Gordon and Prins 2008 - Ecology of Grazers and Browser PDF

Ecological Studies 195
Iain J. Gordon, Herbert H.T. Prins

Editors
The Ecology of
Browsing and Grazing

Ecological Studies, Vol. 195
Analysis and Synthesis
Edited by
M.M. Caldwell, Washington, USA
G. Heldmaier, Marburg, Germany
R.B. Jackson, Durham, USA
O.L. Lange, Würzburg, Germany
H.A. Mooney, Stanford, USA
E.-D. Schulze, Jena, Germany
U. Sommer, Kiel, Germany
Ecological Studies
Volumes published since 2003 are listed at the end of this book.
I.J. Gordon • H.H.T. Prins
Editors
The Ecology of Browsing

and Grazing
Iain J. Gordon Herbert H.T. Prins
CSIRO Davies Laboratory Resource Ecology Group
PMB PO Aitkenvale Wageningen University
Qld 4814 Droevendaalsesteeg 3a
Australia 6708 PB Wageningen
The Netherlands
Cover illustration: Zebras alerted by a predator in Mana Pools National Park in the Zambezi Valley,
Zimbabwe. (Photo Iain J. Gordon)
ISSN 0070-8356
ISBN 978-3-540-72421-6 e-ISBN 978-3-540-72422-3
Library of Congress Control Number: 2007931594
This work is subject to copyright. All rights are reserved, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting,
reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication
or parts thereof is permitted only under the provisions of the German Copyright Law of September 9,
1965, in its current version, and permissions for use must always be obtained from Springer-Verlag.
Violations are liable for prosecution under the German Copyright Law.
Springer is a part of Springer Science+Business Media
springer.com
© Springer-Verlag Berlin Heidelberg 2008
The use of general descriptive names, registered names, trademarks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
Editor: Dr. Dieter Czeschlik, Heidelberg, Germany

Desk editor: Dr. Andrea Schlitzberger, Heidelberg, Germany
Cover design: WMXDesign GmbH, Heidelberg, Germany
Production and typesetting: SPi Publisher Services
Printed on acid-free paper SPIN 11382201 31/3180 5 4 3 2 1 0

Foreword
About fifty years ago, when I, as a young comparative anatomist, first looked at a
wild ruminant – the European roe deer – the basic thinking concerning the ecology,
behaviour, physiology and anatomy of ruminants was based on domesticated grazers,
namely sheep and cattle. I could not believe the customary view that the roe deer was
nothing more than a mini-cow with a choosy predilection for flowering herbs, tender
leaves and shoots. Already my comparison of a red deer stomach with that of a roe
deer caused me to bring to mind the different evolutionary traits of cervids as com-
pared with bovids, of which Europe has but a few wild species left. At that time, there
was no thought of integrated management of vegetation and herbivores: hunters
aimed for higher game densities, foresters considered (and still do) browsers a pest,
to be reduced if not eliminated from their planted forests, and advocates of animal
welfare agitated against hunting. All this has negatively influenced any serious
attempt to develop sustained yield concepts, certainly in Central Europe.
Thus I was overwhelmed by the living demonstration of bovid evolutionary
‘explosion’ and niche separation between extant species, when I came to study
large herbivores in East Africa for ten years prior to decolonisation (the ‘Uhuru’ of
1963). When I first presented some of my morphological findings on African
herbivores at a London Symposium in 1966, the audience encouraged me to extend
and deepen my observations systematically. This lead in 1972, initially in collabo-
ration with the British botanist and wildlife researcher Don Stewart, to a classification
of ruminants into three feeding types – first recognising a dichotomic evolution
with numerous intermediate forms, a system in common use today.
We have to remember that mammalian digestive tracts (of carnivores, omnivores
and herbivores) are extremely set and conservative – the result of evolution; this is
especially the case with foregut-fermenting herbivores. Ruminant evolution beyond
tragulids proceeded over more than 25 million years apparently not step by step
(like a ladder), but frequently in parallel fashion, like the growth pattern of a bush
or baobab tree. This is why we find extant frugivorous and browsing concentrate
selectors (both large and small) in dominant numbers in three of the four ruminant
families, but almost no true bulk and roughage grazers (except the Père David’s
deer) amongst the cervids. In contrast, we see most of the grazers, stimulated by
changing climate and following the spread of the grasses, amongst the Bovidae.
v
vi Foreword
Browsers, irrespective of the family they belong to, have retained their archaic
morphophysiological features, which evolved before grasses became the dominant
plants under the then-prevailing climatic conditions. Browsers are poorly adapted to
digesting the structural carbohydrates within grasses, yet browsers have successfully
remained within the large herbivore spectrum for more than 10 million years. If it is
the browsers that, according to the elaborate analyses and conclusions of this stimulat-
ing book, will be the prime winners of the global future, one can only hope and pray
that the type of collaboration between scientists and ecosystem managers (including
foresters and agronomists) which the editors appeal for will come to pass.
After a long and active life in the field of basic and applied herbivore wildlife
research, I feel honoured and encouraged by the authors and especially by the
editors of this future-oriented volume to contribute a foreword, with all my good
wishes for a worldwide positive reception not only of this book but also of the
fascinating animals which, I strongly believe, must remain the gentle modifiers
of our landscapes and perhaps even of our anthropocentric view of this world’s
nature.
Berlin, Baruth Reinhold R. Hofmann
Dr.med.vet.
Professor emeritus
Contents
1 Introduction: Grazers and Browsers

in a Changing World . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Herbert H.T. Prins and Iain J. Gordon
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Dominance of Domesticated Grazers and Browsers. . . . . . . . . . . . . . 3
1.3 The Last 30 Years and the Immediate Future . . . . . . . . . . . . . . . . . . . 5
1.4 Societal Relevance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2 An Evolutionary History of Browsing

and Grazing Ungulates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Christine Janis
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.1.1 Ungulate Phylogeny and Evolutionary History . . . . . . . . . . . 22
2.1.2 Determination of Feeding Adaptations . . . . . . . . . . . . . . . . . . 26
2.1.3 Cenozoic Changes in Climate . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.2 Fossil Record Evidence of Dietary Evolution in Ungulates. . . . . . . . 32
2.2.1 Early Archaic Ungulates and Ungulate-like
Mammals (65 to 40 Ma). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.2.2 The Eocene Emergence of Modern Ungulates
(55 to 34 Ma) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.2.3 The Lull Before the Storm: Oligocene
and Early Miocene Times (34 to 20 Ma) . . . . . . . . . . . . . . . . 34
2.2.4 The Rise of the Grasslands (20 to 10 Ma) . . . . . . . . . . . . . . . 35
2.2.5 The Rise of Grazing Ungulates (10 to 2 Ma) . . . . . . . . . . . . . . 37
2.2.6 The Late Cenozoic Dawn of the Modern
Ungulate Fauna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.3 Discussion and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
vii
viii Contents
3 The Morphophysiological Adaptations of Browsing

and Grazing Mammals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Marcus Clauss, Thomas Kaiser, and Jürgen Hummel
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.2 Grass and Browse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.3 Predictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.4 Testing the Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.5 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.6 Conclusion and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
4 Nutritional Ecology of Grazing and Browsing Ruminants . . . . . . . . . . 89

Alan J. Duncan and Dennis P. Poppi
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.2 Nutritive Value of Plant Material for Ruminant Herbivores. . . . . . . . 90
4.2.1 Digestion of Plant Material by Herbivores . . . . . . . . . . . . . . . 90
4.2.2 How do Browse and Grass Differ in Nutritive Value? . . . . . . 90
4.2.3 Plant Secondary Metabolites . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.3 Do Browsers and Grazers Differ in the Way They
Process Their Food? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
4.3.1 Reticulo-Rumen Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.3.2 Retention Time and Passage Rate . . . . . . . . . . . . . . . . . . . . . . 100
4.3.3 Mean Particle Size Escaping Rumen. . . . . . . . . . . . . . . . . . . . 101
4.3.4 Absorptive Surface Area of Rumen. . . . . . . . . . . . . . . . . . . . . 102
4.3.5 Saliva Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.3.6 Fermentation Rate and Fibre Digestibility
in the Rumen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.3.7 Rumen Microbial Genetic Profiles
and Digestion End Products. . . . . . . . . . . . . . . . . . . . . . . . . . . 105
4.4 Problems with the Meta-Analysis Approach. . . . . . . . . . . . . . . . . . . . 106
4.5 Detoxification of Plant Secondary Metabolites:
Do Browsers and Grazers Differ? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
4.5.1 Salivary Tannin-Binding Proteins . . . . . . . . . . . . . . . . . . . . . . 107
4.5.2 Ruminal Detoxification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
4.5.3 Post Absorptive Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . 110
4.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
5 The Comparative Feeding Bahaviour of Large Browsing

and Grazing Herbivores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Kate R. Searle and Lisa A. Shipley
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.2 The Functional Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.2.1 Components of the Functional Response . . . . . . . . . . . . . . . . 120
Contents ix
5.3 Foraging in Patches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126

5.3.1 Definition of Patches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.4 Spatial Variation in the Quality of Grass. . . . . . . . . . . . . . . . . . . . . . . 128
5.5 Spatial Variation in the Quality of Browse . . . . . . . . . . . . . . . . . . . . . 130
5.6 Perception of Patches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5.6.1 Evidence for Patch Perceptions
of Grazers and Browsers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.6.2 Patch Perceptions by Grazers . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.6.3 Patch Perceptions by Browsers . . . . . . . . . . . . . . . . . . . . . . . . 136
5.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
6 The Comparative Population Dynamics of Browsing

and Grazing Ungulates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Norman Owen-Smith
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
6.2 Spatial and Temporal Dynamics of Grass and Browse. . . . . . . . . . . . 150
6.3 Population Density Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
6.4 Demographic Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
6.5 Regulation of Abundance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
6.6 Population Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
6.7 Weather Patterns and Population Fluctuations . . . . . . . . . . . . . . . . . . 168
6.8 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
7 Species Diversity of Browsing and Grazing Ungulates:

Consequences for the Structure and Abundance
of Secondary Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Herbert H.T. Prins and Hervé Fritz
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
7.2 Suggested Causes of Species Richness . . . . . . . . . . . . . . . . . . . . . . . . 181
7.3 The Effect of Species Richness on Ecosystem Functioning:
An Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
7.4 Herbivore Diversity and the Use of Primary Production . . . . . . . . . . 186
7.4.1 Diet Overlap and Feeding Niches . . . . . . . . . . . . . . . . . . . . . . 186
7.4.2 Postulated Advantages of Mixed-Species Feeding. . . . . . . . . 189
7.5 Mammalian Herbivore Species Richness
Links to Secondary Productivity and Biomass . . . . . . . . . . . . . . . . . . 190
7.5.1 Domestic Herbivore Diversity and Secondary
Productivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
7.5.2 Diversity–Biomass Relationship in Wild Assemblages . . . . . 193
7.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
x Contents
8 Impacts of Grazing and Browsing by Large Herbivores

on Soils and Soil Biological Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
Kathryn A. Harrison and Richard D. Bardgett
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
8.2 Herbivore Effects on Nutrient Dynamics. . . . . . . . . . . . . . . . . . . . . . . 201
8.3 Positive Feedback Effects of Above-Ground Herbivory . . . . . . . . . . 202
8.3.1 Urine and Dung Deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
8.3.2 Alterations in Plant C and N Allocation . . . . . . . . . . . . . . . . . 206
8.3.3 Selective Foraging on Less Nutritious Species. . . . . . . . . . . . 208
8.3.4 Litter Deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
8.3.5 Increased Soil Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
8.4 Negative Feedback Effects of Above-Ground Herbivory. . . . . . . . . . 209
8.4.1 Selective Foraging on Nutrient-Rich Tissue . . . . . . . . . . . . . . 210
8.4.2 Production of Secondary Metabolites . . . . . . . . . . . . . . . . . . . 211
8.4.3 Impact on Soil Physical Properties . . . . . . . . . . . . . . . . . . . . . 212
8.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
9 Plant Traits, Browsing and Gazing Herbivores,

and Vegetation Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Christina Skarpe and Alison J. Hester
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
9.2 Plant Architecture and Herbivory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
9.2.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
9.2.2 Trees and Shrubs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
9.2.3 Herbaceous Plants Other Than Graminoids . . . . . . . . . . . . . . 220
9.2.4 Graminoids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
9.3 The Chemistry of Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
9.3.1 Photosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
9.3.2 Energy and Nutrient Reserves . . . . . . . . . . . . . . . . . . . . . . . . . 223
9.4 Plant Resistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
9.4.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
9.4.2 Escape Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
9.4.3 Structural and Chemical Defences . . . . . . . . . . . . . . . . . . . . . 227
9.4.4 Plant Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
9.5 Effects of Herbivory on Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
9.5.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
9.5.2 Reserve Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
9.5.3 Sprouting and Resprouting. . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
9.5.4 Repeated Herbivory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
9.5.5 Compensatory Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
9.5.6 Regeneration and Persistence . . . . . . . . . . . . . . . . . . . . . . . . . 236
9.6 Herbivore Foraging Behaviour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
9.6.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
9.6.2 Hierarchical Foraging. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
Contents xi
9.6.3 Large Herbivores and Predators . . . . . . . . . . . . . . . . . . . . . 238

