we find no evidence for Vanzolini’s (1948) suggestion 

that there is a trend toward nocturnal activity and fossoriality in the cerrado 

herpetofauna. 

Contrasting with the savannas of northern South America (Staton 

and Dixon 1977; Rivero-Blanco and Dixon 1979; Vitt and Carvalho 

1995), the higher density of trees and shrubs in the cerrados supports a 

relatively large number of arboreal squamates. Among lizards, the most 

arboreal species is Polychrus acutirostris, which even possesses a prehensile 

tail (Vitt and Lacher 1981). The presence of a developed arboreal stratum 

allows the coexistence of divergent, congeneric species pairs, such as 

Tropidurus guarani (larger, arboreal) and T. oreadicus (smaller, ground) 

in the same area (Colli et al. 1992). Perhaps the same could be said about 

Anolis chrysolepis brasiliensis and A. meridionalis, but there are no 

detailed studies on their coexistence. The only difference in the distribution 

of lizard species across habitat categories between cerrado and 

neotropical rainforest assemblages concerns the proportion of arboreal 

versus ground species. In neotropical rainforest sites, the mean percentage 

of arboreal species is approximately twice that for cerrado sites, 

whereas the reverse is true concerning species that use primarily the 

ground (table 12.2). That difference stems from the higher number of 

polychrotids in neotropical rainforest sites and of skinks and teiids in cerrado 

sites (table 12.2). Polychrotids are primarily arboreal, whereas teiids 

and skinks are primarily ground-dwellers. 

Life History 

Apparently, most anurans of the cerrado region breed in areas of open 

vegetation (Haddad et al. 1988, Feio and Caramaschi 1995, Brandão and 

Araujo 1998). Conversely, in anuran assemblages from forested biomes, 

such as Amazonia and Atlantic forests, a much lower proportion of 

species reproduce in open vegetation (Crump 1974; Heyer et al. 1990; 

Hödl 1990; Bertoluci 1998). However, Manaus (state of Amazonas) and 

Belém (state of Pará) present percentages that are similar to the cerrados, 

perhaps because of the noticeable dry season they exhibit (Hödl 1990). 

The prevalence of species with foam-nest reproduction in Manaus and 

Belém may be an adaptation to the seasonal environment (Hödl 1990). 

Moreover, in open habitats, as opposed to forest formations, the number 

of species is usually much larger than the number of available microhabitats 

such as calling perches (Cardoso et al. 1989). Consequently, spatial 

overlapping is seemingly more extensive among cerrado anurans 

(Blamires et al. 1997).



In tropical regions, rainfall and temperature are the main variables 

affecting anuran reproduction, and most species breed in a restricted 

period, determining a reproductive cycle (Aichinger 1987; Rossa-Feres 

and Jim 1994; Pombal 1997; Bertoluci 1998). One of us (RPB) and some 

students monitored the breeding activity of one cerrado anuran assemblage, 

between June 1995 and May 1998, at the Estação Florestal de 

Experimentação (EFLEX), in Silvânia, state of Goiás (Blamires et al. 

1997). Even though most species reproduced in the rainy season, between 

October and April, some were active even during the dry season, such as 

Hyla albopunctata, H. biobeba, H. goiana, and Odontophrynus cultripes. 

Like several other neotropical anurans (Pombal et al. 1994; Bastos and 

Haddad 1996, 1999), these species can be regarded as prolonged breeders. 

At EFLEX, males of most species (e.g., Hyla albopunctata, H. 

biobeba, H. cruzi, and H. goiana) aggregate during several nights, forming 

choruses and leks (exhibition areas) and defending territories with 

resources to oviposition. 

A common cerrado leptodactylid, Physalaemus cuvieri, breeds exclusively 

in the rainy season throughout its range (Barreto and Andrade 

1995; Moreira and Barreto 1997). A detailed study on the reproductive 

ecology of Scinax centralis was also carried out at EFLEX (Bugano 1999). 

This species belongs to the catharine group, typical of the Atlantic forest, 

and restricted to gallery forest in the cerrado region (Pombal and Bastos 

1996). Leks occur from November to July; males always outnumber 

females, showing a uniform spatial pattern maintained by acoustic interactions 

(three call types were observed) or fights; and oviposition occurs 

away from calling perches. 

With the exception of species of Mabuya and Ophiodes striatus, all 

lizard species of the cerrado are oviparous. All South American Mabuya 

have a long gestation period, ovulate small ova, form a chorioallantoic 

placenta, and have placental provision of practically all the energetic 

requirements for development (Blackburn and Vitt 1992). Clutch size is 

practically fixed in some lineages, with Anolis producing predominantly 

one egg, whereas gekkonids and gymnophthalmids produce clutches of 

usually two eggs (Vitt 1991). In the remaining species, clutch size shows 

a positive correlation with body size. The species so far studied are cyclic 

breeders, including Ameiva ameiva (Colli 1991), Amphisbaena alba (Colli 

and Zamboni 1999), Mabuya frenata (Vrcibradic and Rocha 1998; Pinto 

1999), M. nigropunctata (Pinto 1999), Tropidurus itambere (Van Sluys 

1993), and T. torquatus (Wiederhecker et al. 2002). With the exception 

of A. alba, all other species studied to date reproduce primarily during the 

rainy season. Furthermore, with the exception of the species of Mabuya,



the remaining species produce multiple clutches during the breeding season. 

It has been repeatedly suggested that reproductive seasonality in the 

neotropical region is largely determined by fluctuations in arthropod 

abundance which, in their turn, are associated with fluctuations in rainfall. 

However, by using fat body mass as an indicator of food availability, 

Colli et al. (1997) concluded that lizard reproductive seasonality is not 

constrained by food limitation in the cerrado. Long-term studies and comparisons 

among populations of the same species in distinct regions of the 

Cerrado Biome, however, are still lacking. 

SUGGESTIONS FOR FUTURE RESEARCH 

The knowledge about the cerrado herpetofauna is in its early infancy. 

New species are still being described, and large tracts of the biome have 

never been adequately sampled. Meaningful biogeographic analyses of 

the herpetofauna will depend heavily on museum data. Most large collections 

throughout the world have few specimens from the cerrado 

region, while several small collections in Brazilian institutions, housing 

potentially useful material, usually go unnoticed due to lack of integration 

and institutional support. The organization of such collections and 

their integration with larger institutions would be very helpful in future 

systematic and ecological work. 

Basic natural history information is currently lacking for most species 

of the cerrado herpetofauna. This kind of descriptive work, often undervalued 

by funding agencies and reviewers of scientific journals, is essential 

for the formulation of generalizations and testing of hypotheses. 

Species which are locally abundant and/or have broad distributions in the 

cerrados, such as Bufo paracnemis, Hyla albopunctata, H. minuta, Leptodactylus 

ocellatus, Physalaemus cuvieri, Cnemidophorus ocellifer, and 

Tropidurus oreadicus, provide model organisms for basic studies of natural 

history. Finally, detailed demographic and community-level studies 

on the cerrado herpetofauna are badly needed. Most importantly, they 

should be analyzed in concert with phylogenetic hypotheses to answer 

questions adequately such as, “To what extent do current environmental 

conditions and lineage effects determine the natural history characteristics 

of the herpetofauna?’’ A promising group to be investigated is the 

genus Tropidurus, with several species in open and forested biomes, for 

which well-resolved phylogenetic hypotheses are available. Another interesting 

venue of research is the investigation of the relative roles of structural 

features of the environment and fire history versus past geological



and climatic events on the distribution and composition of amphibian and 

reptile assemblages. For example, are species often associated with rock 

outcrops, such as Leptodactylus syphax and Phyllopezus pollicaris, 

dependent on some kind of resource only available at those sites, or do 

they represent a relictual distribution on ancient and dissected plateaus? 

The predictive power that could be gained with this kind of research 

would be invaluable in assisting conservation efforts for the biome, considering 

the accelerated pace of its destruction. 


We thank R. A. Brandão, W. Ronald Heyer, R. F. Juliano, D. Melo e Silva, 

C. C. Nogueira, P. S. Oliveira, P. H. Valdujo, and L. J. Vitt for insightful 

criticisms on earlier versions of the manuscript. We also thank IBAMA 

for granting access to the EFLEX study site. This study was partially supported 

by grants from the Brazilian Research Council (CNPq) to GRC 

(no. 302343/88-1), and RPB (no. 400381-97.4). 

REFERENCES 

Aichinger, M. 1987. Annual activity patterns of anurans in a seasonal 

neotropical environment. Oecologia 71:583–592. 

Araujo, A. F. B. 1992. “Estrutura Morfométrica de Comunidades de Lagartos 

de Áreas Abertas do Litoral Sudeste e Brasil Central.” Ph.D. thesis, 


Ávila-Pires, T. C. S. 1995. Lizards of Brazilian Amazonia (Reptilia: Squamata). 

Zool. Verh. 1995:3–706. 

Barreto, L. and G. V. Andrade. 1995. Aspects of the reproductive biology of 

Physalaemus cuvieri (Anura: Leptodactylidae) in northeastern Brazil. 

Amphibia-Reptilia 16:67–76. 

Bastos, R. P. and C. F. B. Haddad. 1996. Breeding activity of the neotropical 

treefrog Hyla elegans (Anura, Hylidae). J. Herpetol. 30:355–360. 

Bastos, R. P. and C. F. B. Haddad. 1999. Atividade reprodutiva de Scinax rizibilis 

(Bokermann) (Anura, Hylidae) na Floresta Atlântica, sudeste do 

Brasil. Rev. Bras. Zool. 16:409–421. 

Bertoluci, J. 1998. Annual patterns of breeding activity in Atlantic rainforest 

anurans. J. Herpetol. 32:607–611. 

Blackburn, D. G. and L. J. Vitt. 1992. Reproduction in viviparous South 

American lizards of the genus Mabuya. In W. C. Hamlett, ed., Reproductive 

Biology of South American Vertebrates, pp. 150–164. Berlin: 

Springer-Verlag.



Blamires, D., J. A. O. Motta, K. G. Sousa, and R. P. Bastos. 1997. Padrões de 

distribuição e análise de canto em uma comunidade de anuros no Brasil 

central. In L. L. Leite and C. H. Saito, eds., Contribuição ao Conhecimento 

Ecológico do Cerrado, pp. 185–190. Brasília: ECL/Universidade 

de Brasília. 

Brandão, R. A. and A. F. B. Araujo. 1998. A herpetofauna da Estação Ecológica 

de Águas Emendadas. In J. S. Marinho-Filho, ed., Vertebrados da 

Estação Ecológica de Águas Emendadas: História Natural e Ecologia de 

um Fragmento de Cerrado do Brasil Central, pp. 9–21. Brasília: Instituto 

de Ecologia e Meio Ambiente do Distrito Federal. 

Bugano, M. A. 1999. “Ecologia Reprodutiva de Scinax centralis (Anura: Hylidae) 

em uma Região do Brasil Central.’’ Master’s thesis, Universidade 

Federal de Goiás, Goiânia, Brazil. 

Cardoso, A. J., G. V. Andrade, and C. F. B. Haddad. 1989. Distribuição espacial 

em comunidades de anfíbios (Anura) no sudeste do Brasil. Rev. Bras. 

Biol. 49:241–249. 

Colli, G. R. 1991. Reproductive ecology of Ameiva ameiva (Sauria: Teiidae) 

in the Cerrado of central Brazil. Copeia 1991:1002–1012. 

Colli, G. R., A. F. B. de Araújo, R. da Silveira, and F. Roma. 1992. Niche partitioning 

and morphology of two syntopic Tropidurus (Sauria: 

Tropiduridae) in Mato Grosso, Brazil. J. Herpetol. 26:66–69. 

Colli, G. R., A. K. Péres, Jr., and M. G. Zatz. 1997. Foraging mode and reproductive 

seasonality in tropical lizards. J. Herpetol. 31:490–499. 

Colli, G. R., and D. S. Zamboni. 1999. Ecology of the worm-lizard Amphisbaena 

alba in the Cerrado of central Brazil. Copeia 1999:733–742. 

Colli, G. R., M. G. Zatz, and H. J. da Cunha. 1998. Notes on the ecology and 

geographical distribution of the rare gymnophthalmid lizard, Bachia 

bresslaui. Herpetologica 54:169–174. 

Crump, M. L. 1974. Reproductive strategies in a tropical anuran community. 

Misc. Publ. Mus. Nat. Hist. Univ. Kansas 61:1–68. 

Duellman, W. E. 1979. The South American herpetofauna: A panoramic view. 

In W. E. Duellman, ed., The South American Herpetofauna: Its Origin, 

Evolution, and Dispersal, pp. 1–28. Lawrence: The Museum of Natural 

History, The University of Kansas. 

Duellman, W. E. 1990. Herpetofaunas in neotropical rainforests: Comparative 

composition, history, and resource use. In A. H. Gentry, ed., Four 

Neotropical Rainforests, pp. 455–505. New Haven: Yale University 

Press. 

Duellman, W. E. 1993. Amphibians in Africa and South America: Evolutionary 

history and ecological comparisons. In P. Goldblatt, ed., Biological 

Relationships Between Africa and South America, pp. 200–243. New 

Haven: Yale University Press. 

Feio, R. N. and U. Caramaschi. 1995. Aspectos zoogeográficos dos anfíbios 

do médio Rio Jequitinhonha, nordeste de Minas Gerais. Revista Ceres 

42:53–61.



Haddad, C. F. B., G. V. Andrade, and A. J. Cardoso. 1988. Anfíbios anuros 

no Parque Nacional da Serra da Canastra, estado de Minas Gerais. Brasil 

Florestal 64:9–20. 

Heringer, E. P., G. M. Barroso, J. A. Rizzo, and C. T. Rizzini. 1977. A flora 

do Cerrado. In M. G. Ferri, ed., IV Simpósio sobre o Cerrado, pp. 

211–232. Belo Horizonte and São Paulo: Livraria Itatiaia Editora and 


Heyer, W. R. 1988. On frog distribution patterns east of the Andes. In P. E. 

Vanzolini and W. R. Heyer, eds., Proceedings of a Workshop on Neotropical 

Distribution Patterns, pp. 245–273. Rio de Janeiro: Academia 

Brasileira de Ciências. 

Heyer, W. R. 1994. Variation within the Leptodactylus podicipinus-wagneri 

complex of frogs (Amphibia: Leptodactylidae). Smithsonian Contrib. 

Zool. 546:1–124. 

Heyer, W. R. 1999. A new genus and species of frog from Bahia, Brazil 

(Amphibia: Anura: Leptodactylidae) with comments on the zoogeography 

of the Brazilian campos rupestres. Proc. Biol. Soc. Washington 

112:19–39. 

Heyer, W. R., A. S. Rand, C. A. G. d. Cruz, O. L. Peixoto, and C. E. Nelson. 

1990. Frogs of Boracéia. Arq. Zool. São Paulo 31:231–410. 

Hödl, W. 1990. Reproductive diversity in Amazonian lowland frogs. Fortschr. 

D. Zool. 38:41–60. 

Lamotte, M. 1983. Amphibians in savanna ecosystems. In F. Bourlière, ed., 

Ecosystems of the World: 13. Tropical Savannas, pp. 313–323. Amsterdam: 

Elsevier Scientific Publishing Company. 

Magnusson, W. E., C. R. Franke, and L. A. Kasper. 1986. Factors affecting 

densities of Cnemidophorus lemniscatus. Copeia 1986:804–807. 

Moreira, G. and L. Barreto. 1997. Seasonal variation in nocturnal calling 

activity of a savanna anuran community in central Brazil. Amphibia- 

Reptilia 18:49–57. 

Motta, J. A. O. 1999. “A Herpetofauna no Cerrado: Composição de Espécies, 

Sazonalidade e Similaridade.” Master’s thesis, Universidade Federal 

de Goiás, Goiânia, Brazil. 

Pianka, E. R. 1989. Desert lizard diversity: Additional comments and some 

data. Amer. Nat. 134:344–364. 

Pinto, M. G. M. 1999. “Ecologia das Espécies de Lagartos Simpátricos 

Mabuya nigropunctata e Mabuya frenata (Scincidae), no Cerrado de 

Brasília e Serra da Mesa (GO).’’ Master’s thesis, Universidade de Brasília, 


Poe, S. 1996. Data set incongruence and the phylogeny of crocodilians. Syst. 

Biol. 45:393–414. 

Pombal, J. P., Jr. 1997. Distribuição espacial e temporal de anuros (Amphibia) 

em uma poça permanente na Serra de Paranapiacaba, sudeste do Brasil. 

Rev. Bras. Biol. 57:583–594. 

Pombal, J. P., Jr. and R. P. Bastos. 1996. Nova espécie de Scinax Wagler, 1830



do Brasil central (Amphibia, Anura, Hylidae). Bol. Mus. Nac., nova sér., 

Zool., Rio de Janeiro 371:1–11. 

Pombal, J. P., Jr., I. Sazima, and C. F. B. Haddad. 1994. Breeding behavior of 

the pumpkin toadlet, Brachycephalus ephippium (Brachycephalidae). J. 

Herpetol. 28:516–519. 

Rivero-Blanco, C. and J. R. Dixon. 1979. Origin and distribution of the herpetofauna 

of the dry lowland regions of northern South America. In W. 

E. Duellman, ed., The South American Herpetofauna: Its Origin, Evolution, 

and Dispersal, pp. 281–298. Lawrence: The Museum of Natural 

History, The University of Kansas. 

Rodrigues, M. T. 1988. Distribution of lizards of the genus Tropidurus in 

Brazil (Sauria, Igaunidae). In W. R. Heyer and P. E. Vanzolini, eds., Proceedings 

of a Workshop on Neotropical Distribution Patterns, pp. 

305–315. Rio de Janeiro: Academia Brasileira de Ciências. 

Rossa-Feres, D. C. and J. Jim. 1994. Distribuição sazonal em comunidades 

de anfíbios anuros na região de Botucatu, São Paulo. Rev. Bras. Biol. 

54:323–334. 

Sartorius, S. S., L. J. Vitt, and G. R. Colli. 1999. Use of naturally and anthropogenically 

disturbed habitats in Amazonian rainforests by the teiid 

lizard Ameiva ameiva. Biol. Conserv. 90:91–101. 

Schluter, D., and R. E. Ricklefs. 1993. Convergence and the regional component 

of species diversity. In R. E. Ricklefs and D. Schluter, eds., Species 

Diversity in Ecological Communities: Historical and Geographical Perspectives, 

pp. 230–240. Chicago: The University of Chicago Press. 

Silva, J. M. C. 1995. Birds of the Cerrado Region, South America. Steenstrupia 

21:69–92. 

Silva, N. J., Jr. and J. W. Sites, Jr. 1995. Patterns of diversity of neotropical 

squamate reptile species with emphasis on the Brazilian Amazon and the 

conservation potential of indigenous reserves. Conserv. Biol. 9:873–901. 

Staton, M. A. and J. R. Dixon. 1977. The herpetofauna of the central llanos 

of Venezuela: Noteworthy records, a tentative checklist and ecological 

notes. J. Herpetol. 11:17–24. 

Van Sluys, M. 1993. The reproductive cycle of Tropidurus itambere (Sauria: 

Tropiduridae) in southeastern Brazil. J. Herpetol. 27:28–32. 

Vanzolini, P. E. 1948. Notas sôbre os ofídios e lagartos da Cachoeira de Emas, 

no município de Pirassununga, estado de São Paulo. Rev. Bras. Biol. 

8:377–400. 

Vanzolini, P. E. 1974. Ecological and geographical distribution of lizards in 

Pernambuco, northeastern Brazil (Sauria). Pap. Avul. Zool., S. Paulo 

28:61–90. 

Vanzolini, P. E. 1976. On the lizards of a Cerrado-caatinga contact, evolutionary 

and zoogeographical implications (Sauria). Pap. Avul. Zool., 

S. Paulo 29:111–119. 

Vanzolini, P. E. 1988. Distribution patterns of South American lizards. In 

P. E. Vanzolini and W. R. Heyer, eds., Proceedings of a Workshop on



Neotropical Distribution Patterns, pp. 317–343. Rio de Janeiro: Academia 

Brasileira de Ciências. 

Vitt, L. J. 1991. An introduction to the ecology of Cerrado lizards. J. Herpetol. 

25:79–90. 

Vitt, L. J. 1995. The ecology of tropical lizards in the Caatinga of northeast 

Brazil. Occ. Pap. Oklahoma Mus. Nat. Hist. 1:1–29. 

Vitt, L. J. and J. P. Caldwell. 1993. Ecological observations on Cerrado lizards 

in Rondônia, Brazil. J. Herpetol. 27:46–52. 

Vitt, L. J. and C. M. Carvalho. 1995. Niche partitioning in a tropical wet season: 

Lizards in the Lavrado area of northern Brazil. Copeia 

1995:305–329. 

Vitt, L. J. and G. R. Colli. 1994. Geographical ecology of a neotropical lizard: 

Ameiva ameiva (Teiidae) in Brazil. Can. J. Zool. 72:1986–2008. 

Vitt, L. J. and T. E. Lacher, Jr. 1981. Behaviour, habitat, diet and reproduction 

of the iguanid lizard Polychrus acutirostris in the Caatinga of northeastern 

Brazil. Herpetologica 37:53–63. 

Vrcibradic, D., and C. F. D. Rocha. 1998. Reproductive cycle and life-history 

traits of the viviparous skink Mabuya frenata in southeastern Brazil. 

Copeia 1998:612–619. 

Warming, E. 1892. Lagoa Santa. Et bidrag til den biologiske plantegeographi. 


Webb, S. D. 1978. A history of savanna vertebrates in the New World: Part 

II, South America and the Great Interchange. Annu. Rev. Ecol. Syst. 

9:393–426. 

Wiederhecker, H. C., A. C. S. Pinto, and G. R. Colli. 2002. Reproductive ecology 

of Tropidurus torquatus (Squamata: Tropiduridae) in the highly seasonal 

Cerrado biome of central Brazil. J. Herpetol. 36:82–91.


Appendix Checklist of the Cerrado Amphibians and Reptiles 

LISSAMPHIBIA: APODA: Caeciliidae: Siphonops annulatus, S. paulensis. Total 

species of caecilians: 2; total endemics: 0. ANURA: Bufonidae: Bufo crucifer, B. 

granulosus, B. guttatus, B. ocellatus, E B. paracnemis, B. rufus, B. typhonius; 

Centrolenidae: Hyalinobatrachium eurygnathum; Dendrobatidae: Colostethus 

goianus, E Epipedobates braccatus, E,C2 E. flavopictus, C2 E. pictus C2 ; Hylidae: 

Aplastodiscus perviridis, Corythomantis greeningi, Hyla albopunctata, H. alvarengai, 

E H. anataliasiasi, E H. biobeba, E H. branneri, H. cipoensis, E H. circumdata, H. 

crepitans, H. faber, H. ibitiguara, H. melanargyrea, H. microcephala, H. minuta, H. 

multifasciata, H. nana, H. nanuzae, E H. pardalis, H. pinima, H. polytaenia, H. 

pseudopseudis, E H. pulchella, H. punctata, H. raniceps, H. rubicundula, E H. saxicola, 

E H. sazimai, E H. tritaeniata, E Phasmahyla jandaia, E Phrynohyas venulosa, 

Phyllomedusa burmeisteri, P. centralis, E P. hypochondrialis, P. megacephala, Scinax 

acuminatus, S. canastrensis, E S. centralis, E S. duartei, S. fuscomarginatus, S. fuscovarius, 

S. luizotavioi, S. machadoi, E S. maracaya, E S. nebulosus, S. squalirostris, 

Trachycephalus nigromaculatus; Leptodactylidae: Adenomera bokermanni, A. martinezi, 

Barycholos savagei, E Crossodactylus bokermanni, C. trachystomus, 

Eleutherodactylus dundeei, E. fenestratus, E. juipoca, Hylodes otavioi, 

Leptodactylus camaquara, E L. chaquensis, L. cunicularius, E L. furnarius, L. fuscus, 

L. jolyi, E L. labyrinthicus, L. mystaceus, L. mystacinus, L. ocellatus, L. petersii, L. 

podicipinus, L. pustulatus, L. syphax, L. troglodytes, L. tapiti, E Odontophrynus 

americanus, O. cultripes, O. moratoi, E O. salvatori, E Physalaemus albonotatus, P. 

centralis, P. cuvieri, P. deimaticus, E P. evangelistai, E P. fuscomaculatus, P. nattereri, 

Proceratophrys cururu, E P. goyana, E Pseudopaludicola boliviana, P. falcipes, P. 

mineira, E P. mystacalis, P. saltica, P. ternetzi, Thoropa megatympanum, 

Microhylidae: Chiasmocleis albopunctata, C. centralis, E C. mehelyi, Dermatonotus 

muelleri, Elachistocleis bicolor, E. ovalis; Pseudidae: Lysapsus caraya, L. limellus, 

Pseudis paradoxa; Ranidae: Rana palmipes. Total species of anurans: 113; total 

endemics: 32. 

REPTILIA: TESTUDINES: Pelomedusidae: Podocnemis expansa, C2 Podocnemis 

unifilis C2 ; Chelidae: Chelus fimbriatus, Phrynops geoffroanus, P. gibbus, P. vanderhaegei, 

Platemys platycephala; Kinosternidae: Kinosternon scorpioides; Testudinidae: 

Geochelone carbonaria, C2 G. denticulata. C2 Total species of turtles: 10; 

total endemics: 0. CROCODYLIA: Alligatoridae: Caiman crocodilus crocodilus, C2 

C. crocodilus yacare, C2 C. latirostris, C1 C. niger, C1 Paleosuchus palpebrosus, C2 P. 

trigonatus. C2 Total species of crocodilians: 5; total endemics: 0. SQUAMATA: 

Amphisbaenidae: Amphisbaena alba, A. anaemariae, E A. crisae, E A. fuliginosa, A. 

leeseri, A. miringoera, E A. neglecta, E A. sanctaeritae, E A. silvestrii, E A. talisiae, E A. 

vermicularis, Bronia kraoh, E Cercolophia roberti, C. steindachneri, Leposternon 

infraorbitale, L. microcephalum. Total amphisbaenian species: 16; total endemics: 8. 

Hoplocercidae: Hoplocercus spinosus E ; Iguanidae: Iguana iguana C2 ; Polychrotidae: 

Anolis chrysolepis brasiliensis, A. meridionalis, E Enyalius bilineatus, E. brasiliensis, 

E. catenatus, Polychrus acutirostris; Tropiduridae: Stenocercus caducus, Tropidurus 

etheridgei, T. hispidus, T. itambere, E T. montanus, E T. oreadicus, T. gnarani, T. 

torquatus; Gekkonidae: Coleodactylus brachystoma, E C. meridionalis, Gonatodes 

humeralis, Gymnodactylus geckoides, Hemidactylus mabouia, Phyllopezus 

pollicaris; Teiidae: Ameiva ameiva, Cnemidophorus ocellifer, Kentropyx calcarata, 

K. paulensis, E K. striata, K. vanzoi, E K. viridistriga, Teius teyou, Tupinambis 

duseni, C2 T. merianae, C2 T. quadrilineatus, E,C2 T. teguixin C2 ; Gymnophthalmidae: 

Bachia bresslaui, E B. scolecoides, E B. cacerensis, E Cercosaura ocellata, Colobosaura 

modesta, Micrablepharus atticolus, E M. maximiliani, Pantodactylus schreibersii,



Vanzosaura rubricauda; Scincidae: Mabuya dorsivittata, M. frenata, M. guaporicola, 

M. nigropunctata (= bistriata); Anguidae: Ophiodes striatus. Total species of lizards: 

47; total endemics: 12. Anomalepididae: Liotyphlops beui, L. ternetzii; Leptotyphlopidae: 

Leptotyphlops albifrons, L. koppesi, L. munoai; Typhlopidae: Typhlops 

brongersmianus; Aniliidae: Anilius scytale; Boiidae: Boa constrictor, C2 Corallus caninus, 

C2 C. hortulanus, C2 Epicrates cenchria, C2 Eunectes murinus C2 ; Colubridae: 

Apostolepis assimilis, A. dimidiata, A. flavotorquata, E A. goiasensis, E A. lineata, E A. 

vittata E , Atractus pantostictus, A. reticulatus, A. taeniatus, Boiruna maculata, Chironius 

bicarinatus, C. exoletus, C. flavolineatus, C. laurenti, C. quadricarinatus, Clelia 

bicolor, C. plumbea, C. quimi, C. rustica, Dipsas indica, Drymarchon corais, Drymoluber 

brazili, Echinanthera occipitalis, Erythrolamprus aesculapii, Gomesophis 

brasiliensis, Helicops angulatus, H. carinicaudus, H. gomesi, H. leopardinus, H. 

modestus, Hydrodynastes bicinctus, H. gigas, C2 Imantodes cenchoa, Leptodeira 

annulata, Leptophis ahaetulla, Liophis almadensis, L. cobellus, L. dilepis, L. longiventris, 

L. maryellenae, L. meridionalis, L. miliaris, L. paucidens, L. poecilogyrus, 

L. reginae, L. typhlus, Lystrophis dorbignyi, L. histricus, L. mattogrossensis, L. nattereri, 

Mastigodryas bifossatus, M. boddaerti, Oxybelis aeneus, Oxyrhopus guibei, 

O. petola, O. rhombifer, O. trigeminus, Phalotris concolor, E P. lativittatus, E P. 

mertensi, P. multipunctatus, E P. nasutus, E P. tricolor, Philodryas aestivus, P. livida, E 

P. mattogrossensis, P. nattereri, P. olfersii, P. psammophideus, P. patagoniensis, Phimophis 

guerini, Pseudablabes agassizii, Pseudoboa neuwiedii, P. nigra, Pseudoeryx 

plicatilis, Psomophis genimaculatus, P. joberti, Rhachidelus brazili, Sibynomorphus 

mikanii, S. turgidus, Simophis rhinostoma, Spilotes pullatus, Tantilla melanocephala, 

Thamnodynastes rutilus, T. strigilis, Waglerophis merremi, Xenopholis undulatus; 

Elapidae: Micrurus brasiliensis E , M. frontalis, M. lemniscatus; Viperidae: Bothrops 

alternatus, B. itapetiningae, E B. moojeni, B. neuwiedi, Crotalus durissus. Total 

species of snakes: 107; total endemics: 11. 

Note: We included in the list all species known to occur within the borders of the Cerrado 

Biome as defined by the maps prepared during the workshop “Ações Prioritárias para a Conservação 

da Biodiversidade do Cerrado e Pantanal,” held in Brasília, Distrito Federal in 1999 

(see chapter 18). C1 = listed in Appendix I of CITES, C2 = listed in Appendix II of CITES, E 

= endemic.


13 

The Avifauna: Ecology, 

Biogeography, and Behavior 

Regina H. F. Macedo 

One of the earliest studies of the birds in the cerrado 

region was that of Sick (1955), whose records date to an expedition to 

Rio das Mortes in central Brazil in 1944. Subsequently, Sick’s (1965, 

1966) field work in Goiás, Mato Grosso and Pará, at the time almost 

untouched by civilization, led him to conclude that it was difficult to 

describe a “typical’’ avifauna for the cerrado region, and that it was relatively 

poor (in this chapter “cerrado region’’ or “Brazilian cerrados’’ refer 

to the biome itself, while “cerrado’’ refers to the sensu stricto vegetation 

type. See chapter 6 for detailed descriptions of cerrado physiognomies). 

He estimated a diversity of only 200 species for the biome, of which 

approximately 11% was endemic. Since then, other studies gradually 

increased the number of species to 837 (see table 13.1), and also identified 

typical species for the different cerrado region formations. Most of 

these species (759, 90.7%) are known or assumed to breed in the region 

(see appendix). Considering that Brazil has 1,590 resident bird species, 

the conservation of the Brazilian cerrados should be considered a top priority 

(chapter 18), since the region harbors approximately 48% of the 

country’s avifauna. Of the region’s species, approximately 30 are considered 

endemic. The habitat transformations currently underway in the 

Brazilian cerrados (chapter 5) will undoubtedly limit the breeding ranges 

of most of these species, while reducing or completely eliminating endemic 

bird populations. 

Avian community structure results from several complex factors, all 

interacting at various levels. I have grouped information on cerrado 

region birds into sections that reflect some of these elements: community 

242


The Avifauna: Ecology, Biogeography, and Behavior 243 

Table 13.1 Records of Bird Species Richness 

in the Cerrado Region 

Number Increase over 

Reference of species previous total (%) 

Sick 1955 245 

Fry 1970 263 18 (7) 

Müller 1973 357 94 (36) 

Negret et al. 1984 429 72 (20) 

Vuilleumier 1988 a 454 25 (5.8) 

Silva 1995 837 383 (84.4) 

a Includes birds from cerrado, chaco, and caatinga regions (see chapter 6). 

composition and biogeography, vegetation structure, seasonality effects, 

foraging and food resources, and impacts of fire. Additionally, I have 

focused upon the breeding patterns of a few species, including some 

socially breeding birds. In conclusion, I briefly discuss promising avenues 

for future work. 

COMMUNITY COMPOSITION AND BIOGEOGRAPHY 

The most species-rich families in the breeding avifauna are the Tyrannidae 

(111 species), Emberizidae (87 species), Formicariidae (58 species: in recent 

treatments subdivided into two families, with most of the cerrado region 

species included in the Thamnophilidae), Furnariidae (41 species), Trochilidae 

(36 species), and Psittacidae (33 species) (see appendix). Levels of 

endemicity range from a low estimate of 3.8% (Silva 1995) to higher estimates 

of 12% (Müller 1973) and 11% (Sick 1965; see table 13.2). 

Naturally, some species exhibit an abundance of individuals, while 

most species are represented by fewer individuals. In a study by Negret 

(1983) the 10 most abundant species in the Brazilian cerrados were: 

Aratinga aurea, Streptoprocne zonaris, Colibri serrirostris, Colaptes 

campestris, Elaenia flavogaster, Suiriri suiriri, Camptostoma obsoletum, 

Cyanocorax cristatellus, Neothraupis fasciata, and Ammodramus humeralis. 

Some species are generalists and can be found in most habitats within 

the cerrado region, for example: Colibri serrirostris, Milvago chimachima, 

Sporophila nigricollis, Piaya cayana, Nyctidromus albicollis, 

Galbula ruficauda, Chelidoptera tenebrosa, Turdus amaurochalinus, 

Basileuterus flaveolus, and Ramphocelus carbo (Fry 1970; Negret 1983).



Table 13.2 Endemic Birds of the Cerrado Region 

by Habitat and Range 

Family Species Habitat a Range 

Tinamidae Nothura minor Cerrado W 

Taoniscus nanus Cerrado W 

Cracidae Penelope ochrogaster Forest W 

Columbidae Columbina cyanopis Cerrado W 

Psittacidae Pyrrhura pfrimeri Forest R 

Amazona xanthops Cerrado W 

Caprimulgidae Caprimulgus candicans Cerrado W 

Trochilidae Augastes scutatus Cerrado R 

Furnariidae Geobates poecilopterus Cerrado W 

Philydor dimidiatus Forest W 

Synallaxis simoni Forest R 

Asthenes luizae Cerrado R 

Automolus rectirostris Forest W 

Formicariidae Herpsilochmus longirostris Forest W 

Cercomacra ferdinandi Forest R 

Rhinocryptidae Melanopareia torquata Cerrado W 

Scytalopus novacapitalis Forest R 

Tyrannidae Phyllomyias reiseri Forest W 

Polystictus superciliaris Cerrado R 

Knipolegus franciscanus Forest R 

Pipridae Antilophia galeata Forest W 

Emberizidae Poospiza cinerea Cerrado W 

Embernagra longicauda Cerrado R 

Sporophila melanops Cerrado R 

Charitospiza eucosma Cerrado W 

Paroaria baeri Forest/Cerrado R 

Saltator atricollis Cerrado W 

Porphyrospiza caerulescens Cerrado W 

Conothraupis mesoleuca Forest R 

Tachyphonus nattereri Forest R 

Parulidae Basileuterus leucophrys Forest W 

Corvidae Cyanocorax cristatellus Cerrado W 

Sources: Data from Sick (1997), Silva (1995, 1997), and Cavalcanti (1999). 

Note: W = widespread; R = restricted range. 

a “Cerrado” refers to the following open physiognomies: cerrado sensu stricto, 

campo cerrado, campo sujo, campo limpo, vereda, and campo rupestre. “Forest” 

includes gallery forest, the forest-like cerradão, and the transitional zone between 

these formations. See chapter 6 for description of vegetation physiognomies. 

However, many species (86 in Negret’s 1983 study) are specialists and 

occur in only one habitat. The greatest number of these exclusive species 

are restricted to gallery forests, followed by the cerrado. Negret (1983) 

considered that of the 118 species he found in the cerrado, 102 were treedwelling 

birds, and 13 were essentially terrestrial. The latter include mem-



bers of the family Tinamidae and also the very conspicuous Cariama 

cristata. In the veredas (chapter 6), the Mauritia palm groves are dominated 

by species such as Thraupis palmarum, Reinarda squamatta and 

Phacellodomus ruber. The Psittacidae are well represented: in the veredas 

of the region lives the world’s largest macaw, Anodorhynchus hyacinthinus, 

as well as Ara ararauna, with a wider distribution (Sick 1965). 

Most studies concerning the biogeography of the Brazilian cerrados 

avifauna were published in the last five years by Silva (1995a, b 1996, 

1997) and thus constitute the basis for any discussion on the subject (see 

Nores 1992, 1994, for alternative discussions regarding biogeography of 

South American birds). Species diversity results from two processes: speciation 

within a region and biotic interchange between regions (Ricklefs 

1990; Ricklefs and Schluter 1993). Species production in the Brazilian cerrados 

seems to have been low during the last two million years, and most 

of the endemic species are very old, dating from the middle and late Tertiary 

(Sick 1966; Silva 1995b). As for biotic interchanges, biogeographic 

analyses suggest that many species were received from neighboring biomes 

during the Pleistocene-Holocene cyclic climatic fluctuations (Silva 

1995b) and were able to maintain viable populations because their habitats 

did not disappear entirely through successive changes in the vegetation 

(Silva 1997; see also chapters 3, 6). 