9.6.4 The Importance of Neighbours . . . . . . . . . . . . . . . . . . . . . . 238
9.7 Large Herbivore Effects on Vegetation . . . . . . . . . . . . . . . . . . . . . . 239
9.7.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
9.7.2 Herbivory and Composition of Plant Populations . . . . . . . 239
9.7.3 Herbivory and the Composition
of Plant Communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
9.8 The Theory of Vegetation Change . . . . . . . . . . . . . . . . . . . . . . . . . 244
9.8.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
9.8.2 Succession Theories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
9.8.3 The State-and-Transition Model. . . . . . . . . . . . . . . . . . . . . 245
9.8.4 State-and-Threshold and Catastrophe Theories . . . . . . . . . 246
9.9 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
10 The Impact of Browsing and Grazing Herbivores

on Biodiversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
Spike E. van Wieren and Jan P. Bakker
10.1 Biodiversity and Large Mammalian Herbivores . . . . . . . . . . . . . . . 263
10.2 Effects on Traits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
10.3 Effects on Plant Communities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
10.4 Effects on Invertebrates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
10.5 Effects on Birds. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
10.6 Effects on Mammals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
10.7 Large Herbivores and Biodiversity . . . . . . . . . . . . . . . . . . . . . . . . . 283
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
11 Managing Large Herbivores in Theory and Practice: Is the

Game the Same for Browsing and Grazing Species? . . . . . . . . . . . . . . 293
Jean-Michel Gaillard, Patrick Duncan, Sip E. Van Wieren,
Anne Loison, François Klein, and Daniel Maillard
11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
11.2 The Dynamics of Grazer and Browser Populations . . . . . . . . . . . . 295
11.3 Monitoring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
11.4 Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
11.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
12 Grazers and Browsers in a Changing World: Conclusions . . . . . . . . . 309

Iain J. Gordon and Herbert H.T. Prins
12.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
12.2 Responses in Plant Species Composition . . . . . . . . . . . . . . . . . . . . 310
12.3 Responses in Plant Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
12.4 Responses in Terms of Population Dynamics . . . . . . . . . . . . . . . . . 312
12.5 Responses in Herbivore Community Structure . . . . . . . . . . . . . . . . 313
xii Contents
12.6 Ways of Managing Browsers and Grazers. . . . . . . . . . . . . . . . . . . . 316

12.6.1 Managing for Variability Rather than Stability . . . . . . . . 316
12.6.2 Gardening Versus Laisser Faire . . . . . . . . . . . . . . . . . . . . . 317
12.7 Where Do We Go From Here? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
12.7.1 Landscape Scale Experiments . . . . . . . . . . . . . . . . . . . . . . 318
12.8 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
Subject Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
Species Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327

Contributors
Jan P. Bakker
Community and Conservation Ecology Group, University of Groningen, P.O. Box
14, 9750 AA, Haren, The Netherlands
Marcus Clauss
Division of Zoo Animals, Exotic Pets and Wildlife, Vetsuisse Faculty, University
of Zurich, Winterthurerstr. 260, 8057 Zurich, Switzerland, mclauss@vetclinics.
unizh.ch
Alan J. Duncan
Macaulay Institute, Craigiebuckler, Aberdeen AB15 8QH, Scotland, UK, a.duncan@
macaulay.ac.uk
Patrick Duncan
Centre d’Etdudes Biologiques de Chizé, Centre National de la Recherche
Scientifique, Villiers-en-Bois, 79360 Beauvoir-sur-Niort, France
Hervé Fritz
Centre d’Études Biologiques de Chizé, CNRS UPR 1934, 79360
Beauvoir-sur-Niort, France, fritzh@cebc.cnrs.fr
Jean-Michel Gaillard
Laboratoire de Biométrie et Biologie Évolutive (Unité Mixte de Recherche
N° 5558), Centre National de la Recherche Scientifique, Université
Lyon 1, 43 Boulevard du 11 Novembre, 69622, Villeurbanne Cedex, France,
capreolus@wanadoo.fr
Iain J. Gordon
CSIRO - Davies Laboratory, PMB PO Aitkenvale, Qld 4814, Australia, iain.gordon@
csiro.au
Kathryn A. Harrison
Institute of Environmental and Natural Sciences, Soil and Ecosystem Ecology
Laboratory, Department of Biological Sciences, Lancaster University,
Lancaster LA1 4YQ UK, k.a.harrison@lancs.ac.uk
xiii
xiv Contributors
Alison J. Hester
Macaulay Institute, Craigiebuckler, Aberdeen, AB15 8QH, UK
Jürgen Hummel
Institute of Animal Science, Department of Animal Nutrition, University of Bonn,
Endenicher Allee 15, 53115 Bonn, Germany
Christine Janis
Department of Ecology and Evolutionary Biology, Brown University, Providence,
RI 02912, USA, Christine_Janis@Brown.edu
Thomas Kaiser
University of Hamburg, Biozentrum Grindel and Zoological Museum, Martin-
Luther-King-Platz 3, 20146 Hamburg, Germany
François Klein
Office National de la Chasse et de la Faune Sauvage, Centre National d’Étude et
de Recherche Appliquée, 1 Place Exelmans, 55000 Bar-le-Duc, France
Anne Loison
Laboratoire de Biométrie et Biologie Évolutive (Unité Mixte de Recherche
N° 5558), Centre National de la Recherche Scientifique, Université Lyon 1,
43 boulevard du 11 novembre, 69622, Villeurbanne Cedex, France
Daniel Maillard
Office National de la Chasse et de la Faune Sauvage, Centre National d’Étude et
de Recherche Appliquée, 95 rue Pierre Flourens, BP 74267, 32098 Montpellier
Cedex 05, France
Norman Owen-Smith
Centre for African Ecology, School of Animal, Plant and Environmental Sciences,
University of the Witwatersrand, Wits 2050, South Africa,
norman@gecko.biol.wits.ac.za
Dennis P. Poppi
Schools of Animal Studies and Veterinary Science, University of Queensland,
St Lucia 4072, Brisbane, Australia
Herbert H.T. Prins
Resource Ecology Group, Wageningen University, Droevendaalsesteeg 3a, 6708
PB Wageningen, The Netherlands, herbert.prins@wur.nl
Kate R. Searle
CSIRO - Sustainable Ecosystems, Davies Laboratory, University Drive,
Annandale, QLD 4814, Australia, kate.searle@csiro.au
Lisa A. Shipley
Department of Natural Resources, Washington State University, Pullman,
Washington 99163, USA
Contributors xv
Christina Skarpe
Hedmark University College, Faculty of Forestry and Wildlife Management,
2480 Koppang, Norway, Christina.Skarpe@nina.no
Sip E. Van Wieren
Resource Ecology Group, Wageningen University, Droevendaalsesteeg 3a, 6708
PB Wageningen, The Netherlands, Sip.vanWieren@wur.nl
Chapter 1
Introduction: Grazers and Browsers
in a Changing World
Herbert H.T. Prins and Iain J. Gordon
1.1 Introduction
“During the second half of the 20th century, the global population explosion was the big
demographic bogey … Now that worry has evaporated, and this century is spooking itself
with the opposite fear: the onset of demographic decline. The shrinkage of Russia and
eastern Europe is familiar, though perhaps not the scale of it: Russia’s population is
expected to fall by 22% between 2005 and 2050, Ukraine by a staggering 43%. Now the
phenomenon is creeping into the rich world: Japan has started to shrink and others, such
as Italy and Germany, will soon follow. Even China’s population will be declining by the
early 2030s, according to the UN which projects that by 2050 populations will be lower
than they are today in 50 countries.… People should not mind, though. What matters for
economic welfare is GDP per person ….. The new demographics that are causing
populations to age and to shrink are something to celebrate [not in Russia or Ukraine
though HP & IG]. Humanity was once caught in the trap of high fertility and high
mortality. Now it has escaped into the freedom of low fertility and low mortality.
Women’s control over the number of children they have is an unqualified good – as is the
average person’s enjoyment, in rich countries, of ten more years of life than they had in
1960. Politicians [and companies] may fear the decline of their nations’ economic prowess,
but people should celebrate the new demographics as heralding a golden age”.
(the Economist Editorial 7 January 2006 p 12).
If humanity is at the brink of a ‘Golden Age’, what then is the divination for
nature? In this book we will try to foretell how wild herbivores will react to the
changes that take place in the world in which they live. That world has been
changing since it formed, and for millions of years its plants have been consumed
by herbivores. This has led to adaptations in plants in reaction to herbivory, and
our present-day species assemblages and landscapes are a manifestation of the
forces of natural selection that have been in operation for a very long time. Over
a much shorter time, these landscapes have been heavily impacted upon by
humans; firstly, by accidentally burning patches of the landscape, but later as a
tool to modify that landscape either for capturing or luring game or even for
changing the species composition towards a modified vegetation that yields
desired produce (e.g., cultivation of hazel in the Mesolithic; Simmons et al.
1981). It is now increasingly clear that the ‘primordial’ Amazonian, Middle
American and other rainforests have been strongly modified by local people
I.J. Gordon and H.H.T. Prins (eds.), The Ecology of Browsing and Grazing. 1
© Springer 2008
2 H.H.T. Prins and I.J. Gordon
(Noble and Dirzo 1997); indeed when the first European colonists came to what
is now called New England on the eastern seaboard of the USA the forests were
so open and ‘Arcadian’ that settlers thought it had been created especially for
them (Cronon 1983; Prins 1994; cf. Motzkin and Foster 2002; Rusell and Davis
2001). In a way that was true since the indigenous people were decimated through
diseases involuntarily brought to them by the newcomers, but their hunter-
gatherer imprint was still there (e.g., Douglas and Hoover 1988). The same is true
for what was first named New Holland and later Australia: non-agricultural
people that had settled there some 40,000 years ago had been using fire to modify
that landscape for an uncounted number of generations (e.g., Hughes 1987 p 3;
Lewis 2002). Wild grazers and browsers that have survived to the present day
must have found a way to cope with many of these man-induced changes.
Indigenous plants that had been shaped by natural selection reacted to the new
forces, and new types of vegetation emerged. Really big changes started with
the emergence of agriculture; not only did plant communities get modified because
the competitive interaction between the domesticated and native plant species
changed, but land was cleared and often alien species were introduced for food
consumption or involuntarily. Many of these species are now considered part of the
native flora and only palynological and archaeological research can reveal their
non-local provenance and their alien roots. This can be done where the soil archive
is good enough to yield the necessary information, and especially where there are
sufficient sources and academics to unravel the arcane history of local plant
communities, as for example in Europe (e.g., Godwin 1975; Knörzer 1971, 1975;
Opravil 1978; Pennington 1969; Van Zeist 1980). In other places, it is the scourge
of introduced species that became an economic threat which has fostered a desire
to find out their origin (e.g., McFadyen and Skarratt 1996).
Major changes started when early farmers domesticated sheep, goats, cattle,
onagers, and donkeys in the Middle East, dromedaries in the Arabian Peninsula,
horses in southern Russia, camels and yaks in Central Asia, lamas in South
America, water buffalo and zebu in South Asia, gaur, gayal, and banteng in
Southeast Asia, and perhaps the latest to be domesticated, reindeer in northern
Scandinavia and northern Siberia (e.g., Legge 1996; Köhler-Rollefson 1996;
Zeuner 1967). In many of the centres of origin, the use of domestic grazers and
browsers in the already modified landscapes led to a landscape that would have
been unrecognisable to earlier generations of Man. Superimposed on the changes
brought about by climatic change, fire, and felling, the impact of domesticated
indigenous browsers and grazers caused forests to disappear first from the Zagros
Mountains (e.g., Hole 1996) or in the Lebanon, and then from across that whole
range of landscapes of much of Europe (Prins 1998), the high altitude areas of
South America, China, and Japan.
As people moved across the landscape in prehistoric times they took with them
their domesticated browsers and grazers. The rate of this spread was about 20 km per
generation; the spread of farming (or farmers) from the Levant across Europe was
1 km.yr−1 (Cavalli-Sforza 1996) while the rate for pastoralism across Africa was
0.9 km.yr−1 (Prins 2000). The result was that by the beginning of our Common Era
1 Introduction: Grazers and Browsers in a Changing World 3
across most of the Old World domesticated grazers and browsers (including so-called
mixed feeders) were modifying indigenous vegetation communities, but two
continents stayed free of domestic livestock, namely Australia and North America.
1.2 Dominance of Domesticated Grazers and Browsers
With the advent of European colonialism, Australia and the Americas became an
open access area to the grazers and browsers from the Old World. In North America
the successful newcomers were cattle, horses, and sheep; in South America cattle,
zebu, water buffalo, and sheep took over much of the grasslands, while in Australia
introduced sheep, cattle, water buffalo, banteng, camels, donkeys, and horses lived
side-by-side, while kangaroos expanded their range as a reaction to surface water
becoming available through dams and boreholes. In other places, other non-native
grazers and browsers, such as deer and thar (New Zealand) or reindeer (South
Georgia) were introduced for sport. The newcomers invaded niches of local mam-
mals (e.g., Breebaart et al. 2002; Dawson et al. 1992; Edwards et al. 1996; Escobar
and Gonzalez 1976; Fritz et al. 1996; Genin et al. 1994; Hubbard and Hansen
1976; Lightfoot and Posselt 1977; Prins 2000; Schwartz and Ellis 1981; Thill and
Martin 1986).
By the end of the 19th century, the world as we know it took further shape:
railway lines opened up the vast prairies of Canada and the United States, and
steamships made it possible to start commercially transporting agricultural produce
to metropolises where an industrial revolution spurred human population growth.
In a series of Homestead Acts, pioneers staked out 110 million hectares of land
(Anon. 2005c), which is about 30–50 times the size of countries like the
Netherlands, Belgium, Denmark, or Switzerland. As part of the same industrial
revolution, railway lines and steamships led to a massive exodus of people from the
Ukraine, Poland, Germany, Scandinavia, Scotland, Italy, England, France, Ireland,
the Low Countries, and Spain to the New World and Australia. Many of these
people went to the cities, but many went also to become farmhands or farmers,
cowboys or ranchers. In Argentina, Uruguay, and Paraguay native people were
pushed off their land, just as they were in the United States and Canada, Australia,
South Africa, and Zimbabwe. Along the Trans-Siberia railway, the Russian Far
East was ‘opened up’. As a consequence the human population burgeoned,
demanding more and more meat, milk, and fibre. Australia and New Zealand
became world leaders for the production of mutton and wool, Chicago became the
beef capital of the world, and horses and cattle were exported in enormous numbers
from the southern part of South America for meat. Horses were likewise exported
from Poland to Western Europe for meat and draught power. The end of the 19th
and the beginning of the 20th century saw an enormous demand for horses, not only
for transport but also for the war machinery of clashing empires; some 8 million
horses were killed at only the Western Front during the Great War. The net result
of all these transformations was an enormous growth of domestic browsers and graz-
ers, while indigenous wild ungulates declined or even went extinct, such as the
Bluebuck (Hippotragus leucophaeus) and Quagga (Equus quagga) in southern
Africa, the Aurochs (Bos primigenius) and the Wild (forest) horse (E. caballus) in
Europe, the wild Dromedary (Camelus dromedarius) in Asia, and in Australia the
Eastern hare wallaby (Lagorchestes leporides), Central hare wallaby (L. asoma-
tus), Toolache (Macropus greyi), and Crescent nailtail wallaby (Onychogalea
lunata). Ungulates that reached the brink of extinction during the last century were
the European bison (Bison bonasus), American bison (Bison bison); Père David’s
deer (Elaphurus davidianus) in China; Swamp deer (Cervus duvaucellii) in India;
Camel (Camelus bactrianus) and Przewalski horse (Equus przewalskii) in Central
Asia; Wild Asian buffalo (Bubalus bubalis), Gayal (Bos frontalis), Kouprey (Bos
sauveli) in South East Asia; Black wildebeest (Connochaetus gnou) and White rhi-
noceros (Ceratotherium simum) in South Africa. In 2005 the northern form of the
white rhino was declared extinct. Many grazers and browsers became very rare and
are presently listed as endangered on the IUCN’s Red List. Heavy hunting is cited
as the cause of these (near) extinctions, in other cases it has been clearing of indige-
nous vegetation, or the introduction of new predators such as Red fox (Vulpes
vulpes) in Australia. A new cause is lawlessness associated with failing nation
states or civil war, but competition with domestic stock is a seriously underrated
cause. So, what about this heralded ‘Golden Age’?
Natural grasslands have nearly been wiped off the face of the earth. For exam-
ple, in the Ukraine only about 4% of the steppes were found to remain in the origi-
nal state (Goriup 1998), the South African low veld is natural in about 11% of its
extent only (Low and Rebello 1996), the North American tall-grass prairie is all but
gone (Packard and Mutel 1997), while in SE Australia a stunning 99.5% of its
native grasslands have been converted for agricultural use (Taylor 1998). World-
wide, forests have decreased too, from about 6.5 billion hectares to 3.5 billion since
the rise of agriculture-based civilizations (Noble and Dirzo 1997). At present, about
15% of the Earth’s land surface is occupied by row-crop agriculture or by urban-
industrial areas, and another 6–8% has been converted to pastureland (Vitousek et al.
1997). Estimates of the fraction of land transformed or degraded by humanity, and
the fraction of the land’s biological production that is used by Man is about 40–50%
(Vitousek et al. 1997). Managed grazings cover more than 25% of the global land
surface and has a larger geographical extent than any other form of land use (Asner
et al. 2004). In the beginning of the 21st century CE, nearly half the nitrogen atoms
in the protein of an average human being’s body came at some time or another
through an ammonia factory, often using the Haber-Bosch process, invented in
1909, combining nitrogen from the air with hydrogen from coal (Anon. 2005a).
So even though landscapes were getting transformed from an often rather forested
state to a landscape dominated by agriculture, the potential for increased oppor-
tunities were garnered by a small subset of browsers and especially grazers, namely
by those few species that were useful to man. When artificial fertilisers were
introduced on a massive scale in a number of countries, grassland productivity
increased so much that now agricultural economies in the rich countries are reliant on
only three domestic species, all three grazers, namely cattle, sheep, and horses.
In the United States, the European Union and also the former Eastern bloc,
southern Africa, Australia and New Zealand agricultural policies then started to
distort markets at a massive scale in the 1960s. World markets would have dictated
a shift away from agriculture in these countries because unsubsidised agriculture
became, to a large extent, not profitable if farmers in these areas had to provide their
products at world market prices but had to pay labour costs as dictated by the local
society’s norms. Governments then started subsidising agriculture for two major
reasons, namely, to maintain adequate national self-sufficiency and to guarantee
farmers (and their dependent communities) a sufficiently high income so that they
could continue to participate in society at large. These countries maintained a land-
scape geared towards maximum agricultural production and very high numbers of
domestic grazers. However, there has been a substantial reduction in the numbers
of horses in the world over the 20th century as horse power became transplanted by
engine power. The first tractors had few advantages over the best horses, but they
did not eat hay or oats. The replacement of draft animals by machines released
about 25% more land for growing food for human consumption in the middle of the
20th century (Anon. 2005a). Cattle and sheep remained the dominant ungulates in
temperate, semi-arid, or arid regions of the globe. In most of Africa and the Middle
East goats continued to play an important role in addition to sheep, but for drome-
daries there was less of a place as they too were replaced by engine power. South
Asia remained dominated by zebu cattle and to a lesser extent water buffalo, while
in Southeast Asia it is the other way around; but water buffalo are decreasing in
numbers in Indonesia at a high rate. Worldwide, fewer and fewer ungulate species
dominate the herbivore communities that modify the vegetation.
1.3 The Last 30 Years and the Immediate Future
Since about 1980 major changes in the non-urban landscapes of Europe and
North America have taken place because arable- and livestock-based agricul-
ture is no longer cost-effective; the countryside is becoming devoid of people
who use the local resources while an ever increasing number of people move to
urban metropolises or peri-urban neighbourhoods (Fig. 1.1). The same trend is
visible in other developed countries, such as South Africa, Japan, Thailand, and
Australia. Nowadays proportionally more and more people live in cities and
towns: in Australia about 85% of the people are living in urban areas (Anon.
2006a), in France it is over 80% (Anon. 2005b), in the most developed part of
the world it is on average 76%, and worldwide it is now 40% (United Nations
2006). Fewer people realise, however, that the absolute numbers of people living
in rural landscapes is decreasing and, as a result, the number of abandoned villages
in France, Italy, Spain, and Portugal is increasing. Even when British or Dutch
pensioners take over property in these villages, the surrounding country side
remains unused. According to preliminary results of the first national Census
since 1989, more than half of Russia’s 155,290 villages are abandoned or
Developed (urban)
Developed (rural)
Developing (urban)
4,500,000
Developing (rural)
4,000,000
3,500,000
Population size
3,000,000
2,500,000
2,000,000
1,500,000
1,000,000
500,000
1940 1960 1980 2000 2020 2040
Fig. 1.1 Estimated number of people since 1940 in different parts of the world, and the predicted
number up to 2040. (Population Division of the Department of Economic and Social Affairs of the
United Nations Secretariat, World Population Prospects: The 2004 Revision and World
Urbanization Prospects: The 2003 Revision, http://esa.un.org/unpp)
populated by 50 or fewer stragglers (Anon. 2006b). In Wyoming, Nebraska, or

Colorado one can drive through ghost towns: “Virtually all of the 20 poorest
counties in America, in term of wages, are on the eastern flank of the Rockies
or on the western Great Plains… The area does include several pockets of
wretched Native American poverty, but in most of [these] areas the poor are as
white as a prairie snowstorm. …We are seeing a re-acceleration of the population
decline. After the kids left in the 1980s and 1990s, things sort of leveled off.
Now it’s just plain attrition. People are dying. It is fairly common nowadays for
rural counties across America to lose people: roughly one in four did in the
1990s” (Anon. 2005c). In upstate New York and Maine people have voted with
their feet: if one walks through the dense forests one stumbles over stone
fences that, only a century ago, separated the potato fields of local crofters (Flinn
et al. 2005; Litvaitis 1993; Motzkin and Foster 2002; Ramankutty and Foley
1999). In the drier parts of South Africa sheep farms are transformed into wildlife
sanctuaries because of the high risks of stock theft or because of other factors that
make it uneconomic to continue farming (e.g., Davies 2000). In Australia, one
can drive through villages and towns that have been abandoned (cf. Newman 2005).
In “Old Europe” (the European Union of 10 countries), the utilized agricultural
area diminished by more than 3 million hectares between 1975 and 1987; between
1980 and 1990 this was 1,251,000 ha in Germany, 1,000,000 ha in France, and
307,000 ha in Italy (EC Commission data). The surface area of permanent
meadows and grassland have also strongly declined in “Old Europe” (between
1980 and 1990 1 million hectares; EC Commission data). Also the number of
domestic stock declines. Between 1991 and 2001, the number of cattle in
Western Europe stabilised at about 90 million head, in Eastern and Central
Europe it declined from 50 to 30 million head, and in the EECCA countries (the
12 countries from the Caucasus up to Russia) it showed a massive decline from
about 115 to 60 million head. Similar trends are reported for sheep and goats
(Anon. 2003; see also FAO 2006).
The reasons for this abandonment are not the same in every situation, but nearly
always the ultimate cause is that more jobs and an easier life is available in the
cities. Additional factors include lack of children, stock theft, and lack of schools,
hospitals, or other services. Demography, market and neo-liberalism all lead to the
same result: shops close, services disappear, jobs disappear in the villages but are
offered in towns. With the ever-increasing costs of labour as compared to assets,
farmers lay-off farm-hands and use machinery to do the work, which further
decreases the local job markets. Farmers then get more and more dependent on
banks, fodder producers, or equipment sellers for loans, which makes them more
vulnerable in times of adversity. Ultimately this frequently results in bankruptcy.
Politicians have tried to stop this tide by doling out hundreds of billions of dollars,
pounds, liras, guilders, pesetas, yen, kronor, euros, francs, and marks in subsidies
to maintain the countryside in a productive posture. For example, “the [American
Great Plains] region has plainly failed to adapt to a world in which grain and cattle
are cheap. … The crops go out by the truckload to provide jobs elsewhere. How
does the region survive? For all the brouhaha about independence, it leans heavily
on federal government. In 2003, the government spent an average of US$ 10,200
per person in Judith Basin county. North Dakota counties averaged $ 9,000. Most
of it comes in the form of farm subsidies: federal price supports and disaster
payments” (Anon. 2005c).
This has led to an ever-increasing cost for food for the urban people, and an
ever-increasing call for reducing toll barriers around ‘Fortress Europe’.
Agricultural subsidies have decreased already, and will decrease further. To save
the last of the uneconomic farmers, governments dole out subsidies for maintaining
the countryside for the sake of biodiversity, which often is a gotzpe because the
type of biodiversity-rich agricultural landscapes of the early 1900s have been
eradicated by these very same farmers when they were subsidised to meet pro-
duction goals at the cost of the environment. Land abandonment leads to changes
in the landscape to which wild grazers and browsers react, however, the decreas-
ing numbers of domestic stock inhabiting the former extensive farms in developed
countries offers opportunities for wild herbivores to exploit the unconsumed
vegetation resource. White-tailed deer (Odocoileus virginianus) numbers, for
example, have increased dramatically between 1980 and 2000 in many areas of
North America (Riley et al. 2002).
Apart from economic drivers, sociological and demographic changes, and