Two of the three ancient savanna corridors that purportedly linked 

the central cerrado region with savannas of northern South America are 

congruent with present-day ranges of savanna birds. This led Silva (1995) 

to determine that the distribution patterns of these savanna-adapted birds 

do not conform to one of the most important assumptions of the refuge 

theory positing the existence of a savanna corridor right through central 

Amazonia during the dry and cold periods of the Quaternary (see Haffer 

1969, 1974). The patterns suggest that connections between the avifaunas 

of the northern Amazonia savannas and those of central Brazil probably 

occurred along the eastern borders, and not through the center of the 

Amazon rainforest. 

The headwaters of some major South American rivers (e.g., São Francisco, 

Tocantins, Araguaia, Paraguay) are located in central Brazil. Thus, 

the cerrado region is criss-crossed by gallery forests that run along rivers 

connecting the central Brazilian savanna with the Amazon or the Atlantic 

forests. Several researchers have suggested that these gallery forests may 

constitute mesic corridors through which forest-dependent organisms 

could have colonized the central savanna (chapters 6, 11, 14). Silva (1996) 

evaluated the influence of altitude and distance from the source areas on 

the distribution of birds within the Brazilian cerrados. Included in the 

analysis were 278 species, of which 200 were considered as Amazonian



and 78 as Atlantic forest elements. Most Amazonian elements (86%) do 

not extend more than 250 km into the Brazilian cerrados region, and no 

species extends more than 750 km. In contrast, only 50% of Atlantic forest 

birds are restricted to less than 250 km, and 14% extend more than 

1,000 km into the cerrado region. Silva (1996) also found that the altitudes 

of Amazonian elements are significantly lower than those of 

Atlantic forest elements. Thus, both distance from their centers of distribution 

as well as altitude appear to determine the distribution of birds in 

the gallery forest system of central Brazil. 

VEGETATION STRUCTURE 

The association between vegetation structure and number of species has 

been studied intensively (Orians 1969), especially since the significant 

contribution by MacArthur and MacArthur (1961), who postulated that 

bird species diversity could be predicted by the height profile of foliage 

density. Sick (1966) suggests that a large proportion of the typical species 

inhabiting the cerrado region are forest birds, living in trees, and cannot 

be considered savanna birds. In fact, approximately 70% of the breeding 

avifauna of the Brazilian cerrados is composed of species that are partially 

or totally dependent upon forests (gallery forests or tropical dry forests), 

a habitat that covers less than 15% of the region (Silva 1995a). 

In Negret’s (1983) study, 215 species were censused in all major vegetation 

types of the cerrados. Of these, 168 (64%) were observed within 

gallery forests; the poorest habitat was that of campo limpo, with only 31 

species and, predictably, the simplest structural complexity (chapter 6). 

The greatest abundance of individuals, however, was registered for the 

cerrado sensu stricto. In a previous, smaller-scale study in the northeast 

of the state of Mato Grosso, only 78 species were recorded for gallery 

forests, of which 33 occurred exclusively in that vegetation (Fry 1970). 

Parrots were poorly represented, while kingfishers, toucans, woodcreepers, 

and manakins were all abundant in gallery forests, which boasted, 

however, many fewer flycatchers than the cerrado. Fry (1970) also 

recorded a higher species total for the cerrado than for gallery forest, an 

incongruous finding relative to other studies, probably due to the short 

period of the study and the difficulty of sampling middle storey and 

canopy birds in gallery forests. 

A comparison of the more altered open physiognomies of the cerrado 

region (chapter 6) showed that, as expected, there is greater richness, 

diversity, and abundance of birds in the areas containing more shrubs and



trees (Tubelius 1997). Canopy-dwelling species appear more vulnerable 

to changes in their habitat. The simplification of the habitat from cerrado 

sensu stricto to campo sujo and eventually campo limpo apparently favors 

granivorous species such as Volatinia jacarina, Sicalis citrina, Nothura 

maculosa, and Myospiza humeralis. 

Some birds occupy different strata of gallery forests, thereby increasing 

species diversity. For example, three sympatric and partially syntopic 

Basileuterus warblers inhabit gallery forests of central Brazil (Marini and 

Cavalcanti 1993): B. leucophrys, B. hypoleucus, and B. flaveolus. The 

two species (B. flaveolus and B. leucophrys) that forage below 3 m height 

are not syntopic. Basileuterus hypoleucus, however, which occupies strata 

above 3 m, occurs syntopically with each of the other two species. 

A structural component of landscapes that is often overlooked, and 

which may be important in the organization of grassland bird communities, 

is the existence of forest edges. In the cerrado region, gallery forest 

edges may provide crucial resources that decline in the more open habitats 

during the dry season. Some tyrannid flycatchers, which breed during 

the rainy season, migrate elsewhere during the dry season; however, 

others (e.g., Tyrannus savana) disappear during the dry season from the 

cerrado but may be found feeding on fruits along the edges of gallery 

forests (Cavalcanti 1992). This pattern of gallery forest edge use is evident 

for other families, particularly for hummingbirds. The reverse situation 

occurs during the rainy season, when forest birds (e.g., Turdus spp. 

and Saltator similis) forage in adjacent cerrado areas, particularly rich in 

alates of termites and ants (Cavalcanti 1992). Thus, in addition to harboring 

its own avian community, gallery forests provide cover and 

resources for cerrado birds in periods of stress (e.g., grassland fires, dry 

season, cover from predators). 

SEASONALITY 

The climate of the cerrado region is highly seasonal, with marked dry 

(May–August) and rainy (September–March) seasons. This seasonality is 

evident in the differences exhibited by the vegetation as well as by insect, 

fruit, and flower abundance, all of which profoundly affect bird communities. 

The rigor and periodicity of moisture gradients influence all aspects 

of community composition and organization in the Brazilian cerrados, 

and condition such phenomena as the timing of breeding, flock occurrence 

and composition, migration, shifts in foraging behavior, and competition 

for resources.



Studies in other tropical areas have shown that spatially and temporally 

variable resources are important factors that induce seasonal bird 

movements as well as changes in bird abundances and in the composition 

of local avian communities. For Costa Rican birds, diet and habitat are 

strongly related to seasonal movements (Levey 1988; Levey and Stiles 

1992; Stiles 1983), and the patchy and ephemeral nature of resources such 

as fruits affects the structure of bird communities (Blake and Loiselle 

1991). In the humid tropical forest of Barro Colorado Island, smallseeded 

fruits ripen rather evenly throughout the year, whereas largeseeded 

fruits tend to be seasonal, affecting seed dispersal patterns (Smythe 

1970). In the cerrado region, it appears that the seasonal abundance of 

insects contributes significantly to the variation of richness and abundance 

of birds throughout the year. The maximum peak of insect abundance, 

in the rainy season, coincides with the arrival of migrants and also 

seems to affect patterns of movement within the region. The most abundant 

and well-known migratory species is Tyrannus savanna. This species’ 

range encompasses Mexico, Central America, and all of South America 

except Chile (Sick 1997). It arrives in the cerrado region in August and 

September, reproduces in the region in October and November, and leaves 

in January and February, probably returning to the Amazon region 

(Negret and Negret 1981). 

Likewise, flowering and fruiting patterns affect the movements of 

birds in the region. Ornithophilous plants flower almost continuously 

within gallery forests, providing a source of nectar during the dry season, 

while in campo sujo flowering happens only during the rainy season 

(Oliveira 1998). Thus, a local movement of hummingbirds into gallery 

forests is probably a general phenomenon during the dry season. In some 

cases, specific plants flower only during very circumscribed periods and 

provide much-needed resources. Such plants may be regarded as pivotal 

species in their habitats but have been discussed only occasionally in the 

literature. The leguminous tree Bowdichia virgilioides occurs in cerrado 

habitat in various areas of Brazil, and flowers only at the end of the dry 

season (Rojas and Ribon 1997). A small but diverse group of birds dependent 

upon this tree includes: six species of hummingbirds (Colibri serrirostris, 

Chlorostilbon aureoventris, Eupetomena macroura, Calliphlox 

amethystina, Amazilia lactea, and perhaps A. fimbriata); two Coerebidae 

(Dacnis cayana and Coereba flaveola); as well as some species that eat the 

flowers (Aratinga aurea, Psittacidae; Tangara cayana, Thraupis sayaca, 

Thraupidae; and D. cayana). A similar pivotal species phenomenon 

was described for the rain forest tree Casearia corymbosa in the La Selva 

Biological Station in Costa Rica: fruiting was restricted to an annual



period of fruit scarcity, thus supporting three obligate frugivores in the 

area (Howe 1977). 

Another seasonal effect is that of mixed-species flock formation. 

Although these can be found year-round, their highest occurrence is early 

in the dry season, the non-reproductive period, with a sharp decline during 

the breeding season (Davis 1946; Silva 1980; Alves and Cavalcanti 

1996). Not only the frequency of flocks, but also their composition, is 

affected by the seasonality in the region. Because of migrants that arrive 

during the rains, species such as Elaenia chiriquensis, Myiarchus swainsoni, 

and Tyrannus savanna participate in flocks only during the period 

from August to March (Alves and Cavalcanti 1996). 

Breeding periods worldwide are associated with food availability for 

the young, which means that in temperate regions breeding occurs during 

the spring and summer. In the cerrados region, as in other savannas and 

areas with sharply defined rainy seasons, breeding coincides with peak 

vegetation growth and insect abundance. Thus, insectivores should breed 

somewhat earlier than granivorous species or those that depend upon tall 

grasses for nesting sites (Lack 1968; Young 1994). For the cerrado region, 

there have been only a few in-depth studies recording specific reproductive 

periods, courtship displays, nestling growth patterns, and reproductive 

success rates. The manakin Antilophia galeata, an inhabitant of 

gallery forests, has its peak sexual activity from August to November, 

coinciding with the beginning to middle of the rainy season, although 

males maintain their territory all year (Marini 1992b). For Neothraupis 

fasciata, nesting occurs from September to January or later (Alves and 

Cavalcanti 1990). The macaw Ara ararauna, a cavity nester which uses 

Mauritia palm trees and initiates courtship behavior in May, nests 

between August and December (Bianchi 1998). A few species, such as 

Rhynchotus rufescens, Nothura maculosa, and Columbina talpacoti, 

seem to breed year-round, although peaking during the rains (Soares 

1983; Couto 1985; Setubal 1991). Most species appear to have their 

breeding periods restricted to the rainy season, and descriptive data (but 

few in-depth studies) are available for most (Sick 1997). 

FORAGING AND FOOD RESOURCES 

Foraging studies for tropical species have been conducted mostly for 

forest birds, where foraging specialization may promote coexistence of 

many species, leading to complex communities (Orians 1969). Less is 

known about the diet and foraging patterns of savanna birds. Food may



be superabundant during breeding periods in some habitats and systems 

but limited during other periods, thus reducing the reproductive success 

of some individuals and affecting survival of both adults and nestlings 

(Martin 1987). In nearly all diet categories (insectivory, granivory, frugivory, 

and nectarivory), the cerrado region avifauna offers interesting 

possibilities for studies. Detailed descriptions of the diet of most species 

are not available, but some studies have described foraging specializations 

(see Alves 1991; Marini 1992a). 

In the cerrado region communities there is an elevated number of 

insectivores (113 species of flycatchers alone), and some studies have 

examined how food availability allows the coexistence of so many species 

with broadly overlapping diets. For example, Negret (1978) considered 

eight flycatchers at Distrito Federal (central Brazil) to determine their ecological 

niche and trophic relations: Xolmis cinerea, Xolmis velata, 

Knipolegus lophotes, Tyrannus savanna, Megarhynchus pitangua,Pitangus 

sulphuratus, Suiriri suiriri, and Leptopogon amaurocephalus. These 

species vary in their habitats, which range from urban areas to gallery forest. 

The study clarified the relation between bill morphology and prey 

items. Megarhynchus pitangua, which has the widest bill, for instance, 

foraged upon larger insects. This characteristic also corresponded to the 

“flycatching’’ and “leaf-snatching’’ foraging modes. The shortest bills 

corresponded to the “leaf-gleaning’’ and “ground-feeding’’ behaviors. No 

tendency was found for predation of a particular group of insects. Instead, 

the study suggests an association between the feeding behavior of each 

species and the habitat occupied by its prey. 

Depending on the habitat, from 50% to 90% of tropical shrubs and 

trees have their seeds dispersed by vertebrates. However, knowledge of the 

diet of tropical birds is still limited, and it is often difficult to determine 

which species are frugivorous (Herrera 1981; Fleming et al. 1987). The 

morphology of birds and plants affects which species are essentially frugivorous 

(Moermond and Denslow 1985), with additional important 

characteristics including spatio-temporal distribution and nutritional characteristics 

of the fruits. In the Brazilian cerrados, fruit-consuming birds are 

not primarily frugivorous, using fruits instead to complement their diets. 

Roughly about 50% to 60% of plants in the cerrado region are dispersed 

by animals (Gottsberger and Silberbauer-Gottsberger 1983; Pinheiro 

1999). In the Brazilian cerrados, birds constitute the greatest 

proportion of frugivorous animals, with 75 species that can be classified 

as frugivorous or partially frugivorous (Bagno 1998). A surprisingly large 

number of cerrado plants are potentially bird-dispersed, large even when 

compared to wet forests. For example, in the La Selva Biological Station,



Costa Rica, at least 137 plant species were found to be bird-dispersed, 

although the study only included plants that had fruits below 10 m (i.e., 

plants in the mid-canopy and canopy were not included; Loiselle and Blake 

1991). Thus, this most likely is a low estimate of the number of plants dispersed 

by birds in that area. In the cerrado region, over 179 plants may be 

bird-dispersed. Important birds for seed dispersal include Rhea americana, 

Neothraupis fasciata, Rhynchotus rufescens, Tinamus solitarius, some 

species of Nothura, pigeons and doves (e.g., Columbina cyanopis and Uropelia 

campestris), parakeets, parrots, macaws, toucans, blackbirds of the 

family Icteridae, and several Tyrannidae and Furnariidae (Gottsberger and 

Silberbauer-Gottsberger 1983; Alves 1991; Paes 1993). 

The frugivore guild associated with the Mauritia flexuosa palm is 

composed not only of birds from various families, but also of birds 

exhibiting a wide range of bill gape widths. Foraging behaviors also vary 

greatly: there are terrestrial birds that pick up fruits dropped at the base 

of the tree by canopy foragers; birds that consume the fruit on the tree; 

those that consume the fruit only partially; and birds that take it away in 

flight (Villalobos 1994). Over 90% of the fruit consumption is by three 

psittacids, Ara manilata, A. ararauna, and Amazona aestiva. They usually 

eat only part of the fruit pulp and then drop it at the base of the tree. 

Once the macaws drop the fruit, a number of mammals feed upon it, 

including some rodents, tapirs, maned wolves, and opossums. Other birds 

that also feed on the fruits from this palm include Schistochlamys 

melanopis and Thraupis palmarum (Thraupidae), Gnorimopsar chopi 

(Icteridae), Polyborus plancus (Falconidae), Cyanocorax cristatellus 

(Corvidae), and Porzana albicollis (Rallidae). 

Nectarivory in the tropics has also attracted much interest. There are 

more than 300 species of hummingbirds (Trochilidae) distributed in the 

New World, with peak diversity occurring in the tropics. The combination 

of ecological constraints, usually in the form of interspecific competition 

and mutualistic coevolution with flowers, may be responsible for 

the diversity of morphologies and species of hummingbirds (Brown and 

Bowers 1985). 

Stiles (1985) suggests the existence of a relation between the number 

of plant species with adequate resources and the total number of hummingbird 

species that can be supported in an area. He cites as an example 

the existence of about 20 hummingbird species and a community of 

50 plants they pollinate in the La Selva Biological Station (Costa Rica). In 

contrast, in a southeastern site in Puerto Rico there are only three hummingbird 

species and 13 plants they pollinate (Kodric-Brown et al. 1984). 

These observations would lead one to expect but a small hummingbird



community in the cerrado region, since the number of plants that present 

the ornithophily syndrome (tubular corollas, conspicuous colors, etc.) is 

small (Silberbauer-Gottsberger and Gottsberger 1988; see chapter 17). 

However, that is not the case, as the cerrado region hummingbird community 

is extremely diverse, with at least 36 species listed (most common 

include: Amazilia fimbriata, Chlorostilbon aureoventris, Colibri serrirostris, 

Eupetomena macroura, Phaethornis pretrei, and Thalurania furcata; 

Negret et al. 1984; Oliveira 1998). It is important to keep in mind that 

the elevated number of species listed to date is for the whole Cerrado Biome, 

an area of approximately 2 million km 2 , while hummingbirds recorded at 

La Selva are within a 1,000 ha area. Nonetheless, the elevated number of 

hummingbirds in the region may reflect a generalistic and opportunistic foraging 

behavior: namely, the use of flowers that do not have the floral attributes 

commonly associated with hummingbird pollination (Oliveira 1998). 

Several of the flowers visited by hummingbirds (e.g., species of Inga, 

Vochysia, Qualea, Bauhinia and Caryocar) are regularly pollinated by other 

animals, ranging from moths and bees to bats (Oliveira 1998; chapter 17). 

Moreover, an experimental study with artificial feeders suggests that cerrado 

region hummingbirds may not have fixed preferences for certain floral 

characteristics, but instead a large adaptive capacity to exploit resources 

that are regionally advantageous (Carvalho and Macedo in prep.). 

IMPACT OF FIRE 

Fire may destroy or damage individuals, affect growth forms or reproduction, 

or in some way alter the environment, providing new opportunities 

for some organisms (Frost 1984; chapters 4, 9). The impact of fire 

upon bird communities has been studied primarily in temperate regions; 

fewer studies have occurred in the tropics, and these mostly in Africa 

(Frost 1984; Trollope 1984) and Australia (Luke and McArthur 1977). 

In the cerrado region, both natural and intentional fires occur more 

frequently toward the end of the dry season. Cavalcanti and Alves (1997) 

conducted a study in a cerrado area to examine the impact of fire on the 

avian community and to test some concepts proposed by Catling and 

Newsome (1981): (1) the existence of fire specialists; (2) a prevalence of 

ecological generalists; and (3) an expected tendency toward low species 

diversity in areas subjected to frequent burning. They sampled the avifauna 

before and after burning occurred, and recorded parameters involving 

population changes, site fidelity of marked individuals, foraging 

behavior, and the identification of specialists. Their results were partly 

consistent with Catling and Newsome’s (1981) arguments. They identi-



fied a fire specialist, Charitospiza eucosma, which was seen only sporadically 

before the fire but was captured freely after the fire. However, the 

authors also point out that the predictions regarding a fire-adapted avifauna 

do not exclude other explanations for the low species diversity. A 

study comparing burned and unburned open savanna sites during a fiveyear 

period showed similar species’ numbers and abundances, which does 

not support the suggestion of fire-adapted avifaunas (Figueiredo 1991). 

Although particular traits characterizing a fire-adapted avifauna may 

not be obvious, birds exhibit some typical behaviors during fires. For 

example, small birds may seek refuge in protected sites (termite mounds) 

or simply avoid the area by moving to unaffected areas. In various 

instances birds are attracted to fires to seek out prey trying to escape. 

Migrant species are often associated with fires that provide them with easy 

prey. Raptors, the most frequent fire-followers, include: Heterospizias 

meridionalis, Polyborus plancus, Buteogallus urubitinga, Ictinia plumbea, 

and Cathartes aura (Sick 1983). Among terrestrial birds, fires attract 

Cariama cristata and Rhea americana, and often other birds such as Streptoprocne 

zonaris and Tyrannus melancholicus (Sick 1965). The incidence 

of fire may also be associated with a populational increase in some species 

such as Rhynchotus rufescens and Nothura maculosa (Setubal 1991). 

Most studies concerning the impact of fire upon the cerrado region 

avifauna have been restricted to birds of more open cerrado physiognomies 

(chapter 6). Community changes after a fire in a gallery forest have 

been described only in Marini and Cavalcanti (1996). They found that 

similarity coefficients for the avian communities differed little before and 

after the fire, considering the total community. However, considering 

groups of species related to the types of habitat most frequently used, they 

found that species more dependent upon gallery forests were more 

severely affected, with community structure differing between periods. 

The authors speculate that possibly cerrado species have evolved behavioral 

and perhaps physiological adaptations for fire (e.g., feeding on 

arthropods after the fire), whereas forest birds may not have the same 

responses. Therefore, fires may decrease bird populations in gallery forest 

while having very little impact on cerrado bird community structure. 

BREEDING PATTERNS 

Very few tropical birds have been studied in detail and over a long-term 

period; consequently, a vast number of questions concerning their reproduction 

remain unanswered. Additionally, because of the contrast between 

temperate and tropical regions in temperature extremes, availability of



resources, predator pressure and other crucial factors, we can expect to 

find important differences in birds’ breeding characteristics, such as the 

role of territories, investment in parental care, male strategies through 

singing and displays, and the evolution of social systems, to name a few. 

The literature is prolific in speculative concepts, sometimes based on studies 

of temperate birds and occasionally resulting from analyses of general 

data from a limited number of tropical birds. For instance, although Ricklefs 

(1976) advanced several hypotheses that might explain the slower 

growth rate of nestlings of tropical species despite higher nest predation 

rates, these remain largely unverified. 

In this section I have highlighted some facets of the reproductive biology 

of a few species, ranging from very common species, adapted to disturbed 

habitats, to those requiring specific habitats. While a few studies 

exist concerning behavior and breeding biology of a limited number of 

species in the cerrado region, in-depth studies over long-term periods for 

even the most common birds are virtually nonexistent. 

Columbina talpacoti, for example, a widely distributed and common 

bird, has been studied mostly in disturbed areas, although it also occurs 

in urban areas and undisturbed cerrado. In the region it reproduces all 12 

months of the year (Couto 1985; Cintra and Cavalcanti 1997), peaking 

during the rainy period (Couto 1985). The ability to maintain reproduction 

even during the dry season may be due to the capacity of producing 

“pigeon milk’’ to nurture the young, as well as to the plasticity allowing 

occupation of disturbed areas. The main food source that probably 

allowed year-round reproduction in the study population was the natural 

grain from the surrounding cerrado as well as from neighboring farms, 

which included peanuts, sunflowers, wheat, rice, beans, corn, etc. Nests 

were found mostly in coffee bushes, and the state of the plants was important 

in determining reproductive activity: when the foliage decreased, so 

did the number of nests. 

Another very common, though essentially unstudied bird in disturbed 

cerrado sensu stricto and campo sujo areas is Volatinia jacarina. This 

species is widely distributed, and its range includes areas other than the 

cerrado region. However, its reproductive behavior has been virtually 

ignored, although it presents interesting questions relevant to the understanding 

of the evolution of mating systems in general. In this species, 

males form loose aggregations and execute displays, in the form of vertical 

leaps from perches, that resemble traditional leks. Almeida and 

Macedo (2001) found that the number of displaying males declined 

throughout the season, although the intensity in displays showed no variation. 

Focal males observed had significantly different display rates and



also defended territories of different sizes. The size of territories, ranging 

from 13.0 m 2 to 72.5 m 2 , was not associated with the average display rate 

of their owners. Additionally, no association was found between the display 

of males and the vegetation structure of territories. Observations of 

parental care indicate that nests are commonly built within male territories, 

and that both sexes feed chicks. These results suggest that the mating 

system of this species, despite the aggregations of displaying males, 

does not fit the traditional lek system, as had been previously proposed in 

Murray’s (1982) and Webber’s (1985) brief observations. However, the 

underlying reasons for the aggregations of males and their territories are 

still obscure, as well as mating strategies for males and females. 

The availability of nesting sites and specificity of their characteristics 

are important factors regulating breeding in many birds. Cavity nesting is 

common in the tropics, as it allows a certain degree of freedom from 

weather extremes and predation. However, natural cavities may constitute 

a limited resource. The macaw Ara ararauna, an obligatory cavitynesting 

species distributed from southern Central America to the central 

region of Brazil (Sick 1997), is an obligate cavity-nesting species, with 

very specific habitat requirements. The breeding season for this species 

occurs between August and December, although courtship is initiated in 

May (Bianchi 1998). In the Brazilian cerrados these birds nest within cavities 

of the buriti-palm Mauritia flexuosa in the vereda formations (chapter 

6). They lay from two to four eggs that hatch asynchronously, and 

nestlings develop during approximately 78 days. The availability of cavities 

found during the study was low (about 5% within a palm tree field), 

and their occupancy much reduced, ranging from 10–15%, that is, less 

than 1% of palms per field. In disturbed areas the availability of cavities 

was higher (43.5%), but the occupancy was even lower (5.1%). Analyses 

suggested that the important parameters that determine cavity choice are: 

(1) total trunk height; (2) cavity height; and (3) cavity depth. 

Antilophia galeata is an atypical manakin of the dry and flooded 

gallery forests of central Brazil, in contrast to most others of its family, 

which inhabit the Amazonian and Atlantic forests (Marini 1992a, b; Sick 

1997). Reproduction coincides with the rainy season. It resembles other 

manakins in that it is highly dichromatic and essentially frugivorous, but 

differs in that it establishes long term pairbonds, whereas other known 

members of the Pipridae are lekking species and polygynous (Marini 

1992b). Nest building and nestling care appear to be performed only by 

females. Its nesting biology, in general, conforms to what is known about 

other manakins, including details concerning nest architecture, male 

gonadal development, and molting patterns (Marini 1992b).



SOCIALLY BREEDING SPECIES 

Patterns of sociality have long interested biologists, and the number of 

studies of social animals has been increasing steadily (Brown 1987). 

Because the most common breeding pattern in temperate birds is that of 

pairs in territories, it was not until Davis (1940a,b, 1941) and Skutch 

(1959) described year-round social groups of birds in tropical America 

that social organization in birds became a topic of interest. These birds 

were dubbed cooperative breeders, and can be subdivided loosely into 

species where: (1) only a pair reproduces, but with the aid of several nonparental 

helpers; and (2) several co-breeders share breeding opportunities 

within the group, using one or several nests within the same territory 

(Brown 1987; Stacey and Koenig 1990). Cooperative breeding is relatively 

rare, and is known to occur in only about 220 of the roughly 9,000plus 

species of birds (Brown 1987). However, because the geographic 

distribution of cooperative breeders is biased toward lower latitudes, 

reaching its greatest abundance in the neotropics (Brown 1987), it is likely 

that many more such species will be described in the future. 

The Guira guira (Cuculidae) system has been under investigation for 

approximately 10 years, revealing a complex and intriguing social system 

briefly described here (Macedo 1992, 1994; Quinn et al. 1994; Macedo 

and Bianchi 1997a,b; Melo and Macedo 1997; Macedo and Melo 1999). 

Guira cuckoos breed primarily during the period from August to March, 

with groups renesting as many as five times. Groups average approximately 

six individuals, but membership may reach up to 15 birds occupying 

a single territory and using the same nest. Communal clutch size ranges 

from a couple of eggs to as many as 26 eggs laid by the various females. 

One of the most intriguing aspects of breeding in this species is the practice 

of egg ejection by group members, with nests averaging a loss of about 

four eggs. Although an average of 10 eggs are laid per nest, due to the ejection 

of eggs only about five survive to hatching. The brood that hatches is 

then further reduced through infanticide practiced by group members. The 

genetic relations among group members and chicks has been studied in a 

preliminary way, indicating that: (1) the mating system is polygynous as 

well as polyandrous; (2) several individuals contribute to the communal 

brood; (3) reproductive monopoly by a single pair does not occur; and (4) 

some individuals may be excluded from a breeding bout. 

The social system and the helping behavior of the tanager Neothraupis 

fasciata have been described in Alves and Cavalcanti (1990) and 

Alves (1990). This species forms groups of up to six members, which 

actively participate in mixed-species flocks. Nesting occurs from Septem-



ber to at least January. The most common social structure recorded was 

that of a pair with nestlings and young from former nesting bouts. Groups 

were generally stable in composition and remained so between years. Pairs 

are socially monogamous and, while males helped care for nestlings, only 

the females were seen building nests. Average clutch size is 2.8 eggs, but 

ranges from two to five eggs, and incubation lasts for about 15 to 17 days. 

Nests are frequently parasitized by Molothrus bonariensis. Of the seven 

nests monitored in these studies, male and female helpers occurred in 

three, and assisted through feeding of nestlings, acting as sentinels and in 

territorial defense. 

Although the study of these cooperative breeders provides some 

detailed information, additional research is needed to answer many 

remaining questions. For example, for Guira guira, the genetic identity of 

infanticidal adults, hierarchical organization within groups, and the distribution 

of reproductive opportunities among group members are 

promising lines of investigation. There are also studies comparing this 

species with the other crotophagine in the region, Crotophaga ani. Several 

research questions remain for Neothraupis fasciata as well; it would 

be of interest not only to determine the genetic identity of helpers, but also 

to quantify their help and compare the success rates of groups with and 

without auxiliaries, as well as dispersal patterns of the young. 

RESEARCH POSSIBILITIES 

In presenting a general overview of the information available for the cerrado 

region avifauna, I hope to have provided a catalyst for future 

research. Although descriptive work has provided meaningful background 

data, a thorough understanding of the Brazilian cerrados bird 

community is mostly lacking, and there are extensive opportunities for 

research in most areas. At the community level, bird abundance and distribution 

information remain scanty for the vast areas of uncensused cerrados 

of central and northern Brazil. Additionally, there is a serious gap 

in our knowledge of gallery forest birds, because they are difficult to 

observe and capture. Studies of habitat partitioning patterns, dependent 

upon resource abundance (which is different in each cerrado region habitat), 

would also provide important data on community structure. In the 

area of conservation it is likely that the study of natural patches of forest 

within the cerrado region (e.g., gallery forests) will provide important 

information concerning the likely consequences of anthropogenic fragmentation. 

Also, studies on seed dispersal and forest regeneration would



be pertinent to conservation issues. Autoecological and behavioral studies 

of cerrado region birds are almost nonexistent, and information is 

needed on the social structure, mating system, and reproductive biology 

of almost all species. Other topics of interest include: competition and predation, 

differential resistance to parasites, long-term impact of fire upon 

bird communities, habitat fragmentation and its effects (e.g., expansion 

in host species for Molothrus bonariensis; changes in community composition), 

timing of breeding as it relates to nest predation and/or weather 

and food abundance, and social behavior (e.g., in breeding and flocking) 

in relation to behavioral and morphological attributes of the species 

involved. In short, the topics listed above, a far from comprehensive list, 

include broad categories for which understanding is scanty at best. 


I thank J. M. C. Silva and two anonymous reviewers for their comments 

and suggestions on previous drafts of the manuscript. I also thank the 

Brazilian Research Council (CNPq) for a research grant during the time 

this manuscript was produced. 

REFERENCES 

Almeida, J. B. and R. H. Macedo. 2001. Lek-like mating system of the monogamous 

blue-black grassquit. Auk 118:404–411. 

Alves, M. A. S. 1990. Social system and helping behavior in the white-banded 

tanager (Neothraupis fasciata). Condor 92:470–474. 

Alves, M. A. S. 1991. Dieta e táticas de forrageamento de Neothraupis fasciata 

em cerrado no Distrito Federal, Brasil (Passeriformes: Emberizidae). 

Ararajuba 2:25–29. 

Alves, M. A. S. and R. B. Cavalcanti. 1990. Ninhos, ovos e crescimento de 

filhotes de Neothraupis fasciata. Ararajuba 1:91–94. 

Alves, M. A. S. and R. B. Cavalcanti. 1996. Sentinel behavior, seasonality, and 

the structure of bird flocks in a Brazilian savanna. Ornitol. Neotropical 

7:43–51. 

Bagno, M. A. 1998. As aves da Estação Ecológica de Águas Emendadas. In 

J. Marinho-Filho, F. Rodrigues, and M. Guimarães, eds., Vertebrados da 

Estação Ecológica de Águas Emendadas, pp. 22–33. Brasília: Governo 

do Distrito Federal. 

Bianchi, C. A. 1998. “Biologia Reprodutiva da Arara-canindé (Ara ararauna, 

Psittacidae) no Parque Nacional das Emas, Goiás.” Master’s thesis, Universidade 

de Brasília, Brasília, Brazil.



Blake, J. G. and B. A. Loiselle. 1991. Variation in resource abundance affects 

capture rates of birds in three lowland habitats in Costa Rica. Auk 

108:114–130. 

Brown, J. H. and M. A. Bowers. 1985. Community organization in hummingbirds: 

Relationships between morphology and ecology. Auk 102:251–269. 

Brown, J. L. 1987. Helping and Communal Breeding in Birds. Princeton, N.J.: 

Princeton University Press. 

Carvalho, C. B. V. and R. H. Macedo. In prep. Memory, learning and floral 

characteristics in foraging patterns of neotropical hummingbirds. 

Catling, P. C. and A. E. Newsome. 1981. Responses of the Australian vertebrate 

fauna to fire: An evolutionary approach. In A. M. Gill, R. H. 

Groves, and I. R. Noble, eds., Fire and the Australian Biota, pp. 

273–310. Canberra: Australian Academy of Science. 

Cavalcanti, R. B. 1992. The importance of forest edges in the ecology of open 

country cerrado birds. In P. A. Furley, J. Proctor, and J. A. Ratter, eds., 

The Nature and Dynamics of Forest-Savanna Boundaries, pp. 513–517. 

London: Chapman and Hall. 

Cavalcanti, R. B. and M. A. S. Alves. 1997. Effects of fire on savanna birds 

in central Brazil. Ornitol. Neotropical 8:85–87. 

Cintra, R. and R. B. Cavalcanti. 1997. Intrapopulational variation in growth 

rates of nestling ruddy ground-doves (Columbina talpacoti, Aves: 

Columbidae) in Brazil. Rev. Bras. Ecol. 1:10–14. 

Couto, E. A. 1985. “O Efeito da Sazonalidade na População da Rolinha 

(Columbina talpacoti) no Distrito Federal.” Master’s thesis, Universidade 


Davis, D. E. 1940a. Social nesting habits of the smooth-billed ani. Auk 

57:179–218. 

Davis, D. E. 1940b. Social nesting habits of Guira guira. Auk 57:472–484. 

Davis, D. E. 1941. Social nesting habits of Crotophaga major. Auk 58:179–183. 

Davis, D. E. 1946. A seasonal analysis of mixed flocks of birds in Brazil. Ecology 

27:168–181. 

Figueiredo, S. V. 1991. “Efeito do Fogo Sobre o Comportamento e Sobre a 

Estrutura da Avifauna de Cerrado.’’ Master’s thesis, Universidade de 


Fleming, T. H., R. Breitwisch, and G. H. Whitesides. 1987. Patterns of tropical 

vertebrate frugivore diversity. Ann. Rev. Ecol. Syst. 18:91–109. 

Frost, P. G. H. 1984. The responses and survival of organisms in fire-prone 

environments. In P. de V. Booysen and N. M. Taiton, eds., Ecological 

Effects of Fire in South African Ecosystems, pp. 273–309. Berlin: 


Fry, C. H. 1970. Ecological distribution of birds in north-eastern Mato 

Grosso State, Brazil. An. Acad. Bras. Ciênc. 42:275–318. 

Gottsberger, G. and I. Silberbauer-Gottsberger. 1983. Dispersal and distribution 

in the cerrado vegetation of Brazil. Sonderbd. naturwiss. Ver. Hamburg 

7:315–352.



Haffer, J. 1969. Speciation in Amazonian forest birds. Science 165:131–137. 

Haffer, J. 1974. Avian Speciation in South America. Cambridge: Publ. Nuttall 

Ornith. Club 14. 

Herrera, C. M. 1981. Are tropical fruits more rewarding to dispersers than 

temperate ones? Amer. Nat. 118:896–907. 

Howe, H. F. 1977. Bird activity and seed dispersal of a tropical wet forest tree. 


Kodric-Brown, A., J. H. Brown, G. S. Byers, and D. F. Gori. 1984. Organization 

of a tropical island community of hummingbirds and flowers. 


Lack, D. 1968. Ecological Adaptations for Breeding in Birds. London: 

Meuthuen. 

Levey, D. J. 1988. Tropical wet forest treefall gaps and distributions of understory 

birds and plants. Ecology 69:1076–1089. 

Levey, D. J. and F. G. Stiles. 1992. Evolutionary precursors of long-distance 

migration: Resource availability and movement patterns in neotropical 

landbirds. Amer. Nat. 140:447–476. 

Loiselle, B. A. and J. G. Blake. 1991. Temporal variation in birds and fruits 

along an elevational gradient in Costa Rica. Ecology 72:180–193. 

Luke, R. H. and A. G. McArthur. 1997. Bushfires in Australia. Canberra: 

Australian Government Publishing Service. 

MacArthur, R. H. and J. W. MacArthur. 1961. On bird species diversity. Ecology 

42:594–600. 

Macedo, R. H. 1992. Reproductive patterns and social organization of the communal 

guira cuckoo (Guira guira) in central Brazil. Auk 109:786–799. 

Macedo, R. H. 1994. Inequities in parental effort and costs of communal 

breeding in the guira cuckoo. Ornitol. Neotropical 5:79–90. 

Macedo, R. H. and C. A. Bianchi. 1997a. Communal breeding in tropical 

guira cuckoos (Guira guira): Sociality in the absence of a saturated habitat. 

J. Avian Biology 3:207–215. 

Macedo, R. H. and C. A. Bianchi. 1997b. When birds go bad: Circumstantial 

evidence for infanticide in the communal South American guira 

cuckoo. Ethol. Ecol. and Evolution 9:45–54. 

Macedo, R. H. and C. Melo. 1999. Confirmation of infanticide in the communally-breeding 

guira cuckoo. Auk 116:847–851. 

Marini, M. A. 1992a. Foraging behavior and diet of the helmeted manakin. 

Condor 94:151–158. 

Marini, M. A. 1992b. Notes on the breeding and reproductive biology of the 

helmeted manakin. Wilson Bull. 104:168–173. 

Marini, M. A. and R. B. Cavalcanti. 1993. Habitat and foraging substrate use 

of three Basileuterus warblers from central Brazil. Ornitol. Neotropical 

4:69–76. 