market distortions, there is a fourth insidious cause of change, namely climate
change, especially the increasing concentrations of carbon dioxide in the atmos-
phere. This change may lead to changed competitive interactions between different
plant groups; carbon dioxide is used in horticulture as an artificial “fertiliser” and
it allows plants to grow for more hours in the day so that more carbohydrates are
produced. It appears as if a number of woody species surpassed a threshold of
starch deposition in their roots, making them much more fire-resistant, because
with sufficiently high carbohydrate reserves these woody species can resprout and
grow quickly out of the fire-trapping zone. This in turn leads to an altered balance
between grasses and woody species, favouring the latter (Bond et al. 2003). This
leads to changes in species composition creating decreasing opportunities for
grazers and increased ones for browsers.
Under a mode of production that encourages high efficiency, land abandonment
and land use intensification are two sides of one coin. “With the [human] popula-
tion growth rates falling sharply while [agricultural] yields continue to rise, even
the acreage devoted to wheat may now begin to decline for the first time since the
stone age.” (Anon. 2005a). Wheat leads the field in world crops. In 2004, the world-
wide area (in millions of hectares) under wheat was about 220, rice about 150,
maize some 140, other cereals about 160, soybeans 90, pulses 70, roots and tubers
50 (FAO 2006). This intensification is much easier to achieve in flat areas than in
montane or steeply hilly areas (Anon. 2005d). The number of farms decreases fast
in OECD-countries. In Europe, most changes take place in mountainous areas
(Batzing et al. 1997; MacDonald et al. 2000). For example in France the number of
farms declined from about 880,000 in 1950 to 220,000 in 2005 (Anon 2005d). Poor,
hilly areas in France, Italy, Spain, and Portugal lost more than half their active pop-
ulation since the 1970s (EC Commission data; Torrano and Valderrabano 2004). In
the north of former East Germany, between 1991 and 1997 the number of people
working in agriculture declined from 296,000 to 164,000 (Shaphiro and Stankevich
2000). Also in Australia the rural population is declining (Bennett et al. 2004), as
it is in Patagonia where sheep stock halved between 1980 and 2000 and hundreds
of farms were abandoned here (Borrelli and Cibils 2005) and in northwestern
Argentina ( Montero and Villalba 2005). In New Zealand’s hill country woody
encroachment takes place because of economic marginalisation (Popay et al. 2002),
and Japan’s countryside also is experiencing depopulation (Liu et al. 2003, Shimoda
2005). In a number of countries, even the absolute population numbers are falling.
Russia’s population is expected to fall by 22% between 2005 and 2050, the Ukraine
by a staggering 43%. Japan has started to shrink in 2005 and by 2050 a 12% fall is
expected, the same fall as is expected for Italy. Germany is expected to have
declined in 2050 by 5%. Even China’s population will be declining by the early
2030s, according to the UN, which projects that by 2050 populations will be lower
than they are today in 50 countries (the Economist, Editorial, 7 January 2006).
Russia’s Far East is quickly depopulating (Kontorovich 2000) but so is Nova Scotia
and parts of Canada’s hinterland (Millward 2005).
The net results of these changes in land use, are to a large extent the focus of
this book. In Portugal, for example, rural depopulation and land abandonment has
led to a significant decline in agricultural land and low shrub and an increase in
tall shrublands and forest, with an increase of 20–40% in fuel accumulation at a
landscape level (Moreira et al. 2001; see also Debussche et al. 1999; Tinner et al.
1998; Preiss et al. 1997; Valderrabano and Torrano 2000). In Table 1.1 we have
summarised different scenarios as we see them, focussing on the balance between
woody species and grasses or herbs. Grazers and browsers react to these changes
(Table 1.2), but they can also be used to prevent changes in vegetation composi-
tion, or to facilitate them. Herbivores, in contrast to many other organisms, can
act as landscape modifiers, but they can only do this to a limited extent. Since
browsers concentrate on woody species and herbs, while grazers focus on grass,
the questions we are asking are:
1. What adaptations do large herbivores have to consuming browse- or grass-
dominated diets?
2. What are the consequences of consuming these diets for large herbivore popula-
tion ecology?
3. Are there difference between grazers and browsers in their impact on ecosystem
structure and functioning?
4. How can we use this information to manage large herbivores in landscapes that
are changing due to anthropogenic and climatic effects?
Land abandonment and woody encroachment are prevailing trends in the rich
countries or in countries of the temperate zones. In the developing countries more
typical for the tropics and sub-tropics, land abandonment is not prevalent yet
although the level of urbanization is still increasing. However, after the expected
peak in human numbers before the middle of the 21st century (Fig. 1.1), it is also
likely that in these areas there will be an expansion of bush and forest.
1.4 Societal Relevance
Are changes in the landscape relevant? Should the public or the natural resource
managers worry about a shift in the balance between woody species and grasses, or
about the rapid increase of wildlife species, or about land abandonment? Is it
important to know whether these shifts in the landscape mosaic can be controlled,
either facilitated or suppressed? There are a number of reasons why we think this
is important:
Water. A shift towards more woody cover often leads to a reduced amount of
water infiltration into the soil, because trees evaporate more water than grasses, and
pine trees or firs continue this evapo-transpiration even in winter (e.g., Persson
1997). This means less water is available for agricultural production and less
aquifer replenishment, so less drinking water and/or process water for industry.
Table 1.1 Different combinations of change lead to different effects on herbivores and the landscape 10
Increased CO2- No depopulation No land aban- Market distor- Very strong No place for wild Very intense use of
levels lead to of rural areas donment tions and intensification herbivores. fertilisers, irrigation,
shift in balance farm subsi- but no bigger Domestic herbivore grass cutting;
between woody dies continue farms; intensi- species richness woody species only
species and fication also far decreases and if necessary (e.g.,
herbaceous layer from markets will only be kept fodder)
indoors
Market distor- Very strong intensi- Place for wild herbiv- More interregional
tions and farm fication and also ores on game farms. variation; some areas
subsidies stop bigger farms; Domestic herbivore very intensively used,
intensification species richness others much less so.
only where it can even increase if Woody encroachment
pays there is a market for in less intensively
novelty products used areas
Land abandon- Market distor- Intensification on Increasing place for Governments can demand
ment tions and bigger farms; wild herbivores. environmental
farm subsi- intensification ‘Wild game’ schemes for
dies continue also far from tolerated if incenti- biodiversity.
markets ves are paid to Discouraging woody
maintain them. encroachment costly
Domestic herbivore
species richness
decreases
Market distor- Bigger farms at the Increasing place for Interregional variation
tions and farm cost of others; wild herbivores. in farm profitability
subsidies stop intensification ‘Wild game’ will increase, and so
only where it tolerated if they will on-farm variation.
pays provide economic Woody encroachment
return. Place for will be tolerated in
wild herbivores on many places within
game farms. the (farm) landscape
H.H.T. Prins and I.J. Gordon
even if they have no

function.
Domestic herbivore
species richness
may increase if
there is a market for
novelty products
No land aban- Market Increased land hold- No place for wild Intermediate use of
donment distortions ings, leading to herbivores. fertilisers, irrigation,
and farm bigger farms; Domestic herbivore grass cutting; woody
subsidies further intensifi- species richness species maintained on
continue cation decreases farm only if necessary
(e.g., fodder).
Increased CO2- Depopulation of Land abandon- Market Increased and inten- No place for wild her- More interregional varia-
levels lead to rural areas ment distortions sified holdings bivores on farms. tion; some areas inten-
shift in balance and farm only where it Domestic herbivore sively used, others
between woody subsidies pays (increased species richness much less so. Woody
species and her- stop abandonment decreases encroachment in less
baceous layer elsewhere) intensively used areas
Market Much bigger farms Little place for Importance of fertilisers
1 Introduction: Grazers and Browsers in a Changing World
distortions are possible wild herbivores. decreases, but local

and farm but hardly any Domestic herbivore irrigation, grass cut-
subsidies intensification species richness ting; woody species
continue taking place; decreases, ample maintained on farm
land holding size scope for mega only if necessary (e.g.,
will also increase dairy projects and as windbreak)
far from markets feedlots
and hence land
abandonment
will be countered
(continued)
11
12
Table 1.1 (continued)
Market distor- Much bigger farms Increasing place for Interregional variation
tions and are possible but wild herbivores. in farm profitability
farm subsi- hardly any inten- ‘Wild game’ will increase, and so
dies stop sification will tolerated if cost- will on-farm variation.
take place, and effective. Place for Woody encroachment
only then where wild herbivores will be tolerated in
it pays (increased in game parks. many places within
abandonment Domestic herbivore the (farm) landscape
elsewhere) species richness even if they have no
may increase if function
there is a market for
novelty products
H.H.T. Prins and I.J. Gordon
Table 1.2 The effect of the scenarios from Table 1.1 on domestic herbivores and on wild
herbivores
Scenario elements from Effect on domestic Effect on wild herbivores in the
Table 1.1 (CO2-levels are herbivores in the context of the ecology of
assumed to increase in all context of the ecology grazers and browsers
possible scenarios) of grazers and browsers
Rural areas without land aban- Very few browsers, if any, Small wild grazers, like migra-
donment or depopulation, will find a place in this tory geese, can benefit but
while market distortions system. Animals will otherwise the landscape will
and farm subsidies do be kept indoors, and become devoid of wild graz-
not allow much scope for be fed with imported ers and browsers.
increased land holding size food and silage from
due to mergers, and farm- intensively used (parts
ers rely heavily on further of) farms.
farm intensification
Rural areas without land aban- Market for novelty There will be an effect outside
donment or depopulation, products can become the region towards other
but cessation of market important, such as regions: there abandonment
distortions and farm sub- lamas in Europe, and will be promoted leading
sidies lead to a moderate farmers may switch to to woody encroachment
increase in size of land game farming as and build-up of browser
holdings in places where in South Africa or populations.
it is economically profit- New Zealand.
able to merge. Cessation
of agriculture in regions
where it is no longer via-
ble, but in regions where
it is profitable most farms
will further intensify
Land abandonment but no Very few browsers, if any, Little opportunity for wild
depopulation of rural will find a place in this species in the landscape
areas, while market distor- system. The industry because abandonment is
tions and farm subsidies would like to keep counteracted upon by gov-
continue leading to much animals indoors, and ernment and farmers. Nature
increased land holding fed on imported food management will ask for
size due to mergers even and silage from inten- woody species suppression
in areas where it does not sively used farms, but by browsing, mowing, and
make sense economically. the public will demand cutting. Increased options
Intensification is driven by ‘romantic’ landscapes for small wild browsers; if
subsidies even in places with domestic stock wild grazers are available,
far removed from markets. kept outdoors they will be kept fenced
There will be little land
abandonment because this
trend will be counteracted
by subsidies
(continued)

Land abandonment but no ‘Traditional’ grazers (cat- Reasonable opportunities for
depopulation of rural areas, tle, sheep) will be wild species in the landscape
but cessation of market used on big farms, because pockets of land will
distortions and farm sub- but market for novelty be abandoned by agriculture.
sidies lead to an increase products could become The opportunities are there
in size of land holdings in important, and farmers for small and not so small
places where it is economi- may switch to game browsers. Little scope
cally profitable but cessa- farming. for 20th century nature
tion of agriculture where management. Increased
it is not. Intensification is options for large herds of
not necessary on farms that wild grazers if they can be
stay in operation made economically; they
will be kept fenced.
Depopulation of rural Very few browsers, if any, Small wild grazers, like
areas but without land will find a place in this migratory geese, can benefit
abandonment, while system. Animals will but otherwise the landscape
market distortions and be kept both indoors will become devoid of wild
farm subsidies continue and outdoors, and be grazers and browsers
leading to increased land fed on imported food
holding size due to merg- and silage from inten-
ers and further farm intensively used (parts of)
sification farms
Depopulation of rural areas Very few browsers, if any, There will be a strong effect
but without land abandon- will find a place in this outside the region towards
ment, but cessation of mar- system. The industry other regions: there
ket distortions and farm would like to see ani- abandonment will be
subsidies lead to a moder- mals kept indoors but promoted leading to woody
ate increase in size of land to be outdoors as much encroachment and build-up
holdings in places where as feasible, and partly of browser populations
it is economically profit- fed on silage from
able to merge. Cessation intensively used parts
of agriculture where it is of the farms. Feedlot
no longer viable, but on farming
most farms intensification
necessary
Land abandonment and Very few browsers, if any, Small wild grazers, such as
depopulation of rural areas, will find a place in this migratory geese, can benefit
while market distortions system. The industry to a limited extent because
and farm subsidies con- would like to see ani- of decreased fertiliser
tinue to lead to increased mals kept indoors but to input. The landscape will
land holding size due to be outdoors as much as become devoid of graz-
mergers. Intensification is feasible, and partly fed ers and browsers, because
not necessary on farms that on silage from inten- farmers will suppress woody
stay in operation. There will sively used parts of the encroachment and will see
be little land abandonment farms. Good options game as a competitor for
because this trend will be for mega-dairy projects, forage
counteracted by subsidies feedlot farming and
extensive ranching
(continued)

Land abandonment and Domestic herbivore species Increasing place for wild her-
depopulation of rural richness may increase bivores. ‘Wild game’ toler-
areas, but with cessation if there is a market ated if cost-effective. Place
of market distortions and for novelty products. for wild herbivores on game
farm subsidies will lead Low fertilizer input, ranches and private game
to an increase in size of coupled with large farm parks. Burning necessary
land holdings in places sizes will require the to encourage grazers. There
where it is economically suppression of woody will be a strong effect out-
profitable but cessation of encroachment. Burning side the region towards other
agriculture where it is not. and stumping necessary regions: there abandonment
Intensification is not nec- to maintain grazing will be promoted leading
essary on farms that stay in lands to woody encroachment
operation and build-up of browser
populations
Fire. More trees and less herbaceous vegetation can result in large build ups of
combustible material, changing landscapes that were fire resistant in their agricul-
tural state into fire-prone landscapes (cf. Bonazountas et al. 2005; Finney 2005;
Haight et al. 2004; O’Laughlin 2005; Sturtevant et al. 2004). This leads to loss of
life and property (e.g., Chen and McAneney 2004), and increased transaction costs
for society because of increasing insurance premiums. Uncontrolled forest succes-
sion as consequence of the accelerated socioenvironmental change in the
Mediterranean forested landscapes has shown to lead to critically enhanced risk for
fires (Tabara et al. 2004).
Accidents. Land abandonment together with higher wildlife densities increase as
chances for traffic accidents increase (e.g., Doerr et al. 2001). A person driving an ordi-
nary car at 80 km/hr who collides with a wild boar has a very high chance of being
killed; tall-legged moose smash through wind screens in collisions. Apart from sorrow
and suffering, increased insurance premiums lead to increased costs of transport.
Diseases. Increased wildlife densities, especially of deer, can lead to the closure
of the life cycle of tick-borne diseases. For example, Lyme’s disease needs small
intermediary hosts such as mice or blackbirds but also large ones, such as roe
deer or red deer. An increased number of deer leads to an increased incidence of
life-endangering parasitic diseases (Jensen and Jespersen 2005; Randolph 2004;
Zavaleta and Rossignol 2004).
Cultural heritage. Particularly in Japan and Europe, societies and governments have
gone to great effort to stop the countryside from getting clogged-up with ungainly
industry by applying zoning laws and heritage protection. In these regions, national
parks have even been equated with mediaeval or pre-industrial landscapes, and not, like
in Africa or North America, with pristine nature (where, admittedly, ‘native peoples’
had had an often unrecognised impact). Land abandonment and depopulation threaten
this medium-intensity agricultural countryside that is still highly valued by the public.
By asking ourselves how grazers and browsers function in the landscape, we

believe that we are able to help developing awareness of these often unwanted
changes, but also that we are finding tools for the management of these landscapes.
The negative aspects of landscape change that we spelled out are, however, offset
by a very positive trend:
Nature. Changes in the balance between grasses and trees can have profound
impact on the many other wild species that make use of the landscape. That local
plant communities change is self-evident, but linked to that are the changes at lower
and higher trophic levels (e.g., Litvaitis 1993; Preiss et al. 1997). Depopulation and
land abandonment is leading to enormous opportunities for wildlife to re-establish
itself in the countryside, and for a new natural system to develop. In many of the
areas that have been intensively used for decades or even centuries, wildlife was on
the brink of extinction, but now it is bouncing back. The number of roe deer in
Europe or white-tailed deer in North America have not been so high for centuries,
and that is true for many others species as well.
Many of the indigenous wild species have been wiped out, especially wild
grazing species. Now browsers and intermediate feeders are doing well in the
increasingly forested landscapes of Europe and North America. Also feral
browsers do exceedingly well in some areas, for example goats and dromedaries
in Australia. In the absence of wild grazers, horses and donkeys have the ability
to thrive, but not sheep. Perhaps in coming years in Europe there will as much
opportunity for wisent (European bison) as there is for bison in America.
Landscapes all over the world have been changing, and herbivores whether they
are wild, feral or domestic, live and die at the hand of man. The different chapters
in this book give clues about why some species live and others die, why some
species or classes of species have selective advantages over others, and how an
understanding of their ecology in the past or under ‘pristine conditions’ may give
educated guesses about how they will fare in future. In the Third World or in the
rapidly developing new economies the coming decades will be bleak, but in coun-
tries in the temperate zone we forecast a ‘Golden Age’ for wild browsers and to a
lesser extent for wild grazers.
References
Anonymous (2003) Europe’s Environment: The Third Assessment. Environmental assessment