Marini, M. A. and R. B. Cavalcanti. 1996. Influência do fogo na avifauna do 

sub-bosque de uma mata de galeria do Brasil central. Rev. Bras. Biol. 

56:749–754.



Martin, T. E. 1987. Food as a limit on breeding birds: A life-history perspective. 

Ann. Rev. Ecol. Syst. 18:453–487. 

Melo, C. and R. H. Macedo. 1997. Mortalidade em ninhadas de Guira guira 

(Cuculidae): Competição por recursos. Ararajuba 5:49–56. 

Moermond, T. C. and J. S. Denslow. 1985. Neotropical avian frugivores: Patterns 

of behavior, morphology, and nutrition, with consequences for fruit 

selection. Ornithol. Monogr. 36:865–897. 

Müller, P. 1973. The Dispersal Centers of Terrestrial Vertebrates in the 

Neotropical Realm: A Study in the Evolution of the Neotropical Biota 

and its Native Landscapes. Haag: Dr. W. Junk. 

Murray Jr., B. G. 1982. Territorial behaviour of the blue-black grassquit. 

Condor 84:119. 

Negret, A. J. 1983. “Diversidade e Abundância da Avifauna da Reserva 

Ecológica do IBGE, Brasília—D.F.” Master’s thesis, Universidade de 


Negret, A. J. and R. A. Negret. 1981. As Aves Migratórias do Distrito Federal. 

Boletim Técnico No. 6. Brasília: Ministério da Agricultura, Instituto 

Brasileiro de Desenvolvimento Florestal. 

Negret, A. J., J. Taylor, R. C. Soares, R. B. Cavalcanti and C. Johnson. 1984. 

Aves da Região Geopolítica do Distrito Federal. Brasília: Ministério do 

Interior, Secretaria Especial do Meio Ambiente. 

Negret, R. A. 1978. “O Comportamento Alimentar Como Fator de Isolamento 

Ecológico em Oito Espécies de Tyrannidae (Aves) do Planalto 

Central, Brasil.’’ Master’s thesis, Universidade de Brasília, Brasília, 

Brazil. 

Nores, M. 1992. Bird speciation in subtropical South America in relation to 

forest expansion and retraction. Auk 109:346–357. 

Nores, M. 1994. Quaternary vegetational changes and bird differentiation in 

subtropical South America. Auk 111:499–503. 

Oliveira, G. M. 1998. “Disponibilidade de Recursos Florais para Beija-flores 

no Cerrado de Uberlândia/MG.’’ Master’s thesis, Universidade de 


Orians, G. H. 1969. The number of bird species in some tropical forests. Ecology 

50:783–801. 

Paes, M. M. N. 1993. “Utilização de Frutos por Aves em uma Área de Cerrado 

do Distrito Federal.’’ Master’s thesis, Universidade de Brasília, 


Pinheiro, F. 1999. “Síndromes de Dispersão de Sementes de Matas de Galeria 

do Distrito Federal.’’ Master’s thesis, Universidade de Brasília, 


Quinn, J. S., R. H. Macedo, and B. White. 1994. Genetic relatedness of communally-breeding 

guira cuckoos. Anim. Behav. 47:515–529. 

Ricklefs, R. E. 1976. Growth rates of birds in the humid New World tropics. 

Ibis 118:179–207. 

Ricklefs, R. E. 1990. Speciation and diversity: The integration of local and



regional processes. In D. Otte and J. A. Endler, eds., Speciation and its 

Consequences, pp. 599–622. Chicago: University of Chicago Press. 

Ricklefs, R. E. and D. Schluter. 1993. Species diversity: Regional and historical 

influences. In R. E. Rcklefs and D. Schluter, eds., Species Diversity 

in Ecological Communities, pp. 350–363. Chicago: University of 


Rojas, R. and R. Ribon. 1997. Guilda de aves em Bowdichia virgilioides 

(Fabaceae: Faboideae) em área de cerrado de Furnas, Minas Gerais. 

Ararajuba 5:189–194. 

Setubal, S. S. 1991. “Biologia e Ecologia dos Tinamídeos Rhynchotus 

rufescens (Temminck, 1815) e Nothura maculosa (Temminch, 1815) na 

Região do Distrito Federal, Brasil.’’ Master’s thesis, Pontifícia Universidade 

Católica do Rio Grande do Sul, Porto Alegre, Brazil. 

Sick, H. 1955. O aspecto fitofisionômico da paisagem do médio Rio das 

Mortes, Mato Grosso e a avifauna da região. Arq. Mus. Nac. XLII 

2:541–576. 

Sick, H. 1965. A fauna do cerrado. Arq. Zool. 12:71–93. 

Sick, H. 1966. As aves do cerrado como fauna arborícola. An. Acad. Bras. 

Ciênc. 38:355–363. 

Sick, H. 1983. Migrações de aves na América do Sul continental (tradução: 

Walter A. Voss). Publicação Técnica No. 2. Brasília: CEMAVE—Centro 

de Estudos de Migrações de Aves. 

Sick, H. 1997. Ornitologia Brasileira. Rio de Janeiro: Ed. Nova Fronteira. 

Silberbauer-Gottsberger, I. and G. Gottsberger. 1988. A polinização de plantas 

do cerrado. Rev. Bras. Biol. 48:651–663. 

Silva, E. M. D. 1980. “Composição e Comportamento de Grupos Heteroespecíficos 

de Aves em Área de Cerrado, no Distrito Federal.’’ Master’s 


Silva, J. M. C. 1995a. Avian inventory of the cerrado region, South America: 

implications for biological conservation. Bird. Conserv. Int. 5:291–304. 

Silva, J. M. C. 1995b. Biogeographic analysis of the South American cerrado 

avifauna. Steenstrupia 21:49–67. 

Silva, J. M. C. 1995. Birds of the Cerrado Region, South America. Steenstrupia 

21:69–92. 

Silva, J. M. C. 1996. Distribution of Amazonian and Atlantic birds in gallery 

forests of the Cerrado Region, South America. Ornitol. Neotropical 

7:1–18. 

Silva, J. M. C. 1997. Endemic bird species and conservation in the Cerrado 

Region, South America. Biodiv. and Conserv. 6:435–450. 

Skutch, A. F. 1959. Life history of the groove-billed ani. Auk 76:281–317. 

Smythe, N. 1970. Relationships between fruiting seasons and seed dispersal 

methods in a neotropical forest. Amer. Nat. 104:25–35. 

Soares, R. C. 1983. “Taxas de Crescimento de Filhotes de Rolinha, 

Columbina talpacoti (Aves: Columbidae), em Relação a Fatores Ecológi-



cos e Populacionais no Planalto Central.’’ Master’s thesis, Universidade 


Stacey, P. B. and W. D. Koenig (eds.). 1990. Cooperative Breeding in Birds. 

Cambridge: Cambridge University Press. 

Stiles, F. G. 1983. Birds: introduction. In D. H. Janzen, ed., Costa Rican Natural 

History, pp. 502–530. Chicago: University of Chicago Press. 

Stiles, F. G. 1985. Seasonal patterns and coevolution in the hummingbirdflower 

community of a Costa Rican subtropical forest. In P.A. Buckley, 

M. S. Foster, E. S. Morton, R. S. Ridgely, and F. G. Buckely, eds., 

Neotropical Ornithology, pp. 757–787. Lawrence: Allen Press. 

Trollope, W. S. W. 1984. Fire in savanna. In P. de V. Booysen and N. M. 

Taiton, eds., Ecological Effects of Fire in South African Ecosystems, pp. 


Tubelius, D. P. 1997. “Estrutura de Comunidades de Aves em Habitats Preservados 

e Alterados de Cerrado, na Região do Distrito Federal.’’ Master’s 


Villalobos, M. P. 1994. “Guilda de Frugívoros Associada com o Buriti (Mauritia 

flexuosa: Palmae) em uma Vereda no Brasil Central.’’ Master’s thesis, 


Webber, T. 1985. Songs, displays, and other behaviour at a courtship gathering 

of blue-black grassquits. Condor 87:543–546. 

Young, B. E. 1994. The effects of food, nest predation and weather on the 

timing of breeding in tropical house wrens. Condor 96:341–353.



Appendix Avifauna Classified by Breeding Status, Habitat, 

and Endemicity (as Reported in Silva 1995) 

Breeding Habitat 

No. In Un- 

Family species region known Migrants Cerrado a Forest Both Endemics 

Rheidae 1 1 0 No 1 0 0 No 

Tinamidae 16 16 0 No 5 11 0 Yes: 2 

Podicipedidae 2 2 0 No 2 0 0 No 

Phalacrocoracidae 1 1 0 No 1 0 0 No 

Anhingidae 1 1 0 No 1 0 0 No 

Ardeidae 16 16 0 No 13 1 2 No 

Ciconiidae 3 3 0 No 3 0 0 No 

Threskiornithidae 6 6 0 No 5 0 1 No 

Anhimidae 2 2 0 No 2 0 0 No 

Anatidae 11 9 2 No 10 0 1 No 

Cathartidae 4 4 0 No 3 0 1 No 

Pandionidae 1 0 0 1 NA 1 0 0 No 

Accipitridae 33 29 2 2 NA 14 11 8 No 

Falconidae 13 12 0 1 NA 5 5 3 No 

Cracidae 11 11 0 No 0 9 2 Yes: 1 

Aramidae 1 1 0 No 1 0 0 No 

Rallidae 16 11 5 No 8 0 8 No 

Heliornithidae 1 1 0 No 1 0 0 No 

Eurypygidae 1 1 0 No 0 0 1 No 

Cariamidae 1 1 0 No 1 0 0 No 

Jacanidae 1 1 0 No 1 0 0 No 

Recurvirostridae 1 1 0 No 1 0 0 No 

Charadriidae 4 3 0 1 NA 4 0 0 No 

Scolopacidae 13 2 0 11 NA 13 0 0 No 

Laridae 3 2 1 No 3 0 0 No 

Rhynchopidae 1 1 0 No 1 0 0 No 

Columbidae 18 18 0 No 7 7 4 Yes: 2 

Psittacidae 33 33 0 No 4 15 14 Yes: 2 

Opisthocomidae 1 1 0 No 0 0 1 No 

Cuculidae 14 13 0 1 NA 3 5 6 No 

Tytonidae 1 1 0 No 1 0 0 No 

Strigidae 14 14 0 No 3 7 4 No 

Nyctibiidae 3 3 0 No 0 1 2 No 

Caprimulgidae 15 13 1 1 NA 9 3 3 Yes: 1 

Apodidae 9 4 5 No 4 1 4 No 

Trochilidae 36 36 0 No 5 17 14 Yes: 1 

Trogonidae 8 8 0 No 0 8 0 No 

Alcedinidae 5 5 0 No 1 0 4 No 

Momotidae 4 4 0 No 0 4 0 No 

Galbulidae 5 5 0 No 0 4 1 No 

Bucconidae 13 13 0 No 1 10 2 No 

Capitonidae 1 1 0 No 0 1 0 No 

Ramphastidae 10 10 0 No 0 9 1 No 

Picidae 25 25 0 No 2 17 6 No 

Dendrocolaptidae 23 23 0 No 1 20 2 No




Breeding Habitat 

No. In Un- 

Family species region known Migrants Cerrado a Forest Both Endemics 

Furnariidae 41 41 0 No 9 25 7 Yes: 5 

Formicariidae 58 58 0 No 2 49 7 Yes: 2 

Conopophagidae 1 1 0 No 0 1 0 No 

Rhinocryptidae 2 2 0 No 1 1 0 Yes: 2 

Tyrannidae 122 111 2 3 AB, 31 65 26 Yes: 3 

6 SA 

Pipridae 17 17 0 No 0 16 1 Yes: 1 

Cotingidae 10 6 0 4 AB 0 9 1 No 

Hirundinidae 14 8 1 5 NA 12 1 1 No 

Motacillidae 2 2 0 No 2 0 0 No 

Troglodytidae 9 9 0 No 3 4 2 No 

Mimidae 2 2 0 No 2 0 0 No 

Muscicapidae 11 7 0 1 AB, 1 7 3 No 

1 NA, 

2 SA 

Emberizidae 103 87 12 4 SA 40 47 16 Yes: 9 

Parulidae 12 11 0 1 NA 1 10 1 Yes: 1 

Vireonidae 7 7 0 No 0 6 1 No 

Icteridae 18 16 1 1 NA 11 3 4 No 

Corvidae 4 4 0 No 1 1 2 Yes: 1 

Total 837 758 33 (3.9) 46 258 411 168 32 

(%) (90.6) (5.6) (30.8) (49.1) (20.1) (3.8) 

a Includes all open formations (e.g., cerrado sensu stricto, campo cerrado, campo sujo, campo limpo, 

vereda, and campo rupestre). See chapter 6 for description of vegetation physiognomies. 

Abbreviations: NA and SA = long-distance migrants from North and South America, respectively; AB 

= altitudinal migrant within Brazil.


14 

The Cerrado Mammals: Diversity, 

Ecology, and Natural History 

Jader Marinho-Filho, Flávio H. G. Rodrigues, 

and Keila M. Juarez 

The first formal records of cerrado mammals were made 

by one of the first Brazilian scientists, Alexandre R. Ferreira, who from 

1783 to 1792 explored the provinces of Grão-Pará, Rio Negro, Mato 

Grosso, and Cuiabá (Hershkovitz 1987). However, only in the second half 

of the 20th century have Brazilian zoologists made the transition from a 

merely taxonomic treatment of the fauna towards a more naturalistic and 

ecological approach. Herein we present a review and analysis of the available 

information on natural history and geographical ranges of species 

and groups of the cerrado mammalian fauna, delineating patterns, making 

comparisons with other tropical savannas, and indicating lacunas and 

lines of investigation remaining to be explored. 

DATABASE OF CERRADO MAMMALS 

We followed Wilson and Reeder (1993) as a guide for the taxonomic status 

and distribution of mammals. Since the limits of the distributional 

ranges of most Brazilian mammals are far from well defined, we established 

the database on the distribution of Brazilian mammals from a number 

of different sources, including comprehensive works such as Vieira 

(1942), Moojen (1952), Cabrera (1957, 1961), Alho (1982), Koopman 

(1982, 1993), Streilein (1982), Emmons and Feer (1990), and Eisenberg 

and Redford (1999); published compilations with analyses on local 

and/or regional faunas, such as Schaller (1983), Redford and Fonseca 

266


The Cerrado Mammals 267 

(1986), Marinho-Filho and Reis (1989), Medellin and Redford (1992), 

Fonseca et al. (1996), Marinho-Filho and Sazima (1998), and references 

cited therein; technical reports and local/regional inventories such as Naturae 

(1996), PCBAP (1997), Marinho-Filho (1998); and unpublished 

original information from scientific collections at the Universidade de 

Brasília, the Museu de História Natural da Universidade Estadual de 

Campinas, Museu Nacional, and the authors’ personal observations. 

This database generated a checklist of mammalian species for the 

entire cerrado region (see table 14.1). We considered as endemics those 

species occurring only within this biome in Brazil. Those species whose 

entire known range is contained within a circular area of 300 km of diameter 

were considered as restricted. There are still many gaps to be filled, 

controversies about the status of some species, and many species remaining 

to be described. Genera such as Nectomys, Dasyprocta, Oryzomys, 

Galea, and Cavia are in urgent need of extensive revision. Even with these 

potential sources of error, the general picture presented here seems adequate 

for our purposes. 

The limits of the Cerrado Biome are those presented by the Brazilian 

Institute of Geography and Statistics (IBGE 1993; chapter 6; and fig. 6.1), 

with minor modifications. The maps showing these limits were produced 

by the Biodiversity Conservation Data Center of the Fundação Biodiversitas 

(Belo Horizonte, Brazil). Our analysis considers all habitat types 

within the Cerrado biome, including more dense formations such as 

gallery and dry forests. 

CHARACTERIZATION OF FAUNA: 

COMPOSITION, SPECIES RICHNESS, AND ABUNDANCE 

In all, 194 mammalian species from 30 families and 9 orders are recognized 

from the Cerrado Biome (table 14.1), making this biome the third 

most speciose in Brazil, after the Amazon and the Atlantic Forest, and followed 

by the caatinga and the pantanal (see Fonseca et al. 1999 for a better 

understanding of the Brazilian mammalian fauna). The largest groups 

are bats and rodents, represented by 81 and 51 species, respectively, 

including the notably speciose families Phyllostomidae and Muridae. 

Likewise, carnivores, didelphimorph marsupials, and xenarthrans are 

rather diversified groups, the last two being distinctive elements of the 

neotropical mammalian fauna. 

In general, this fauna is essentially composed of small-sized animals: 

85% of the species have body masses no greater than 5 kg, and only five


Table 14.1 Checklist of Cerrado Mammals with Endemic 

Species, and Species Included in the Brazilian 

Official List of Species Threatened with Extinction (Thr) 

Habitat 

Scientific Abun- Weight 

Name Endemic Thr. dance Range Open Forest (g) Diet 

DIDELPHIMORPHIA 

Family Didelphidae 

Caluromys lanatus R w x 310–410 om 

Caluromys philander R w x 140–270 om 

Chironees minimus R w x 600–700 om 

Didelphis albiventris A w x x 500–2000 om 

Didelphis marsupialis A w x x 500–2000 om 

Gracilinanus agilis A w x x 20–30 om 

Lutreolina R w x x 200–540 om 

crassicaudata 

Marmosa murina A w x x 45–60 om 

Metachirus R w x 300–450 om 

nudicaudatus 

Micoureus R w x x 80–150 om 

demerarae 

Monodelphis A w x x 11–35 om 

americana 

Monodelphis A w x 35–100 om 

domestica 

Monodelphis kunsi R w x 8.5–14 om 

Monodelphis rubida x R r x 45–46 om 

Philander opossum R w x 200–600 om 

Thylamys pusilla A w x 12–30 om 

Thylamys velutinus R w x 16–32 om 

XENARTHRA 

Family Myrmecophagidae 

Myrmecophaga x r w x x 22000–40000 in 

tridactyla 

Tamandua a w x x 3500–8500 in 

tetradactyla 

Family Bradypodidae 

Bradypus variegatus r w x 2300–5500 fo 

Family Dasypodidae 

Cabassous tatouay r w x 3400–6400 in 

Cabassous unicinctus r w x x 1500–5000 in 

Dasypus r w x x 2500–6300 om 

novemcinctus 

Dasypus a w x x 1500–2000 om 

septemcinctus 

Euphractus sexcintus a w x 3000–7000 om 

Priodontes maximus x r w x 30000–60000 in 

Tolypeutes matacus r w x 1000–1150 in 

Tolypeutes tricinctus x r w x x 1000–1800 in



Habitat 



CHIROPTERA 

Family Emballonuridae 

Centronycteris r w x x in 

maximiliani 

Peropteryx kappleri r w x x 6–11 in 

Peropteryx macrotis r w x x 4–8 in 

Rhinchonycteris naso a w x x 4–7 in 

Saccopteryx a w x x 7–12 in 

bilineata 

Saccopteryx r w x x 4–7 in 

leptura 

Family Noctilionidae 

Noctilio albiventris a w x x 21–55 in 

Noctilio leporinus a w x x 60–90 fi 

Family Mormoopidae 

Pteronotus parnellii a w x x 11–28 in 

Pteronotus r w x x 6–9 in 

personatus 

Pteronotus r w x x 10–16 in 

gymnonotus 

Family Phyllostomidae 

Anoura caudifer a w x x 10–13 ne 

Anoura geoffroyi a w x x 13–19 ne 

Artibeus cinereus a w x x 12–14 fr 

Artibeus concolor a w x x 18–20 fr 

Artibeus a w x x 50–65 fr 

jamaicensis 

Artibeus lituratus a w x x 60–85 fr 

Artibeus planirostris a w x x 50–65 fr 

Carollia perspicillata a w x x 12–25 fr 

Chiroderma r w x x 13–15 fr 

trinitatum 

Chiroderma r w x x 44–50 fr 

villosum 

Choeroniscus r w x x 10 ne 

minor 

Chrotopterus r w x 60–95 ca 

auritus 

Desmodus rotundus a w x x 25–50 sa 

Diaemus youngi a w x x 27–35 sa 

Diphylla ecaudata r w x x 20–35 sa 

Glossophaga a w x x 10–14 ne 

soricina 

Lonchophylla a r x x 9–12 ne 

bokermanni 

Lonchophylla de x a a x x 10–12 ne 

keyseri



Habitat 



Lonchorhina aurita a w x 12–22 in 

Macrophylum r w x x 7–11 in 

macrophylum 

Micronycteris behni r r x



Habitat 



Lasiurus cinereus r w x x 20 in 

Lasiurus ega a w x x 10–15 in 

Myotis albescens r w x x 7–11 in 

Myotis nigricans a w x x 4–8 in 

Myotis riparius r w x x 4–7 in 

Rhogeessa tumida r w x x 3–5 in 

Family Molossidae 

Eumops a w x x 62–66 in 

auripendulus 

Eumops bonariensis r w x x 11–20 in 

Eumops glaucinus r w x x 22–28 in 

Eumops hansae r w x x in 

Eumops perotis w x x 60–76 in 

Molossops abrasus r w x x 25–42 in 

Molossops w x x 7–9 in 

mattogrossensis 

Molossops w x x 5–9 in 

planirostris 

Molossops a w x x 4–9 in 

temminckii 

Molossus ater a w x x 21–43 in 

Molossus molossus a w x x 12–28 in 

Nyctinomops w x x in 

aurispinosus 

Nyctinomops a w x x 8–16 in 

laticaudatus 

Nyctinomops w x x 16–20 in 

macrotis 

Promops nasutus r w x x 14–25 in 

Tadarida r w x x 9–19 in 

brasiliensis 

PRIMATES 

Family Callithrichidae 

Callithrix jacchus a w x x 250–325 om 

Callithrix melanura r w x 380–500 om 

Callithrix a w x x 250–350 om 

penicillata 

Family Cebidae 

Alouatta caraya a w x 3000–10000 fo/fr 

Aotus infulatus r w x 600–1000 fr 

Cebus apella a w x 1700–4500 in/fr 

CARNIVORA 

Family Canidae 

Cerdocyon thous a w x x 4000–9000 om 

Chrysocyon x r w x 20,000–30,000 om 

brachyurus



Habitat 



Pseudalopex vetulus x r w x 3000–4500 in/fr 

Speothos venaticus x r w x x 5000–7000 ca 

Family Procyonidae 

Nasua nasua a w x x 3000–7500 om 

Potos flavus r w x 2000–3500 fr 

Procyon a w x x 3500–7500 om 

cancrivorus 

Family Mustelidae 

Conepatus r w x 1500–3500 in 

semistriatu 

Eira barbara r w x x 2700–7000 om 

Galictis cuja a w x x 1000–2500 om 

Galictis vittata a w x x 1500–2500 om 

Lontra longicaudis x r w x 5000–15,000 fi 

Pteronura x r w x 24,000–34,000 fi 

brasiliensis 

Family Felidae 

Herpailurus r w x x 4000–9000 ca 

yaguaroundi 

Leopardus x r w x x 8000–15,000 ca 

pardalis 

Leopardus tigrinus x r w x x 1300–3000 ca 

Leopardus wiedii x r w x x 3000–9000 ca 

Oncifelis colocolo x r w x 1700–3650 ca 

Panthera onca x r w x x 30,000–150,000 ca 

Puma concolor x r w x x 30,000–120,000 ca 

PERISSODACTYLA 

Family Tapiridae 

Tapirus terrestris r w x x 200,000–250,000 fo/fr 

ARTIODACTYLA 

Family Tayassuidae 

Pecari tajacu r w x x 17,000–30,000 om 

Tayassu pecari r w x x 25,000–40,000 om 

Family Cervidae 

Blastoceros x r w x x 100,000–150,000 fo 

dichotomus 

Mazama americana r w x 24,000–50,000 fo/fr 

Mazama a w x x 13,000–23,000 fo/fr 

gouazoupira 

Ozotoceros x r w x 28,000–35,000 fo 

bezoarticus 

RODENTIA 

Family Muridae 

Akodon cursor a w x 20–65 fr/gr/in 

Akodon montensis a w x 20–65 fr/gr/in



Habitat 



Akodon lindberghi x r r x 16–19 fr/gr/in 

Bibimys labiosus x r r x 20–35 fr/gr/in 

Bolomys lasiurus a w x 20–58 fr/gr/in 

Calomys callosus a w x 22–30 fr/gr/in 

Calomys laucha r w x 30–38 fr/gr/in 

Calomys tener x a w x 15–23 fr/gr/in 

Holochilus r w x 275–455 fo 

brasiliensis 

Holochilus sciureus r w x 144–177 fo 

Juscelinomys x x r r x fr/gr/in 

candango 

Kunsia fronto r r x 110–400 fr/gr/in 

Kunsia tomentosus x r r x 200–600 fr/gr/in 

Microakodontomys x r r x x 

transitorius 

Nectomys rattus a w x 200–450 om 

Oecomys bicolor r w x 21–41 fr/gra 

Oecomys cleberi x r ? x 19–25 fr/gra 

Oecomys concolor a r w x 45–96 fr/gra 

Oligoryzomys r r x 14–25 fr/gr/in 

chacoensis 

Oligoryzomys x r w x x 10–25 fr/gr/in 

eliurus 

Oligoryzomys r w x 14–25 fr/gr/in 

nigripes 

Oryzomys a w x 30–65 fr/gr/in 

megacephalus 

Oryzomys lamia x r r x fr/gr/in 

Oryzomys ratticeps r w 120–157 fr/gr/in 

Oryzomys a w x x 60–140 fr/gr/in 

subflavus b 

Oxymycterus x a r x 40–105 fr/gr/in 

delator 

Oxymycterus x a w x 40–105 fr/gr/in 

roberti 

Pseudoryzomys x r w x x 30–56 fr/gr/in 

simplex 

Rhipidomys emiliae w x 40–100 om 

Rhipidomys a w x 40–100 om 

macrurus 

Thalpomys x r r x fr/gr/in 

cerradensis 

Thalpomys lasiotis x r r x fr/gr/in 

Wiedomys r w x x 24–32 fr/gr/in 

pyrrhorhinos



Habitat 



Family Erethizontidae 

Coendou r w x 3200–5300 fr/fo 

prehensilis 

Family Caviidae 

Cavia aperea a w x 500–1000 fo 

Galea spixii a w x 300–600 fo 

Kerodon rupestris r w x 900–1000 fo 

Family Hydrochaeridae 

Hydrochaeris a w x x 35000–65000 fo 

hydrochaeris 

Family Agoutidae 

Agouti paca r w x 5000–13000 fr/fo 

Family Dasyproctidae 

Dasyprocta a w x x fr/gr 

leporina 

Dasyprocta azarae a w x x 2500–3200 fr/gr 

Family Ctenomyidae 

Ctenomys x r r x fo 

brasiliensis 

Family Echimyidae 

Carterodon x r r x fo 

sulcidens 

Clyomys bishopi a r x 100–300 fr/gr/in 

Clyomys laticeps r w x 100–300 fr/gr/in 

Dactylomys r w x 600–700 fo 

dactylinus 

Echimys r r x fr/gr 

braziliensis c 

Proechimys roberti a w x 160–500 fr/gr/in 

Proechimys a w x 160–500 fr/gr/in 

longicaudatus 

Trinomys moojeni x r r x fr/gr 

Thrichomys a w x x 247–500 fr/gr/in 

apereoides 

LAGOMORPHA 

Family Leporidae 

Sylvilagus a w x x 450–1200 fo 

brasiliensis 

TOTAL: 194 species 19 17



species weigh more than 50 kg (table 14.1), strongly contrasting with the 

mammalian fauna of African savannas, where large mammals abound. 

There are also differences concerning species composition, number of 

species, and biomass of mammals between African savannas and other 

savannas in the world. In Africa there are practically as many bovid 

species as there are murid rodents (Sinclair, 1983). Almost 100 species of 

ungulates inhabit Africa (Sinclair 1983; Ojasti 1983), compared to a little 

over 20 species in South America. The average number of ungulates 

for eight African savanna locations is 14.5 ± 3.7 (range = 11–20), whereas 

in the savannas of southern Asia, this number is only 7.3 ± 1.8 (range = 

6–10; n = 6; see data compilation in Bourlière 1983a). In South America, 

this number may be even smaller: only two ungulate species inhabit the 

Llanos in Masaguaral, Venezuela (Einsenberg et al. 1979). In the Brazilian 

cerrados, the number of ungulates in a given location varies between 

6 and 7 species, depending on the presence of the marsh deer, Blastocerus 

dichotomus. There are no native ungulates in Australia, and the niche of 

the large herbivores is filled by kangaroos (Freeland 1991). Only six 

species of large Macropodidae occur in the Australian savannas, and only 

3 to 4 may be found per location (Freeland 1991). Another difference 

between the ungulates of the African savannas and those of the Brazilian 

cerrados is feeding behavior. The great herbivores of the cerrados are 

browsers (Rodrigues and Monteiro-Filho 1999; Tomas and Salis 2000), 

unlike their African counterparts, the majority of which are grazers. In 

the cerrados, the role of the grazers is carried out by rodents, like the capybara 

(Hydrochaeris hydrochaeris), cavies (Caviidae), and the lagomorph 

Sylvilagus brasiliensis (see table 14.1). Aside from the ungulates, carnivores 

represent another prominent group in African savannas. Between 8 

and 27 (average 14.7 ± 10.7; n = 3) species of carnivores can be found at 

Table Note: Categories of abundance: r = rare, a = abundant; distributional range: w 

= widely distributed, r = restricted distribution. Categories of feeding habit: om = 

omnivore, in = insectivore, fo = folivore, fi = fish specialist, ne = nectarivore, fr = frugivore, 

as = sanguinivore, ca= carnivore, gr = grainivore. 

a Musser and Carleton (1993) state that Oecomys concolor is restricted to localities 

north of the Amazon. However, this specific name has been frequently used in the 

literature referring to large-bodied Oecomys forms in the cerrado range, and that is 

why we decided to keep it here. 

b Oryzomys subflavus has been split into a number of species, four of which occur in 

the cerrado domain (Percequillo 1998). However, these species are not decribed yet 

and were not considered herein. 

c Emmons and Feer (1990) recognize Echimys braziliensis as Nelomys sp., stating 

that the correct specific name is not clear. This species has also been called Phylomys 

brasiliensis, which Cabrera (1961:540) considers a nomem nudum.



a given location in the African savannas (Bourlière 1983b). The number 

of carnivores per location in the cerrado is not much less (13.6 ± 1.8; range 

12–16), considering five sampling areas: Emas National Park (GO) (Silveira, 

1999), the Distrito Federal (DF) (Fonseca and Redford, 1984; Marinho-Filho 

et al., 1998, and records of the Brasília Zoo), Grande Sertão 

Veredas National Park (MG), Serra da Mesa (GO), and the Jatobá Ranch 

(BA) (unpublished data). Actually this average may be even greater, 

because some species may be very difficult to sample, especially the small 

felines, which may go undetected even when present in an area. 

The great majority of cerrado mammalian species have wide distributions, 

and, although the total number of individuals for a given species may 

be considered high throughout the entire range of the biome, most species 

tend to be locally rare. A comparative analysis of communities of small, 

non-flying mammals from 11 cerrado areas of central Brazil shows great 

variation among areas for the abundance of 39 species of marsupials and 

rodents (Marinho-Filho et al. 1994). Approximately one third of the individuals 

captured in all study areas were Bolomys lasiurus. Though the 

dominant species in most areas, it represented only 2.2% of the captures 

in one area and was absent from two additional areas. Similar patterns 

were found for other species, high in number at a given site, but rare or 

even absent in another. Thus, another third of the total number of captured 

individuals was represented by five species, and the remaining 33 species 

corresponded approximately to 30% of the total number of individuals of 

small mammals in the sampling areas. (Marinho-Filho et al. 1994). 

In the same study, the beta diversity and distances between each of 

the 11 areas in relation to all others were calculated. For the 55 location 

pairs examined, Marinho-Filho et al. (1994) found a high mean beta 

diversity (mean = 0.58, SD = 0.13; range = 0.29–0.80), but there was no 

strong association between the distance between areas (5 km to 1,300 km) 

and beta diversity. 

HABITAT UTILIZATION AND ENDEMICITY 

The mammalian fauna of the cerrado region consists mainly of elements 

inhabiting a great variety of environments (table 14.1). About 54% of the 

mammalian species occupy forest environments as much as open areas, 

whereas 16.5% are exclusive to open areas and 29% exclusive to forests. 

The mammalian fauna of the cerrado appears to be derived primarily from 

a set of forest species (Redford and Fonseca 1986; Marinho-Filho and Sazima 

1998). Gallery forests appear to play an important role as mesic corridors 

that allow for the establishment of elements not adapted to the



conditions found in dry, open cerrado areas (Mares et al. 1985; Redford 

and Fonseca 1986). This results in the low endemism found in the cerrado: 

only 18 species (9.3%) may be considered exclusive to this biome. 

Endemism rates for plants are considered high (see chapters 6, 7). In 

contrast, the cerrado fauna shares many elements with other open formations 

in tropical South America and is strongly influenced by two adjacent 

forest biomes, the Atlantic forest and the Amazonian rainforest. For 

vertebrates, the degree of endemism is low (Vanzolini 1963; Sick 1965; 

Silva 1995a, 1995b; see also chapters 11–13), and these animals exhibit 

no specific adaptations for life in the cerrado. 

Most (56%) of the endemic mammalian species of the cerrado inhabit 

exclusively open areas. Of the remaining 44%, four species are forest 

inhabitants, and four occur in forests and open areas (table 14.1). Of the 

open area species, four (Ctenomys brasiliensis, Carterodon sulcidens, 

Juscelinomys candango, and Oxymycterus roberti) are semi-fossorial and 

thus avoid the environmental extremes that savanna inhabitants must 

confront. 

Considering the 18 endemic species, five are restricted to a type locality: 

Bibimys labiosus, Juscelinomys candango, Microakodontomys transitorius, 

Oecomys cleberi, and Carterodon sulcidens. Of these, J. 

candango, M. transitorius, and O. cleberi were described from the Distrito 

Federal at the core of the cerrado region. The other two are known 

from the Lagoa Santa area (state of Minas Gerais), in the southeastern 

portion of the cerrado. Of the remaining 13 species, five are distributed 

in the central and central-southeastern portion of the cerrado; four are 

restricted to the southern and two to the western-southwestern portion; 

one species is found in all the Cerrado range (Pseudalopex vetulus); and 

the only endemic bat is found in the central-northern region. There are no 

known endemics restricted to the northern portion, but this picture may 

only represent the greater concentration of studies in the south-central 

region, which is more accessible and closer to important scientific centers. 

The analysis of habitat utilization by the cerrado mammals (fig. 

14.1A) confirms the predominance of generalists over specialists, except 

for the primates, which are predominantly forest specialists, and rodents, 

which have as many specialist species for forests as for open areas. 

Xenarthra is the only taxon with a predominance of open-area species. 

DIETS 

The mammals of the cerrados are grouped here according to feeding 

habits resulting in 12 diet categories. The insectivorous feeding habit




ranks as the most frequent, including around 27% of the species. Among 

the mammals that basically feed on insects, the orders Chiroptera and 

Xenarthra stand out, along with a single member of the Carnivora (table 

14.1). More than 80% of the species with insectivorous feeding habits 

exploit open areas as much as forests (fig. 14.1B). The second most frequent 

group, comprising about 18% of the species, consists of omnivorous 

mammals, which feed on items of both animal and plant origin. 

Despite representing a rather diverse group, including species from several 

mammalian orders, 49% are from the order Didelphimorphia. The 

great majority are small-sized animals (54% of the omnivorous species 

weight less than 500 g), and use open areas as much as forests (fig. 14.1B). 

Fruits, representing a highly important food resource for cerrado mammals, 

are consumed by 55% of the species, ranging in size from small to 

large in several mammalian orders. Primarily frugivorous mammals 

include many bat species, one primate and one carnivore, representing 

about 9% of the species total. 

Diets represented by only one food category are relatively frequent 

among cerrado mammals. In contrast to common observations in relation 

to habitat, most of the mammalian fauna consists of dietary specialists. 

Carnivores, frugivores, insectivores, folivores, piscivores, sanguivores, 

and nectarivores account for 54% of the total number of species. Orders 

with a greater number of species also tend to have more feeding categories 

(r 2 = 0.58; P = .017, fig. 14.1D), reflecting a possible niche segregation 

among species, as observed for canids, which present very little overlap 

among diets (Juarez 1997). 

Figure 14.1 (preceding page) (A) Pattern of habitat utilization by cerrado 

mammalian orders. Did = Didelphimorpha; Xen = Xenarthra; Chi = Chiroptera; 

Pri = Primates; Car = Carnivora; Per = Perissodactyla; Art = Artiodactyla; 

Rod = Rodentia; Lag = Lagomorpha. (B) Feeding habits of cerrado 

mammals associated with habitat types; om = omnivorous; ca = carnivorous; 

in = insectivorous; fo = folivorous; fi = fish; bl = blood; ne = nectarivorous; fr 

= frugivorous; fo/fr = folivorous/frugivorous; in/fr = insectivorous/frugivorous; 

fr/gr = frugivorous/granivorous; fr/gr/in = frugivorous/granivorous/ 

insectivorous. (C) Relative frequency of Brazilian cerrado mammalian species 

in the four categories of rarity: locally abundant and widespread; locally rare 

and widespread; locally abundant with restricted distribution; and locally 

rare with restricted distribution. (D) Number of feeding categories in relation 

to the number of species in each mammal order in the Brazilian cerrados. 

Twelve feeding categories are recognized, as indicated in (B).



CONSERVATION STATUS OF THE 

CERRADO MAMMALIAN FAUNA 

Based on the Official Brazilian List of Species Threatened with Extinction 

(Bernardes et al. 1989), 17 species that are confirmed to occur in the Cerrado 

Biome are threatened (table 14.1). Following Arita (1993), who 

made a similar analysis of Central American bats, we classified the cerrado 

mammalian species into four categories, considering their distributional 

ranges and relative abundances from data in the literature and our 

own experience: species (a) locally abundant with restricted distribution; 

(b) locally abundant and widespread; (c) locally scarce and widespread; 

and (d) locally scarce with restricted distribution. Even considering the 

variation in numbers of species with wide distributions, this classification 

seems adequate to delineate a general pattern. Species with broad distributions 

are less threatened than restricted species, and locally scarce 

species tend to be more vulnerable than locally abundant ones. Rare 

species with narrow distributions face the highest risks. 

Of the species here analyzed (see table 14.1), 47.6% possess wide distributions 

and are locally rare; 42.7% are locally abundant and widely 

distributed; 1.1% are locally abundant but have restricted distributions; 

and 8.6% are locally rare and have a restricted distribution. The latter 

categories are principally found in the order Rodentia, along with a representative 

from the order Chiroptera and another from Didelphimorphia 

(fig. 14.1C). Most of the cerrado species that are considered threatened 

with extinction (83%; see Bernardes et al. 1989) fall in the category of 

locally rare with widespread distribution, and only two species are in the 

higher risk category (locally rare and restricted distribution). 

This discrepancy may reflect the lack of knowledge regarding mammals 

of the cerrado region (which can be expanded to all of South America). 

Few data are available for the true status of many species, especially 

small, rare, and geographically restricted species. The larger species, 

which tend to have a greater emotional appeal, are more frequently listed, 

as are those with wider distribution. In fact, 80% of the cerrado species 

weighing over 50 kg are listed as threatened; whereas 36.4% of the species 

between 5 and 50 kg, 17.4% of the species between 0.5 and 5 kg, and just 

0.8% of the species less than 0.5 kg are also considered at risk. One should 

also consider that certain abundant or widely distributed species, theoretically 

not at high risk for extinction, are nonetheless threatened by factors 

extrinsic to their biology, related instead to anthropogenic pressures 

(chapters 5, 18). The pattern currently recognized here is probably the 

sum of both situations.




We thank Marina Anciães, Nana Rocha, Ludmilla Aguiar, and Patrícia S. 

de Oliveira for helping with the data compilation and elaboration of the 

mammal list. Ricardo “Pacheco” Machado produced the species distribution 

maps used in our analysis. We also thank the Fundação Biodiversitas 

and Conservation International for the GIS treatment and analysis. 

Marc Johnson helped with the English version. The Brazilian Research 

Council (CNPq) provided financial support to JMF (Proc. 300591/86-1). 

REFERENCES 

Alho, C. J. R. 1982. Brazilian rodents: Their habitats and habits. In M. A. 

Mares and H. H. Genoways, eds., Mammalian Biology in South America, 

pp. 143–166. Pittsburgh: Pymatuning Laboratory of Ecology, University 

of Pittsburgh. 

Arita, H. T. 1993. Rarity in Neotropical bats: Correlations with phylogeny, 

diet, and body mass. Ecol. Appl. 3:500–517. 

Bernardes, A. T., A. B. M. Machado, and A. B. Rylands. 1989. Fauna Brasileira 

Ameaçada de Extinção. Belo Horizonte: Fundação Biodiversitas. 

Bourlière, F. 1983a. The Savanna mammals: Introduction. In F. Bourlière, ed., 

Tropical Savannas, pp. 359–361. Amsterdam: Elsevier. 

Bourlière, F. 1983b. Mammals as secondary consumers in savanna ecosystems. 

In F. Bourlière ed., Tropical Savannas, pp. 463–475. Amsterdam: 

Elsevier. 

Cabrera, A. 1957. Catalogo de los Mamiferos de America del Sur (Metatheria-Unguiculata-Carnivora). 

Rev. Mus Arg. Cienc. Nat. Bernardino 

Rivadavia 4:1–307. 

Cabrera, A. 1961. Catálogo de los Mamíferos de America del Sur (Sirenia- 

Perissodactyla-Artiodactyla-Lagomorpha-Rodentia-Cetacea). Rev. Mus 

Arg. Cienc. Nat. Bernardino Rivadavia 4:1–732. 

Eisenberg, J. F., M. A. O’Connel, and P. V. August. 1979. Density, productivity, 

and distribution of mammals in two Venezuelan habitats. In J. F. 

Eisenberg, ed., Vertebrate Ecology in the Northern Neotropics, pp. 

187–207. Washington, D.C., Smithsonian Institution Press. 

Eisenberg, J. F. and K. H. Redford, 1999. Mammals of the Neotropics: The 

Central Neotropics, Ecuador, Peru, Bolivia, Brazil. Chicago: University 

of Chicago Press. 

Emmons, L. H. and F. Feer. 1990. Neotropical Rainforest Mammals: A Field 

Guide. Chicago: University of Chicago Press. 

Fonseca, G. A. B. and K. H. Redford. 1984. The mammals of IBGE’s Ecological 

Reserve, Brasília, and an analysis of the role of gallery forests in 

increasing diversity. Rev. Bras. Biol. 44:517–523.



Fonseca, G. A. B., G. Herrmann, Y. L. R. Leite, R. A. Mittermeier, A. B. 

Rylands, and J. L. Patton. 1996. Lista anotada dos mamíferos do Brasil. 

Occas. Pap. Cons. Biol. 4:1–38. 

Fonseca, G. A. B., G. Herrmann, and Y. L. R.Leite. 1999. Macrogeography 

of Brazilian mammals. In J. Eisenberg and K. H. Redford, eds., Mammals 

of the Neotropics: The Central Neotropics, Ecuador, Peru, Bolivia, 

Brazil , pp. 549–563. Chicago: University of Chicago Press. 

Freeland, W. J. 1991. Large herbivorous mammals: Exotic species in northern 

Australia. In: P. A. Werner, ed., Savanna Ecology and Management: 

Australian Perspectives and Intercontinental Comparisons, pp. 

101–105. Oxford: Blackwell Scientific Publications. 

Hershkovitz, P. 1987. A history of the recent mammalogy of the Neotropical 

Region from 1492 to 1850. In B. D. Patterson and R. M. Timm, eds., 

Studies in Neotropical Mammalogy: Essays in Honor of Philip Hershkovitz. 

Fieldiana: Zoology 39:11–98 

IBGE. 1993. Mapa de Vegetação do Brasil. Brasília: Fundação Instituto 

Brasileiro de Geografia e Estatística. 

Juarez, K. M. 1997. “Dieta, Uso de Habitat e Atividade de Três Especies de 

Canídeos Simpátricas do Cerrado.” Master’s thesis, Universidade de 


Koopman, K. 1982. Biogeography of bats of South America. In M. A. Mares 

and H.H. Genoways (eds). Mammalian Biology in South America, pp. 

273–302. Pittsburgh: Pymatuning Laboratory of Ecology, University of 

Pittsburgh. 

Koopman, K. F. 1993. Order Chiroptera. In D. E. Wilson and D. M. Reeder, 

eds., Mammal Species of the World: A Taxonomic and Geographic Reference, 

pp. 137–241. Whashington, D.C.: Smithsonian Institution Press. 

Mares, M. A., M. R. Willig, and T. E. Lacher Jr. 1985. The Brazilian caatinga 

in South American zoogeography: Tropical mammals in a dry region. 

J. Biogeogr. 12:57–69. 

Marinho-Filho, J. 1998. Informações prévias para o grupo temático “Mastozoologia”. 

In Ações Prioritárias para a Conservação da Biodiversidade 

do Cerrado e Pantanal. Brasília: Conservation International, Funatura, 

Fundação Biodiversitas, and Universidade de Brasília. 

Marinho-Filho, J. and M. L. Reis. 1989. A fauna de mamíferos associada as 

matas de galeria. In L. M. Barbosa, ed., Simpósio sobre Mata Ciliar, 

Anais, pp. 43–60. Campinas: Fundação Cargill. 

Marinho-Filho, J., M. L. Reis, P. S. Oliveira, E. M. Oliveira, and M. N. Paes, 

1994. Diversity standards, small mammal numbers and the conservation 

of the cerrado biodiversity. An. Acad. Bras. Ciênc. 66 (supp.):149–157. 

Marinho-Filho, J., F. H. G. Rodrigues, M. M. Guimarães, and M. L. Reis. 

1998. Os mamíferos da Estação Ecológica de Águas Emendadas, 

Planaltina, DF. In J. Marinho-Filho, F.H.G. Rodrigues and M. 

Guimarães, eds., Vertebrados da Estação Ecológica de Águas Emendadas: 

História Natural e Ecologia em um Fragmento de Cerrado do



Brasil Central. pp. 34–63. Brasília: Secretaria de Meio Ambiente e Tecnologia 

e Instituto de Ecologia e Meio Ambiente. 

Marinho-Filho, J. and I. Sazima 1998. Brazilian bats and conservation biology: 

A first survey. In T. H. Kunz and P. A. Racey, eds., Bat Biology and 

Conservation, pp. 282–294. Washington D.C.: Smithsonian Institution 

Press. 

Medellin, R. A. and K. H. Redford. 1992. The role of mammals in forestsavanna 

boundaries. In P. A. Furley, J. Proctor, and J. Ratter, eds., Nature 

and Dynamics of Savanna Boundaries, pp. 519–548. London: Chapman 


Moojen, J. 1952. Os Roedores do Brasil. Rio de Janeiro, Instituto Nacional 

do Livro. 

Musser, G. G. and M. D. Carleton 1993. Family Muridae. In D. E. Wilson 

and D. M. Reeder, eds., Mammal Species of the World: A Taxonomic and 

Geographic Reference. 2nd. ed. Washington, D.C.: Smithsonian Institution 

Press. 

Naturae, 1996. Inventário da Fauna da Área sob Influência da UHE Serra da 

Mesa. Relatório Técnico. Rio de Janeiro: Furnas Centrais Elétricas. 

Ojasti, J. 1983. Ungulates and large rodents of South America. In F. Bourlière, 

ed., Tropical Savannas, pp. 427–439. Amsterdam: Elsevier. 

PCBAP, 1997. Plano de Conservação da Bacia do Alto Paraguai (Pantanal). 

Vol II, Tomo III. Brasília: Ministério do Meio Ambiente Recursos Hídricos 

e da Amazônia Legal e Secretaria de Coordenação dos Assuntos de 

Meio Ambiente. 

Percequillo, A. R. 1998. “Sistemática de Oryzomys Baird 1858 do Leste de 

Brasil (Muroidea, Sigmodontinae).” Master’s thesis, Universidade de São 

Paulo, São Paulo, Brazil. 

Redford, K. H., and G. A. B. Fonseca. 1986. The role of gallery forests in the 

zoogeography of the Cerrado’s non-volant mammalian fauna. Biotropica 

18:126–135. 

Rodrigues, F. H. G. and E. L. A. Monteiro-Filho. 1999. Feeding behaviour of 

the pampas deer: A grazer or a browser? IUCN Deer Specialist Group 

News 15:12–13. 

Sarmiento, G. 1983. The savannas of Tropical America. In F. Bourlière, ed., 

Tropical Savannas, pp. 245–288. Amsterdam: Elsevier. 

Schaller, G. B. 1983. Mammals and their biomass on a Brazilian ranch. Arq. 

Zool. 33:1–36. 

Sick, H., 1965. A fauna do cerrado. Arq. Zool. 12:71–93. 

Silva, J. M. C. 1995a. Biogeographic analysis of the South American Cerrado 

avifauna. Steenstrupia 21:47–67. 

Silva, J. M. C. 1995b. Birds of the Cerrado region, South America. Steenstrupia 

21:69–92. 

Silveira, L. 1999. “Ecologia e Conservação dos Mamíferos Carnívoros do 

Parque Nacional das Emas, Goiás.” Master’s thesis, Universidade Federal 

de Goiás, Goiás, Brazil.



Sinclair, A. R. E. 1983. The adaptations of African ungulates and their effects 

on community function. In F. Bourlière ed., Tropical Savannas, pp. 

401–426. Amsterdam: Elsevier. 

Streilein, K. E. 1982. Behavior, Ecology and distribution of the South American 

marsupials. In M. .A. Mares and H. H. Genoways eds., Mammalian 

Biology in South America. pp. 231–250. Pittsburgh: Pymatuning Laboratory 

of Ecology, University of Pittsburgh. 

Tomas, W. M. and S. M. Salis. 2000. Diet of the marsh deer (Blastocerus 

dichotomus) in the Pantanal wetland, Brazil. Stud. Neotrop. Fauna and 

Environ. 35:165–172. 

Vanzolini, P. E. 1963. Problemas faunísticos do cerrado. In M. G. Ferri, ed., 

Simpósio Sobre o Cerrado, pp. 307–320. São Paulo: Editora da Universidade 


Vieira, C. C. 1942. Ensaio monográfico sobre os quirópteros do Brasil. Arq. 

Zool. São Paulo, 3:219–471. 

Wilson, D. E. and D. M. Reeder. (eds). 1993. Mammal Species of the World: 

A Taxonomic and Geographic Reference. 2nd. ed. Washington, D.C.: 

Smithsonian Institution Press.


15 

Ant Foraging on Plant 

Foliage: Contrasting Effects 

on the Behavioral Ecology 

of Insect Herbivores 

Paulo S. Oliveira, André V. L. Freitas, 

and Kleber Del-Claro 

Ants are dominant organisms whose individual colonies 

may contain several million workers. Their numerical dominance in terrestrial 

habitats is combined with a broad taxonomic diversity and a widespread 

distribution throughout the Globe (Hölldobler and Wilson 1990). 

The ecological success of ants is attributed to their eusocial mode of life, 

local abundance, and diversity of adaptations, among other things (Wilson 

1987). Such traits result in a wide variety of feeding habits and foraging 

strategies, including the use of plant foliage as a foraging substrate 

(Carroll and Janzen 1973). Intense foraging on vegetation appears to have 

set the scenario for a multitude of interactions with many plant species 

worldwide, ranging from facultative to obligate ant-plant associations 

(reviewed by Davidson and McKey 1993; Bronstein 1998). Incidentally, 

by frequently foraging on the plant surface, ants often affect the life of a 

particular trophic group: the herbivores. 

Why are ants so common on foliage? First, ants may nest in plant 

structures, and therefore the plant itself is part of the colony’s immediate 

patrolled area (Janzen 1967). Second, ground-nesting ants may extend 

their foraging areas by climbing on plants to search for food (Carroll and 

Janzen 1973). A predictable food source can reinforce ant visitation to a 

particular plant location, and plant-derived food products such as 

287


288 insect-plant interactions 

extrafloral nectar and/or food bodies are known to promote ant activity 

on foliage (Bentley 1977; Koptur 1992). Additionally, some insect herbivores 

may also produce food secretions that are highly attractive to a variety 

of ant species (Way 1963; Malicky 1970; DeVries and Baker 1989). 

Whatever the factor promoting their activity on plants, ants may affect 

the life of insect herbivores in different ways, resulting in positive, negative, 

or neutral consequences (Bronstein 1994). Most studies on ant-plant interactions, 

however, have focused on the deterrence of insect herbivores by 

ants and the possible influence of such activity on plant fitness (Bronstein 

1998, and included references). Rarely has this interface been studied from 

the herbivore’s standpoint (Heads and Lawton 1985). In this chapter we 

illustrate how intense ant activity on plant foliage can strongly affect the 

behavioral ecology of insect herbivores in the cerrado. We first present the 

factors that likely promote ant foraging on cerrado plants, and then 

describe two case studies that demonstrate a close link between the behavior 

of insect herbivores and their encounters with ants on the plant surface. 

ANT FORAGING ON CERRADO PLANTS 

Several factors contribute to the ubiquity of ants on cerrado plant foliage. 

First, the stems of many plants are hollowed out by boring beetles, and 

the galleries are then used as nesting sites by numerous arboreal ant 

species. Morais (1980) recorded a total of 204 arboreal ant colonies in 

1,075 m 2 of campo cerrado (scattered shrubs and trees; see chapter 6), 

and within this area 136 live woody individuals and 17 dead standing 

trunks were found to house stem-nesting ants. Such a high occurrence of 

ant nests in the vegetation likely results in intensive foraging on cerrado 

foliage (Morais 1980; Morais and Benson 1988) and rivals similar censuses 

undertaken in tropical forests (Carroll 1979). Second, plants bearing 

extrafloral nectaries are abundant among local woody floras (Oliveira 

and Leitão-Filho 1987; Oliveira and Oliveira-Filho 1991; Oliveira and Pie 

1998), and such glands have been shown to be important promoters of 

ant activity on the cerrado foliage (see fig. 15.1A, B; see also Oliveira et 

al. 1987, Oliveira and Brandão 1991; Costa et al. 1992; Oliveira et al. 

1995). Third, insect herbivores that produce food secretions play a key 

role in attracting ants to leaves, and both honeydew-producing homopterans 

and lycaenid butterfly larvae are known to induce ant foraging on 

cerrado plants (fig. 15.1C, D; Dansa and Rocha 1992; Lopes 1995; 

Del-Claro and Oliveira 1996, 1999; Diniz and Morais 1997). 

Given that ants are dominant components of the insect fauna found 

on the cerrado foliage, experimental investigation of ant-herbivore inter-


Ant Foraging on Plant Foliage 289 

Figure 15.1 Liquid food sources for ants on cerrado foliage. (A) Accumulated 

extrafloral nectar in Qualea grandiflora (Vochysiaceae). (B) Camponotus 

sp. collecting extrafloral nectar at Q. grandiflora. (C) Camponotus blandus 

collecting honeydew from a Guayaquila xiphias treehopper. (D) Synargis (Riodininae) 

caterpillar being tended by Camponotus ants. 

actions in this vegetation type should be particularly profitable for testing 

hypotheses concerning the impact of ants on herbivore survival and behavior. 

Recent experimental work with two distinct systems has provided 

strong evidence that the occurrence of ant-herbivore encounters on the 

host plant can be largely mediated by behavioral patterns of the herbivore. 

Results show that, depending on the nature of the impact of the ants (negative 

or positive), herbivore behavior can promote either the breakage or 

the reinforcement of the relationship, thereby decreasing or increasing the 

chance of encountering an ant on the host plant (see fig. 15.2). 

ANT-BUTTERFLY INTERACTIONS 

Ant effects on butterfly biology and behavior have been investigated for 

decades, with most studies focusing on myrmecophilous lycaenid species 

whose larvae are protected against natural enemies by tending ants (Malicky 

1970; Pierce and Mead 1981; Pierce and Elgar 1985; DeVries 1984, 

1991). By living in close proximity to ants, however, butterfly larvae risk



Figure 15.2 Diagram illustrating how behavioral traits of insect herbivores 

can mediate contact with ants on the host plant. Depending on the nature of 

the impact from foraging ants—negative or positive—herbivores can mediate 

the interaction either by avoiding (Eunica butterflies) or promoting 

(Guayaquila treehoppers) contact with ants on the host plant. 

attack by the latter (but see DeVries 1991). This risk is minimized/avoided 

by lycaenid larvae via traits that decrease physical damage from ant 

attacks, reduce ant aggressiveness, and/or incite tending behavior. Such 

traits include a protective thick cuticle, the production of sweet appeasing 

substances, ant-mimicking vibration calls, and reduction of the beat 

reflex upon disturbance (Malicky 1970; DeVries 1990, 1991). The energetic 

costs to lycaenid larvae of feeding associated ants may include a prolongation 

of larval development (Robbins 1991) and a sex-related loss of 

pupal weight (Fiedler and Hölldobler 1992). However, no measurable 

cost to larvae has been found in other ant-lycaenid systems (DeVries and 

Baker l989; Cushman et al. 1994; Wagner and Martinez del Rio, 1997). 

On the other hand, larvae of non-myrmecophilous butterflies are frequently 

preyed upon or removed from host plants by foraging ants (Jones 

1987; Freitas and Oliveira 1992, 1996; Freitas 1999). Caterpillars of 

many lepidopteran species have evolved traits to escape ant predation, 

especially on ant-visited plants (reviewed by Heads and Lawton 1985; 

Salazar and Whitman 2001). Few studies, however, have been conducted



on these systems, and most were with Heliconius butterflies (Benson et al. 

1976; Smiley 1985, 1986). To date only one ant-butterfly system has been 

documented in greater detail in the cerrado (Freitas and Oliveira 1992, 

1996; Oliveira 1997). We here summarize the negative effects of ants on 

a non-myrmecophilous butterfly, Eunica bechina, and show that both the 

larval and adult stages possess traits that result in decreased contact with 

ants on a highly ant-visited host plant (fig. 15.2). 

Negative Impact of Ants: Eunica Butterflies vs. 

Ants on a Nectary Plant 

Eunica bechina (Nymphalidae) is a non-myrmecophilous butterfly whose 

larvae feed on Caryocar brasiliense (Caryocaraceae). This host plant bears 

extrafloral nectaries on the sepals and leaf buds, and is visited day and 

night by 34 species of nectar-gathering ants in an area of cerrado sensu 

stricto (see chapter 6) near Itirapina, SE Brazil (Oliveira and Brandão 

1991). Controlled ant-exclusion experiments revealed that visiting ants 

decrease the infestation levels of three common herbivores of Caryocar, 

including eggs and larvae of E. bechina (see Oliveira 1997). 

Females lay eggs singly on young leaves, on which the caterpillars 

preferably feed (Oliveira and Freitas 1991). As also recorded for a number 

of other nymphalids (DeVries 1987), Eunica larvae rest on stick-like 

frass chains constructed at leaf margins (see fig. 15.3B). A series of field 

observations and experiments on the system involving Eunica and ants 

(Freitas and Oliveira 1992, 1996) has demonstrated that the behavioral 

biology of the butterfly is closely linked with ant activity on the host plant, 

and can be summarized as follows. 

Ants and butterfly eggs. Although ants are known to prey on or 

remove insect eggs from plants (Letourneau 1983), they do not consume 

or remove Eunica bechina eggs from the host plant (Freitas and Oliveira 

1996). Field observations indicated that foraging ants frequently walk in 

the vicinity of Eunica eggs but ignore them. On plants other than Caryocar, 

we have observed Pheidole ants preying on eggs of the nymphalid 

butterflies Actinote pellenea and Dione juno, whereas Crematogaster ants 

prey on eggs of Placidula euryanassa. Both Pheidole and Crematogaster 

also consume eggs of Anaea otrere (J. M. Queiroz and P. S. Oliveira, 

unpublished data). Such butterfly eggs (all non-euryteline Nymphalidae) 

consumed by foraging ants have a soft chorion and are weakly attached 

to the host plants. Features like toughness and firm attachment to leaves 

possibly account for the lack of attractiveness of Eunica eggs to the ants 

that forage on Caryocar. 

Ant activity and caterpillars. Foraging ants frequently found and



Figure 15.3 Interaction between Eunica bechina and ants. (A) Camponotus 

sp. attacking a third-instar caterpillar. (B) Second-instar caterpillar resting 

motionless on the tip of its stick-like frass chain, as a Camponotus ant forages 

nearby. Note a previously used frass chain at upper left. (C) Rubber ants and 

(D) control rubber circles used in field experiments to test whether adult Eunica 

visually avoid ovipositing on ant-occupied plant locations. See also fig. 15.4. 

attacked Eunica caterpillars on the host plant (fig. 15.3A), and field 

experiments revealed that larval mortality is affected by the rate of ant 

visitation to the host plant (see fig. 15.4C). Larval vulnerability to ant predation, 

however, varies with the ant species and size of the caterpillar (Freitas 

and Oliveira 1992, 1996). If touched by ants, larvae usually display 

the beat reflex (curling and wriggling; see Malicky 1970) and/or also jump 

off the leaf and hang by a silken thread. When an ant bites a caterpillar, 

the latter vigorously bends its body towards the ant and frequently regurgitates, 

eventually inhibiting further ant attacks. Moreover, field experiments 

have demonstrated that the stick-like frass chains built by 

caterpillars at leaf margins (fig. 15.3B) constitute a safe refuge against ant 

predation on the host plant (fig. 15.4D). Although frass chains have long 

been described by naturalists, and their function has been assumed to be 

predator avoidance (DeVries 1987, and included references), the field 

experiment on ant-Eunica interactions demonstrated their relevance for 

larval survival on a host plant with high rates of ant visitation. 

Ant activity and ovipositing females. Female butterflies avoid



Figure 15.4 Field experiments on the interaction between Eunica bechina 

and ants on shrubs of Caryocar brasiliense. (A) Oviposition by Eunica females 

on egg-free experimental branch pairs during a 24 h period. Ant presence negatively 

affects butterfly oviposition, but the effect is significant only under 

high rates of ant visitation (mean > 0.5 ants per branch in six previous censuses). 

(Mann-Whitney U-tests; ranges are given inside bars). (B) In 2-choice 

experiments using egg-free plants, butterflies laid more eggs (after 24 hours) 

on plant branches bearing rubber circles than on neighboring branches with 

rubber ants (Mann-Whitney U-tests; ranges are given inside bars). (C) Ant 

foraging negatively affects caterpillar survival on the host plant, but mortality 

after 24 hours is significant only on branches with high ant density (G 

tests). (D) During 10-min trials, foraging ants attack live termites in significantly 

greater numbers on Caryocar leaves than on the frass chains constructed 

by Eunica caterpillars (G tests). (After Freitas and Oliveira 1996). 

See also fig. 15.3. 

ovipositing on plant locations with high ant densities (fig. 15.4A). 

Although chasing by ants can have an inhibitory effect on the oviposition 

behavior of female insects (Janzen 1967; Schemske 1980), this was not 

detected in our observations of E. bechina. Since ants do not chase egglaying 

Eunica, and the oviposition event lasts only 1–3 seconds, the differential 

occurrence of butterfly eggs on ant-visited and ant-excluded 

Caryocar plants (Oliveira 1997) presumably results from the discriminating 

abilities of the ovipositing female. This hypothesis was tested by



simultaneously placing artificial rubber ants and rubber circles at neighboring 

branches of the host plant (fig. 15.3C, D). The results unequivocally 

indicate that branches with rubber ants were less infested than those 

with rubber circles (fig. 15.4B) and that visual cues (i.e., ants) likely mediated 

egg-laying decisions by the butterfly. Although ant presence per se was 

shown to produce an avoidance response by E. bechina females, ant behavior 

and/or chemical cues could also potentially affect female oviposition. 

In conclusion, E. bechina, in both immature and adult life stages, possesses 

traits that facilitate living in an ant-rich environment. Although 

such traits are probably more clear-cut in the larvae (i.e., jumping off the 

leaf, construction of frass chains) than in the adults (selection of plants 

with low ant densities), the correct decision of the egg-laying female can 

be crucial for the survival chances of her offspring. 

Ants can inhibit herbivore occupation of host plants and have been 

thought to provide a consistent defense system relatively immune to evolutionary 

changes by the herbivore (Schemske, 1980). One may expect 

that lepidopteran larvae bearing ant-avoiding traits would have an advantage 

in the cerrado ant-rich environment. Even if larvae-constructed frass 

chains did not evolve as a direct response to the risk of ant predation, they 

may have initially facilitated the use of ant-visited plants by increasing larval 

safety against ant attacks. Data from field experiments strongly suggest 

that such stick-like structures at leaf margins provide protection 

against walking predators (Freitas and Oliveira 1996; Machado and Freitas 

2001). 

Butterflies are known to use visual cues prior to oviposition to evaluate 

both plant quality and the presence of conspecific competitors 

(Rausher 1978; Williams and Gilbert 1981; Shapiro 1981). The field 

study of Eunica bechina demonstrated that visual detection of ant presence 

can also mediate egg-laying decision by female butterflies (Freitas 

and Oliveira 1996). Although the influence of ants on oviposition decisions 

of butterflies has been documented in species with myrmecophilous 

larvae (Pierce and Elgar 1985), the precise cues eliciting the oviposition 

response have never been determined. Although our work has shown that 

visual detection of ant presence can inhibit butterfly oviposition, there is 

likely an array of ant-avoiding traits still to be discovered. 

ANT-HOMOPTERA INTERACTIONS 

The honeydew produced by phloem-feeding Homoptera (primarily 

aphids, membracids, and scales) is an ant attractant consisting of a mix-



ture of sugars, amino acids, amides, and proteins (Auclair 1963). Associations 

between ants and such homopteran groups have been commonly 

considered mutualistic (Way 1963). Tending ants may harvest the energyrich 

fluid around the clock (fig. 15.1C) and in turn provide a range of benefits 

to the homopterans, including protection from predators and 

parasitoids, and increased fecundity (Bristow 1983; Buckley 1987). Honeydew 

can be a relevant item in the diet of many ant species (Tobin 1994; 

Del-Claro and Oliveira 1999), and intra- and interspecific competition 

among homopteran aggregations for the services of ants can negatively 

affect homopteran fitness through reduced tending levels (Cushman and 

Whitham 1991). Ant-derived benefits to honeydew-producing Homoptera 

can also vary with factors such as the species of ant partner, size of 

homopteran group, developmental stage of homopterans, frequency of 

ant attendance, and predator abundance (Cushman and Whitham 1989; 

Breton and Addicott 1992; Del-Claro and Oliveira 2000; Queiroz and 

Oliveira 2000). Therefore, the outcomes of ant-homopteran associations 

are strongly dependent upon the ecological conditions in which they occur 

(Cushman and Addicott 1991; Bronstein 1994). 

Although experimental research on ant-plant-homopteran interactions 

has increased markedly over the past two decades, most studies 

come from temperate areas (e.g., Bristow 1983, 1984; Buckley 1987; 

Cushman and Whitham 1989, 1991). Only recently have these associations 

been studied in tropical habitats, including the Brazilian cerrados 

(Dansa and Rocha 1992; Del-Claro and Oliveira 1999, 2000). We report 

here on the system involving the treehopper Guayaquila xiphias (Membracidae) 

and ants, and show that ant-tending can positively affect both 

homopteran survival and fecundity, and that the homopterans’ capacity 

to attract ants early in life is a crucial behavioral trait reinforcing this relationship 

(fig. 15.2). 

Positive Impact of Ants: Guayaquila Treehoppers 

and Honeydew-Gathering Ants 

The honeydew-producing treehopper Guayaquila xiphias feeds on shrubs 

of Didymopanax vinosum (Araliaceae) in the cerrado vegetation (sensu 

stricto, see chapter 6) near Mogi-Guaçu, SE Brazil, and occurs in aggregations 

of 1 to 212 individuals near the flowers or the apical meristem 

(see fig. 15.5A, C; Del-Claro and Oliveira 1999). Guayaquila females 

exhibit parental care and guard both the egg mass and young nymphs (fig. 

15.5A, B). Nymphs develop into adults in 20–23 days, and then disperse 

from the natal aggregations. Treehopper aggregations are tended day and



Figure 15.5 Interaction between ants and honeydew-producing Guayaquila 

xiphias treehoppers. (A) Brood-guarding Guayaquila female being tended by 

Camponotus blandus ants. (B) Gonatocerus parasitoid wasp (arrow) near an 

untended brood-guarding female. (C) Camponotus rufipes tending a 

Guayaquila aggregation. (D) Larvae of predatory Ocyptamus syrphid fly 

(arrow) near untended treehopper nymphs. (E) Scattered droplets of flicked 

honeydew on leaves beneath an untended Guayaquila aggregation. (F) Antconstructed 

shelter for Guayaquila. See also fig. 15.6. 

night by an assemblage of 21 honeydew-gathering ant species, which may 

construct shelters as satellite nests to house the homopterans (fig. 15.5F; 

Del-Claro and Oliveira 1999). The attractiveness of Guayaquila’s honeydew 

to ants is high enough to maintain tending activities unchanged, even 

after the ants have discovered an alternate sugar source on the host plant 

(Del-Claro and Oliveira 1993). 

A series of field observations and controlled experiments has revealed 

that the treehoppers can receive a range of benefits from ant-tending and



that their behavior can promote contact with ants on the host plant (Del- 

Claro and Oliveira 1996, 2000). The ecology of the system can be summarized 

as follows. 

Ant effects on Guayaquila’s natural enemies. Due to continuous honeydew-gathering 

activity, ant density at any given time is higher near the 

treehoppers than at other plant locations, and this can markedly affect the 

spatial distribution and foraging behavior of Guayaquila’s natural enemies, 

such as parasitoid wasps, salticid spiders, and syrphid flies (fig. 

15.5B, D). For instance, parasitoid distribution on the plant was shown 

to be significantly affected by increased ant activity near brood-guarding 

Guayaquila, and parasitization of treehopper ovipositions was more successful 

in the absence of ants (fig. 15.5A, B; Del-Claro and Oliveira 2000). 

Aggressive toward intruding predators and parasitoids, tending ants not 

only ward off such enemies from the vicinity of the treehoppers, but may 

also attack and kill the intruders. Controlled ant-exclusion experiments 

revealed that ant presence decreases the abundance of Guayaquila’s natural 

enemies on the host plant (see fig. 15.6A). 

Ant-derived benefits to Guayaquila xiphias. Ant-exclusion experiments 

have demonstrated that tending ants can have a positive impact on 

treehopper survival (fig. 15.6 B). Moreover, ants can also confer a direct 

reproductive benefit to Guayaquila (see also Wood 1977; Bristow 1983). 

By transferring parental care to ants, ant-tended brood-guarding females 

(fig. 15.5A) have a higher chance of producing an additional clutch than 

untended females (91% vs. 54% of the cases; P = 0.018, χ 2 = 5.61; N = 

22 females in each experimental group). Two years of experimental 

manipulations, however, have shown that ant-derived benefits related to 

protection and fecundity can vary with time and/or with the species of 

tending ant (Del-Claro and Oliveira 2000). Several other studies have also 

shown that species of ants may differ greatly in the protection they afford 

to homopterans, and this may depend on the ants species-specific traits 

such as size, promptness to attack intruders, morphological and chemical 

weapons, as well as recruitment behavior (e.g., Addicott 1979; Messina 

1981; Buckley 1987; Buckley and Gullan 1991). 

Attraction of ants through honeydew flicking. Ant-tending unequivocally 

plays a crucial role in the survival of developing brood of 

Guayaquila xiphias in the cerrado, as also shown for other temperate antmembracid 

systems (e.g., Bristow 1983; Cushman and Whitham 1989). 

It is therefore reasonable to predict that any behavior promoting early 

contact with ants would be advantageous for ant-tended treehoppers (see 

also DeVries 1990; DeVries and Baker 1989, on ant-tended caterpillars). 

Guayaquila xiphias females, as well as developing nymphs, frequently 

flick away the accumulated honeydew if it is not promptly collected by



Figure 15.6 Field experiments on the interaction between honeydew-producing 

Guayaquila xiphias and tending ants on shrubs of Didymopanax 

vinosum. (A) Ant presence significantly reduces the number of Guayaquila’s 

natural enemies (spiders, syrphid flies, and parasitoid wasps) on the host plant 

(Treatment: F = 11.54, df = 1, P = .0015). (B) Ant-tending positively affects 

treehopper survival through time (Treatment × Time: F = 4.33, df = 7, P = 

.0001). (C) After finding scattered droplets of flicked honeydew on the 

ground beneath untended treehoppers, the number of ants involved with 

tending activities increases with time due to recruitment behavior (F = 2.44, 

df = 5, P = .04). (D) Pieces of honeydew-soaked filter paper placed beneath 

treehopper-free plants induce significantly more ground-dwelling ants to 

climb onto the plant than control papers with water (Treatment: F = 15.89, 

df = 1, P = .001). All tests performed with repeated-measures ANOVA. Data 

from (A) and (B) after Del-Claro and Oliveira (2000); (C) and (D) after Del- 

Claro and Oliveira (1996). See also fig. 15.5. 

tending ants; this results in the occurrence of scattered honeydew droplets 

below untended or poorly tended treehopper aggregations (fig. 15.5E). 

Field experiments have shown that honeydew flicking by untended 

Guayaquila can provide cues to ground-dwelling ants, which climb onto 

the plant and start tending activities (fig. 15.6C, D; Del-Claro and 

Oliveira 1996). Groups of untended Guayaquila nymphs start secreting 

honeydew soon after introduction on previously unoccupied host plants.



Upon encountering the droplets on the ground, alerted ants climb onto 

the plant and eventually find the homopterans. The number of ants 

engaged in tending activities increases with time due to recruitment to the 

newly discovered food source (fig. 15.6C). Honeydew-soaked filter 

papers placed beneath unoccupied host plants further confirmed that 

flicked honeydew provides cues to ants and induces them to climb onto 

the plant (fig. 15.6D). 

Attraction of Ants by Ant-Tended Insects 

The presence of honeydew on lower foliage or on the ground beneath 

untended homopterans is well documented (Buckley 1987; Hölldobler 

and Wilson 1990). Douglas and Sudd (1980) discounted the possibility 

that scattered aphid honeydew attracted Formica ants since they had seen 

these ants ignoring fallen droplets. In the Guayaquila-ant association, 

however, we have shown that flicking accumulated honeydew can mediate 

this ant-homopteran system by promoting contact between potentially 

interacting species. Honeydew accumulated on the bodies or in the vicinity 

of untended homopterans may result in increased mortality due to fungal 

infections (Buckley 1987). It is therefore possible that ant attraction 

through honeydew flicking has evolved as a by-product of a primarily 

defensive behavior against fungi-induced damage. 

Ant-tending may also confer a range of benefits to butterfly larvae in 

the family Lycaenidae (Pierce and Mead 1981; DeVries 1991). Some adult 

butterflies promote contact with ants by choosing ant-occupied plant 

individuals (Pierce and Elgar 1985). Myrmecophilous butterfly larvae and 

pupae produce substrate-borne vibrational calls, which have been demonstrated 

to attract nearby ants (DeVries 1990, 1992; Travassos and Pierce 

2000). Therefore, for myrmecophilous butterflies, contact with tending 

ants can be promoted by both adults and immatures. Cocroft (1999) has 

recently shown that substrate-borne vibration calls are used in offspringparent 

communication by Umbonia treehoppers. DeVries (1991b) has 

speculated that vibrational communication by ant-tended membracids as 

well as by other myrmecophilous insects could be used to maintain ant 

association. 