report 10. European Environment Agency, Copenhagen
Anonymous (2005a) Ears of plenty: the story of man’s staple food. Economist 24 December 2005,
p 26–30
Anonymous (2005b) Economist 24 December 2005, p 75
Anonymous (2005c) The poorest part of America: not here surely? Economist 10 December 2005,
p 37–38
Anonymous (2005d) Europe’s farm follies: why the European Union remains its strange fondness
for farm subsidies. Economist 10 December 2005, p 25–27
Anonymous (2006a) http://earthtrends.wri.org/pdf_library/country_profiles/pop_cou_036.pdf
Anonymous (2006b) http://www.cdi.org/russia/johnson/7246-17.cfm
Asner GP, Elmore AJ, Olander LP, Martin RE, Harris AT (2004) Grazing systems, ecosystem
responses, and global change. Annu Rev Env Resour 29:261–299
Batzing W, Perlik M, Dekleva M (1997) Urbanization and depopulation in the Alps (with 3 col-
oured maps). Mt Res Dev 16:335–350
Bennett J, van Bueren M Whitten S (2004) Estimating society’s willingness to pay to maintain
viable rural communities. Aust J Agr Resour Ec 48:487–512
Bonazountas M, Kallidromitou D, Kassomenos PA, Passas (2005) Forest fire risk analysis. Hum
Ecol Risk Assess 11:617–626
Bond W J, Midgley GF, Woodward WI (2003) The importance of low atmospheric CO2 and fire
in promoting the spread of grasslands and savannas. Global Change Biol 9:973–982
Borrelli P, Cibils A (2005) Rural depopulation and grassland management in Patagonia. In:
Reynolds SG, Frame J (eds) Grasslands, developments, opportunities, perspectives Science
Publishers, London, pp 461–487
Breebaart L, Brikraj R, O’Connor TG (2002) Dietary overlap between Boer goats and indigenous
browsers in a South African savanna. Afr J Range Forage Sci 19:13–20
Cavalli-Sforza LL (1996) The spread of agriculture and nomadic pastoralism: insights from
genetics, linguistics and archaeology. In: Harris DR (ed) The origins and spread of agriculture
and pastoralism in Eurasia, UCL Press, London, p 51–69
Chen KP McAneney J (2004) Quantifying bushfire penetration into urban areas in Australia.
Geophys Res Lett 31:L2212, 1–4
Cronon W (1983) Changes in the land: Indians, colonists, and the ecology of New England. Hill
and Wang, New York
Davies R (2000) Madikwe Game Reserve: a partnership for conservation. In: Prins HHT,
Grootenhuis JG, Dolan TT (eds) Conservation of wildlife by sustainable use. Kluwer
Academic, Boston, pp 439–458
Dawson TJ, Tierney PJ, Ellis BA (1992) The diet of the bridled nailtail wallaby (Onychogalea
fraenata): II. Overlap in dietary niche breadth and plant preferences with the black-striped
wallaby (Macropus dorsalis) and domestic cattle. Wildl Res 19:79–87
Debussche M, Lepart J, Devieux A (1999) Mediterranean landscape changes: evidence from old
post cards. Global Ecol Biogeogr 8:3–15
Doerr ML, McAninch JB Wiggers EP (2001) Comparison of four methods to reduce white-tailed
deer abundance in an urban community. Wildlife Soc B 29:1105–1113
Douglas JE, Hoover MD (1988) History of Coweeta. In: Swank WT, Crossley DA (eds) Forest
hydrology and ecology of Coweeta, Springer, Berlin Heidelberg New York, pp 17–34
Edwards GP, Croft DB, Dawson TJ (1996) Competition between red kangaroo (Macropus
rufus) and sheep (Ovis aries) in the arid rangelands of Australia. Aust J Ecol
21:165–172
Escobar A, Gonzalez JE (1976) Study on the competitive consumption of large herbivores of the
flooded area of the Llanos with special reference to the capybara (Hydrochoerus hydro-
chaeris). Agron Trop 26:215–277
FAO (2006): http://www.fao.org/es/ess/census/default.asp/ [accessed 30/1/06]
Finney MA (2005) The challenge of quantitative risk analysis for wildland fire. Forest Ecol
Manag 211:97–108
Flinn KM, Vellend M, Marks PL (2005) Environmental causes and consequences of forest
clearance and agricultural abandonment in central New York, USA. J Biogeogr 32, 439–452
Fritz H, Degarinewichatitsky M Letessier G (1996) Habitat use by sympatric wild and domestic
herbivores in an African savanna woodland: the influence of cattle spatial behaviour. J App
Ecol 33:589–598
Genin D, Villca Z, Abasto P (1994) Diet selection and utilization by llama and sheep in high-
altitude arid rangeland of Bolivia. J Range Manage 47, 245–248
Gil Montero R, Villalba R (2005) Tree rings as a surrogate for economic stress: an example from
the Puna of Jujuy, Argentina in the 19th century. Dendrochronologia 22:141–147
Godwin, H (1975) History of the British flora: a factual basis for phytogeography, 2nd edn.
Cambridge Univ Press, Cambridge
Goriup, P (1998) The pan-European biological and landscape diversity strategy: integration of
ecological agriculture and grassland conservation. Parks 8:37–46
Haight RG, Cleland DT, Hammer RB, Radeloff VC, Rupp TS (2004) Assessing fire risk in the
wildland-urban interface. J Forestry 102:41–48
Hole, F (1996) The context of caprine domestication in the Zagros region. In: DR Harris (ed)
The origins and spread of agriculture and pastoralism in Eurasia. UCL Press, London,
pp 263–281
Hubbard RE, Hansen RM (1976) Diets of wild horses, cattle, and mule deer in the Piceance basin,
Colorado. J Range Manage 29:389–392
Hughes R (1987) The fatal shore: a history of the transportation of convicts to Australia
1787–1868. Pan Books, London
Jensen PM, Jespersen JB (2005) Five decades of tick-man interaction in Denmark: an analysis.
Exp Appl Acarol 35:131–146
Knörzer KH (1971) Urgeschichtler Unkräuter im Rheinland: Ein Beitrage zur Entstehungsgeschichte
der Segetalgesellschaften. Vegetatio 23:89–11
Knörzer KH (1975) Entstehung und Entwicklung der Grünlandvegetation im Rheinland.
Decheniana 127:195–214
Köhler-Rollefson I (1996) The one-humped camel in Asia: origin, utilization and mechanisms of
dispersal. In: Harris DR (ed) The origins and spread of agriculture and pastoralism in Eurasia.
UCL Press, London, pp 282–294
Kontorovich V (2000) Can Russia resettle the Far East? Post-Communist Econ 12:365–384
Legge T (1996) The beginning of caprine domestication in Southwest Asia. In: Harris DR (ed)
The origins and spread of agriculture and pastoralism in Eurasia, UCL Press, London,
pp 238–262
Lewis D (2002) Slower than the eye can see: environmental change in Northern Australia’s cattle
lands; a case study from the Victoria River district, Northern Territory. Tropical Savannas
CRC, Darwin
Lightfoot CJ, Posselt J (1977) Eland (Taurotragus oryx) as a ranching animal complementary to
cattle in Rhodesia. 2. Habitat and diet selection. Rhod Agr J 74:53–61
Litvaitis JA (1993) Response of early successional vertebrates to historic changes in land-use.
Conserv Biol 7:866–873
Liu YB, Nishiyama S, Kusaka T (2003) Examining landscape dynamics at a watershed scale using
Landsat TM imagery for detection of wintering hooded crane decline in Yashiro, Japan.
Environ Manage 31:365–376
Low B. Rebello AG (1996) Vegetation of South Africa, Lesotho and Swaziland. Department of
Environmental Affairs and Tourism, Pretoria
MacDonald D, Crabtree JR, Wiesinger G, Dax T, Stamou N, Fleury P, Lazpita JG, Gibon A
(2000) Agricultural abandonment in mountain areas of Europe: environmental consequences
and policy response. J Environ Manage 59:47–69
McFadyen RC, Skarratt B (1996) Potential distribution of Chromolaena odorata (Siam weed) in
Australia, Africa and Oceania. Agr Ecosyst Environ 59:89–96
Millward H (2005) Rural population change in Nova Scotia, 1991–2001: bivariate and multivari-
ate analysis of key drivers. Can Geogr–Geogr Can 49:180–197
Motzkin G, Foster DR (2002) Grasslands, heathlands and shrublands in coastal New England:
historical interpretations and approaches to conservation. J Biogeogr 29:1569–1590
Moreira F, Rego FC, Ferreira PG (2001) Temporal (1958–1995) pattern of change in a cultural land-
scape of northwestern Portugal: implications for fire occurrence. Landscape Ecol 16:557–567
Newman P (2005) The city and the bush: partnerships to reverse the population decline Australia’s
wheatbelt. Aust J Agr Res 56:527–535
Noble IR, R (1997) Forests as human-dominated ecosystems. Science 277:522–525
O’Laughlin J (2005) Policies for risk assessment in federal land and resource management deci-
sions. Forest Ecol Manag 211:15–27
Opravil E (1978) Synanthrope Pflanzengesellschaften aus der Burgwallzeit (8 - 10 Jh) in der
Tschechoslowakei. Berichte der deutsche Botanische Gesellschaft 91:97–106
Packard S Mutel CF (1997) The tallgrass restoration handbook for prairies, savannas and
woodlands. Island Press, Washington, DC
Pennington W (1969) The history of the British vegetation. English Univ Press, London
Persson G (1997) Comparison of simulated water balance for willow, spruce, grass ley and barley.
Nord Hydrol 28:85–98
Popay AI, Rahman A, James TK (2002) Future changes in New Zealand’s hill country pasture
weeds. New Zeal Plant Prot (Zydenbos SM ed) 55:99–105
Preiss E, Martin JL, Debussche M (1997) Rural depopulation and recent landscape changes in a
Mediterranean region: consequences for the breeding avifauna. Landscape Ecol 12:51–61
Prins HEL (1994) Children of Gluskap: Wabanaki Indians on the eve of the European invasion.
In: Baker EW, Churchill EA, D’Abate RS, Jones KL, Konrad VA, Prins HEL (eds) American
beginnings: exploration, culture and cartography in the land of Norumbega. U Nebraska Press,
Lincoln, pp 165–211
Prins HHT (1998) The origins of grassland communities in northwestern Europe. In: Wallis de
Vries MF, Bakker JP, van Wieren SE (eds) Grazing and conservation management. Kluwer,
Boston pp 55–105
Prins HHT (2000) Competition between wildlife and livestock. In: Prins HHT, Grootenhuis
JG Dolan TT (eds) Conservation of wildlife by sustainable use. Kluwer, Boston,
pp 51–80
Ramankutty N, Foley JA (1999) Estimating historical changes in land cover: North American
croplands from 1850 to 1992. Global Ecol Biogeogr 8:381–396
Randolph SE (2004) Evidence that climate change has caused ‘emergence’ of tick-borne diseases
in Europe? Int J Med Microbiol 293(Suppl):5–15
Riley SJ, Decker DJ, Enck JW, Curtis PD, Lauber TB, Brown TL (2002) Deer populations up,
hunter populations down: implications for interdependence of deer and hunter population
dynamics on management. Ecoscience 10:455–461
Rusell EWB, Davis RB (2001) Five centuries of changing forest vegetation in the northeastern
United States. Plant Ecol 155:1–13
Schwartz CC, Ellis JE (1981) Feeding ecology and niche separation in some native and domestic
ungulates on the shortgrass prairie. J Appl Ecol 18:343–353
Shaphiro S Stankevich B (2000) Structural changes in agriculture in northeastern Germany
(in Russian). Vestsi Akademii Agrarnykh Navuk Respubliki Belarus 2:40–43
Shimoda M (2005) Emerged shore vegetation of irrigation ponds in western Japan. Phytocoenologia
35:305–325
Simmons IG, Dimbley GW, Grigson C (1981) The Mesolithic. In: Simmons IG, Tooley MJ (eds),
The environment in British prehistory. Duckworth, London, pp 82–124
Sturtevant BR, Zollner PA, Gustafson EJ, Cleland DT (2004) Human influence on the abundance
and connectivity of high-risk fuels in mixed forests of northern Wisconsin, USA. Landscape
Ecol 19:235–253
Tabara D, Sauri D, Cerdan R (2004) Forest fire risk management and public participation in
changing socioenvironmental conditions: a case study in a Mediterranean region. Risk Anal
23:249–260
Taylor SC (1998) South-eastern Australian temperate lowland native grasslands: protection levels
and conservation. Parks 8: 21–26
Thill RE, Martin A (1986) Deer and cattle diet overlap on Louisiana pine-bluestem range.
J Wildlife Manage 50:707–713
Tinner W, Conedura M, Amman B, Gaggeler HW, Gedye S, Jones R, Sagesser B (1998) Pollen
and charcoal in lake sediments compared with historically documented forest fires in southern
Switzerland since AD 1920. Holocene 8:31–42
Torrano L, Valderrabano J (2004) Impact of grazing on plant communities. Span J Agric Res
2:93–105
United Nations (2006) World urbanization prospects, 1999 revision. http://www.prb.org/
Valderrabano J, Torrano L (2000) The potential for using goats to control Genista scorpis shrubs
in European black pine stands. Forest Ecol Manag 126:377–387
Van Zeist W (1980) Prehistorische cultuurplanten, ontstaan, verspreiding, verbouw. In: Chamaulan
M, Waterbolk HT (eds), Voltooid Verleden Tijt: Een hedendaagse kijk op de prehistorie.
Intermediair, Amsterdam, pp147–165
Vitousek PM, Mooney HA, Lubchenco J, Melillo JM (1997) Human domination of Earth’s
ecosystems. Science 277:494–499
Zavaleta JO, Rossignol PA (2004) Community-level analysis of risk of vector-borne disease.
T Roy Soc Trop Med H 98:610–618
Zeuner FE (1967) Geschichte der Haustiere. Bayerische Landwirtschaftsverlag, Muenchen
Chapter 2
An Evolutionary History of Browsing
and Grazing Ungulates
Christine Janis
2.1 Introduction
Browsing (i.e., eating woody and non-woody dicotyledonous plants) and grazing
(i.e., eating grass) are distinctively different types of feeding behaviour among
ungulates today. Ungulates with different diets have different morphologies (both
craniodental ones and in aspects of the digestive system) and physiologies, although
some of these differences are merely related to body size, as grazers are usually larger
than browsers. There is also a difference in the foraging behaviour in terms of the
relationship between resource abundance and intake rate, which is linear in browsers
but asymptotic in grazers. The spatial distribution of the food resource is also different
for the different types of herbage, browse being more patchily distributed than grass,
and thus browsers and grazers are likely to have a very different perception of food
resources in any given ecosystem (see Gordon 2003, for review).
Grass is a relatively recent type of food resource: extensive grasslands only
emerged during the later Cenozoic, within the past 25 million years (Ma), while the
first ungulates date back to the early Cenozoic, around 55 Ma. Browsing, probably
with the incorporation of a fair amount of fruit, is thus the primitive diet of
ungulates (Bodmer and Ward 2006). The general evolutionary perspective is that
when the grazing species evolved later they eclipsed the browsers (Kowalevsky
1873; Matthew 1926; Pérez-Barbería et al. 2001). While this view of the later
Cenozoic rise of grazing ungulates, and corresponding demise of browsing ones, is
broadly correct, the actual evolutionary picture is of course much more complex.
While the most familiar ungulates today (horses, cows, elephants, etc.) are mainly
grazers, in fact specialised grazing is a fairly recent evolutionary adaptation: the
first true grazers (i.e., animals subsisting primarily on grass year round) are no more
than 10 million years old, and a predominance of tropical grazers as is familiar to
us in Africa today has a history of only a couple of million years.
In order to understand the evolutionary history of diets and feeding adaptations
in ungulates we need to first pose several questions. Firstly, what exactly are
ungulates, and how are different ungulate groups related to each other? Secondly,
how can we deduce the feeding behaviour of extinct species from their fossilised
remains? And thirdly, how have differences in the earth’s climate and environment
© Springer 2008
22 C. Janis
in past times influenced the distribution and abundance of ungulates of different

feeding behaviours?
2.1.1 Ungulate Phylogeny and Evolutionary History
While ungulates have long been considered as a clade, united by the possession of
hooves (which is in fact not strictly true in any case), recent molecular phylogenies have
refuted the monophyly of the ‘Ungulata’ (see Springer et al. 2005 for review). The
paenungulates (elephants, hyraxes, and sea cows) are now united with other African
endemic mammals in the basal placental clade Afrotheria. Although artiodactyls and
perissodactyls are fairly closely related, they are not sister taxa: the perissodactyls are
now considered to be more closely related to a clade consisting of carnivorans plus
pangolins, and artiodactyls are now included with whales in the Cetartiodactyla. The
name of the clade uniting this diverse assemblage is the Ferungulata.
It is difficult to state how extinct ‘ungulate’ groups, such as the condylarths and
the endemic South American ungulates, fit within this new phylogenetic scheme.
The South American orders Notoungulata and Litopterna paralleled northern
ungulates in the later Cenozoic in their evolution of body morphologies (with types
resembling rhinos, rodents, hyraxes, horses, and camels), and in similar evolution-
ary transitions to animals with skulls and teeth adapted for grazing, and limbs
adapted for cursoriality (see Prothero and Schoch 2002). See the Glossary for terms
of extinct ungulate groups (Box 2.1).
Box 2.1 Glossary Guide to extinct taxa. See chapters in Janis et al. (1998) and
Rose (2006) for more details
Anchitheriinae. A subfamily of horses (family Equidae), known from
the late Eocene to late Miocene of North America and the Miocene of
Eurasia, ranging from sheep-sized (earlier forms) to the size of a modern
horse (later forms). More derived than the Hyracotheriinae in having more
lophed teeth, indicative of a more folivorous diet (but all were brachydont,
presumed browsers), and somewhat longer legs, with the loss of the fourth
toe in the front foot, and no evidence of a foot pad.
Brontotheriidae. A family of perissodactyls (also known as titanotheres),
all members brachydont (indicative of browsing), known from the Eocene
of North America and Eurasia. Early members were pig-sized and hornless,
later forms became larger and grew forked bony nasal horns, and some of
the latest forms were of a comparable size to a white rhino.
Chalicotheroidea. A superfamily of perissodactyls, all members brachydont
(indicative of browsing). The earlier Eomoropidae, known from the Eocene
of Asia and North America, were small (dog-sized), unspecialised forms.
The later Chalicotheriidae, known from the Oligocene to Pleistocene of
2 An Evolutionary History of Browsing and Grazing Ungulates 23
Eurasia, the Miocene of North America, and the Miocene to Pleistocene