CONCLUSIONS AND RESEARCH DIRECTIONS 

Ant-plant-herbivore interactions offer numerous promising avenues for 

future research in the cerrado, with ramifications for different areas of



experimental field biology and applied ecology. The uniqueness of the cerrado 

for this type of research relies on the prevalence of ants on the plant 

substrate, and on the abundance of predictable liquid food sources in the 

form of extrafloral nectar and insect-derived secretions. Moreover, arboreal 

ants commonly nest inside hollowed-out stems of cerrado plants 

(Morais 1980), and this per se promotes intense ant patrolling activity on 

leaves, regardless of the presence of liquid food rewards on the plant. The 

prevalence of ants on foliage makes ant-herbivore-plant interactions especially 

pervasive in the cerrado, as revealed by the high abundance of 

extrafloral nectary-mediated interactions (Oliveira and Oliveira-Filho 

1991; Oliveira 1997), as well as the large number of ant-tended treehoppers 

(Lopes 1995) and lycaenids (Brown 1972) occurring in this biome. 

The data summarized in this chapter illustrate how foraging by ants on 

cerrado plants can affect herbivore biology in contrasting ways, and at 

the same time point to a number of facets in ant-herbivore systems that 

have not yet been investigated. For instance, although it is clear that both 

butterfly adults and larvae can either avoid (as in Eunica) or promote (as 

in ant-tended lycaenids) encounters with ants on the host plant, we are 

only beginning to understand the mechanisms through which such interactions 

can be behaviorally mediated by the herbivore. Although visual 

stimuli play an important role for Eunica females to avoid ants, the cues 

used by lycaenids to lay eggs on ant-occupied plants are still unknown. 

Similarly, we know virtually nothing of the decision mechanisms used by 

ant-tended treehoppers in selecting individual host plants. Is ant presence 

somehow perceived by treehopper females, and can this mediate oviposition? 

Can ant-tended treehoppers use vibrational communication to 

attract ant partners? Moreover, since the negative/positive impact of ants 

on a given herbivore species can vary among different ant species, can the 

herbivore tell ants apart and behave/respond differently to them depending 

on the intensity of their harmful/beneficial effects? Finally, the cerrado 

savanna is unique for the study of ant-plant-herbivore systems because in 

most cases the researcher can have full visual access to the foliage. Field 

work under this situation permits not only a more accurate description of 

the behavioral traits mediating the interactions, but also the development 

of controlled field experiments to identify the selective forces operating 

within such multitrophic systems. 


We thank R.J. Marquis, P.J. DeVries, S. Koptur, T.K. Wood, R.K. Robbins, 

G. Machado, T. Quental, and H. Dutra for helpful suggestions on



the manuscript. P.J. DeVries also provided information on butterflies. Our 

studies in the cerrado were supported by Brazilian Federal agencies 

(CNPq, CAPES), and by research grants from the Universidade Estadual 

de Campinas (FAEP). 

REFERENCES 

Addicott, J. F. 1979. A multispecies aphid-ant association: Density dependence 

and species-specific effects. Can. J. Zool. 57:558–569. 

Auclair, J. L. 1963. Aphid feeding and nutrition. Ann. Rev. Entomol. 8:439–490. 

Benson, W. W., K. S. Brown, and L. E. Gilbert. 1976. Coevolution of plants 

and herbivores: Passion flower butterflies. Evolution 29:659–680. 

Bentley, B. L. 1977. Extrafloral nectaries and protection by pugnacious bodyguards. 


Breton, L. M. and J. F. Addicott. 1992. Density-dependent mutualism in an 

aphid-ant interaction. Ecology 73:2175–2180. 

Bristow, C. M. 1983. Treehoppers transfer parental care to ants: A new benefit 

of mutualism. Science 220:532–533. 

Bronstein, J. L. 1994. Conditional outcomes in mutualistic interactions. 

Trends Ecol. Evol. 9:214–217. 

Bronstein, J. L. 1998. The contribution of ant-plant protection studies to our 

understanding of mutualism. Biotropica 30:150–161. 

Brown Jr., K. S. 1972. Maximizing daily butterfly counts. J. Lep. Soc. 

26:183–196. 

Buckley, R. C. 1987. Interactions involving plants, Homoptera, and ants. 


Buckley, R. C. and P. Gullan. 1991. More aggressive ant species (Hymenoptera: 

Formicidae) provide better protection for soft scales and mealybugs 

(Homoptera: Coccidae, Pseudococcidae). Biotropica 23:282–286. 

Carroll, C. R. 1979. A comparative study of two ant faunas: The stem-nesting 

ant communities of Liberia, West Africa, and Costa Rica, Central 

America. Amer. Nat. 113:551–561. 

Carroll, C. R. and D. H. Janzen. 1973. Ecology of foraging by ants. Ann. Rev. 

Ecol. Syst. 4:231–257. 

Cocroft, R. B. 1999. Parent-offspring communication in response to predators 

in a subsocial treehopper (Hemiptera: Membracidae: Umbonia crassicornis). 

Ethology 105:553–568. 

Costa, F. M. C. B., A. T. Oliveira-Filho, and P. S. Oliveira. 1992. The role of 

extrafloral nectaries in Qualea grandiflora (Vochysiaceae) in limiting 

herbivory: An experiment of ant protection in cerrado vegetation. Ecol. 

Entomol. 17:362–365. 

Cushman, J. H. and J. F. Addicott. 1991. Conditional interactions in antplant-herbivore 

mutualisms. In C. R. Huxley and D. F. Cutler, eds., Ant- 

Plant Interactions, pp. 92–103. Oxford: Oxford University Press.



Cushman, J. H., V. K. Rashbrook and A. J. Beattie. 1994. Assessing benefits to 

both participants in a lycaenid-ant association. Ecology 75:1031–1041. 

Cushman, J. H. and T. G. Whitham. 1989. Conditional mutualism in a membracid-ant 

association: Temporal, age-specific, and density-dependent 

effects. Ecology 70:1040–1047. 

Cushman, J. H. and T. G. Whitham. 1991. Competition mediating the outcome 

of a mutualism: Protective services of ants as a limiting resource 

for membracids. Amer. Nat. 138:851–865 

Dansa, C. V. A. and C. F. D. Rocha. 1992. An ant-membracid-plant interaction 

in a cerrado area of Brazil. J. Trop. Ecol. 8:339–348. 

Davidson, D. W. and D. McKey. 1993. The evolutionary ecology of symbiotic 

ant-plant relationships. J. Hym. Res. 2:13–83. 

Del-Claro, K. and P. S. Oliveira. 1993. Ant-homoptera interaction: Do alternative 

sugar sources distract tending ants? Oikos 68:202–206. 

Del-Claro, K. and P. S. Oliveira. 1996. Honeydew flicking by treehoppers provides 

cues to potential tending ants. Anim. Behav. 51:1071–1075. 

Del-Claro, K. and P. S. Oliveira. 1999. Ant-Homoptera interactions in a 

neotropical savanna: The honeydew-producing treehopper Guayaquila 

xiphias (Membracidae) and its associated ant fauna on Didymopanax 

vinosum (Araliaceae). Biotropica 31:135–144. 

Del-Claro, K. and P. S. Oliveira. 2000. Conditional outcomes in a neotropical 

treehopper-ant association: Temporal and species-specific variation 

in ant protection and homopteran fecundity. Oecologia 124:156–165. 

DeVries, P. J. 1984. Of crazy-ants and Curetinae: Are Curetis butterflies 

tended by ants? Zool. J. Linn. Soc.79:59–66. 

DeVries, P. J. 1987. The Butterflies of Costa Rica and Their Natural History. 

Princeton: Princeton University Press. 

DeVries, P. J. 1990. Enhancement of symbiosis between butterfly caterpillars 

and ants by vibrational communication. Science 248:1104–1106. 

DeVries, P. J. 1991a. Mutualism between Thisbe irenea butterflies and ants, 

and the role of ant ecology in the evolution of larval-ant associations. 

Biol. J. Linn. Soc. 43:179–195. 

DeVries, P. J. 1991b. Call production by myrmecophilous riodinid and 

lycaenid butterfly caterpillars (Lepidoptera): Morphological, acoustical, 

functional, and evolutionary patterns. Amer. Mus. Nov. 26:1–23. 

DeVries, P. J. 1992. Singing caterpillars, ants and symbiosis. Scient. Amer. 

267:76–82. 

DeVries, P. J. and I. Baker. 1989. Butterfly exploitation of a plant-ant mutualism: 

Adding insult to herbivory. J. N. Y. Entomol. Soc. 97:332–340. 

Diniz, I. R. and H. C. Morais. 1997. Lepidopteran caterpillar fauna of cerrado 

host plants. Biodiv. and Conserv. 6:817–836. 

Douglas, J. M. and J. H. Sudd. 1980. Behavioral coordination between an 

aphid and the ant that tends it: An ethological analysis. Anim. Behav. 

28:1127–1139.



Fiedler K. and B. Hölldobler. 1992. Ants and Polyommatus icarus immatures 

(Lycaenidae): Sex-related developmental benefits and costs of ant attendance. 

Oecologia 91:468–473. 

Freitas, A. V. L. 1999. An anti-predator behavior in larvae of Libytheana carinenta 

(Nymphalidae, Libytheinae). J. Lep. Soc. 53:130–131. 

Freitas, A. V. L. and P. S. Oliveira. 1992. Biology and behavior of Eunica 

bechina (Lepidoptera: Nymphalidae) with special reference to larval 

defense against ant predation. J. Res. Lepid. 31:1–11. 

Freitas, A. V. L. and P. S. Oliveira. 1996. Ants as selective agents on herbivore 

biology: effects on the behaviour of a non-myrmecophilous butterfly. J. 

Anim. Ecol. 65:205–210. 

Heads, P. A. and J. H. Lawton. 1985. Bracken, ants and extrafloral nectaries: 

III. How insect herbivores avoid predation. Ecol. Entomol. 10:29–42. 

Hölldobler, B. and E. O. Wilson. 1990. The Ants. Cambridge, Mass.: The 

Belknap Press of Harvard University Press. 

Janzen, D. H. 1967. Interaction of the bull’s horn acacia (Acacia cornigera L.) 

with an ant inhabitant (Pseudomyrmex ferruginea F. Smith) in Eastern 

Mexico. Univ. Kansas Sci. Bull. 47:315–558. 

Jones, R. E. 1987. Ants, parasitoids, and the cabbage butterfly Pieris rapae. 

J. Anim. Ecol. 56:739–749. 

Koptur, S. 1992. Extrafloral nectary-mediated interactions between insects 

and plants. In E. Bernays, ed., Insect-Plant Interactions, Vol. 4, pp. 

81–129. Boca Raton: CRC Press. 

Letourneau, D. K. 1983. Passive aggression: An alternative hypothesis for the 

Piper-Pheidole association. Oecologia 60:122–126. 

Lopes, B. C. 1995. Treehoppers (Homoptera: Membracidae) in the Southeast 

Brazil: Use of host plants. Rev. Bras. Zool. 12:595–608. 

Machado, G. and A. V. L. Freitas 2001. Larval defence against ant predation 

in the butterfly Smyrna blomfildia. Ecol. Entomol. 26:436–439. 

Malicky, H. 1970. New aspects of the association between lycaenid larvae 

(Lycaenidae) and ants (Formicidae, Hymenoptera). J. Lep. Soc. 

24:190–202. 

Messina, F. J. 1981. Plant protection as a consequence of ant-membracid mutualism: 

Interactions on Goldenrod (Solidago sp.). Ecology 62:1433–1440 

Morais, H. C. 1980. “Estrutura de uma comunidade de formigas arborícolas 

em vegetação de campo cerrado.” Master’s thesis, Universidade Estadual 

de Campinas, Campinas, Brazil. 

Morais, H. C. and W. W. Benson 1988. Recolonização de vegetação de cerrado 

após queimada, por formigas arborícolas. Rev. Bras. Biol. 48:459–466. 

Oliveira, P. S. 1997. The ecological function of extrafloral nectaries: Herbivore 

deterrence by visiting ants and reproductive output in Caryocar 

brasiliense (Caryocaraceae). Funct. Ecol. 11:323–330. 

Oliveira, P. S. and C. R. F. Brandão. 1991. The ant community associated with 

extrafloral nectaries in Brazilian cerrados. In C. R. Huxley and D. F. Cut-



ler, eds., Ant-Plant Interactions, pp. 198–212. Oxford: Oxford University 

Press. 

Oliveira, P. S. and A. V. L. Freitas. 1991. Hostplant record for Eunica bechina 

magnipunctata (Nymphalidae) and observations on oviposition sites and 

immature biology. J. Res. Lepid. 30:140–141. 

Oliveira, P. S., C. Klitzke, and E. Vieira. 1995. The ant fauna associated with 

the extrafloral nectaries of Ouratea hexasperma (Ochnaceae) in an area 

of cerrado vegetation in Central Brazil. Ent. Mo. Mag. 131:77–82. 

Oliveira, P. S. and H. F. Leitão-Filho. 1987. Extrafloral nectaries: Their taxonomic 

distribution and abundance in the woody flora of cerrado vegetation 

in southeast Brazil. Biotropica 19:140–148. 

Oliveira, P. S. and A. T. Oliveira-Filho. 1991. Distribution of extrafloral nectaries 

in the woody flora of tropical communities in western Brazil. In P. 

W. Price, T. M. Lewinsohn, G. W. Fernandes, and W. W. Benson, eds., 

Plant-Animal Interactions: Evolutionary Ecology in Tropical and Temperate 

Regions, pp. 163–175. New York: John Wiley and Sons. 

Oliveira, P. S. and M. R. Pie. 1998. Interaction between ants and plants bearing 

extrafloral nectaries in cerrado vegetation. An. Soc. Entomol. Brasil 

27:161–176. 

Oliveira, P. S., A. F. da Silva, and A. B. Martins. 1987. Ant foraging on extrafloral 

nectaries of Qualea grandiflora (Vochysiaceae) in cerrado vegetation: 

Ants as potential antiherbivore agents. Oecologia 74:228–230. 

Pierce, N. E. and M. A. Elgar. 1985. The influence of ants on host plant selection 

by Jalmenus evagora, a myrmecophilous lycaenid butterfly. Behav. 

Ecol. Sociobiol. 16:209–222. 

Pierce, N. E. and P. S. Mead. 1981. Parasitoids as selective agents in the symbiosis 

between lycaenid butterfly larvae and ants. Science 211:1185–1187. 

Queiroz, J. M. and P. S. Oliveira. 2000. Tending ants protect honeydewproducing 

whiteflies (Homoptera: Aleyrodidae). Environ. Entomol. 

30:295–297. 

Rausher, M. D. 1978. Search image for leaf shape in a butterfly. Science 

200:1071–1073. 

Robbins, R. K. Cost and evolution of a facultative mutualism between ants 

and lycaenid larvae (Lepidoptera). Oikos 62:363–369. 

Salazar, B. A. and D. W. Whitman. 2001. Defensive tactics of caterpillars 

against predators and parasitoids, pp. 161–207. In T. N. Ananthakrishnan 

(ed.), Insect and Plant Defense Dynamics. Enfield: Science Publisher 

Inc. 

Schemske, D. W. 1980. The evolutionary significance of extrafloral nectar 

production by Costus woodsonii (Zingiberaceae): An experimental 

analysis of ant protection. J. Ecol. 68:959–967. 

Shapiro, A. M. 1981. The pierid red-egg syndrome. Amer. Nat. 117:276–294. 

Smiley, J. T. 1985. Heliconius caterpillars mortality during establishment on 

plants with and without attending ants. Ecology 66:845–849.



Smiley, J. T. 1986. Ant constancy at Passiflora extrafloral nectaries: Effects 

on caterpillar survival. Ecology 67:516–521. 

Tobin, J. E. 1994. Ants as primary consumers: Diet and abundance in the 

Formicidae. In J. H. Hunt and C. A. Nalepa, eds., Nourishment and Evolution 

in Insect Societies, pp. 279–308. Oxford: Westview Press. 

Travassos, M. A. and N. E. Pierce. 2000. Acoustics, context and function of 

vibrational signalling in a lycaenid butterfly-ant mutualism. Anim. 

Behav. 60:13–36. 

Wagner, D. and C. Martinez del Rio. 1997. Experimental tests of the mechanism 

for ant-enhanced growth in an ant-tended lycaenid butterfly. 

Oecologia 112:424–429. 

Way, M. J. 1963. Mutualism between ants and honeydew-producing 

Homoptera. Ann. Rev. Entomol. 8:307–344 

Williams, K. S. and L. E. Gilbert. 1981. Insects as selective agents on plant 

vegetative morphology: Egg mimicry reduces egg laying by butterflies. 

Science 212:467–469. 

Wilson E. O. 1987. Causes of ecological success: The case of the ants. J. Anim. 

Ecol. 56:1–9. 

Wood, T. K. 1977. Role of parent females and attendant ants in maturation 

of the treehopper, Entylia bactriana (Homoptera: Membracidae). Sociobiology 

2:257–272.


16 

Interactions Among Cerrado 

Plants and Their Herbivores: 

Unique or Typical? 

Robert J. Marquis, Helena C. Morais, 

and Ivone R. Diniz 

A long dry season, frequent fires, and very low soil 

nutrient quality are stress factors that make life difficult for cerrado plants 

(see chapters 2, 4, 9). Perhaps as a result, ecological studies of plant adaptation 

to cerrado environments have emphasized the role of abiotic factors 

in shaping plant adaptation and plant distribution in these 

environments (Lewinsohn et al. 1991). Investigation of biotic factors such 

as herbivory, and of plant-animal interactions in general, have lagged 

behind such efforts. 

Initial studies of plant-herbivore interactions in cerrado have been 

descriptive, focusing on basic information. Questions addressed by such 

studies include: how many herbivore species are involved, and what are 

their abundances and diets? What is the relationship between herbivorous 

insect abundance and plant phenology? How much damage do herbivores 

cause, and which plant traits account for interspecific differences in damage 

level? Here we provide at least partial answers to these questions, 

highlighting our own work at the Fazenda Água Limpa (FAL), a reserve 

in central Brazil (Brasília, DF), while drawing on studies conducted at 

other sites when available. 

When tackling an understudied ecosystem, it is useful to ask how different 

that system is from others already studied. Besides describing our 

current understanding of cerrado plant-herbivore interactions, we make 

comparisons when appropriate with data for other tropical savanna systems 

and tropical forests in general. 

306


Interactions Among Cerrado Plants and Their Herbivores 307 

COMPOSITION OF A LOCAL LEPIDOPTERA FAUNA 

Over the last 10 years, HCM and IRD and their students have focused on 

the surface-feeding caterpillars found on leaves of cerrado plants at FAL. 

Rainfall at FAL is approximately 1,400 mm annually, with a strong dry 

season from May to September. Average annual temperature is 22°C. 

Fifteen individuals of each of 40 plant species, representing 21 

families, were censused weekly for at least one year per plant species. 

Collected caterpillars were then reared in the laboratory on leaves of the 

host species on which they were encountered. A total of 3,347 individuals 

of 415 species were successfully reared. A smaller sample of caterpillars 

(147 total reared from 33 plant species of 17 families) were 

collected from flowers and developing fruits (see also Clark and Martins 

1987; Del Claro et al. 1997 for studies of beetle and thrips attack of flowers, 

respectively). 

These rearing efforts demonstrated that most cerrado folivorous 

species were rare. Though 38 families were represented in the collection, 

50% of all reared species were from just five families: Elachistidae (formerly 

the Oecophoridae), Gelechiidae, Pyralidae, Geometridae, and the 

Arctiidae. For 42% of these reared species, a single individual was reared, 

and in 51% of the total species, only 2–5 individuals were reared. Only 

4% of the species were represented by more than 50 individuals (see Diniz 

and Morais 1997). 

Not only were most caterpillar species rare, but caterpillars as a whole 

were rare. Occurrence by one or more leaf-feeding caterpillars on plants 

censused ( = occupancy; a total of 30,000 individual plants were censused 

over the 10-year period) was only 10%, varying from 0.7% to 34% per 

plant species. Similar low rates of occupancy by caterpillars have been 

reported for tropical wet forest plants, including Piper arieianum shrubs 

in Costa Rican wet forest (occupancy equaled 0.07%, Marquis 1991), 

seedlings of various tree species in French Guiana wet forest (Basset 

1999), and from sweep samples in the understory of tropical wet forest 

(Elton 1973; Boinski and Fowler 1989). 

In a detailed study at FAL, the cumulative number of species yielded 

no asymptotic level of richness for three species of Erythroxylum (E. 

deciduum, E. suberosum, and E. tortuosum) over three time periods of 6 

mo, 7 mo, and 23 mo (Price et al. 1995). The caterpillar fauna of these 

three tree species was strikingly different from that of a sample of woody 

plants from a temperate savanna (Arizona, U.S.) at the same altitude. Estimated 

richness per plant species was 2–3 times higher in cerrado, abundance 

per plant was 11-fold higher in the temperate site, and the number



of plants with at least one caterpillar was 12% in cerrado versus 49% in 

the temperate savanna (Price et al. 1995). 

Caterpillar species averaged 19 per plant species but varied greatly 

from four to 53 species (N = 40 plant species). Previous studies have shown 

that interspecific variation in local species richness of herbivorous insects 

is related to plant size, taxonomic isolation, and both local and regional 

estimates of abundance (reviewed by Strong et al. 1984; Marquis 1991). 

We used detrended correspondence analysis (DCA) to determine the plant 

traits that might account for this interspecific variation in cerrado Lepidoptera 

faunal diversity per plant species (Morais et al. in prep.). Analysis 

was based on a matrix of 40 species and 7 variables: growth form (herb, 

shrub, or tree), foliar phenology (after Morais et al. 1995), the presence of 

latex and extrafloral nectaries, leaf pubescence, the number of species per 

family in the Distrito Federal (an estimate of taxonomic isolation), and the 

number of censuses, an estimate of sampling intensity. The first axis 

explained 33.6% of the interspecific variation in lepidopteran species richness, 

while the cumulative variance explained by the two first axes was 

61.8% (eigenvalues for axis 1 and 2 were 0.095 and 0.079, respectively). 

Richness was negatively related to the presence of latex, extrafloral nectaries, 

and leaf pubescence, with the first two variables accounting for 60% 

of the separation among host plant species (see fig. 16.1). 

Local species richness of another guild of herbivorous insects, galling 

species, likewise does not show a relationship with host plant geographic 

range for cerrado. Rather, galling insect richness is related to plant growth 

form and habitat quality, with the number of galling species greatest on 

shrubs rather than on trees, contrary to the usual pattern. The most 

important factor for galling species richness in cerrado seems to be 

hygrothermic stress, as richness is higher in xeric than in mesic sites and 

declines with altitude (Fernandes and Price 1991). 

Contrary to expectation, 29% of all caterpillar species (55 species 

from 12 families) reared from reproductive structures had been found and 

reared previously on leaves (Diniz and Morais 1995; Diniz and Morais 

submitted). The separate and combined impacts on plant fitness of herbivores 

feeding on both leaves and reproductive parts has not been considered 

(Strauss 1997), perhaps because rarely are insect herbivores 

recognized to feed on both plant part types. Three families represented 

71% of all species eating flowers and young fruits: Gelechiidae (19 

species), Tortricidae (12 species), and Pyralidae (eight species). 

The reasons for the observed rarity of caterpillars in cerrado are not 

well understood. Presumably low abundance is due to some combination 

of the effect of low relative atmospheric humidity, especially at the end of



Figure 16.1 Similarity analysis (Sorensen) showing groupings, and plant 

traits associated with those groupings, of plant species in terms of the number 

of associated caterpillar species. Sixty percent of the variability is 

accounted for by the first two groupings, which in turn were associated with 

the presence of latex in the leaves and extrafloral nectaries (efn). 

dry season; low nutrient content of plant material (see below); and predation 

pressure from natural enemies. Elton (1973; see also Coley and 

Barone 1997) hypothesized that high predation and parasitism were the 

factors that maintained low abundance of herbivorous insects at the 

Brazilian and Panamanian wet forest sites that he sampled. High parasitism 

levels (see below) would suggest that parasitoids may have an 

important influence on the abundance of cerrado caterpillars. 

SEASONALITY AND INSECT HERBIVORE ABUNDANCE 

Previous research throughout the tropics, both in the Old and New World, 

demonstrates in general that insect herbivore abundance becomes more 

seasonal as the distribution of rainfall becomes more seasonal. With a 

more seasonal distribution of rainfall, and increasing deciduousness of 

the foliage, abundance and activity of herbivores is limited to a smaller 

portion of the year: namely, the narrowing wet season. Further, a peak in 

herbivore abundance is associated with the flushing of new leaves with 

the coming of the wet season. In contrast, in the wettest and most aseasonal 

of forests, abundance changes little over the year (e.g., Barlow and



Woiwod 1990) and may sometimes be highest in the drier portion of the 

year (e.g., Wong 1984). 

Rainfall is highly seasonal in distribution in cerrado of central Brazil, 

and as a result leaf flushing at the community level peaks just prior to the 

beginning of the wet season (Morais et al. 1995). Flowering and fruiting 

occur year-round, with a peak in flowering activity at the end of dry season 

(August to September) and a peak in fruiting at the beginning of the wet season 

(October to November) (Oliveira 1998). In contrast to other sites with 

highly seasonal rainfall, cerrado plants retain leaves throughout the dry season. 

Plant species show one of three patterns of leaf flush and leaf loss 

(Morais et al. 1995): (1) all leaves are abscised just before the major flush 

of new leaves (deciduous species); (2) leaves are steadily abscised beginning 

at the end of the dry season, throughout leaf flush, and on into the wet season 

(semideciduous species); or (3) leaves are produced and abscised all year 

round (evergreen species) and live approximately one year (most evergreen 

species) or, rarely, up to three years (a few evergreen species). 

Herbivorous insect abundance does not fit well into the expected pattern 

based on seasonal distribution of rainfall alone. Based on one year of 

data from window, pitfall, and malaise trapping by IRD and her students, 

the abundance of immature and adult Homoptera, immature and adult 

Orthoptera, and adult Lepidoptera were not strongly seasonal, occurring 

in relatively high numbers throughout the year, although the peak in 

abundance of Homoptera and Lepidoptera adults did coincide with the 

beginning of the wet season (see fig. 16.2A). Distributions for all were not 

different from random across the year (circular statistic S 0 = 96.9, 98.0, 

and 88.4, r = 0.24, 0.23, 0.30, for Homoptera, adult Lepidotera, and 

Orthoptera, respectively) (Pinheiro et al. submitted). Most strikingly, in 

contrast to what has been reported for many other seasonal forest sites in 

which insect herbivore abundance is greatest at leaf flush (e.g., Frith and 

Frith 1985), or increases as the rainy season progresses (e.g., Wolda 1978; 

Murali and Sukumar 1993), caterpillar abundance FAL peaked 7–8 

months after the main leaf flush period, coinciding instead with the early 

dry season (Price et al. 1995; Pinheiro et al. 1997; Morais et al. 1999; fig. 

16.2B). Only Coleoptera peaked in abundance with the beginning of the 

wet season and just after the time of highest new leaf availability (Pinheiro 

et al. submitted; fig. 16.2A). This general lack of seasonality in some taxa 

clearly shows that herbivorous insects can persist during a harsh dry season 

if their food plants are available. 

Morais et al. (1999) suggest escape from parasitism as a possible 

reason for the end of the wet season peak in caterpillar abundance. Parasitism 

at FAL peaked in September–October, just after the time of leaf 

flush (fig. 16.2B). If escape from natural enemies has been the selective


Figure 16.2 Seasonal variation in cerrado plant-herbivore-parasitoid interactions. 

(A) Monthly number of individuals to four herbivore orders estimated 

by three collection methods (pitfall, window, and Malaise traps) from 

May 1997 to April 1998 in cerrado sensu stricto, Distrito Federal, Brazil. The 

dry season during the period of study is indicated by the shading. (B) Seasonal 

pattern of new leaf production, caterpillar abundance, and parasitism of 

reared caterpillars in an area of cerrado in the Distrito Federal of Brazil (after 

Morais et al. 1995, 1999). The dry season during the period of study is indicated 

by the shading.



factor resulting in caterpillars feeding on mature leaves, then this suggests 

that selection to avoid mortality by the third trophic level has outweighed 

the cost of feeding on less nutritious, mature leaves. 

Within this general overall pattern of caterpillar abundance are idiosyncratic 

patterns that do not conform to the overall trends. For example, 

some species can be found throughout the entire year on their host 

plants, such as Cerconota achatina, Stenoma cathosiota (Elachistidae: 

Stenomatinae), and Fregela semiluna (Arctiidae), whereas others are 

found principally during the rainy season, such as Siderone marthesia 

(Nymphalidae) and Aucula munroei (Noctuidae) (Andrade et al. 1995; 

Morais et al. 1996; Morais et al. 1999; Diniz et al. 2000; Morais and 

Diniz 2001). 

The described seasonal patterns for central Brazil in particular may 

not apply to all of Brazilian cerrado. For example, there is a marked reduction 

in herbivorous insect abundance in general (Cytrynowicz 1991), and 

for Membracidae (Homoptera) in particular (Lopes 1995), during the dry 

and cold period between May and September in southeastern Brazil. These 

data suggest that herbivorous insects cannot withstand the combined 

stresses of low temperatures and low humidity experienced in that region. 

HOST SPECIFICITY OF LEPIDOPTERAN CATERPILLARS 

Herbivorous insects show a wide range of host specificity, from those that 

specialize on an individual plant part of only one plant species, to those 

that include literally hundreds of host species in their diet list. Most, however, 

are specialists, and initial studies in the tropics suggested that a high 

degree of specificity is characteristic of tropical herbivorous insects. 

Median host range for 12 taxa (weevils, flea beetles, hispine beetles, butterflies, 

and one subfamily of Geometridae) ranged from 1–4 host plant 

species per insect species (see review in Marquis 1991). 

Since this initial list was compiled, a number of other studies have 

been published on the host specificity of herbivore faunas in general. In 

cerrado, HCM and IRD successfully reared 3,347 caterpillars from the 

leaves of 86 host plant species, representing 40 plant families at FAL. Of 

those reared, 174 species were reared only once and are not considered 

for this analysis. For the remainder, results show that the majority of 

reared caterpillar species were host-plant specialists (155 species or 64% 

feed within one host plant family) (see table 16.1). However, this level of 

specificity was lower than that reported for a Panamanian forest and a 

Costa Rican forest (see table 16.2), but not as low as that for a set of wet



Table 16.1 Host Specificity of Lepidopteran 

Caterpillars in the Cerrado Sensu Stricto 

of Fazenda Água Limpa, Brasília 

No. of 

caterpillar No. of host No. of 

species plant families adults reared 

155 1 1766 

45 2 604 

11 3 203 

8 4 55 

6 5 90 

3 6 29 

2 7 42 

4 8 89 

2 9 140 

1 11 109 

1 12 20 

2 13 138 

1 23 62 

Total 241 38 3347 

forest trees in Papua New Guinea (Bassett 1996) and one tree species in 

a Queensland rain forest of Australia (Bassett 1992; table 16.2). Within 

the cerrado sample, families of Lepidoptera vary greatly in their degree of 

specialization, from approximately 83% restricted to a single plant family 

(Pyralidae) to none (Megalopygidae) (Diniz and Morais in prep.; see 

table 16.3). Cerrado Membracidae (sapsucking treehoppers), like other 

tropical Membracidae (Wood and Olmstead 1984), demonstrate a high 

degree of polyphagy (Lopes 1995; table 16.2). Together, these data support 

the view that tropical herbivorous insect species are greatly variable 

in their degree of specificity, both within taxa across sites, and across taxa 

within sites. 

The degree of specialization of the herbivore fauna of a particular 

plant is important from the plant’s point of view. Plants attacked mainly 

by generalists may be able to escape those herbivores in evolutionary time 

by evolving chemically novel deterrents as toxins to distinguish themselves 

from their neighbors, with whom they share herbivores. In contrast, 

an herbivore fauna consisting of mostly specialists may be much more difficult 

to evade. For 17 species of plants the degree of herbivore specialization 

varied from 36% to 90%. Such a variable degree of specialization 

among herbivores of individual host plant species was also found for a 

sample of 10 Panamanian tree species (Barone 1998). The proportion of



Table 16.2 Host Specificity of Insect Herbivores in Tropical Regions 

Santa 

Papua Rosa Cerrado 

New BCI Costa Distrito Cerrado 

STUDY SITE Australia Guinea Panama Rica Federal São Paulo 

Reference Basset Basset Barone Janzen Diniz Lopes 1995 

1992 1996 1998 1988 and Morais, 

this paper 

Insect groups Chewing Chewing Chewing Lepidoptera Lepidoptera Membracidae 

and sucking herbivores herbivores 

herbivores 

Methodology Field and Feeding Feeding Rearing Rearing Field 

literature trial trial to maturity to maturity observations 

observations 

No. insect 283 340 46 400 241 26 

species 

No. of host — 10/10 10/6 725 a 86/40 40/20 

plant species/ 

families 

Feeding on just 11% 54% 85% 90% 64% 31% 

one plant family 

a 725 = number of species of vascular plants in Santa Rosa. 

specialists in the fauna was not related to the number of species involved 

for either the cerrado or the Panamanian sample (Barone 1998). The next 

step should be to assess the amount of damage to a given host plant species 

caused by generalists versus specialists, as Barone (1998) has done. 

INTERSPECIFIC VARIATION IN LEAF DAMAGE 

AND LEAF QUALITY TRAITS 

One approach for understanding the traits that influence the interactions 

between plants and their herbivores is a comparative one (i.e., to measure 

a number of putative defensive traits on various plant species in the same 

habitat, and then determine if the variation in traits among plant species 

accounts for differences in levels of herbivore attack among those plant 

species). Classic work by Coley (1983) used this approach to explain variation 

in leaf chewing damage for a set of 42 tree species in Panama. Similar 

studies involved interspecific variation in leaf chewing damage in 

Mexico (Filip et al. 1995) and in Australia (Lowman and Box 1983), and 

attack by sucking insects in Indonesia (Hodkinson and Casson 1987). 

Results show that young leaves are often the most vulnerable to attack



Table 16.3 Host Specificity of Families of 114 

Lepidopteran Species Reared to Maturity 

No. of % of polyphagous 

Family species species 

Arctiidae 8 87.5 

Geometridae 12 50.0 

Hesperiidae 9 22.2 

Limacodidae 7 87.5 

Megalopygidae 6 100.0 

Mimallonidae 11 18.8 

Oecophoridae 8 75.0 

Pyralidae 29 17.2 

Riodinidae 8 87.5 

Saturniidae 9 44.0 

Tortricidae 7 28.6 

Note: A polyphagous species is one that feeds on a host species from 

more than one plant family. 

(Coley and Kursar 1996), and that depending on the study, leaf toughness, 

and nitrogen and phenolic content, can explain some of the observed 

interspecific variation. An added benefit of these types of studies is that 

they begin to indicate how rates of herbivore attack vary among locations, 

and which factors are responsible for such variation. 

Marquis et al. (2001) undertook a study at FAL to determine the plant 

traits that contribute to variation among plant species in damage by 

insects and pathogens. For the 25 species studied (10 trees, 10 shrubs, and 

5 herbs), they found that damage by herbivorous insects at the end of the 

leaf life (one year old at the end of the dry season) ranged between 0.5% 

and 14.3% per plant species (mean = 6.8%). Pathogen attack was much 

higher, ranging from 2.0% to 52.8% (mean = 17.3%). Damage by insects 

peaked at two points in the life of the leaves: during the leaf expansion 

period at the beginning of the wet season, and again sometime in the second 

half of the wet season. These two peaks in abundance coincided with 

known peaks in abundance of herbivorous insects in this system: the 

annual peak in abundance of leaf-chewing Coleoptera during the study 

year occurred during the leaf expansion period, while peaks in abundance 

of leaf-chewing Lepidoptera larvae in the years 1991–1993 occurred at 

the beginning of the dry season (Morais et al. 1999). The seasonal pattern 

of attack by pathogens was quite different, in that almost no damage 

occurred during the leaf expansion period, but fully expanded leaves continued 

to accrue damage throughout their lives. 

Various traits were measured to test their impact on interspecific



variation in insect and pathogen attack (leaf toughness, total phenolics, 

protein-binding capacity, nitrogen and water content, leaf pubescence, 

time to full leaf expansion, the presence of extrafloral nectaries and latex, 

plant size, and local species abundance). Of these, only protein-binding 

capacity was significantly negatively related to the amount of insect damage. 

Although cerrado leaves are very tough compared to those of more 

mesic sites (see table 16.4), and toughness was the most important predictor 

of interspecific variation in attack on Panamanian trees (Coley 

1983), it was not a significant predictor for this set of cerrado plants. Further, 

leaf pubescence was found to provide protection for flushing leaves 

in a deciduous forest of Ghana (Lieberman and Lieberman 1984), while 

there was no influence of leaf pubescence on herbivore attack at the Brazil 

site. In fact, at least two lepidopteran herbivores of Byrsonima crassa and 

B. verbascifolia (Malpighiaceae) use leaf hairs to build shelters (Diniz and 

Morais 1997; Andrade et al. 1999). 

Protein availability (nitrogen content/protein-binding capacity) and 

plant height were significantly positively correlated with pathogen attack. 

It may be that taller plant species are more susceptible to pathogen colonization 

because they are on average more likely to intercept windborne 

pathogen spores. Rate of leaf expansion was negatively correlated with 

pathogen attack. This result is consistent with the following scenario. 

Young leaves are first colonized and their tissues invaded during the leaf 

expansion period. The presence of the pathogen only becomes obvious 

after the fungi have further developed in the wet season. Plant species 

whose leaves rapidly pass through the vulnerable stage (period of leaf 

expansion) are less susceptible to attack. 

The amount of pathogen and insect attack were uncorrelated for any 

time period or leaf age group, suggesting that insect herbivores do not 

contribute to infection levels, either as carriers of spores or by creating 

sites for infection, both activities shown to occur in other systems (e.g., 

de Nooij 1988). The lack of a negative correlation between herbivore and 

pathogen attack also suggests that herbivores were not avoiding 

pathogen-damaged leaves. 

CROSS-SITE COMPARISONS IN INSECT DAMAGE 

AND LEAF QUALITY TRAITS 

As a point of comparison for cerrado sites with other tropical terrestrial 

ecosystems, we suggest a unimodal relationship between total rainfall and 

length of the dry season, on the one hand, and rates of leaf area loss to

Table 16.4 Comparison of Leaf Quality Factors for Woody Plants of Three Different Neotropical Sites 

Total Pubescence 

Percent Percent phenolics (bottom leaf Toughness Days to full 

Site N nitrogen water (mg/g) surface only) (g/mm 2 ) expansion 

BCI Pioneers 22 3.2 a /2.4 a 74 a /70 a 127.0 a /80.2 a 5.9 a /5.2 a 11.4 a /20.8 a 35.1 a 

BCI Persistents 24 3.3 a /2.2 b 76 a /62 b 173.4 a /95.6 a 1.4 a /0.5 a 41.6 b /33.8 b 45.1 b 

Chamela 16 3.3 a /2.7 a 80 a /71 a NA 10.1 a /7.6 a 8.4 a /12.9 a NA 

Brasília 20 1.7 b /1.5 c 60 b /52 c 289.8 b /261.4 b 159.1 c /149.1 c 91.2 c /130.8 c 37.6 a 

Note: First and second values are for young and mature leaves, respectively. N = number of species. Different superscript letters indicate significantly 

different means by ANOVA. BCI = Coley 1983, Chamela = Filip et al. 1995, Brasília = Marquis et al. 2001. NA = not available. 




insect folivores on the other. Available data from closed canopy habitats 

are consistent with a pattern of decreasing leaf area loss to insect herbivores 

with increasing rainfall, within the rainfall range of 750–5000 

mm/year (see fig. 16.3). Lower damage levels in wet evergreen forests are 

probably due to a combination of factors, including low host plant density, 

low seasonality in new leaf production, high natural enemy abundance, 

and the negative effects of high rainfall both on flight times of 

adults and feeding and survival of larvae. At sites with less rainfall, the 

physical effects of rainfall are ameliorated. Further, lower annual rainfall 

in the tropics inevitably means a greater seasonal distribution of that rain- 

Figure 16.3 Relationship between annual rainfall and folivory by insects 

across various Neotropical sites. Solid circles are for relatively high soil nutrient 

locations, and stars are for low soil nutrient sites. BC = Barro Colorado 

Island (persistent trees species only (Coley 1982); CH = Chamela, Mexico 

(Filip et al. 1995); SR = Santa Rosa, Costa Rica (Stanton 1984); PR = El Verde, 

Puerto Rico (Angulo-Sandoval and Aide 2000); LS = La Selva, Costa Rica 

(Hartshorn et al. in Marquis and Braker 1994); LT = Los Tuxtlas, Mexico (de 

la Cruz and Dirzo 1987); CO = Corumbatai, São Paulo, Brazil (Fowler and 

Duarte 1991); BR = Fazenda Agua Limpa, Brasília, Brazil (Marquis et al. 

2001). Fitted regression line: y = –0.0000018x 2 + 0.007x + 16.732; r 2 = 0.66.



fall. As a result, leaf production tends to be more synchronous not only 

within a plant species, but across many plant species, and timed to occur 

just before or at the onset of rains following the dry season. The availability 

of rains as a cue for herbivorous insects to emerge from diapause 

may allow for greater synchrony with sudden flushes of new leaves 

(Tauber et al. 1986). Even more so, this emerging herbivore fauna has the 

possibility of escaping control by their natural enemies, whose populations 

have also been reduced in activity during the dry season (Janzen 

1993). Together, lower negative impact of rainfall and natural enemies, 

and greater synchrony (and predictability) of new leaf production, may 

lead to higher leaf area loss in dry deciduous (typhoon) forests than in wet 

evergreen forests. 

Two estimates of end of the dry season leaf area loss by herbivores 

are available for cerrado (fig. 16.3; Fowler and Duarte 1991; Marquis et 

al. 2001). Both are approximately one-third that predicted by the relationship 

between rainfall and insect-caused leaf area loss described above. 

Differences could be due to site differences in seasonality, plant quality 

factors, or natural enemies. The cerrado sites are no more seasonal in their 

distribution of rainfall, as the length of dry season at both sites (Corumbatai 

= 6 mo, Brasília = 5–6 mo; Marquis et al. 2001) falls within the range 

found at Santa Rosa in Costa Rica (6 mo; Janzen 1993) and Chamela in 

Mexico (6–7 mo; Bullock and Solís-Magallanes 1990), where damage levels 

are much higher. Furthermore, damage levels are lower in cerrado 

compared to other seasonal deciduous forests, such as Santa Rosa and 

Chamela, despite the fact that leaves are maintained on the plant essentially 

year-round and thus are potentially exposed to attack for a longer 

time period. One other low-nutrient site in the neotropics also has correspondingly 

low leaf damage (Vasconcelos 1999). 

One important aspect in which cerrado plants are unique compared 

to other sites is their phenology of leaf production. Cerrado contains an 

unusually high number of species that flush leaves before the rains begin 

(fig. 16.2B), so that the most vulnerable stage of development (the new 

leaf stage: Coley and Kursar 1996) has already passed by the time the herbivorous 

insect activity begins (at least for Coleoptera, Homoptera, and 

Orthoptera: fig. 16.2A). When comparisons have been made for other systems 

between species that flush leaves before the start of the rainy season 

and after, those species producing leaves prior to rains suffer less damage 

(Aide 1993; Murali and Sukumar 1993). 

In addition to the effect of phenology of leaf flushing, initial data on 

plant quality factors show that leaves of cerrado plants at Brasília are 

tougher, and lower in nitrogen and water, than those at both Barro Col-



orado Island (BCI), Panama (Coley 1983) and Chamela (Filip et al. 1995; 

see table 16.4). Also, central Brazil plants are higher in phenolics than 

those on BCI (no comparative data are available from Chamela). Thus, 

differences in leaf quality may account for lower than expected (based on 

rainfall) levels of herbivory at the cerrado sites. 

Finally, there are few comparative data on natural enemy composition 

or attack levels among savanna or tropical forest sites. Available data 

suggest that parasitism levels are highest sometime after the onset of the 

rainy season in seasonal forests, and that rates of parasitism are quite 

high. At the central Brazil site, parasitism of leaf-feeding Lepidoptera 

peaks at about 35% during the onset of the rainy season (Morais et al. 

1999). A similar peak in caterpillar-feeding ichneumonids also occurs 

after the return of the rainy season in Kampala, Uganda (Owen and 

Chanter 1970). Olson (1994) found that attack by parasitoids on caterpillars 

of the saturniid moth Rothschildia lebeau was greatest (82%) during 

the middle rainy season, compared to early season (56%) values, at 

Santa Rosa in Costa Rica. In either case, rates of attack were very high. 

Rates of parasitism for the semideciduous forest of BCI, Panama are 15% 

to 25% (P. D. Coley and T. Kursar, pers. comm.). 

The third trophic level also has an impact in the form of predatory 

ants visiting extrafloral nectaries. Specifically, ant activity at EFNs has 

been shown to reduce herbivory in two cerrado plants, Qualea grandiflora 

and Caryocar brasiliense (Costa et al. 1992; Oliveira 1997). Percentage 

of species and plants with EFNs at eight different cerrado sites 

varied from 15% to 26% and 8% to 31%, respectively (Oliveira and 

Leitão-Filho 1987; Oliveira and Oliveira-Filho 1991). These values all fall 

within the range reported for various other tropical forested habitats (5% 

to 80%; Coley and Aide 1991), so it does not appear that an unusually 

high abundance of EFNs in cerrado can explain the low rate of herbivory 

found there. 

PATHOGEN ATTACK 

A reasonable expectation for the relationship between pathogen attack 

and rainfall across sites is that higher pathogen attack would be found at 

sites with higher rainfall. The high levels of pathogen attack observed at 

FAL (mean = 17.3%, range = 2%–52.8%) would appear to contradict this 

result. In fact, leaf area loss to pathogen attack at the wet forest site of 

Los Tuxtlas in Mexico was on average only about 1% (G. Garcia-Guzman 

and R. Dirzo, pers. comm.). At the semideciduous forest on Barro



Colorado Island, Panama, pathogen damage to young leaves was higher 

than that in Mexico and more in accord with levels in cerrado (mean = 

11.3%; range = 0 to 40.5%) (Barone 1998). 

CONCLUSIONS AND FUTURE RESEARCH DIRECTIONS 

In sharp contrast to neotropical savannas, the savannas of Africa, India, 

and Southeast Asia are (or until recently) dominated by ungulate herbivores. 

Studies in Africa have shown that these megaherbivores can have 

major impacts on plant species composition, particularly on the abundance 

and composition of woody biomass. This is not to say that vertebrates, 

both as browsers and seed predators, do not have impacts on 

vegetation structure in cerrado. Studies of the impacts of vertebrates as 

browsers, grazers, and seed predators in cerrado have not gone beyond 

diet studies and descriptions of community composition (Rodrigues and 

Monteiro-Filho 1999; chapter 14). Studies of seed predation by cerrado 

mammals and insects are still solely descriptive (Macedo et al. 1992; Lombardi 

and Toledo 1997; but see Scariot 1998). 

Unfortunately, comparative data for plant-herbivorous insect interactions 

in Old World savannas do not appear to be available (Andersen 

and Lonsdale 1990; Werner 1991). This leaves us to make comparisons 

with other tropical forested sites. Such comparisons suggest that cerrado 

ecosystems differ from forests in a number of important ways. First, lepidopteran 

larval abundance is greatest at the end of the wet season and 

beginning of the dry season, a pattern unlike that reported for any other 

tropical site. However, as for other sites in the Old World, abundance of 

Coleoptera coincides with the onset of the rainy season, diet breadth of 

Lepidoptera is larger than once expected, and abundance of individual 

caterpillar species and caterpillars in general is low. Second, damage levels 

by herbivorous insects are much lower than expected based on our tentative 

model relating rainfall and annual leaf area loss to insect herbivores. 

Damage levels fall within the range found at low nutrient sites; however, 

both phenological escape (many plants produce new leaves before the 

increase in abundance of herbivorous insects) and low-quality tissues may 

account for relatively low damage levels in cerrado. Third, pathogen damage 

levels, at least at one cerrado site, are much higher than predicted a 

priori, given the level of rainfall that occurs there. How these data on 

pathogen attack conform with what happens at other locations, both in 

cerrado and out, awaits data collection from other sites. 

We expect that herbivory has an important impact in cerrado, not



only through its influence per se on plant growth and reproduction, but 

also through its potential interacting effect with abiotic factors. Although 

abiotic factors have great influence in cerrado, studies from other systems 

suggest that the impact of herbivory often modifies the influence of soil 

nutrient quality and water availability (Franco 1998) and interacts with 

fire frequency to determine plant population dynamics. Cerrado vegetation 

provides an excellent opportunity for future experimental work to 

determine the main and interactive effects of fire and herbivory on individual 

plant fitness, plant population dynamics, and plant succession. 

Previous work in seasonally deciduous forest in Panama has shown 

that leafing phenology at the population level is an important determinant 

of early attack by chewing insects (Aide 1992, 1993). These results, 

together with our observation that leaf expansion rate influences attack 

by pathogens, suggest that a very profitable study would be an investigation 

of the potential role of intraspecific variation in leafing phenology on 

early attack by both chewing insects and pathogens in cerrado. This information 

then could be linked to correlated changes in plant physiology that 

allow plants to anticipate the coming of the rains, producing new leaves 

before the wet season begins. Such links between plant physiology and 

herbivore attack remain virtually unexplored in tropical habitats. The situation 

is further complicated in that the timing of the wet season is 

extremely variable from year to year. The consequences of such variability 

on plant phenology, plant reserves that allow leafing, and herbivorous 

insect population dynamics are unknown. 

Studies of factors that control herbivory levels in tropical systems are 

beginning to examine influences beyond plant traits alone, including the 

potential role of the third trophic level. Initial data from central Brazil suggest 

that the third trophic level, in the form of parasitoids, may have a 

strong influence on the timing of lepidopteran life histories. The degree to 

which natural enemies maintain low population levels of caterpillars 

should be explored. The relative importance of top-down vs. bottom-up 

forces in influencing plant-insect herbivore interactions in this system is 

unknown. 

Our focus thus far in the study of plant-insect herbivore interactions 

has been on cerrado vegetation itself. However, the cerrado landscape is 

one of savanna intermixed with gallery forests associated with streams 

and rivers. Because herbivores and their natural enemies may move from 

cerrado to gallery forests and back (e.g., Camargo and Becker 1999), the 

interactions that we have studied may depend on the surrounding landscape 

context. Gallery forests may provide shelter and alternative food 

sources for adults (both of herbivores and their natural enemies), and



alternative food plants for larvae of herbivores. Distance to gallery forests 

would seem to be an important factor to consider in such a study. 

In summary, there are a plethora of unanswered questions, many very 

basic to our understanding of cerrado plant-herbivore interactions. More 

detailed studies of basic natural history, including identification of the 

herbivores and their diet breadths, are essential. But sufficient information 

is now available to suggest feasible experiments that would reveal the 

impact of the herbivores on plant community structure and plant population 

dynamics in the context of the important abiotic factors. Similarly, 

manipulation of plants and the third trophic level can begin to reveal the 

factors that structure herbivore communities and drive insect population 

dynamics in cerrado ecosystems. 


We thank the many students of the project “Herbivores and Herbivory in 

Cerrado,’’ who assisted in the collection and rearing of the caterpillars. 

This study was supported by FAPDF, FINATEC, CNPq, which also provided 

support for scientific initiation (PIBI-CNPq-UnB). We thank Katerina 

Aldás, Karina Boege, Phyllis Coley, Rebecca Forkner, Nels Holmberg, 

Damond Kyllo, John Lill, Eric Olson, and an anonymous reviewer for 

valuable comments on earlier versions. 

REFERENCES 

Aide, T. M. 1992. Dry season leaf production and escape from herbivory. 


Aide, T. M. 1993. Patterns of leaf development and herbivory in a tropical 

understory community. Ecology 74:455–466. 

Andersen, A. N. and W. M. Lonsdale. 1990. Herbivory by insects in Australian 

tropical savannahs: A review. J. Biogeogr. 17:433–444. 

Andrade, I., I. R. Diniz, and H. C. Morais 1995. A lagarta de Cerconota 

achatina (Oecophoridae: Stenomatinae): Biologia e ocorrência em plantas 

hospedeiras do gênero Byrsonima (Malpighiaceae). Rev. Bras. Zool. 

12:735–741. 

Andrade, I., H. C. Morais, I. R. Diniz, and C. van den Berg. 1999. Richness 

and abundance of caterpillars on Byrsonima (Malpighiaceae) species in 

an area of cerrado vegetation in Central Brazil. Rev. Biol. Trop. 

47:691–695. 

Angulo-Sandoval, P. and T. M. Aide. 2000. Leaf phenology and leaf damage



of saplings in the Luquillo Experimental Forest, Puerto Rico. Biotropica 

32:425–422. 

Barlow, H. S. and I. P. Woiwod. 1990. Seasonality and diversity of Macrolepidoptera 

in two lowland sites in the Dumoga-Bone National Park, 

Sulawesi Utara. In W. J. Knight and J. D. Holloway, eds., Insects of the 

Rain Forests of South East Asia (Wallacea), pp. 167–172. London: Royal 

Entomological Society of London. 

Barone, J. A. 1998. Host-specificity of folivorous insects in a moist tropical 

forest. J. Anim. Ecol. 67:400–409. 

Basset, Y. 1992. Host specificity of arboreal and free-living insect herbivores 

in rain forests. Biol. J. Linn. Soc. 47:115–133. 

Basset, Y. 1996. Local communities of arboreal herbivores in Papua New 

Guinea: Predictors of insect variables. Ecology 77:1906–1919. 

Basset, Y. 1999. Diversity and abundance of insect herbivores foraging on 

seedlings in a rainforest in Guyana. Ecol. Entomol. 24:245–259. 

Boinski, S. and N. L. Fowler. 1989. Seasonal patterns in a tropical lowland 

forest. Biotropica 21:223–233. 

Bullock, S. H. and J. A. Solís-Magallanes. 1990. Phenology of canopy trees 

of a tropical deciduous forest in Mexico. Biotropica 22:22–35. 

Camargo, A. J. A. and V. O. Becker. 1999. Saturniidae (Lepidoptera) from the 

Brazilian cerrado: Composition and biogeographic relationships. Biotropica 

31:696–705. 

Clark, W. E. and R. P. Martins. 1987. Anthonomus biplagiatus Redtenbacher 

(Coleoptera: Curculionidae): A Brazilian weevil associated with 

Kielmeyera (Guttiferae). Coleopt. Bull. 41:157–164. 

Coley, P. D. 1982. Rates of herbivory on different tropical trees. In E. G. 

Leigh, Jr., A. S. Rand, and D. M. Windsor, eds., The Ecology of a Tropical 

Forest: Seasonal Rhythms and Long-Term Changes, pp. 123–132. 

Washington, D.C.: Smithsonian Institution Press. 

Coley, P. D. 1983. Herbivory and defensive characteristics of tree species in a 

lowland tropical forest. Ecol. Monogr. 53:209–233. 

Coley, P. D. and T. M. Aide. 1991. Comparison of herbivory and plant 

defenses in temperate and tropical broad-leaved forests. In P. W. Price, 

T. M. Lewinsohn, G. W. Fernandes, and W. W. Benson, eds., Plant-Animal 

Interactions: Evolutionary Ecology in Tropical and Temperate 

Regions, pp. 25–50. New York: John Wiley. 

Coley, P. D. and J. A. Barone. 1996. Herbivory and plant defenses in tropical 

forests. Ann. Rev. Ecol. Syst. 27:305–335. 

Coley, P. D. and T. A. Kursar. 1996. Anti-herbivore defenses of young tropical 

leaves: Physiological constraints and ecological tradeoffs. In S. S. 

Mulkey, R. L. Chazdon, and A. P. Smith, eds., Tropical Forest Plant Ecophysiology, 

pp. 305–336. New York: Chapman and Hall. 

Costa, F. M. C. B., A. T. Oliveira-Filho, and P. S. Oliveira. 1992. The role of 

extrafloral nectaries in Qualea grandiflora (Vochysiaceae) in limiting



herbivory: An experiment of ant protection in cerrado vegetation. Ecol. 

Entomol. 17:363–365. 

Cytrynowicz, M. 1991. Resource size and predictability, and local herbivore 

richness in a subtropical Brazilian cerrado community. In P. W. Price, T. 

M. Lewinsohn, G. W. Fernandes, and W. W. Benson, eds., Plant-Animal 

Interactions: Evolutionary Ecology in Tropical and Temperate Regions, 

pp. 561–590. New York: John Wiley. 

De la Cruz, M. and R. Dirzo. 1987. A survey of the standing levels of herbivory 

in seedlings from a Mexican rain forest. Biotropica 19:98–106. 

Del Claro, K., R. Marullo, and L. A. Mound. 1997. A new Brazilian species of 

Heteropthrips (Insecta: Thysanoptera) co-existing with ants in the flower 

of Peixotoa tomentosa (Malpighiaceae). J. Nat. Hist. 31:1307–1312. 

De Nooij, M. P. 1988. The role of weevils in the infection process of the fungus 

Phomopsis subordinaria in Plantago lanceolata. Oikos 52:51–58. 

Diniz, I. R. and H. C. Morais. 1995. Larvas de Lepidoptera e suas plantas 

hospedeiras em um cerrado de Brasília, DF. Rev. Bras. Ent. 39:755–770. 

Diniz, I. R. and H. C. Morais. 1997. Lepidopteran caterpillar fauna on cerrado 

host plants. Biodiv. and Conserv. 6:817–836. 

Diniz, I. R. and H. C. Morais. Submitted. Lepidopteran larvae associated with 

reproductive structures of cerrado plants. An. Soc. Entomol. Brasil. 

Diniz, I. R. and H. C. Morais. In prep. Local diet breadth of lepidopteran larvae 

in cerrado vegetation. 

Diniz, I. R., H. C. Morais, S. Scherrer, and E. O. Emery. 2000. The 

polyphagous caterpillar Fregela semiluna (Lepidoptera: Arctiidae): 

Occurrence on plants in the Central Brazilian cerrado. Bol. Herb. 

Ezechias Paulo Heringer 5:103–112. 

Elton, C. S. 1973. The structure of invertebrate populations inside neotropical 

rain forest. J. Anim. Ecol. 42:55–104. 

Fernandes, G. W. and P. W. Price. 1991. Comparison of tropical and temperate 

galling species richness: The roles of environmental harshness and 

plant nutrient status. In P. W. Price, T. M. Lewinsohn, G. W. Fernandes, 

and W. W. Benson, eds., Plant-Animal Interactions: Evolutionary Ecology 

in Tropical and Temperate Regions, pp. 91–115. New York: John 

Wiley. 

Filip, V., R. Dirzo, J. M. Maass, and J. Sarukhan. 1995. Within- and amongyear 

variation in the levels of herbivory on the foliage of trees from a 

Mexican tropical deciduous forest. Biotropica 27:78–86. 

Fowler, H. G. and L. C. Duarte. 1991. Herbivore pressure in a Brazilian cerrado. 

Naturalia 16:99–102. 

Franco, A. C. 1998. Seasonal patterns of gas exchange, water relations and 

growth of Roupala montana, an evergreen savanna species. Plant Ecol. 

136:69–76. 

Frith, C. B. and D. W. Frith. 1985. Seasonality of insect abundance in an Australian 

upland tropical rainforest. Austr. J. Ecol. 10:237–248.



Hodkinson, I. D. and D. S. Casson. 1987. A survey of food-plant utilization 

by Hemiptera (Insecta) in the understorey of primary lowland rain forest 

in Salawesi, Indonesia. J. Trop. Ecol. 3:75–85. 

Janzen, D. H. 1988. Ecological characterization of a Costa Rican dry forest 

caterpillar fauna. Biotropica 20:120–135. 

Janzen, D. H. 1993. Caterpillar seasonality in a Costa Rican dry forest. In 

N. E. Stamp and T. M. Casey, eds., Caterpillars: Ecological and Evolutionary 

Constraints on Foraging, pp. 448–477. New York: Chapman and 

Hall. 

Lewinsohn, T. M., G. M. Fernandes, W. W. Benson, and P. W. Price. 1991. 

Introduction: Historical roots and current issues in tropical evolutionary 

ecology. In P. W. Price, T. M. Lewinsohn, G. W. Fernandes, and W. W. 

Benson, eds., Plant-Animal Interactions: Evolutionary Ecology in Tropical 

and Temperate Regions, pp. 1–22. New York: John Wiley. 

Lieberman, D. and M. Lieberman. 1984. The causes and consequences of synchronous 

flushing in a dry tropical forest. Biotropica 16:193–201. 

Lombardi, J. A. and F. R. N. Toledo. 1997. Davilla grandiflora St.Hill. et Tul. 

(Dilleniaceae) as a host of larvae of Mandarellus sp. (Coleoptera: Curculionidae) 

in Minas Gerais, Brazil. Ciênc. Cult. 49:211–212. 

Lopes, B. C. 1995. Treehoppers (Homoptera: Membracidae) in southeastern 

Brazil: Use of host plants. Rev. Bras. Zool. 12:595–608. 

Lowman, M. D. and J. D. Box. 1983. Variation in leaf toughness and phenolic 

content among five Australian rainforest trees. Austr. J. Ecol. 

8:17–25. 

Macedo, M. V., T. M. Lewinsohn, and J. M. Kingsolver. 1992. New host 

records of some bruchid species in Brazil with the description of a new 

species of Caryedes (Coleoptera: Bruchidae). Coleopt. Bull. 46:330–336. 

Marquis, R. J. 1991. Herbivore fauna of Piper (Piperaceae) in a Costa Rican 

wet forest: Diversity, specificity, and impact. In P. W. Price, T. M. Lewinsohn, 

G. W. Fernandes, and W. W. Benson, eds., Plant-Animal Interactions: 

Evolutionary Ecology in Tropical and Temperate Regions, pp. 

179–208. New York: John Wiley. 

Marquis, R. J. and H. E. Braker. 1994. Plant-herbivore interactions: Diversity, 

specificity, and impact. L. A. McDade, K. S. Bawa, H. A. Hespenheide, 

and G. S. Hartshorn, eds., La Selva: Ecology and Natural History 

of a Neotropical Rain Forest, pp. 261–281. Chicago: University of 


Marquis, R. J., I. R. Diniz, and H. C. Morais. 2001. Patterns and correlates 

of interspecific variation in foliar insect herbivory and pathogen attack 

in Brazilian cerrado. J. Trop. Ecol. 17:1–23. 

Morais, H. C. and I. R. Diniz. 2001. Immature stage and host plant of the 

Brazilian cerrado moth Aucula munroei (Lepidoptera: Noctuidae: 

Agarastinae). Trop. Lepidop. (in press). 

Morais, H. C., I. R. Diniz, and L. C. Baumgarten. 1995. Padrões de produção



de folhas e sua utilização por larvas de Lepidoptera em um cerrado de 

Brasília, DF. Rev. Bras. Bot. 13:351–356. 

Morais, H. C., I. R. Diniz, B. C. Cabral, and J. Mangabeira. Submitted. 

Stenoma cathosiota (Lepidoptera: Elachistidae) in the cerrado: Temporal 

and spatial variation of caterpillars abundance. Ciênc. Cult. 

Morais, H. C., I. R. Diniz, and J. M. F. Maia. In prep. Richness and abundance 

of lepidopteran caterpillars associated with cerrado plants. 

Morais, H. C., I. R. Diniz, and J. R. Silva. 1996. Larvas de Siderone marthesia 

nemesis (Lepidoptera: Nymphalidae, Charaxinae) em um cerrado de 

Brasília. Rev. Bras. Zool. 13:351–356. 

Morais, H. C., I. R. Diniz, and D. M. S. Silva. 1999. Caterpillar seasonality 

in a central Brazilian cerrado. Rev. Biol. Trop. 47:1025–1033. 

Murali, K. S. and R. Sukumar. 1993. Leaf flushing phenology and herbivory in 

a tropical dry deciduous forest, southern India. Oecologia 94:114–119. 

Oliveira, P. E. 1998. Fenologia e biologia reprodutiva das espécies de cerrado. 

In S. M. Sano and S. P. Almeida, eds., Cerrado: Ambiente e Flora, pp. 


Oliveira, P. S. 1997. The ecological function of extrafloral nectaries: Herbivore 

deterrence by visiting ants and reproductive output in Caryocar 

brasiliense (Caryocaraceae). Funct. Ecol. 11:323–330. 

Oliveira, P. S. and H. F. Leitão-Filho. 1987. Extrafloral nectaries: Their taxonomic 

distribution and abundance in the woody flora of cerrado vegetation 

in southeast Brazil. Biotropica 19:140–148. 

Oliveira, P. S. and A. T. Oliveira-Filho. 1991. Distribution of extrafloral nectaries 

in the woody flora of tropical communities in Western Brazil. In 

P. W. Price, T. M. Lewinsohn, G. W. Fernandes, and W. W. Benson, eds., 

Plant-Animal Interactions: Evolutionary Ecology in Tropical and Temperate 

Regions, pp. 163–175. New York: John Wiley. 

Olson, E. J. 1994. “Dietary Ecology of a Tropical Moth Caterpillar, Rothschildia 

lebeau (Lepidoptera: Saturniidae).’’ Ph.D. thesis, University of 

Pennsylvania, Philadelphia, Pennsylvania, U.S. 

Owen, D. F. and D. O. Chanter. 1970. Species diversity and seasonal abundance 

in tropical Ichneumonidae. Oikos 21:142–144. 

Pinheiro, F., I. R. Diniz, D. Coelho, and M. P. S. Bandeira. Submitted. Temporal 

distribution abundance in the Brazilian cerrado: Effect of climatic 

variation. Austral Ecol. 

Pinheiro, F., H. C. Morais, and I. R. Diniz. 1997. Composição de herbívoros 

em plantas hospedeiras com látex: Lepidoptera em Kielmeyera spp. (Guttiferae). 

In L. L. Leite and C. H. Saito, eds,Contribuição ao Conhecimento 

Ecológico do Cerrado, pp. 101–106. Brasília: ECL/Universidade 

de Brasília. 

Price, P. W., I. R. Diniz, H. C. Morais, and E. S. A. Marques. 1995. The abundance 

of insect herbivore species in the tropics: High local richness of rare 

species. Biotropica 27:468–478.



Rodrigues, F. H. G. and E. Monteiro-Filho. 1999. Feeding behaviour of the 

pampas deer: A grazer or a browser? Deer Specialist Group News 

15:12–13. 

Scariot, A. 1998. Seed dispersal and predation of the palm Acrocomia 

aculeata. Principes 42:5–8. 

Stanton, N. 1975. Herbivore pressure in two types of tropical forests. 


Strauss, S. Y. 1997. Floral characters link herbivores, pollinators, and plant 

fitness. Ecology 78:1640–1645. 

Strong, D. R. Jr., J. H. Lawton, and T. R. E. Southwood. 1984. Insects 

on Plants: Community Patterns and Mechanisms. Oxford: Blackwell 

Scientific. 

Tauber, M. J., C. A. Tauber, and S. Masaki (eds). 1986. Seasonal Adaptation 

of Insects. Cambridge, UK: Oxford University Press. 

Vasconcelos, H. L. 1999. Levels of leaf herbivory in Amazonian trees from 

different stages in forest regeneration. Acta Amazonica 29:615–623. 

Werner, P. A. (ed.). 1991. Savanna Ecology and Management. Oxford, UK: 

Blackwell Scientific. 

Wolda, H. 1978. Fluctuations in abundance of tropical insects. Amer. Nat. 

112:1017–1045. 

Wong, M. 1984. Understory foliage arthropods in the virgin and regenerating 

habitats of Pasoh Forest Reserve, West Malaysia. Malaysian Forester 

47:43–69. 

Wood, T. K. and K. L. Olmstead. 1984. Latitudinal effects on treehopper species 

richness (Homoptera, Membracidae). Ecol. Entomol. 9:109–115.


17 

Pollination and Reproductive 

Biology in Cerrado 

Plant Communities 

Paulo E. Oliveira and Peter E. Gibbs 

Studies on the pollination biology and breeding systems 

of extensive samples of species in plant communities are valuable since 

they allow various conceptual issues to be addressed. With such data, 

comparisons can be made between different communities (e.g., moist versus 

seasonal forest), or between different components of the vertical strata 

(e.g., canopy versus understory taxa). In this way we can determine the 

role of diverse pollinators in different kinds of woodland, or analyze the 

frequency of obligate outbreeding versus self-compatible taxa in different 

communities or subunits of the same community. And with estimations 

of actual mating systems for particular species, usually using molecular 

genetic markers, we can attempt to determine whether self-incompatibility 

mechanisms do indeed result in outbreeding, or whether, due to small 

effective population size, consanguineous mating prevails. In this review 

we will focus on comparisons between community studies for the Brazilian 

cerrados and other neotropical woodlands. 

Most community surveys of the reproductive biology of neotropical 

woodlands have been undertaken in Central America: Costa Rica, 

Panama (Barro Colorado Island) and Mexico. Preeminent among such 

studies are those for moist forest areas at La Selva (Costa Rica) which 

include flowering phenology, pollination biology, and breeding systems 

(Kress and Beach 1994). Other studies in Costa Rica include those at a 

variety of habitats ranging from moist to seasonal woodlands at localities 

near Cañas (Guanacaste province) (see Bawa 1990 for a review). These 

329



and other studies on Barro Colorado Island (Panama) and seasonal forests 

in Mexico were reviewed by Bullock (1995). 

In South America, studies are available for Venezuelan secondary 

deciduous forest (Ruiz and Arroyo 1978), montane cloud forest in the 

Cordillera de la Costa at 1,700 m (Sobrevilla and Arroyo 1982), and 

savanna areas (e.g., Ramírez and Brito 1990). For Brazil, no community 

reproductive biology survey is available for Amazonia, although a number 

of individual species have been studied (e.g., Gribel et al. 1999). Some community 

studies of hummingbird and bat pollination systems have been published 

recently for the Atlantic coastal forest (Sazima et al. 1996, 1999). 

The cerrado has a very rich flora, comprising more than 800 species 

of trees with perhaps three times that number of herbs and hemixyle, subshrubby 

species (Furley and Ratter 1988; Mendonça et al. 1998; see also 

chapters 6, 7). Moreover, the cerrados interdigitate with gallery forest and 

have extensive boundaries with mesophytic and moist forest, with its 

woody flora linked to neighboring tropical rainforest groups (Rizzini 

1963; Sarmiento 1983; Prance 1992; Oliveira-Filho and Ratter 1995). A 

basic question, therefore, is whether the ecological differences between 

savanna and forest areas in the cerrado region involve different pollination 

and breeding systems. 

REPRODUCTIVE BIOLOGY STUDIES IN CERRADO 

Most reproductive biology studies of cerrado vegetation have been undertaken 

in the state of São Paulo, in the various “islands’’ of cerrado sensu 

lato within the SE fringe of the main cerrado area (chapter 6). Following 

in the tradition of Warming (1908) and his pioneer study of a cerrado 

community at Lagoa Santa, Mantovani and Martins (1988) studied the 

flowering and fruiting phenology of an area of cerrado sensu stricto at 

Mogi Guaçu by means of monthly censuses. Silberbauer-Gottsberger and 

Gottsberger (1988) studied the pollination biology of 279 woody and 

herbaceous species in an area of cerrado sensu stricto at Botucatu. These 

authors defined in general terms the floral visitors and aspects of pollination 

syndromes for the community, but without providing a list of species 

and their visitors. Saraiva et al. (1996) investigated the breeding systems 

of selected species in an area of cerrado sensu stricto at Corumbataí. They 

defined the sexual systems for 135 species and studied the breeding systems 

of a subsample of 21 species using controlled hand pollination. 

In a cerrado sensu stricto area of 40 ha near Brasília, central Brazil, 

Oliveira and Gibbs (2000) studied the phenology, pollination biology and


Pollination and Reproductive Biology 331 

sexual systems of 59 woody species, with a subsample of 30 species studied 

for breeding system by means of hand pollination and observations 

on postpollination events in the pistil. In a similar survey, Barbosa (1997) 

investigated these reproductive parameters for 204 herbaceous and 

shrubby species in an area of ca. 3 ha of cerrado (campo sujo) near Uberlândia 

(state of Minas Gerais). Breeding systems for a subsample of 84 

species were undertaken using hand pollinations, but details of the site of 

incompatibility were not studied. 

These community surveys, together with other studies on reproductive 

biology of individual cerrado species (see Oliveira and Gibbs 2000 for a 

review), groups of species (e.g., Gottsberger 1986; Oliveira 1998b) or other 

reproductive aspects (e.g., dioecy, Oliveira 1996), provide a data base for 

pollination biology which includes around 113 species of the woody flora 

(estimated 800 species), around 200 of the understory flora (another 2,500 

species at least), and breeding systems for some 70 woody species and 200 

shrubs and herbs. Although the number of studied species in relation to the 

total flora is fairly modest, it is necessary to bear in mind that the cerrado 

flora as a whole is very heterogeneous. Ratter et al. (1996) encountered 534 

woody and large shrub species in a survey of 98 sites throughout the cerrado 

region, of which 158 species (30%) occurred at only one site. Most 

studied sites will show less than 100 woody species. If we consider the 94 

species which were found at 20 sites or more, and which Ratter et al. (1996) 

considered “a working list of the commonest tree species of the Cerrado,’’ 

then our data for breeding systems represents 52% of these “core’’ cerrado 

species, while our data for pollination biology would include most of these 

species (71 spp., or 76%). Furthermore, data accumulated so far for sexual 

systems comprises 327 cerrado woody species (Oliveira 1996). Therefore, 

generalizations based on the data currently available are likely to give a reasonable 

picture of the reproductive biology of woody cerrado species. Certainly 

the cerrado data base in this respect is comparable with or better than 

that available for humid forest sites. 

POLLINATION BIOLOGY 

Studies of the reproductive biology of cerrado plants have shown a great 

diversity of pollination systems similar to those encountered in other 

neotropical forests (see fig. 17.1). In the cerrado sensu lato, bees are the 

main pollinators, as noted by Silberbauer-Gottsberger and Gottsberger 

(1988), but other pollinators have a role both for the woody and herbaceous 

cerrado flora (see table 17.1). They include pollinators characteristic


Figure 17.1 Diversity of pollination systems in cerrado vegetation: (A) Small 

bee (Paratrigona sp.) visiting a flower of Casearia grandiflora, a species pollinated 

mainly by flies. (B) Dynastidae beetles in a flower of Annona coriacea. 

(C) Hummingbird (Amazilia sp.) visiting flowers of Palicourea rigida (D) Noctuidae 

moth visiting Aspidosperma macrocarpum flowers. (E) Large-bee (Centris 

violascens) visiting a flower of Eriotheca pubescens. (F) Nectarivorous bat 

(Glossophaga soricina) visiting Hymenaea stigonocarpa. Sources: Original 

photographs by P. E. Oliveira, and (F) from Gibbs et al. 1999.



Table 17.1 Pollination Systems in Cerrado Woody Species 

and Other Tropical Plant Formations 

Cerrado sensu 

Gallery forest, stricto, La Selva rainforest, 

Uberlândia (MG), Brasília (DF) Costa Rica 

P. E. Oliveira (Oliveira and (Kress and 

(unpublished) Gibbs 2000) Beach 1994) 

% spp. % spp. % spp. 

System (no.) System (no.) System (no.) 

Wind 1 (1) Wind 0 Wind 2.5 (7) 

Very small 7 (7) Very small insects 5 (3) Small diverse insects 11.2 (31) 

insects 

Small 45 (46) Small insects 44 (26) Flies 1.8 (5) 

insects 

(small bees/ (small bees/flies/ Wasps 2.5 (7) 

flies/wasps) wasps) 

Small bees 14.1 (39) 

Large bees 23 (23) Large bees 32 (19) Large bees 24.3 (67) 

Butterflies 2 (2) Butterflies 0 Butterflies 4.3 (12) 

Moths 12 (12) Moths 12 (7) Moths 8 (22) 

Nonflying 1 (1) Nonflying 0 Nonflying mammals 0 

mammals mammals 

Bats 4 (4) Bats 3 (2) Bats 3.6 (10) 

Humming- 0 Hummingbirds 2 (1) Hummingbirds 14.9 (41) 

birds 

Beetles 6 (6) Beetles 2 (1) Beetles 12.7 (35) 

Note: Percentage of species in each pollination system for a gallery forest in the cerrado region, a cerrado 

sensu stricto area in Brasília, central Brazil, and at La Selva rainforest reserve in Costa Rica. 