of Africa, were large (camel-sized) forms that had substituted claws for
hooves, and likely fed with a bipedal stance clawing down vegetation.
Condylarthra. A paraphyletic assemblage (order) of basal archaic ungulates,
comprising at least eight families, known from the Palaeocene and Eocene
of North America, South America and Eurasia, mostly ranging in size from
the size of a squirrel to the size of a pig (some carnivorous forms reached
bear size). Most forms had bunodont cheek teeth indicative of omnivory;
some had more lophed teeth indicative of herbivory, while some were even
carnivorous. None had cursorial (running-adapted) postcranial morphologies
nor any degree of hypsodonty. The carnivorous mesonychid ‘condylarths’ are
considered to be the sister taxon to the Cetartiodactyla, but the precise rela-
tionship of the other families to extant ungulates is not resolved.
Deinotheriidae. A family of proboscideans, known from the Miocene to
Pleistocene of Africa and Eurasia. Characterised by very large size and the pos-
session of recurved lower tusks only (no upper tusks). Cheek teeth were tapir-
like, bilophodont and brachydont, indicative of a specialised browsing diet.
Dinocerata. An order of ungulate-like (but probably not ungulate-related) mam-
mals known from the Palaeocene and Eocene of North America and Asia,
ranging from sheep-size to rhino-size, with later forms having a bizarre series
of horns on their skulls and sabre-like upper canines. They had brachydont,
bilophodont teeth indicative of a diet of soft browse, and later forms had post-
cranial adaptations indicative of a semi-aquatic mode of life.
Gomphotheriidae. A family of Proboscidea, known from the Miocene to
Pleistocene of North America, Eurasia, and Africa. Ranged from small
rhino-sized to modern elephant-sized. Earlier forms had cheek teeth that
were brachydont and bunodont to bunolophodont, indicative of omnivory
and/or browsing. Included bizarre ‘shovel-tusked’ forms. Some later
(Pliocene) forms developed highly lophed, hypsodont cheek teeth, indica-
tive of mixed feeding and/or grazing.
Hipparionini. Horses belonging to an extinct tribe of the subfamily Equinae
(modern horses and their direct ancestors belong to the tribe Equini), known
from the middle Miocene to Pliocene of North America, and late Miocene
to Pliocene of Eurasia and Africa (survived into mid Pleistocene of Africa).
Earlier hipparionines and equines were similar types of pony-sized animals,
but hipparionines in general remained less hypsodont than later equines, and
they retained a tridactyl foot. Dental microwear suggests mainly mixed-feeding
diets rather than grazing (Hayek et al. 1992; Solounias and Sempebron 2002).
Hypertragulidae. A primitive family of ruminants, known from the Eocene
and Oligocene of North America and Asia. Diminutive forms, about the
size of extant Tragulus (Tragulidae). Teeth range from brachydont to
hypsodont (but not likely to have been grazing).
(continued)
24 C. Janis
Box 2.1 (continued)

Hyracotheriinae. A subfamily of horses, known from the Eocene of North
America. Small forms (cat to small dog-sized), with relatively bunodont
cheek teeth indicative of a folivorous/frugivorous diet, and relatively short
legs retaining three toes in the hind foot and four in the fore foot, with the
likely presence of a tapir-like foot pad.
Indricotheriinae. A subfamily of rhinos (of the extinct family Hyracodontidae),
known from the late Oligocene and early Miocene of western Asia, all with
brachydont cheek teeth (indicative of browsing). Known as ‘giraffe rhinos’
because of the evolution of long neck and long legs, including the largest
ever known land mammal, weighing up to 15 tons.
Litopterna. An extinct order of South American ungulates, known from the
Palaeocene to the Pleistocene, comprising forms with brachydont cheek
teeth (indicative of browsing), or somewhat hypsodont teeth (indicative of
mixed feeding), generally resembling horses or camels.
Mammutidae. A family of Proboscidea (mastodons), known from the
Miocene to Pleistocene of North America, Eurasia, and Africa. Ranged
from small rhino-sized to modern elephant-sized, and had cheek teeth
that were brachydont and bunolophodont, indicative of browsing.
Notoungulata. An order of South American ungulates, known from the
Palaeocene to the Pleistocene, comprising forms with high-crowned or
ever-growing cheek teeth (but were not necessarily all grazers), generally
resembling rodents, hyraxes, or rhinos.
Oreodontoidea. Extinct superfamily of (possibly) tylopod artiodactyls, pig-like
or hyrax-like in appearance, known from the late Eocene to late Miocene of
North America. They combined relatively derived selenodont check teeth
(albeit in a snub-nosed skull), earlier forms all brachydont, with a primitive
postcranium that retained a digitigrade stance. The earlier and more primitive
Agriochoeridae (cat-sized to dog-sized) had replaced their hooves with claws
and were likely secondarily arboreal. The later Merycoidodontidae (hyrax-
sized to tapir-sized) were the commonest mammals in the Oligocene and early
Miocene of North America. Somve later Miocene forms showed a tendency
to evolve somewhat longer legs and more hypsodont teeth.
Palaeomerycidae. A family of deer-like (and deer-related) artiodactyls,
known from the Miocene and Pliocene of North America (subfamily
Dromomerycinae) and Eurasia. Mostly brachydont (indicative of
browsing), some forms slightly hypsodont and may have been mixed
feeders; many forms characterised by a third, median occipital horn in
addition to a pair of frontal horns.
Palaeotheriidae. A family of horse-like, and horse-related, perissodactyls,
known from the Eocene of Eurasia, ranging from dog-size to modern
horse-size. All brachydont (indicative of browsing).
Pantodonta. An order of ungulate-like (but probably not ungulate-related)
mammals, known from the Palaeocene and Eocene of North America,
Eurasia, and South America, ranging from goat-sized to bison-sized, and

later forms resembling ground sloths or hippos. All brachydont (indicative
of browsing), and some forms were likely semi-aquatic.
Sivatheriinae. A subfamily of Giraffidae, known from the Miocene to
Pleistocene of Africa and Eurasia. Differed from extant Giraffinae in being
relatively short legged and heavily-built (often termed ‘moose-like’, but
with shorter legs than a moose), with horns that were more lobate in form
than giraffines. Cheek teeth were generally more hypsodont than giraff-
ines, indicative of a more mixed-feeding type of diet.
Taeniodonta. An order of ungulate-like (but probably not ungulate-related)
mammals, known from the Palaeocene and Eocene of primarily North
America, ranging from cat-sized to pig-sized. Later forms had hypselo-
dont (ever-growing) cheek teeth and showed adaptations for digging,
rather resembling large pigs. Their hypselodont cheek teeth were probably
related to grit found on excavated roots and tubers, as their diet otherwise
appears to have been omnivorous.
Fig. 2.1 Temporal ranges of ungulate families (adapted from range charts in Benton 1993 and
Janis et al. 1998). Not all extinct families of artiodactyls shown; extinct subfamilies of Equidae
included. Dotted lines: omnivorous groups. Solid lines: folivorous groups. Thick solid lines indi-
cate time when there were at least some taxa in this lineage that likely took a fair amount of grass
(>50%) in their diet
This chapter will mainly consider the evolution of feeding adaptations in the two
major orders of extant ungulates: the Perissodactyla, or odd-toed ungulates, and the
Artiodactyla, or even-toed ungulates (see Fig. 2.1). Both orders are of Northern
Hemisphere origin, first appearing at the start of the Eocene (55 Ma) in North
America and Eurasia. Both orders show considerable evolutionary parallelisms,
including the evolution of hypsodont (high-crowned) cheek teeth, and elongated
limbs with an unguligrade foot posture (i.e., standing on the tip of the phalanges,
26 C. Janis
like a ballerina ‘en pointe’), but they differ profoundly in their morphophysiological
adaptations to folivory (the eating of leafy material, whether browse or grass).
Perissodactyls are specialised hindgut fermenters while camelid and ruminant
artiodactyls (in the taxonomic sense) are foregut fermenting ruminants (in the
physiological sense). Such differences in digestive physiology are reflected in
differences in both craniodental morphology and feeding behaviour. For example:
in terms of craniodental morphology, hindgut fermenters process more food per day
than ruminants (all other things such as body size, diet, etc., being equal) so they
have a greater volume of masticatory musculature, which is reflected in a deeper
angle of the mandible. In terms of feeding behaviour, ruminants are more limited
in their daily intake, due to their longer digestive passage time, and so they have to
feed in a more selective fashion than a hindgut fermenter, and use their tongue for
food selection to a much greater extent than other ungulates. Figure 2.1 shows the
distribution of the major families of perissodactyls and artiodactyls through time.
Note that forms that have craniodental adaptations for grazing are relatively few
and appear relatively late in ungulate evolutionary history.
It is worth making a brief mention of the evolutionary history of the proboscideans
and hyracoids (both hindgut fermenters), even though these orders are not closely
related to other ungulates (see above). Present-day hyraxes are a relict specialised clade
of small-bodied forms: the extinct family Pliohyracidae comprised the dominant
small-to-medium sized herbivores in the Eocene and Oligocene of Africa, with a diver-
sity of morphologies, including a long-legged antelope-like form, but none showing any
craniodental adaptations for grazing (see Prothero and Schoch 2002; Turner and Antón
2004). This radiation was largely gone by the Miocene, perhaps related to the influx
of modern ungulates into Africa, but some pliohyracids persisted into the
Pleistocene (including a large, hippo-like form), and were also found across southern
Eurasia in the later Cenozoic. The extant family Procaviidae first appeared in the late
Miocene, and the modern genus Procavia (the rock hyrax) is the only one that takes a
large amount of grass in its diet. Proboscideans also originated in Africa, and comprised
around ten separate families. Only three families (including the Miocene to Recent
Elephantidae) show craniodental morphologies indicative of grazing, such as
hypsodont, complexly-lophed cheek teeth. Other proboscideans have bunodont cheek
teeth indicative of omnivory, or bilophodont cheek teeth indicative of browsing
(see Fig. 2.2 for tooth types; hyracoids and proboscideans are not shown in Fig. 2.1, for
reasons of economy of space). However, as with the other ungulates, there is no evi-
dence of any craniodental morphology indicative of grazing until the later Miocene.
2.1.2 Determination of Feeding Adaptations
The large diversity of extant ungulates of known diet has made it possible to
quantitatively determine features of cranial and dental morphology that correlate
with feeding behaviour (Janis 1995; Mendoza et al. 2002). Ungulate cheek teeth
(molars and premolars) are the most obvious dietary indicators. Omnivores have
low-cusped molars with rounded, bumpy cusps (‘bunodont’), designed to process
non-brittle food such as fruit and roots (Fig. 2.2A). Many different herbivorous
forms have evolved from this type of dentition a more ‘lophed’ or ridged type of
A B C
D E F
Buccal
G H Ant.
I
Pulp cavity
Bone
Enamel
Dentine
Cementum
Fig. 2.2 Types of molar morphologies in herbivorous mammals. A. Bunodont (peccary).