Woody species only were included in the first two areas, whereas La Selva data included the complete 

flora. Small insects in the Brazilian data included fly, wasp, and small bee pollinated species. Very small 

insects or small diverse insects included small, unidentified insects, usually smaller than 5 mm. 

of moist tropical forest, such as beetles and bats. As in other tropical communities 

(Bawa 1990), plant-pollinator relationships in the cerrado seem 

to involve guilds of pollinators associated with a given plant or group of 

plants (Oliveira and Gibbs 2000). There are few one-to-one plant-pollinator 

relationships, although many taxa are still poorly studied. Some cerrado 

plants are dependent on a restricted group of specialized vectors, as 

Byrsonima spp. (Malpighiaceae) with Centridinae bees (Anthophoridae), 

and Annona spp. with Dynastidae beetles (Gottsberger 1986). But most 

species rely on a broader spectrum of pollinators defined more by their 

size and foraging requirements than by specific interaction. Many species 

have small, apparently generalist flowers pollinated by a range of insects



of different groups. Such species may be visited by flies, bees, and wasps, 

and although in some cases the main pollinators are evident, such as 

wasps in Erythroxylum (Barros 1998), or flies in Casearia sylvestris (Barbosa 

1997), in most cases the main pollinators can be defined only on a 

local basis and in quantitative terms. In such cases, it may be better to 

group these less specialized systems as being pollinated by small generalist 

insects (as in Oliveira and Gibbs 2000). 

The diversity of flower-visiting bees observed in different cerrado 

sensu lato studies, some 114–196 species, is similar to that at other tropical 

sites studied so far (Carvalho and Bego 1996). Small to medium-sized 

Apidae bees are the most common flower visitors in different cerrado 

areas (Carvalho and Bego 1996; Oliveira and Gibbs 2000). Eusocial bees 

such as the exotic Apis mellifera scutellata and the almost omnipresent 

Trigona spinipes are very common in the region. Although certainly pollinators 

of many plants, these bee species have also been noted for their 

destructive activity and pollen theft in buds and flowers (Roubik 1989). 

Other Meliponinae, Anthophoridae and Halictidae are small to mediumsized 

bees well represented in cerrado. Halictid bees have been observed 

as common pollinators of understory trees in gallery forests (personal 

observation), and their diversity seems to be higher in forests than in open 

habitats. 

The most conspicuous and diverse group in the cerrado are the large 

anthophorid bee genera, such as Centris and Xylocopa (Carvalho and 

Bego 1996; Oliveira and Gibbs 2000). Some large bee genera of other 

families, such as Eulaema, Bombus, Melipona (Apideae), and Oxaea 

(Oxaeidae), are also common pollinators of cerrado plants (Carvalho and 

Bego 1996; Oliveira and Gibbs 2000), and a guild of large-bee flowers 

may be delimited for the cerrado areas. Many large-bee genera observed 

in cerrado are also cited for wet and dry forests in Costa Rica (Frankie et 

al., 1983; Bawa, Bullock, Perry et al. 1985; Kress and Beach 1994). 

Other pollination systems common in tropical areas are also present 

in cerrado. Beetles are pollinators of cerrado species of Annonaceae 

(Gottsberger 1986, 1989). Hawkmoths are pollinators of some important 

species including Qualea grandiflora (Silberbauer-Gottsberger and Gottsberger 

1975), probably the most widespread cerrado tree species (Ratter 

et al. 1996). Bat pollination is present in some tree species (Sazima and 

Sazima 1975; Gribel and Hay 1993; Gibbs et al. 1999) and also in small 

shrub species of Bauhinia (Barbosa 1997). 

Some pollinators are absent or poorly represented in the cerrado 

sensu lato. Butterflies seem to be important pollinators of herbaceous 

Asteraceae (Silberbauer-Gottsberger and Gottsberger 1988), but in the



study by Oliveira and Gibbs (2000) no woody species was effectively pollinated 

by these vectors. At the same site, hummingbirds were pollinators 

of only one species, although they were common opportunistic visitors of 

more than 30% of the surveyed woody species. However, hummingbirds 

were observed as true pollinators of some shrubs and herbs in open cerrado, 

campo sujo, areas (Barbosa 1997; chapter 6). It is notable that wind 

pollination, which has been associated with seasonally dry areas (Bullock 

1994), is rare among cerrado trees and shrubs, and occurs commonly only 

in the grasses. Vertical stratification of pollination systems observed in 

tropical forests (Bawa, Bullock, Perry et al. 1985; Kress and Beach 1994) 

occurs also in cerrado. Moths and bats are pollinators mostly of trees, 

while wind and hummingbird pollination appear mostly in the herbaceous 

layer. Dominance of animal pollination and virtual absence of wind 

pollination may reflect the rainforest origin of the cerrado sensu lato 

woody elements, whereas the herbaceous layer includes cosmopolitan 

groups of different origins (Rizzini 1963; Sarmiento 1983). 

BREEDING SYSTEMS 

Studies on the breeding systems of Central American forest communities 

indicate that most tree species are obligatory outbreeders (Bawa 1974; 

Bawa, Perry, and Beach 1985; Bullock 1985, 1994). Studying at La Selva, 

Kress and Beach (1994) estimated that 88% of the upper stratum species 

were obligatory outbreeders, and this contrasted with predominant 

(66%) self-compatibility in species of the understory stratum. The Brazilian 

cerrados conform to the Central American pattern in that most 

woody species have obligatory outbreeding mechanisms (see tables 17.2, 

17.3; Oliveira and Gibbs 2000), while self-compatibility is much more 

common in the herbaceous and hemixyle taxa of understory (Saraiva et 

al. 1996) or open cerrado (campo sujo; Barbosa 1997) communities (see 

fig. 17.2). 

Outbreeding cerrado woody species mostly have hermaphrodite 

flowers and self-incompatibility, while dioecy has an incidence of only 

10–15% in this community (Oliveira 1996). This contrasts with the 24% 

dioecious species in the canopy of Central American woodlands (Kress 

and Beach 1994). In fact, in typical cerrado sensu stricto, the incidence of 

dioecy is around half of the 20% or so reported for seasonally dry forests 

by Bullock (1995), and the rather higher frequency of dioecy in dense cerrado 

woodlands (cerradão) seems to be due to the occurrence of some 

moist forest species in these areas (Oliveira 1996). As suggested by Bawa



Table 17.2 Sexual and Breeding Systems of Woody Species 

in Cerrado and in Other Tropical Communities 

Cerrado Secondary Cloud 

sensu Dry forest, forest, forest, Dry forest, Rainforest, 

stricto Costa Rica Venezuela Venezuela Mexico Costa Rica 

Dioecy 15 22 24 31 20 23 

Monoecy 5 10 0 3 10 11 

SI 66 54 64 26 53 53 

Inbreeding 14 14 12 43 17 13 

Outbreeding 81 76 88 57 73 75 

Tested species 30 34 13 36 33 28 

Note: Percentage of species with each feature in cerrado sensu stricto area (Oliveira and Gibbs 2000), 

seasonally dry forest in Costa Rica (Bawa 1974), secondary forest in Venezuela (Ruiz and Arroyo 1978, 

woody species only), cloud forest in Venezuela (Sobrevilla and Arroyo 1982), dry forest in Mexico 

(Bullock 1985), and rainforest in Costa Rica (Bawa et al. 1985a). Estimates for self-incompatible (SI) 

and inbreeding species are based on limited samples of the total surveyed flora (tested species). Inbreeding 

included self-compatible, autogamous, and apomictic species. 

(1980) for tropical moist forests, dioecy in cerrado sensu lato is correlated 

with small, structurally simple unisexual flowers which utilize a broad 

spectrum of small insects capable only of unspecialized pollination interactions. 

Seasonal drought, high temperatures, and distance between conspecific 

trees may limit the efficiency of small insect pollination and 

occurrence of dioecious species in open cerrado areas (Oliveira 1996). 

Where studies of breeding systems of cerrado species have combined 

controlled pollinations with observations of pollen tube growth, a notable 

feature has emerged: the scarcity of taxa with conventional homomorphic 

or heteromorphic self-incompatibility (SI). Homomorphic SI, with inhibition 

of self pollen at the stigma or within the stylar transmitting tract, 

has been reported only for the genera Vochysia (Oliveira and Gibbs 1994) 

and Miconia (Goldenberg and Shepherd 1998), while heteromorphic SI 

in the cerrado occurs only in the genera Erythroxylum (Erythroxylaceae) 

and Palicourea (Rubiaceae), two families in which this breeding system 

predominates (Barros 1998; Oliveira and Gibbs 2000). 

Most of the cerrado species studied for breeding system and postpollination 

events show “late-acting self-incompatibility’’ (sensu Seavey and 

Bawa 1986) or “ovarian’’ sterility (cf. Sage et al. 1994). Despite failure to 

set fruits following selfing, self-pollen tubes grow apparently successfully 

to the ovary where ovule penetration usually occurs, as in Tabebuia 

caraiba and T. ochracea (Gibbs and Bianchi 1993), Vellozia squamata



Table 17.3 Pollination and Breeding Systems of Woody Species 

in Cerrado Plant Formations 

Cerrado Cerrado Cerrado 

Campo sensu sensu sensu 

cerrado stricto stricto stricto Cerradão Cerradão 

CPAC CPAC FAL BBG FAL CPAC 

Pollination 

Small insects 35 (9/26) 38 (22/58) 38 (23/61) 44 (26/59) 45 (27/60) 40 (25/63) 

Large bees 35 (9/26) 33 (19/58) 39 (24/61) 32 (19/59) 37 (22/60) 36 (23/63) 

Bats 8 (2/26) 5 (3/58) 5 (3/61) 3 (2/59) 3 (2/60) 5 (3/63) 

Breeding systems 

Dioecy 12 (3/26) 7 (4/58) 11 (10/89) 15 (9/59) 16 (16/99) 27 (18/67) 

Monoecy 0 (0/26) 5 (3/58) 4 (3/89) 5 (3/59) 3 (3/99) 4 (3/67) 

Hermaphrodite 88 (23/26) 88 (51/58) 85 (76/89) 80 (47/59) 81 (80/99) 69 (46/67) 

SI 77 (14/16) 63 (20/28) 68 (29/36) 66 (25/30) 65 (33/41) 56 (18/22) 

Inbreeding 11 (2/16) 25 (8/28) 17 (7/36) 14 (5/30) 16 (8/41) 12 (4/22) 

Outbreeding 89 70 79 81 81 83 

Note: There is a gradient of woody species density from campo cerrado, to cerrado sensu stricto to cerradão 

that is characteristic of cerrado landscape near Brasília, central Brazil. Data express the percentage 

of species. Number of species showing each feature and the total number of species are given in 

parentheses. Breeding system frequency was estimated from limited samples. Outbreeding included 

dioecious plus self-incompatible (SI) species. Lists of identified species were used for each site. CPAC is 

a cerrado sensu lato reserve in the Centro de Pesquisa Agropecuária de Cerrado (Ribeiro et al. 1985). 

FAL is the University of Brasília experimental reserve (Ratter 1985). BBG is the Brasília Botanic Garden 

area (Oliveira and Gibbs 2000). 

(Oliveira et al. 1991), Eriotheca gracilipes (Oliveira et al. 1992), Dalbergia 

miscolobium (Gibbs and Sassaki 1998), Gomidesia lindeniana (Nic- 

Lughadha 1998), Hymenaea stigonocarpa (Gibbs et al. 1999), Qualea 

multiflora, Q. parviflora (Oliveira 1998b), and Callisthene fasciculata 

(Oliveira 1998b). “Late-acting self-incompatibility’’ (LSI) is a poorly 

understood phenomenon that almost certainly encompasses diverse 

mechanisms, such as early acting lethal recessives (Nic-Lughadha 1998) 

as well as possibly novel recognition systems (see Lipow and Wyatt 

2000). The seemingly widespread occurrence of LSI in woody cerrado 

taxa once again underlines the similarity between the reproductive biology 

of these woodlands and that of moist forest communities, where LSI 

is also common (Bawa, Perry, and Beach 1985). But both communities 

may reflect the widespread but hitherto underestimated occurrence of this 

outbreeding “mechanism’’ (Gibbs and Bianchi 1999). 

Brazilian cerrados also parallel the La Selva situation (Kress and 

Beach 1994), since the understory stratum of herbaceous and hemixyle 

species seem to be dominated by self-compatible species (Barbosa 1997).



Figure 17.2 Frequency of breeding systems in different cerrado formations 

of central Brazil (%). Outbreeding species are the sum of dioecious and selfincompatible 

ones. Inbreeding are self-compatible, autogamous or apomictic 

species. Data for Mauritia flexuosa palm swamp areas (vereda, buritizal, or 

morichal) which occurs all over the cerrado region and also in Venezuela, are 

presented for comparison. Sources: P. E. Oliveira, unpublished data (gallery 

forest); Ramirez and Brito 1990 (vereda); Barbosa 1997 (campo sujo); Saraiva 

et al. 1996 (cerrado sensu stricto at Corumbataí-SP); Oliveira and Gibbs 2000 

(cerrado sensu stricto at Brasília). See chapter 6 for description of vegetation 

physiognomies. 

Although self-compatibility does not necessarily imply a predominantly 

selfing mating system (e.g., Pascarella 1997), self-compatibility has been 

related to situations such as scarcity of pollinators, small population sizes, 

and dominance of herbaceous species (Sobrevilla and Arroyo 1982, Bullock 

1995). However, differences in floristic composition between tree 

versus understory strata may also introduce a family bias that increases 

the incidence of selfing taxa in the latter (Bianchi et al. 2000). Apomixis 

is rare among cerrado sensu lato species (Barbosa 1997; Oliveira and 

Gibbs 2000): the few apomictic taxa include possible apospory in Miconia 

spp. and other Melastomataceae (Goldenberg and Shepherd 1998) 

and adventitious embryony in Eriotheca pubescens (Oliveira et al. 1992). 

Various inconclusive explanations have been offered for the predominantly 

outbreeding behavior of tree species of moist forest and cerrado 

woodlands. In general, long-lived woody taxa in all communities exhibit 

outbreeding (Stebbins 1958), allegedly to maintain genetic heterozygosity 

in species with long reproductive life cycles and high seedling mortal-



ity. Predominant outbreeding has consequences for population structure. 

Genetic variability in the cerrado sensu lato should be similar to that of 

other tropical communities in which intrapopulation variability is greater 

than interpopulation variability (Kageyama 1990). Recent data using 

genetic markers in Caryocar brasiliense, a common bat-pollinated cerrado 

tree, showed high outcrossing rates in this species but indicated that habitat 

fragmentation and disturbance may limit gene flow and create small, relatively 

homogeneous population subunits (Collevatti et al. 2000). 

VICARIANCE AND CERRADO-FOREST BOUNDARIES 

Climatic changes, particularly during the Pleistocene, have caused repeated 

expansions and contractions of the cerrado sensu lato and moist 

forest areas, which have been important for the evolution of the Neotropical 

flora (Prance 1992; Ratter 1992; Oliveira-Filho and Ratter 1995; see 

chapter 3). Given such historical interactions, it is not surprising that 

many pairs of vicariant species between cerrado and forest can be identified 

(Rizzini 1963; Sarmiento 1983). They comprise species pairs with 

morphological similarities but which may show marked differences in 

growth habit. 

Gottsberger (1986) studied the differences between forest and cerrado 

sensu lato species in some taxa. He found that the reproductive biology 

was similar in species of the Bignoniaceae and Malpighiaceae of 

cerrado and moist forest. But in the Annonaceae and other beetle-pollinated 

groups, there seemed to be a more or less distinct fauna of pollinators 

in cerrado and forest plants. Moreover, for Malvaceae, he proposed 

an evolutionary trend from allogamous ornithophilous woody groups in 

the forest to predominantly endogamous and mellitophilous herbaceous 

genera and species in the cerrado. 

In contrast, no such differences could be observed in other more 

recently studied taxa. Vochysia, the largest genus in the neotropical family 

Vochysiaceae, is mostly distributed in rainforest but has some 20% of 

the species occurring in the cerrado region. Six such species were studied 

near Brasília, in central Brazil (Oliveira and Gibbs 1994), where both 

shrub and tree species occur in cerrado sensu lato, while other species are 

large gallery forest trees. Floral morphology, pollination, and xenogamous 

breeding systems were found to be fairly uniform in these Vochysia 

species despite the variation in life form and habitat. 

A similar situation occurs in Hymenaea (Caesalpiniaceae), another 

forest genus with savanna species. In central Brazil Hymenaea courbaril



var. stilbocarpa (sometimes treated as H. stilbocarpa) is a forest tree up 

to 25 m with bat-pollinated flowers and a xenogamous breeding system 

(Bawa, Perry, and Beach 1985; Bawa, Bullock, Perry et al. 1985; for Central 

America trees of Hymenaea courbaril). The smaller cerrado vicariad, 

Hymenaea stigonocarpa (up to 8 m), is also bat-pollinated and self-incompatible 

(Gibbs et al. 1999). 

These studies indicate that the adaptive shift between forest and cerrado 

habitats, accompanied in some taxa by dramatic changes in life form 

from large forest trees to small cerrado shrubs, does not necessarily 

involve changes in floral biology and breeding systems. The same largebee 

and bat species, the main pollinators of forest trees, may be found visiting 

and pollinating congeneric species in neighboring cerrado areas. 

BREEDING BIOLOGY AND CONSERVATION 

IN CERRADO 

The data emerging for the reproductive biology of cerrado plants have 

important consequences for conservation and understanding of the organization 

of cerrado communities. Open cerrado formations have been 

viewed as communities maintained by disturbance and limited in reproductive 

output, in which vegetative reproduction was considered more 

important than sexual reproduction (Rizzini 1965). Other studies, however, 

have indicated a more or less stable mosaic of communities, whose 

primary production is limited by nutritional levels, soil depth, seasonality, 

and even the occurrence of fire (Sarmiento et al. 1985). Cerrado plant 

communities are adapted to these conditions in terms of physiology, phenology, 

and reproductive output (Sarmiento and Monasterio 1983; 

Sarmiento et al. 1985; Oliveira and Silva 1993). Sexual reproduction and 

regeneration via seed is as important in this region as any other tropical 

woodland. 

The reproductive biology data discussed here support this latter point 

of view. The similarities in pollination and breeding systems indicate that 

the cerrado is an environment where the genetic variability provided by 

outcrossing is as important as it is in forest communities. Cerrado species 

seem to present potentially large mating populations and mostly relatively 

generalist pollination systems. These systems rely on a diversity of visitors 

and should be relatively resilient to environmental disturbances. However, 

such theoretical considerations need to be investigated, since to our 

knowledge there have been no studies on the reproductive success of cerrado 

species in disturbed and fragmented areas.



Moreover, despite this potential resilience, anthropomorphic changes 

in the intensity and periodicity of fire and disturbance (chapters 4, 5) may 

represent strong selective factors in a community basically dependent on 

pollinators to reproduce. Dioecious species and other taxa pollinated by 

less mobile vectors and/or dependent on animal dispersal might be particularly 

sensitive. Another consequence of such disturbances may be 

increased vegetative regeneration (Hoffmann 1998). Vegetative regeneration 

by resprouting may also result in islands of clonal plants with consequences 

for sexual reproduction of allogamous species. 

The combined effects of fire and disturbance are likely to cause an 

expansion of open physiognomies and consequently a possible reduction 

of suitable sites for some ground nesting bees, as observed in Costa Rica 

dry forests (Frankie et al. 1990). Such changes may also affect carpenter 

bees, Xylocopa spp., which nest in dry dead timber (Camillo and Garofalo 

1982). Drastic disturbance affecting important groups of pollinators 

could reduce the efficiency of the pollination systems and affect reproduction 

of many species in the community despite the possible resilience 

due to generalist systems. These woody species both depend on and sustain 

a rich fauna of pollinators dependent on them for survival. Herbaceous 

taxa, on the other hand, are mostly capable of selfing, and some are 

wind pollinated (Silberbauer-Gottsberger and Gottsberger 1988; Barbosa 

1997; chapter 7). Many also produce dry autochorous or epizoochorous 

fruits (Silberbauer-Gottsberger 1984), so that open grasslands, which 

result from increasing disturbance, would possibly maintain an impoverished 

sample of the original fauna of the area. 

Diversity of flowering phenology seems to be a characteristic of cerrado 

vegetation (Oliveira 1998a), and so it is possible to find some species 

in flower throughout the year. Such diversity of flowering phenology has 

important consequences for resource availability for pollinators. Biomass 

gradients, which may or may not be accompanied by floristic differentiation, 

are an important feature of cerrado communities (Goodland 1971; 

Ribeiro et al. 1985). Differences in species composition and importance 

along these gradients provoke dramatic changes in resource availability 

in both temporal and spatial terms (Oliveira 1998a). Simple changes in 

species density from place to place may shift the peak of flowering in cerrado 

and cerradão areas (see fig. 17.3) and may result in patchy resource 

availability for pollinators along vegetation gradients. 

The emerging data for the reproductive biology of the woody stratum 

of the cerrado is essentially similar to that established for the moist and 

seasonally dry forests of Central America (Kress and Beach 1994; Bullock 

1995). This applies equally to the spectrum of pollinators, with special



Figure 17.3 Flowering phenology in a cerrado sensu stricto and cerradão 

denser woodland in Brasília, central Brazil. Intensity of flowering as a percentage 

of individuals was obtained from species phenology and relative density 

in each area (redrawn from Oliveira 1998a). See chapter 6 for description 

of vegetation physiognomies. 

importance of medium- to large-bee species, and to the prevalence of outbreeding 

taxa, commonly promoted by “late-acting’’ type mechanisms. 

The reproductive biology of the gallery forests and moist forest areas 

inside the Cerrado Biome are as yet poorly studied. However, it is likely 

that the pollinators of cerrado sensu lato and those of the adjacent forests 

have multiple, mutual interactions. It follows that conservation of biodiversity 

in the cerrados is intimately linked with that of moist forest areas 

(see chapter 18). Each may facilitate the other by providing resources



throughout the year to maintain the local flux of pollinators. Excessive 

perturbation of either component could lead to loss of reproductive efficiency 

and diversity. 

REFERENCES 

Barbosa, A. A. 1997. “Biologia Reprodutiva de uma Comunidade de Campo 

Sujo, Uberlândia-MG.” Ph.D. thesis, Universidade Estadual de Campinas, 

Brazil. 

Barros, M. A. G. 1998. Sistemas reprodutivos e polinização em espécies simpátricas 

de Erythroxylum P.Br. (Erythroxylaceae) do Brasil. Rev. Bras. 

Bot. 21:159–166. 

Bawa, K. S. 1974. Breeding systems of tree species of a lowland tropical community. 

Evolution 28:85–92. 

Bawa, K. S. 1980. Evolution of dioecy in flowering plants. Ann. Rev. Ecol. 

Syst. 11:15–39. 

Bawa, K. S. 1990. Plant-pollinator interactions in tropical rain forests. Ann. 

Rev. Ecol. Syst. 21:399–422. 

Bawa, K. S., S. H. Bullock, D. R. Perry, R. E. Coville, and M. H. Grayum. 

1985. Reproductive biology of tropical lowland rain forest trees: II. Pollination 

systems. Amer. J. Bot. 72:346–356. 

Bawa, K. S., D. R. Perry, and J. H. Beach. 1985. Reproductive biology of tropical 

lowland rain forest trees: I. Sexual systems and incompatibility 

mechanisms. Amer. J. Bot. 72:331–345. 

Bianchi, M. B., P. E. Gibbs, D. E. Prado, and J. L. Vesprini. 2000. Studies on 

the breeding systems of understorey species of a Chaco woodland in NE 

Argentina. Flora 195:339–348. 

Bullock, S. H. 1985. Breeding systems in the flora of a tropical deciduous forest. 


Bullock, S. H. 1994. Wind pollination of Neotropical dioecious trees. 


Bullock, S. H. 1995. Plant reproduction in neotropical dry forests. In S. H. 

Bullock, H. A. Mooney, and E. Medina, eds., Seasonally Dry Tropical 

Forests, pp. 277–303. Cambridge: Cambridge University Press. 

Camillo, E. and C. A. Garófalo. 1982. On the bionomics of Xylocopa 

frontalis (Oliver) and Xylocopa grisescens (Lepeletier) in Southern 

Brazil: I. Nest construction and biological cycle. Rev. Bras. Biol. 

42:571–582. 

Carvalho, A. M. C. and L. R. Bego. 1996. Studies on Apoidea fauna of cerrado 

vegetation at the Panga Ecological Reserve, Uberlândia, MG. Brazil. 

Rev. Bras. Ent. 40:147–156. 

Collevatti, R. G., D. Grattapaglia, and J. D. Hay. 2000. Microsatellite markers 

provide evidence for the effect of fragmentation on the genetic struc-



ture of Caryocar brasiliense, an endangered tropical tree species of the 

Brazilian Cerrado. Mol. Ecol. 10:349–356. 

Frankie, G. W., W. A. Haber, P. A. Opler, and K. S. Bawa. 1983. Characteristics 

and organization of the large bee pollination system in the Costa 

Rican dry forest. In C. E. Jones and R. J. Little, eds., Handbook of Experimental 

Pollination Biology, pp. 411–447. New York: Van Nostrand 

Reinhold Co. 

Frankie, G. W., S. B. Vinson, L. E. Newstrom, J. F. Barthell, W. A. Harber, 

and J. K. Frankie. 1990. Plant phenology, pollination ecology, pollinator 

behaviour and conservation of pollinators in neotropical dry forest. In 

K. S. Bawa and M. Hadley, eds., Reproductive Ecology of Tropical Forest 

Plants, pp. 37–47. Carnforth: Parthenon. 

Furley, P. A. and J. A. Ratter. 1988. Soil resources and plant communities of 

the central Brazilian cerrado and their development. J. Biogeogr. 

15:97–108. 

Gibbs, P. E. and M. Bianchi. 1993. Post-pollination events in species of Chorisia 

(Bombacaceae) and Tabebuia (Bignoniaceae) with late-acting selfincompatibility. 

Bot. Acta 106:64–71. 

Gibbs, P. E. and M. B. Bianchi. 1999. Does late-acting self-incompatibility 

(LSI) show family Clustering? Two more species of Bignoniaceae with 

LSI: Dolichandra cynanchoides and Tabebuia nodosa. Ann. Bot. 

84:449–457. 

Gibbs, P. E., P. E. Oliveira, and M. Bianchi. 1999. Post-zygotic control of selfing 

in Hymenaea stigonocarpa (Leguminosae-Caesalpinoideae), a batpollinated 

tree of the Brazilian cerrados. Int. J. Plant Sci. 160:72–78. 

Gibbs, P. E. and R. Sassaki. 1998. Reproductive biology of Dalbergia miscolobium 

Benth. (Leguminosae-Papilionoideae) in SE Brazil: The effects 

of pistilate sorting on fruit-set. Ann. Bot. 81:735–740. 

Goldenberg, R. S. and G. J. Shepherd. 1998. Studies on the reproductive biology 

of Melastomataceae in “cerrado’’ vegetation. Pl. Syst. Evol. 211: 

13–29. 

Goodland, R. 1971. A physiognomic analysis of the cerrado vegetation of 

Central Brazil. J. Ecol. 59:411–419. 

Gottsberger, G. 1986. Some pollination strategies in neotropical savannas and 

forests. Pl. Syst. Evol. 152:29–45. 

Gottsberger, G. 1989. Beetle pollination and flowering rhythm of Annona 

spp. (Annonaceae) in Brazil. Pl. Syst. Evol. 167:165–187. 

Gribel, R., P. E. Gibbs, and A. L. Queiróz. 1999. Flowering phenology and 

pollination biology of Ceiba pentandra (Bombacaceae) in Central Amazonia. 

J. Trop. Ecol. 15:247–263. 

Gribel, R. and J. D. Hay. 1993. Pollination ecology of Caryocar brasiliense 

(Caryocaraceae) in Central Brazil cerrado vegetation. J. Trop. Ecol. 

9:199–211. 

Heithaus, E. R. 1979. Community structure of neotropical flower visiting 

bees and wasps: diversity and phenology. Ecology 60:190–202.



Hoffmann, W.A. 1998. Post-burn reproduction of in a Neotropical savanna: 

The relative importance of sexual and vegetative reproduction. J. Appl. 

Ecol. 35:422–433 

Kageyama, P. Y. 1990. Genetic structure of tropical tree species of Brazil. In 

K. S, Bawa and M. Hadley, eds., Reproductive Ecology of Tropical Forest 

Plants, pp. 375–387. Paris: UNESCO. 

Kress, W. J., and J. H. Beach. 1994. Flowering plant reproductive systems. In 

L. A. McDade, K. S. Bawa, H. A. Hespenheide, and G. S. Hartshorn, 

eds., La Selva: Ecology and Natural History of a Neotropical Rainforest, 

pp. 161–182. Chicago: University of Chicago Press. 

Lipow, S. R. and R. Wyatt, 2000. Single gene control of postzygotic selfincompatibility 

in poke milkweed, Asclepias exaltata L. Genetics 

154:893–907. 

Mantovani, W. and F. R. Martins. 1988. Variações fenológicas das espécies 

do cerrado da Reserva biológica de Mogi-Guaçú, Estado de São Paulo. 

Rev. Bras. Bot. 11:101–112. 

Mendonça, R. C., J. M. Felfili, B. M. T. Walter, S.-J. M.C., A. V. Rezende, 

T. S. Filgueiras, and P. E. Nogueira. 1998. Flora vascular do cerrado. In 

S. M. Sano and S. P. Almeida, eds., Cerrado: Ambiente e Flora, pp. 

289–556. Brasília: Empresa Brasileira de Pesquisa Agropecuária. 

Nic-Lughadha, E. 1998. Preferential outcrossing in Gomidesia (Myrtaceae) 

is maintained by a post-zygotic mechanism. In S. Owens and P. Rudall, 

eds., Reproductive Biology: In Systematics, Conservation and Economic 

Botany, pp. 363–379. Richmond: Royal Botanic Gardens at Kew. 

Oliveira, P. E. 1996. Dioecy in the cerrado vegetation of Central Brazil. Flora 

191:235–243. 

Oliveira, P. E. 1998a. Fenologia e biologia reprodutiva das espécies de cerrado. 

In S. M. Sano and S. P. Almeida, eds., Cerrado: Ambiente e Flora, 

pp. 169–192. Brasília: Empresa Brasileira de Pesquisa Agropecuária. 

Oliveira, P. E. 1998b. Reproductive biology, evolution and taxonomy of the 

vochysiaceae in Central Brazil. In S. Owens and P. Rudall, eds., Reproductive 

Biology: In Systematics, Conservation and Economic Botany, 

pp. 381–393. Richmond: Royal Botanic Gardens at Kew. 

Oliveira, P. E. and P. E. Gibbs. 1994. Pollination and breeding systems of six 

Vochysia species (Vochysiaceae) in Central Brazil. J. Trop. Ecol. 

10:509–522. 

Oliveira, P. E. and P. E. Gibbs. 2000. Reproductive biology of woody plants 

in a cerrado community of Central Brazil. Flora 195:311–329. 

Oliveira, P. E. and J. C. S. Silva. 1993. Reproductive biology of two species 

of Kielmeyera (Guttiferae) in the cerrados of Central Brazil. J. Trop. Ecol. 

9:67–79. 

Oliveira, P. E., P. E. Gibbs, A. A. Barbosa, and S. Talavera. 1992. Contrasting 

breeding systems in two Eriotheca (Bombacaceae) species of the 

Brazilian cerrados. Pl. Syst. Evol. 179:207–219. 

Oliveira, P. E., P. E. Gibbs, and M. B. Bianchi. 1991. Pollination and breed-



ing biology of Vellozia squamata (Liliales-Velloziaceae): A species of the 

Brazilian cerrados. Bot. Acta 104:392–398. 

Oliveira-Filho, A. T. and J. A. Ratter. 1995. A study of the origin of Central 

Brazilian forests by the analysis of plant species distributions. Edinb. J. 

Bot. 52:141–194. 

Pascarella, J. B. 1997. The mating system of the tropical understorey shrub 

Ardisia escallonoides (Myrsinaceae). Amer. J. Bot. 84:456–460. 

Prance, G. T. 1992. The phytogeography of savanna species of neotropical 

Chrysobalanaceae. In P. A. Furley, J. Proctor, and J. A. Ratter, eds., 

Nature and Dynamics of Forest-Savanna Boundaries, pp. 295–330. London: 

Chapman and Hall. 

Ramirez, N. and Y. Brito. 1990. Reproductive biology of a tropical palm 

swamp community in the Venezuelan llanos. Amer. J. Bot. 77:1260–1271. 

Ratter, J. A. 1985. Notas Sobre a Vegetação da Fazenda Água Limpa. Edinburgh: 

Royal Botanic Garden. 

Ratter, J. A. 1992. Transitions between cerrado and forest vegetation in 

Brazil. In P. A. Furley, J. Proctor, and J. A. Ratter, eds., Nature and 

Dynamics of Forest-Savanna Boundaries, pp. 417–430. London: Chapman 


Ratter, J. A., S. Bridgewater, R. Atkinson, and J. F. Ribeiro. 1996. Analysis of 

the floristic composition of the Brazilian cerrado vegetation: II. Comparison 

of the woody vegetation of 98 areas. Edinb. J. Bot. 53:153–180. 



Ribeiro, J. F., J. C. S. Silva, and G. J. Batmanian. 1985. Fitossociologia de 

tipos fisionômicos de cerrado em Planaltina–DF. Rev. Bras. Bot 

8:131–142. 

Rizzini, C. T. 1963. A flora do cerrado: Análise florística das savanas centrais. 

In M. G. Ferri, ed., Simpósio sobre o Cerrado, pp. 127–177. São Paulo: 




Roubik, D. W. 1989. Ecology and Natural History of Tropical Bees. Cambridge: 

Cambridge University Press. 

Ruiz, T. Z. and M. T. K. Arroyo. 1978. Plant reproductive ecology of a secondary 

deciduous tropical forest. Biotropica 10:221–230. 

Sage, T. L., R. I. Bertin, and E. G. Williams. 1994. Ovarian and other lateacting 

self-incompatibility systems. In E. G. Williams, A. E. Clarke, and 

R. B. Knox, eds., Genetic Control of Self-Incompatibility and Reproductive 

Development in Flowering Plants, pp. 116–140. Netherlands: 

Kluwer Academic Publishers. 

Saraiva, L., O. Cesar, and R. Monteiro. 1996. Breeding systems of shrubs and 

trees of a Brazilian savanna. Arq. Biol. Tecnol. (Brazil) 39:751– 763. 

Sarmiento, G. 1983. The savannas of tropical America. In F. Bouliere, ed.,



Ecosystems of the World: Tropical Savannas, pp. 245–288. Amsterdam: 

Elsevier. 

Sarmiento, G., G. Goldstein, and F. Meinzer. 1985. Adaptative strategies of 

woody species in neotropical savannas. Biol. Rev. 60:315–355. 

Sarmiento, G. and M. Monasterio. 1983. Life forms and phenology. In F. 

Bouliere, ed., Ecosystems of the World: Tropical Savannas, pp. 79–108. 

Amsterdam: Elsevier. 

Sazima, I., S. Buzato, and M. Sazima. 1996. An assemblage of hummingbirdpollinated 

flowers in a montane forest in Southeastern Brazil. Bot. Acta 

109:149–160. 

Sazima, M., S. Buzato, and I. Sazima. 1999. Bat-pollinated flower assemblages 

and bat visitors at two Atlantic forest sites in Brazil. Ann. Bot. 

83:705–712. 

Sazima, M. and I. Sazima. 1975. Quiropterofilia em Lafoensia pacari St. Hil 

(Lythraceae), na Serra do Cipó, Minas Gerais. Ciênc. Cult. 27:406–416. 

Seavey, S. R. and K. S. Bawa. 1986. Late-acting self-incompatibility in 

Angiosperms. Bot. Rev. 52:195–219. 

Silberbauer-Gottsberger, I. 1984. Fruit dispersal and trypanocarpy in Brazilian 

cerrado grasses. Pl. Syst. Evol. 147:1–27. 

Silberbauer-Gottsberger, I. and G. Gottsberger. 1975. Uber sphingophile 

Angiospermen Brasiliens. Pl. Syst. Evol. 123:157–184. 

Silberbauer-Gottsberger, I., and G. Gottsberger. 1988. A polinização das plantas 

do Cerrado. Rev. Bras. Biol. 48:651–663. 

Sobrevilla, C. and M. T. K. Arroyo. 1982. Breeding systems in a montane 

tropical cloud forest in Venezuela. Pl. Syst. Evol. 140:19–37. 

Stebbins, G. L. 1958. Longevity, habitats and release of genetic variability in 

higher plants. Cold Spring Harbor Symp. Quant. Biol. 23:365–378. 

Warming, E. 1908. Lagoa Santa—Contribuição para a Geografia Phytobiologica. 

(Portuguese translation of Lagoa Santa—et Bidrad til den Biologiske 

Plantegeographi, Copenhagen, 1882). Bello Horizonte: Imprensa 

Official do Estado de Minas Gerais.



18 

Biodiversity and Conservation 

Priorities in the Cerrado Region 

Roberto B. Cavalcanti and Carlos A. Joly 

The conservation of the Cerrado Biome and its ecosystems 

has been neglected until very recently, for two related reasons. First, 

central Brazil was very sparsely occupied until the mid-twentieth century, 

and therefore perceived threats to the environment were low (chapter 5). 

Second, the native cerrado had little apparent economic value and was 

often unattractive even in the eyes of specialists, due to the scrubby nature 

of the vegetation, low faunal densities, and a pronounced dry season with 

frequent fires. The region was colonized by Europeans systematically 

since the 18th century, towns were started at strategic points by prospectors 

seeking gold and diamonds, and the countryside was occupied by 

large farms focused primarily on extensive cattle ranching using native 

pasture. The low nutrient content and high acidity of most soils and the 

lack of railroad access restricted agriculture. For an excellent description 

of much of the cerrado in the early 1800s by one of the world’s leading 

botanists of that time, see St. Hilaire (1847–1848). 

The mechanization of Brazilian agriculture after 1950 and construction 

of major highways through central Brazil increased human impact 

dramatically (chapter 5). The use of lime and new fertilization techniques, 

together with development of high yield/drought-resistant varieties of 

soybean, rice and corn, helped open the cerrados as Brazil’s new agricultural 

frontier. The low cost of land, abundant rainfall during the growing 

season, and deep soils on gently rolling terrain suitable for mechanization 

were key factors in the development of large-scale agribusiness operations 

in the region. At least 67% of the cerrado region had been converted to 

intensive human use by the early 1990s, with current estimates placing 

351


352 the conservation of the cerrados 

conversion at 80% (Myers et al. 2000). In the state of São Paulo in southeastern 

Brazil the cerrado vegetation has been reduced from 14% to 

nearly 1.2% of the state’s area (Secretaria do Meio Ambiente de São Paulo 

1997). Further impacts on the cerrado may be an undesirable side effect 

of the pressure by national and international organizations to halt deforestation 

in Amazonia: in the search for alternative sites for agricultural 

development, and especially for planting forests for wood pulp or to act 

as carbon sinks, the Cerrado Biome has often been a target (Ab’Saber et 

al. 1990). A detailed account, including a historical overview, of the 

human-induced transformation of the cerrados is given in chapter 5. 

In this chapter we discuss current methods and results for identifying 

priorities for biodiversity conservation in the cerrado region and provide 

perspectives on strategies for the future of the region. 

THE CERRADO BIOME IS A RECOGNIZED 

GLOBAL BIODIVERSITY “HOTSPOT’’ 

Brazil harbors outstanding biodiversity, sharing with Indonesia the top 

two slots of the world’s richest “Megadiversity” countries (Mittermeier et 

al. 1997). Within Brazil, several biomes have merited individual global 

recognition. The Atlantic forest and the Brazilian cerrados are included in 

the world’s 25 principal “hotspots,” areas with great endemism and less 

than 30% remaining natural vegetation (Myers et al. 2000). Worldwide, 

these 25 “hotspots’’ cover 1.4% of the planet’s land surface and harbor 

44% of the world’s vascular plant species and 35% of all species in four 

vertebrate groups (Myers et al. 2000). 

Cerrado endemism is high for plants, on the order of 44%, but fairly 

low for vertebrates, estimated at 9.2% (Myers et al. 2000). A recent study 

for saturniid lepidopterans also indicated low endemism, 12.6% 

(Camargo and Becker 1999; see also chapter 11). Species richness is high, 

however. The cerrados harbor 13% of the butterfly species from the 

neotropics, 35% of the bees, and 23% of the termites (Brown 1996, Raw 

1998 and references therein). Nearly half of the Brazilian birds occur in 

the cerrado (Silva 1995a; chapter 13). For plants, the cerrado is considered 

the most diverse neotropical savanna (Lenthall et al. 1999). 

The habitat diversity of the cerrado region is exceptionally high. 

Gallery forests, Mauritia palm groves, calcareous caves, mesophytic 

forests, dry forests in the “Mato Grosso de Goiás,” and high-altitude 

rocky campos rupestres, among others, have many endemic species and


Biodiversity and Conservation Priorities 353 

provide a rich mixture of habitats for the region (see chapter 6). The 

gallery forests in particular provide physical connection with the Amazonian 

and Atlantic Forests, in addition to contributing to the maintenance 

of the open cerrado biota in dry periods (Prance 1982, Fonseca and 

Redford 1984; Redford and Fonseca 1986; Cavalcanti 1992). Clearing 

the cerrado is a nonrandom process which focuses on the areas most suitable 

for agriculture. This selective removal destroys the integrity of the 

original habitat matrix and precipitates extinctions due to a variety of 

related effects (Pimm and Raven 2000). 

UNDERSTANDING BIODIVERSITY 


The Cerrado Priority-Setting Workshop in 1997 was developed and organized 

by a consortium consisting of the Fundação Pró-Natureza, Conservation 

International, the Universidade de Brasília, and the Fundação 

Biodiversitas. The pantanal wetlands were included in this Workshop, for 

several reasons: (1) there are extensive similarities in flora and fauna; (2) 

the pantanal headwaters are in the Cerrado; and (3) many scientists have 

expertise in both regions, giving economy of scale and organization 

(Cochrane et al. 1985). Major funding was provided by the Brazilian Ministry 

of the Environment, as part of the National Conservation and Sustainable 

Use of Biodiversity Project supported by the Global Environment 

Facility (Ministério do Meio Ambiente 1999). The design incorporated 

the experience of earlier priority-setting exercises in Amazonia and the 

Northeastern Atlantic Forest (Conservation International et al. 1991). 

Preworkshop data gathering began in 1996 and 1997 with state-ofthe-art 

reviews of knowledge by expert consultants for mammals, birds, 

reptiles and amphibians, invertebrates, fishes and aquatic biota, botany, 

soils and climate, conservation units, coverage of natural vegetation and 

socio-economic factors (Marinho Filho 1998; Silva 1998; Colli 1998; 

Araújo et al. 1998; Raw 1998; Diniz and Castanheira 1998; Britski 1998; 

M. Ribeiro 1998; J. F. Ribeiro 1998; Assad 1998; Pádua and Dias 1998; 

Mantovani and Pereira 1998; Sawyer 1998). Reports, databases, and digital 

maps were combined to generate products for use during the Workshop, 

including distribution maps of endangered and endemic species, 

maps of remaining and present vegetation, collection localities, protected 

areas, human population pressure and land use practices, and classification 

of Cerrado regions according to climate and soils.



PRIORITY AREAS FOR CONSERVATION 


During the workshop, priorities were set in three stages. First, thematic 

groups reviewed the information on all the Cerrado for a single theme, 

defined priority sites for conservation, and documented the relevant data 

for each site. Next, subregional groups took the recommendations from 

all thematic groups and overlaid, merged, enlarged, reduced, or added 

them to consolidate one map of priority areas per subregion. Finally, the 

subregional maps were combined into the overall cerrado biodiversity priority 

map with 87 areas (see fig. 18.1). Priority areas were selected on the 

Figure 18.1 Priority areas for conservation in the cerrado region and adjoining 

pantanal (based on the results of the Priority-setting Workshop; see Brasil 1999). 

List of localities: (1) Roraima savannas; (2) Paru savannas; (3) Monte Alegre 

campos; (4) Ilha de Marajó savannas; (5) Serra dos Carajás; (6) Serra do 

Cachimbo; (7) North Amapá savannas; (8) South Central Amapá savannas;



basis of biological importance using the distribution of endemic, rare, 

threatened, or migratory species; species of economic or cultural value; 

species richness and composition of biological communities; and abiotic 

and landscape features crucial to conserving biodiversity (climate, rivers, 

geomorphology, soils). Priority actions for each site were determined by 

Figure 18.1 (continued) (9) Northeastern Maranhão; (10) Maranhão semideciduous 

forests; (11) Mirador-Uruçuí; (12) Southeastern Maranhão; (13) 

Southeastern Piauí, Maranhão, and Tocantins tablelands; (14) Three biomes; 

(15) Rio Negro, Águas Emendadas, and Rio do Sono basins; (16) Chapada 

do Araripe, (17) Humaitá campos; (18) Pacaás-Guaporé-Ricardo Franco 

Corridor; (19) Middle Araguaia and Rio dos Cocos Basin; (20) Bananal 

Island; (21) Middle Tocantins; (22) Southern Tocantins-Conceição/Manuel 

Alves region; (23) Southeastern Tocantins semideciduous forests; (24) Goiás- 

Bahia hinterlands and São Domingos caves; (25) Posse-Correntina-São 

Domingos region; (26) Chapada Diamantina National Park; (27) Rio Papagaio; 

(28) Paraguay/Sepotuba headwaters; (29) Xingu headwaters; (30) Cascalheira 

and Querência streams; (31) Alto Boa Vista; (32) Araguaia valley and 

Rio das Mortes wetland; (33) Serra da Mesa/Niquelândia/Minaçu; (34) Pouso 

Alto; (35) Chapada dos Veadeiros; (36) Paranã valley and range; (37) Goiás- 

Rio das Almas/upper Tocantins; (38) Federal District and environs; (39) 

Pirenópolis; (40) Goiânia, Silvânia, Aparecida de Goiás, and Serra Dourada; 

(41) Serra Dourada dry forests; (42) Cristalina; (43) Upper Paraná; (44) Emas 

National Park and Araguaia headwaters; (45) Upper São Francisco basin 

(Peruaçu valley); (46) Grão Mogol; (47) Diamantina and Jequitinhonha valley; 

(48) Serra do Cabral; (49) Serra do Cipó; (50) Paracatu/Três Marias; (51) 

Upper Paranaíba/Patrocínio; (52) Minas Gerais Triangle; (53) Serra da Canastra 

National Park; (54) Serra de Santa Bárbara; (55) Serra das Araras; (56) 

Chapada dos Guimarães National Park and Cuiabá lowlands; (57) Nova 

Xavantina monodominant forest; (58) Paraguaizinho; (59) Chapada dos 

Guimarães and Barão de Melgaço; (60) Sucuriú; (61) Piquiri-Itiquira headwaters; 

(62) Pantanal west border A; (63) Rio Negro and Nhecolândia; (64) 

Taquari; (65) Emas/Taquari; (66) Emas-Jauru headwaters; (67) Jauru; (68) 

Pantanal west border B; (69) Taboco; (70) Bodoquena; (71) Chaco (Porto 

Murtinho Pantanal); (72) Mouth of the Ivinhema; (73) São José do Rio Preto; 

(74) Barretos; (75) Northeastern São Paulo; (76) Patrocínio Paulista; (77) 

Araraquara; (78) Campinas; (79) Araçatuba; (80) Presidente Prudente; (81) 

Marília; (82) Bauru; (83) Botucatu; (84) Paraíba valley; (85) Itapeva; (86) 

Itararé; (87) Jaguariaíva and Sengés (Paraná). 

Key to state codes: Amazonas (AM), Bahia (BA), Ceará (CE), Distrito Federal 

(DF), Espírito Santo (ES), Goiás (GO), Maranhão (MA), Minas Gerais 

(MG), Mato Grosso (MT), Mato Grosso do Sul (MS), Pará (PA), Paraná (PR), 

Pernambuco (PE), Piauí (PI), Rio de Janeiro (RJ), Rondônia (RO), São Paulo 

(SP), Tocantins (TO).



analyzing the human pressures at the local level, which included demography, 

vulnerability of natural ecosystems to agriculture, cattle, industries, 

urban expansion, and various other types of economic use (BRASIL 1999). 

Conservation urgency and opportunity were derived from data on 

human occupation and from the estimates of land cover produced by the 

first study to survey the entire cerrado and dry areas in the pantanal using 

satellite imagery (Mantovani and Pereira 1998; see fig. 18.2). Using 

images from 1992 and 1993, with occasional gaps filled with images from 

1987 to 1991, Mantovani and Pereira (1998) estimated that one-third of 

the cerrado remained little or not disturbed. Well-preserved areas were 

found in 3 distinct regions: (a) the border between the states of Piauí, 

Maranhão and Bahia; (b) the border between the states of Tocantins, 

Mato Grosso, and Goiás; and (c) the border between Tocantins, Goiás, 

Figure 18.2 Human disturbance to the cerrado landscape, as assessed by 

satellite imagery (from Brasil 1990; Mantovani and Pereira 1998). State codes 

as in figure 18.1.



and Bahia. The regions of heaviest human impact were in the states of 

Mato Grosso do Sul, Goiás, and São Paulo; the border of São Paulo and 

Paraná; and the “Triangle’’ of Minas Gerais (figs. 18.1, 18.2). In these 

areas, 50% to 92% of the cerrado surface was under heavy human pressure 

(Mantovani and Pereira 1998). 

Presently it is estimated that under 20% of the cerrado remains undisturbed 

(Myers et al. 2000). The agricultural frontier continues to expand. 

Major soybean cultivation projects are underway in regions classified as 

little disturbed by Mantovani and Pereira (1998), particularly at the borders 

of the states of Tocantins, Maranhão and Piauí. Roadbuilding and 

waterway development, and hydroelectric dams on the Tocantins river, 

are opening the last inaccessible regions of the cerrado. 

The cerrado is poorly known for many taxonomic groups. Large 

areas of the states of Bahia and Tocantins are unsurveyed. For birds, 

approximately 70% of the region never has been adequately inventoried 

(Silva 1995a, b, 1998). For reptiles, 97% of the localities recorded were 

insufficiently surveyed (Colli 1998). Plants were arguably the group best 

inventoried in geographical completeness, with an active program to fill 

gaps in the entire region (Felfili et al. 1994; Ratter et al. 1996). For insects, 

coverage varied among groups, with collections generally concentrated 

where researchers are most active, such as Brasília (central Brazil) and the 

state of São Paulo (Diniz and Castanheira 1998). Data for aquatic organisms 

were even more limited. Many species of copepods are known only 

from one locality (Reid 1994). Sampling effort for fishes varied 20-fold 

between major river basins such as the São Francisco, the Tocantins, and 

the Paranaíba (Ribeiro 1998). 

The existing federal and state conservation system in the cerrado 

region is insufficient both in extent and representation, covering approximately 

1.6% of the surface area (Pádua and Dias 1997). Major subdivisions 

of the biome are highly endangered and poorly represented, such as 

the dry forests of the states of Goiás, Minas Gerais, Maranhão and Mato 

Grosso do Sul (fig. 18.1; areas 24, 25, 41, 45, 70). Of the approximately 

100 different environmental units mapped by the Brazilian Institute for 

Agricultural Research (EMBRAPA), 73% have no protected parks or 

reserves (Pádua and Dias 1997). Considering that about 20% of the 

cerrado region is still undisturbed, an interim goal would be to protect 

half of that, or 10% of the biome, which would require expanding sixfold 

the existing units. Workshop participants gave high priority to 17 

locations for creating new major conservation units. Since the Workshop, 

two have been decreed: the Peruaçu valley in the state of Minas Gerais 

and the Serra da Bodoquena in the state of Mato Grosso do Sul (fig. 18.1, 

areas 45, 70).



PRIORITY AREAS AND CONSERVATION ACTION 

IN THE CERRADO/PANTANAL INTERFACE 

The cerrado highlands border the pantanal floodplain on the east and 

north sides and include headwaters of the pantanal’s major rivers. The 

conservation of natural vegetation on these border crests is essential for 

the viability of the world’s largest tropical freshwater wetland. Workshop 

participants recommended creating a biodiversity corridor system that 

would join the priority areas for conservation in the pantanal basin with 

those in the cerrado, following the main river drainages and climbing the 

floodplain escarpments (hatched area of fig. 18.1). Major cerrado habitats 

in this corridor occur in Emas National Park (area 44), Chapada dos 

Guimarães National Park (56), Bodoquena National Park (70), and the 

headwaters of the Paraguay river (28). Despite most of these areas harboring 

parks, the landscape is being rapidly fragmented by clearing for 

pasture and agriculture, threatening the integrity and potential expansion 

of the protected areas system. Since 1999, Conservation International, 

Fundação Emas, IBAMA, the state secretariats for the environment of 

Mato Grosso do Sul and Goiás have been implementing the Emas-Rio 

Negro Biodiversity Corridor (fig. 18.1, areas 63–66) with the support of 

international agencies including USAID. This five year program aims to 

work with private landowners and public agencies to manage the landscape 

to enhance and restore connections between natural habitats 

throughout the corridor (Cavalcanti et al. 1999). 


IN THE CENTRAL AND NORTHEASTERN CERRADOS 

The central region of the cerrados has some of the highest species diversities 

and the best studied sites for various taxa. The Federal District has 

503 of the 820 bee species of the Cerrado, 63 of the 129 social wasps, and 

80% of the lepidopterans (Raw 1998 and data cited therein). The bird list 

of the Federal District and environs includes at least 439 species (Negret 

et al. 1984, updated by Bagno 1996), with 837 listed for all the biome 

(Silva 1995a). 

Priority areas in this region include the Federal District and a group 

of contiguous areas ranging north in the state of Goiás near the border of 

Bahia and into the state of Tocantins (fig. 18.1, areas 22, 24, 25, 34, 35, 

36, 38). In addition to the main cerrado vegetation types, these areas 

include the highly endemic plant communities of the Chapada dos Vead-



eiros (35), the dry forests of the Paranã valley (36), the extensive calcareous 

caves of São Domingos (24), the parkland and sandy savannas of the 

Bahia-Tocantins interface (21–25), and several areas on the Tocantins 

drainage (21–23). On the whole the areas are largely unprotected, with 

the exception of the Federal District, a national park in Chapada dos 

Veadeiros, and a state park in São Domingos. The denser dry forests and 

cerradão (chapter 6) are almost entirely destroyed, together with the habitat 

of several endemic bird subspecies (Silva 1989). 

The Rio Araguaia drainage with its unique wetland habitats, sandy 

beaches with nesting turtles, pink dolphins, and mix of Amazonian and 

cerrado influences is highly threatened by agricultural erosion and pesticide 

runoff, former gold mining tailings, erosion and sedimented mercury 

in its tributaries, and lack of protection of the headwaters. A string of 

areas of high biological importance occurs in the basin, with the only 

notable protected areas in the Ilha do Bananal and a recently established 

state park in Tocantins (fig. 18.1, areas 19, 20, 32). The Xingu headwaters 

and Rio das Mortes (areas 29–31) areas are also highly important, 

studied in the 1950s and 1960s by the Roncador-Xingu and Royal Society 

expeditions (Sick 1955; Fry 1970). 

The extreme northeastern cerrados offer one of the last opportunities 

to set up very large protected areas on the order of 200,000+ hectares. 

The states of Maranhão, Piauí and the Bahia/Tocantins border have six 

high-priority areas, including the Rio do Sono/Formoso do Rio Preto/ 

Jalapão complex (fig. 18.1, areas 13, 15), the Mirador region (11), southwest 

and central Maranhão (10, 12), and the “three biomes’’ region in 

Piauí (14). This hinterland still has few roads and a sparse population. 

Botanical analyses suggest that the northeastern cerrados are a center of 

diversity (Castro and Martins 1998). 


IN THE EASTERN AND SOUTHEASTERN CERRADOS 

This region is the most highly fragmented and occupied for agriculture, 

cattle ranching, and urban expansion. In Minas Gerais Triangle (fig. 18.1, 

area 52) and in the state of São Paulo (SP) the native cerrado is reduced 

to small remnants rarely exceeding 100 hectares. Of the eight high biodiversity 

priority areas in the region, two contain significant national parks, 

Serra da Canastra and Serra do Cipó (fig.18.1, areas 53, 49). There are 

still opportunities for conservation in central and northern state of Minas 

Gerais, in the areas around Diamantina, Grão Mogol, in the valley of the



Jequitinhonha, and in the valley of the São Francisco. The Espinhaço 

range of Minas Gerais and Bahia (fig.18.1, areas 26, 46–48), rich in plant 

and animal endemics, was included in several priority areas of the workshop 

and is recognized as of global importance for bird conservation 

(International Council for Bird Preservation 1992). 

In the state of São Paulo (SP), where cerrados originally covered 14% 

of the area, about 1% still has this vegetation. These 238,400 ha of remnant 

cerrado are fragmented into 8,353 islands, over half of which (4,372) 

are smaller than 10 ha and probably will disappear in the next few years. 

Only 47 fragments exceed 400 ha, many of them (17) in the administrative 

region of Ribeirão Preto (Kronka et al. 1998). The State of São Paulo 

Secretary of the Environment organized a workshop in 1995 to establish 

a policy for the conservation of the cerrado in the state (Secretaria do 

Meio Ambiente de São Paulo 1997). As a result, all permits to deforest 

cerrado remnants are now reviewed on a case by case basis, by a state 

commission coordinated by the São Paulo Biodiversity Conservation Program–PROBIO/SP. 

Several research initiatives are underway with the 

Biota Program of the Fundação de Amparo à Pesquisa do Estado de São 

Paulo, which deals with biodiversity conservation and sustainable use 

(FAPESP 1999). The conservation of cerrado remnants in São Paulo is of 

importance since they occur near the southern limit of the biome and have 

large areas of contact with the Atlantic forest region (see fig. 18.1). Ironically, 

much of the pioneering work on cerrado ecology was done in the 

state of São Paulo, in sites that are now relictual testimonials of the original 

landscape (Labouriau 1966; Ferri 1971). 

The priority areas for conservation extend to the state of Paraná (fig. 

18.1, area 87), where cerrado apparently occurred in five places; the 

largest protected area is a state park with 430 hectares (Straube 1998). 

CONSERVATION POLICY 

The priority setting workshop helped cerrado conservation policy at the 

federal, state, and local levels. The Minister of the Environment instituted 

a working group to develop a cerrado conservation strategy based on the 

workshop recommendations, including setting a target of expanding the 

conservation units to cover 10% of the biome by the year 2002 (Ministério 

do Meio Ambiente 1999). The state of Minas Gerais did its own 

conservation assessment using priority setting methodology and incorporating 

data from the biome level work (Costa et al. 1998). A long-term 

effort to amend the Constitution to include the Cerrado Biome as a



national heritage has been recently revived by a parliamentary front. Several 

new or proposed parks are being implemented in priority areas. 

Many policy recommendations emerged from the workshop, especially 

for incorporating conservation of biodiversity into the land use 

planning instruments, such as river basin management plans, road and 

electricity infrastructure plans, and urban and agricultural expansion initiatives. 

The focus of environmental action must move from environmental 

mitigation to a proactive conservation approach. Specific 

recommendations (BRASIL 1999) include the following: 

• Develop ecological corridors. Federal, state, and local governments 

should develop programs to stimulate restoration and connection 

of natural habitat fragments, combining public and private protected 

areas in biodiversity corridors. 

• Coordinate government agencies and policies. More effective integration 

between Ministries can help develop common environmental 

policies for land tenure and use, energy, waters, and health, 

including agricultural financing and subsidies; management of 

water resources to avoid depletion during the dry season; incentives 

to use the cleared areas more efficiently; restrictions on further 

clearing of native habitat; encouragement of economic use of native 

species; and economic incentives to landowners undertaking ecological 

corridor restoration. 

• Legislation. The current laws, if well applied, could go a long way 

to protect the cerrados. The government should strengthen the cerrado 

technical subcommittee of the National Environmental Council 

and enhance protection of critical habitats within the cerrado 

region such as the dry forests, the Mauritia palm groves (veredas), 

the rocky high altitude campos rupestres, karstic zones, Amazonian 

savannas, and the floodplains of major rivers (chapter 6). The 

Forestry Code, currently under revision, should increase protection 

of cerrado vegetation to exceed current requirements of 20% in 

each property. 

• Consolidation of conservation units. The existing public protected 

areas must be secured, through resolving ousting land claims, 

implementing management plans, and staffing parks and reserves. 

New large parks of 300,000 hectares or more should be created in 

the remaining blocks of natural cerrado. The protected area system 

should be extended to include all landscapes and subregions of the 

biome. The tax abatement Private Reserve Program (RPPN) should 

be expanded at the state and municipal level and provide additional



financial incentives and technical support for landowners to create 

and maintain reserves. 

• Research and inventories. A scientific network for cerrado research 

is needed to support biodiversity inventory and monitoring. A 

rapid assessment program and a detailed inventory project should 

be deployed, focusing on high biodiversity areas and habitats and 

regions underrepresented in existing datasets, using standardized 

protocols for collection and documentation. A fund to finance this 

work should be drawn from the monies paid for environmental 

mitigation of large infrastructure projects. 

• Support for scientific collections. A network of reference collections 

at local institutions should be set up, complemented by holdings 

at major museums. A Cerrado Museum in Brasília (core area 

of cerrados) is highly recommended. Existing collections need electronic 

cataloging to facilitate research and publication of updated 

lists of cerrado fauna and flora. Training of taxonomists and collecting 

in poorly sampled areas is urgently needed. 

• Monitoring. A set of indicator groups should be chosen for continuous 

monitoring, drawn from the endemic biota, from rare 

and/or endangered species, and from species of economic importance. 

The whole biome should be surveyed with satellite data at 

regular intervals to measure fragmentation and expansion of 

human occupation, and to assess effectiveness of corridor and protected 

areas programs. 


We appreciate the help and comments of Gustavo Fonseca, Russel Mittermeier, 

Bráulio Dias, Maria T. J. Pádua, Lauro Morhy, and the coordinators 

and participants of the Cerrado/Pantanal Priority setting Workshop. 

Keith Brown made invaluable suggestions and greatly improved the manuscript. 

We are grateful to Paulo Oliveira for his invitation to write this 

chapter, editorial comments, and patience with deadlines. We would also 

like to recognize the Ministry of the Environment/ Secretariat for Biodiversity 

and Forests, Conservation International, Funatura, Fundação Biodiversitas, 

Universidade de Brasília, the Global Environment Facility 

(GEF), the Brazilian Research Council (CNPq), the World Bank, Unibanco 

Ecologia, and Fundação o Boticário de Proteção à Natureza, among others, 

in developing and supporting biodiversity conservation strategies in 

the cerrado. C. A. Joly was supported by a CNPq grant (520334/99-0).



REFERENCES 

Ab’Saber, A., J. Goldenberg, L. Rodés, and W. Zulauf. 1990. Identificação de 

áreas para o florestamento no espaço total do Brasil. Estudos Avançados 

4:63–119. 

Araújo, A., C. S. Verano, and R. Brandão. 1998. Biodiversidade do cerrado: 

Herpetofauna. Workshop Ações Prioritárias para Conservação da Biodiversidade 

do Cerrado e Pantanal. Brasília: Ministério do Meio Ambiente, 

FUNATURA, Conservation International, Fundação Biodiversitas, 

Universidade de Brasília. Web: www.bdt.org.br/workshop/cerrado/br. 

Assad, E. (ed). 1998. Relatório final do grupo temático fatores abióticos. 

Workshop Ações Prioritárias para Conservação da Biodiversidade do 

Cerrado e Pantanal. Brasília: Ministério do Meio Ambiente, 

FUNATURA, Conservation International, Fundação Biodiversitas, Universidade 

de Brasília. Web: www.bdt.org.br/workshop/cerrado/br. 

Bagno, M. 1996. Atualização da Lista de Aves do Distrito Federal. Relatório 

e Banco de Dados. Departamento de Zoologia, Universidade de Brasília, 

Brasília, Brazil. Web: www.bdt.org.br/zoologia/aves/avesdf/texto. 

BRASIL. 1999. Ações Prioritárias para Conservação da Biodiversidade do Cerrado 

e Pantanal. Brasília: Ministério do Meio Ambiente, FUNATURA, Conservation 

International, Fundação Biodiversitas, Universidade de Brasília. 

Brown Jr., K. S. 1996. Diversity of Brazilian Lepidoptera: History of study, 

methods for measurement, and use as indicator for genetic, specific, and 

system richness. In C. A. Bicudo and N. A. Menezes, (eds.), Biodiversity 

in Brazil: A First Approach, pp. 121–154. São Paulo: CNPq and 

Instituto de Botânica. 

Britski, H. 1998. Peixes do cerrado e pantanal: Informações prévias. Workshop 

Ações Prioritárias para Conservação da Biodiversidade do Cerrado 


International, Fundação Biodiversitas, Universidade de 

Brasília. Web: www.bdt.org.br/workshop/cerrado/br. 

Camargo, A. A. and V. O. Becker. 1999. Saturniidae (Lepidoptera) from the 

Brazilian Cerrado: Composition and biogeographic relationships. 


Castro, A. A. J. F. and F. R. Martins. 1998. Cerrados do Brasil e do nordeste: 

Caracterização, área de ocupação e considerações sobre sua fitodiversidade. 

Workshop Ações Prioritárias para Conservação da Biodiversidade 




Cavalcanti, R. B. 1992. The importance of forest edges in the ecology of open 

country cerrado birds. In P. A Furley, J. Proctor, and J. A. Ratter, (eds.), 

The Nature and Dynamics of Forest-Savanna Boundaries, pp. 513–517. 

London: Chapman and Hall.



Cavalcanti, R. B., L. P. Pinto, J. M. C. da Silva. 1999. Criteria for Establishing 

Protected Areas. 14 pp. International Experts Meeting on Protected 

Forest Areas. San Juan, Puerto Rico. Sponsored by the Governments of 

Brazil and the United States of America. 

Cochrane, T. T., L. G. Sánchez, J. A. Porras, L. G. de Azevedo, and C. L. 

Garver. 1985. Land in Tropical America = La Tierra en América Tropical 

= A Terra na América Tropical, Vol. 1. Cali and Planaltina: Centro 

Internacional de Agricultura Tropical and Empresa Brasileira de Pesquisa 


Colli, G. R. (ed.) 1998. Biogeografia e conservação da herpetofauna no cerrado, 

pantanal e savanas amazônicas: Relatório final do grupo temático 

Herpetofauna. Workshop Ações Prioritárias para Conservação da Biodiversidade 


FUNATURA, Conservation International, Fundação Biodiversitas, 


Conservation International. 1991. Workshop 90. Biological Priorities for 

Conservation in Amazonia. Map. In association with IBAMA and INPA. 

Washington, DC., U.S. 

Conservation International. 1993. Biodiversity Priorities for Papua New 

Guinea. Map. In association with Government of Papua New Guinea 

Dept. of Environment and Conservation, and Biodiversity Support Program. 

Washington, DC., U.S. 

Costa, C. M. R., G. Hermann, C. S. Martins, L. V. Lins, and I. R. Lamas (eds.). 

1998. Biodiversidade em Minas Gerais: Um Atlas para sua Conservação. 

Belo Horizonte: Fundação Biodiversitas. 

Diniz, I. R. and Castanheira, H. C. (eds.). 1998. Relatório final do grupo 

temático Invertebrados. Workshop Ações Prioritárias para Conservação 

da Biodiversidade do Cerrado e Pantanal. Brasília: Ministério do Meio 

Ambiente, FUNATURA, Conservation International, Fundação Biodiversitas, 


Felfili, J. M., T. S. Filgueiras, M. Haridasan, M. C. da Silva Jr., Mendonça, 

R. C., and A. V. Rezende. 1994. Projeto biogeografia do bioma Cerrado: 

Vegetação e solos. Cad. Geoc. 12:75–166. 

Ferri, M. G. (ed.) 1971. Simpósio sobre o Cerrado, pp. 127–177. São Paulo: 


FAPESP. 1999. BIOTA/FAPESP—The Biodiversity Virtual Institute. Web: 

www.biota.org.br. 

Fonseca, G. A. B. and K. H. Redford. 1984. The mammals of IBGE’s Ecological 

Reserve, and an analysis of the role of gallery forests in increasing 

diversity. Rev. Bras. Biol. 44:517–523. 

Fry, C. H. 1970. Ecological distribution of birds in north-eastern Mato 

Grosso State, Brazil. An. Acad. Bras. Ciênc. 42:275–318. 

International Council for Bird Preservation.. 1992. Putting Biodiversity on 

the Map. Cambridge, UK: International Council for Bird Preservation.



Kronka, J. N. F., M. A. Nalon, C. K. Matsukuma, M. Pavão, J.R. Guillaumon, 

A. C. Cavalli, E. Giannotti, M. S. S. Ywane, L. M. P. R. Lima, 

J. Montes, I. H. D. Cali, and P. G. Haack. 1998. Áreas de domínio do 

Cerrado no Estado de São Paulo. São Paulo: Secretaria do Meio Ambiente.Web: 

http://www.bdt.org.br/images/bdt/sma/frag.gif. 

Labouriau, L. G. (ed.) 1966. II Simpósio Sobre o Cerrado. Rio de Janeiro: 

Academia Brasileira de Ciências. 

Lenthall, J. C., S. Bridgewater, and P. Furley. 1999. A phytogeographic analysis 

of the woody elements of new world savannas. Edinb. J. Bot. 

56:293–305. 

Mantovani, J. E. and A. Pereira. 1998. Estimativa da integridade da cobertura 

vegetal do Cerrado/Pantanal através de dados TM/LANDSAT. 

Workshop Ações Prioritárias para Conservação da Biodiversidade do 

Cerrado e Pantanal. Brasília: Ministério do Meio Ambiente, 



Marinho-Filho, J. (ed.) 1998. Relatório final do grupo temático Mastofauna. 

Workshop Ações Prioritárias para Conservação da Biodiversidade do Cerrado 

e Pantanal. Brasília: Ministério do Meio Ambiente, FUNATURA, 

Conservation International, Fundação Biodiversitas, Universidade de 


Ministério do Meio Ambiente. 1999. Termo de Compromisso Visando a Formulação 

e a Implementação do Plano de Ação Integrado para os Biomas 

Cerrado e Pantanal. Brasília: Ministério do Meio Ambiente, SECEX, Secretaria 

de Biodiversidade e Florestas. 

Mittermeier, R., P. R. Gil, and C. G. Mittermeier. 1997. Megadiversity: Earth’s 

Biologically Wealthiest Nations. Mexico: CEMEX. 

Myers, N., R. A. Mittermeier, C. G. Mittermeier, G. A. B. da Fonseca, and 

J. Kent. 2000. Biodiversity hotspots for conservation priorities. Nature 

403:853–858. 

Negret, A., J. Taylor, R. C. Soares e R. B. Cavalcanti. 1984. Aves da Região 

Geopolítica do Distrito Federal. Brasília: Ministério do Interior, Secretaria 

do Meio Ambiente. 

Pádua, M. T. J. and B. F. S. Dias. 1997. Representatividade de unidades de 

conservação do cerrado e do Pantanal Matogrossense. Workshop Ações 

Prioritárias para Conservação da Biodiversidade do Cerrado e Pantanal. 

Brasília: Ministério do Meio Ambiente, FUNATURA, Conservation 

International, Fundação Biodiversitas, Universidade de Brasília. Web: 

www.bdt.org.br/workshop/cerrado/br. 

Pimm, S. and P. Raven. 2000. Extinction by numbers. Nature 403:843–845. 

Prance, G. T. 1982. A review of the phytogeographic evidence for Pleistocene 

climate changes in the neotropics. Ann. Miss. Bot. Gard. 69:594–624. 

Ratter, J.A., S. Bridgewater, R. Atkinson, and J. F. Ribeiro. 1996. Analysis of 

the floristic composition of the Brazilian Cerrado vegetation: II. Comparison 

of the woody vegetation of 98 areas. Edinb. J. Bot. 53:153–180.



Ratter, J.A. and T. C. D. Dargie. 1992. An analysis of the floristic composition 

of 26 Cerrado areas in Brazil. Edinb. J. Bot. 49:235–250. 

Raw, A. 1998. Relatório sobre número de insetos, a riqueza de espécies e 

aspectos zoogeográficos nos cerrados. Workshop Ações Prioritárias para 

Conservação da Biodiversidade do Cerrado e Pantanal. Brasília: Ministério 

do Meio Ambiente, FUNATURA, Conservation International, 

Fundação Biodiversitas, Universidade de Brasília. Web: www.bdt.org.br/ 

workshop/cerrado/br. 

Redford, K. H., and G. A. B. Fonseca. 1986. The role of gallery forests in the 

zoogeography of the Cerrado’s non-volant mammalian fauna. Biotropica 

18:126–135. 

Reid, J. W. 1994. The harpacticoid and cyclopoid copepod fauna in the cerrado 

region of Central Brazil: 1. Species composition, habitats, and zoogeography. 

Acta Limnol. Bras. 6:56–68. 

Ribeiro, J. F. (ed.) 1998. Relatório final do grupo temático Botânica. Workshop 





Ribeiro, M. 1998. Conservação e uso sustentável da biota aquática do cerrado 

e pantanal: Relatório para o grupo temático Biota Aquática. Workshop 





Saint-Hilaire, A. 1847–1848. Voyages Dans l’interieur du Brésil: III part. 

Voyage aux Sources du Rio S. Francisco et Dans la Province de Goyaz. 

Paris: Arthus Bertrand. 

Sawyer, D. (ed.). 1998. Diagnósticos sobre temas econômicos e sociais na 

região do Cerrado/Pantanal: Relatório final, grupo temático Sócio- 

Economia. Workshop Ações Prioritárias para Conservação da Biodiversidade 




Secretaria do Meio Ambiente de São Paulo. 1997. Cerrado: Bases para Conservação 

e Uso Sustentável das Áreas de Cerrado do Estado de São 

Paulo. São Paulo: Secretaria de Estado do Meio Ambiente, SP, Brazil. 

Sick, H. 1955. O aspecto fisionômico da paisagem do médio rio das Mortes, 

Mato Grosso, e a avifauna da região. Arq. Mus. Nac. 42:541–576. 

Silva, J. M. C. 1989. “Análise Biogeográfica da Avifauna de Florestas do 

Interflúvio Araguaia-São Francisco.’’ Master’s thesis, Universidade de 


Silva, J. M. C. 1995a. Birds of the Cerrado Region, South America. Steenstrupia 

21:69–92.



Silva, J. M. C. 1995b. Avian inventory of the Cerrado Region, South America: 

Implications for biological conservation. Bird. Conserv. Int. 

5:291–304. 

Silva, J. M. C. (ed.). 1998. Relatório final do grupo temático Aves. Workshop 

Ações Prioritárias para Conservação da Biodiversidade do Cerrado e 

Pantanal. Brasília: Ministério do Meio Ambiente, FUNATURA, Conservation 

International, Fundação Biodiversitas, Universidade de Brasília. 

Web: www.bdt.org.br/workshop/cerrado/br. 

Straube, F. C. 1998. O cerrado no Paraná: ocorrência original e atual e subsídios 

para sua conservação. Cadernos de Biodiversidade, Instituto 

Ambiental do Paraná 1(2):12–24.

Ecology and Natural History of a Neotropical Savanna - LERF - USP

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?