B. Bilophodont (kangaroo). C. Columnar (warthog). D. Selenodont (deer). E. Lophodont
(rhinoceros). F. Plagiolophodont (horse). G. Bunolophodont (rodent – woodchuck). H. Multilophed
(rodent – capybara). I. Longitudinal sections of molars, showing brachydont (human) form on the
left and hypsodont (horse) form on the right. Modified from Janis and Fortelius 1988, with permis-
sion of Cambridge University Press
28 C. Janis
molar, where the individual cusps are thrown into higher occlusal relief, and are run
together as longitudinal or horizontal ridges or lophs (Jernvall et al. 1996). These
teeth have been evolved to have maximum efficiency after initial wear, so that the
enamel is worn off the top of the lophs exposing a lake of dentine with enamel
edges on either side, and these lophs act to shred more fibrous food, such a leaves.
The different way in which the cusps have been linked into these lophs result in
different patterns of occlusal anatomy: ruminant artiodactyls have ‘selenodont’
teeth (Fig. 2.2D), where the main pattern of the lophs is in an antero-posterior direc-
tion, while perissodactyls have ‘lophodont’ (Fig. 2.2E) or ‘bilophodont’ (Fig. 2.2B)
teeth, where the main pattern of the lophs is in a labio-lingual direction. Other
herbivorous mammals, such as warthogs (Fig. 2.2C) and rodents (Fig. 2.2G, H)
have evolved lophed teeth independently in different fashion.
A more fibrous diet of grass rather than browse is reflected in both the level of
hypsodonty (see below) and in the occlusal pattern. Highly specialised grazing
ruminants and perissodactyls have more complex ‘plagiolophodont’ (Fig. 2.2F)
occlusal enamel patterns, accomplished by cross-linking the enamel ridges. In other
mammals, such as elephants, and also in many rodents and in wombats, the teeth
become ‘multilophed’ with numerous parallel ridges of enamel that can no longer
easily be homologised with the original mammalian tooth cusp pattern (Fig. 2.2H;
Janis and Fortelius 1988). Any tooth in which the full crown is not visible above the
gum line at eruption can be considered as ‘hypsodont’ to some extent, but there are
varying degrees of hypsodonty, and in extreme cases the crown height may be six or
seven times the width of the tooth. Hypsodont teeth usually have a layer of cementum
that coats the tooth and fills in the spaces between the cusps (see Fig. 2.2I).
All dental dimensions scale isometrically: thus a simple ratio, or ‘hypsodonty
index’ of the unworn crown height of the third molar (usually the highest-
crowned tooth) to its width or length can be compared across taxa of different
body sizes (Janis 1988). Fortelius et al. (2002) define a general rule of thumb for
‘hypsodonty classes’, based on the ratio of height to length of the second upper
or lower molar. A brachydont tooth has a ratio of less than 0.8, a mesodont (par-
tially hypsodont) tooth has a ratio of 0.8–1.2, and a hypsodont tooth has a ratio
of greater than 1.2.
The traditional determinant of a browsing versus grazing diet in fossil ungu-
lates has been the level of hypsodonty, or the degree of molar crown height, as
the silica contained in grass tissue results in greater wear on the teeth. A year-
round diet of grass in a (perforce) open habitat necessitates a hypsodont dentition,
but other dietary or environmental factors may lead to hypsodonty. Fortelius
et al. (2002) note that the factors that correlate with hypsodonty are: ‘increased
fibrousness, increased abrasiveness due to intracellular or extraneous dust, and
decreased nutritive value’. Hypsodonty has evolved numerous times within mam-
mals, and probably represents a fairly simple developmental change, involving
delaying the closure of the tooth roots (Janis and Fortelius 1988). For an animal
with an abrasive diet, and hence a high rate of tooth wear, hypsodonty is an
important adaptation as a dentition that is insufficiently durable will result in a
shortened life span, and hence in a reduced reproductive output (Damuth and
Janis 2005). Thus there is a strong evolutionary imperative to make teeth hypsodont
if the diet is abrasive.
While almost all grazers have highly hypsodont cheek teeth, not all hypsod-
ont ungulates are grazers, as the silica contained in grass is far from the sole
abrasive element in a herbivorous diet. The exceptions to a high degree of hyp-
sodonty in grazers include taxa such as the hippo, Hippopotamus amphibius,
and the rock hyrax, Procavia capensis. These animals (both hypsodont, but at
a relatively low level) both have relatively low metabolic rates for their size,
resulting in less food consumed per lifetime and thus overall less dental abra-
sion. Grazing kangaroos are also less hypsodont than grazing ungulates, again
probably due to the relatively lower metabolic rate of marsupials. Fresh grass
grazers, such as the reduncine bovids, are also less hypsodont than grazers sub-
sisting on grass in more dusty habitats. Open-habitat mixed feeders can also be
highly hypsodont, approaching the level of hypsodonty of grazers in the same
habitat, even if including little grass in the diet. Most gazelles fall into this cat-
egory, and the pronghorn, Antilocapra americana, is among the most hypsod-
ont of the ruminants despite having only about 12% of grass in its diet. The
obvious interpretation of this correlation is that grit and dust accumulating on
the food must also contribute to the abrasive nature of the diet (see Janis 1988;
Janis et al. 2002). Hypsodont taxa in the fossil record clearly provide some
form of palaeoecological signal: increasing levels hypsodonty in today’s ungulate
communities show a negative correlation with levels of rainfall (i.e., hypsodont
ungulates are more prevalent in more arid habitats; Damuth et al. 2002;
Fortelius et al. 2002).
A number of aspects of craniodental morphology can be shown to correlate with
dietary behavior (see Solounias and Dawson-Saunders 1988; Janis 1995; Mendoza
et al. 2002). The features distinguishing grazers from browsers relate to the different
physical demands of feeding on grass versus browse. Grass is in general more
fibrous and abrasive than browse, and a more fibrous and abrasive (i.e., lower
quality) diet requires a greater intake and a greater degree of mastication. Grazers
have bigger masseter muscles than browsers, reflected in a larger and deeper angle
of the jaw, and a longer masseteric fossa on the skull. Browsers are selective feeders
(as are most mixed feeders; Gordon and Illius 1988), and both these feeding types
have a narrow muzzle in comparison with grazers, who have a broad muzzle for the
intake of large bites. However, the majority of living ungulates are ruminant
artiodactyls, and quantitative correlations of morphology and behavior derived
from ruminants may not be directly applicable to other types of ungulates, although
general qualitative observations may still hold true.
Dental wear (microwear, mesowear, or macrowear) is another way of deter-
mining past diets. Dental wear records the actual food preparation and masti-
cation events that took place during the life of the animal, and thus holds the
potential for recording the actual ecological history. But dental wear alone
may be an insufficient guide to diet because of the continual abrasion of teeth
during the life of the animal, and especially of the abrasion of the surface
enamel that records microwear patterns. Solounias and Semprebon (2002),
30 C. Janis
and Semprebon et al. (2004) summarise much of the current uses of dental
wear in dietary determination in ungulates.
Another recently developed methodology is the use of carbon isotopes in dental
enamel. Following the shift in photosynthesis in tropical grasses around 7 Ma, from a
C3 carbon cycle to a C4 carbon cycle, the dental enamel of tropical grazers contains
a different composition of enamel isotope from that of browsers, and mixed feeders
have intermediate values between browsers and grazers. While dental isotopes are
only applicable in the rather limited case of fairly recent ungulates (i.e., within the
past 7 Ma or so) in tropical or subtropical settings, they can be extremely useful in
such palaeoecological situations, especially in combination with morphological
features (e.g., Sponheimer et al. 1999).
How does one, then, determine the diet of an extinct ungulate? To a good first
approximation, molar occlusal morphology can distinguish omnivores from
folivores, and hypsodonty (as well as occlusal morphology, to a certain extent) can
distinguish browsers from grazers, but with a degree of caution. A low-crowned
(brachydont) ungulate is extremely unlikely to be anything else but a browser, but
hypsodont taxa may have a variety of diets. Within hypsodont taxa, craniodental
features and/or microwear can be used to distinguish mixed feeders and grazers in
most circumstances, and isotopes are useful for distinguishing diets among later
Cenozoic taxa in dry, lowland tropical habitats.
2.1.3 Cenozoic Changes in Climate
Evolutionary transitions in ungulate diets can only be understood in the context

of climatic changes on planet Earth over the past 60 million years or so. This
environmental history is related to changes in the movement of the continents,
which in turn have resulted in changes in ocean currents, mountain building,
polar glaciation, etc. (see summaries in Janis 1993; Jacobs et al. 1999; Hooker
2000; Zachos et al. 2001). Understanding of the different environmental
patterns on different continents comes from a variety of sources, including pal-
aeotemperatures determined from deep sea foraminifera (Zachos et al. 2001)
and palaeobotany (Jacobs et al. 1999). Figure 2.3 summarises the evolution of
grassland habitats, and the changing global palaeotemperatures. [Note that
grass phytoliths have now been isolated from the latest Cretaceous of India
(Prasad et al. 2005), but there is no evidence for extensive grasslands in Asia
until the later Cenozoic.]
At the start of the Cenozoic, 65 Ma, the entire globe was a fairly equable,
warm place, with evidence of mainly forest cover, and tropical-like forests
extending up into the high latitudes, although herbaceous (non-grassy) open habi-
tats are recorded in places such as Central Asia. Rapid warming at the end of the
Palaeocene was followed by the early Eocene climatic optimum (55–52 Ma; see
Zachos et al. 2001). Following this optimum the climate cooled in the higher
latitudes, with the arrival of winter frosts and more deciduous types of vegetation
AFRICA NORTH SOUTH AUSTRALIA

Ma EURASIA paleotemperature
AMERICA AMERICA
0
Ple. savanna savanna + savanna savanna + prairie
prairie
5 Pli. prairie + prairie savanna
savanna grazers
grazers grassy habitats savanna hypsodont
Miocene
grazers
grazers mammals
15 grassy habitats grassy habitats
hypsodont mammals
25 grassy habitats
Oligo-
grassy habitats
cene
grazers
35
hypsodont hypsodont
mammals mammals
Eocene
grass fossil
45
grass fossil
55
grass fossil
Paleo-
grass fossil
cene
hypsodont
grass fossil mammals
65
colder warmer
Fig. 2.3 Evolution of grasses and grazers on different continents. Bars indicate times of widespread
grasslands: prairie (also equivalent to steppe or pampas) = treeless grassland; savanna = treed
grassland. Closed circles record the first appearance of various events. Grassy habitats = habitats
containing some grasses, probably woodland savanna or brushland. Grazers = mammals with
craniodental adaptations (apart from hypsodonty) indicative of specialist grazing (i.e., > 90% of
grass in the diet on a year-round basis). Palaeobotanical information adapted from data in Jacobs et
al. (1999) and Strömberg (2004). Palaeotemperature curve is from global deep sea oxygen isotopes
(adapted from Zachos et al. 2001); as used here it represents relative temperatures only (for general
comparison, mid latitude mean temperatures during the early Eocene climatic optimum were prob-
ably around 30° C). Modified from Janis et al. 2004
by the late middle Eocene (around 45 Ma). Although grass fossils are known from
the Eocene, and there seem to have been areas of open habitat, such habitats have
been termed ‘woody savannas’, not dominated by grasses, and have no analogues
among modern habitat types (Leopold et al. 1992).
A million years or so after the end of the Eocene there was an episode of extreme
cooling, setting the stage in the higher latitudes for the more temperate world of the
Oligocene, when temperate deciduous woodlands spread in the mid latitudes along
with patches of more arid habitat. Temperatures started to rise again in the late
Oligocene, around 25 Ma, and after a brief fall reached a further peak at around
14 Ma, the ‘mid Miocene optimum’ (Zachos et al. 2001). Palaeobotanical evidence
shows that grasslands spread during the Miocene in the higher latitudes, but tropical
areas such as East Africa had only limited, if any, savanna regions at this time
(Jacobs et al. 1999; see Fig. 2.3). The Antarctic ice cap, which had its inception in
the late Eocene, was firmly in place by the Miocene.
After around 14 Ma temperatures fell in the higher latitudes, with additional
evidence for increasing aridity. By the latest Miocene (around 6 Ma) the short-grass
savannas of North America were replaced by tall-grass prairie (Retallack 2001),
and there is evidence of woodland savannas in East Africa (Jacobs et al. 1999).
Additional higher latitude cooling in the Plio-Pleistocene brought new types of
cold-adapted and/or arid-adapted vegetational habitats: tundra and taiga to the high
32 C. Janis
latitudes, and true deserts to tropical and subtropical regions. An extremely impor-
tant event in establishing the modern grassland habitats was in the middle Pliocene,
at 2.5 Ma. At this time the Isthmus of Panama appeared, linking North and South
America (with subsequent animal migrations) and disrupting circum-equatorial
circulation. This resulted in the aridification of East Africa, and the establishment
of the extensive grasslands that support much of the diversity of grazing ungulates
today. The Arctic ice cap first appeared in the Pliocene, and from the start of the
Pleistocene around 2 Ma the periodic fluctuations in earth’s climate, caused by the
various aspects of the rotation of the earth on its own axis (Milankovitch cycles),
were sufficient to regularly plunge the higher latitudes into periods of extended
glaciation, which we term ‘Ice Ages’.
2.2 Fossil Record Evidence of Dietary Evolution in Ungulates
2.2.1 Early Archaic Ungulates and Ungulate-like

Mammals (65 to 40 Ma)
Palaeocene ‘ungulates’ were various taxa ascribed to the order ‘Condylartha’, that
probably contains the ancestry of both artiodactyls and perissodactyls (see Glossary,
Box 2.1). Other larger ungulate-like taxa (but probably not related to true ungu-
lates) such as taeniodonts, pantodonts, and dinoceratans (see Glossary, Box 2.1)
were also around in the Palaeocene in North America and Asia, persisting into the
mid Eocene. These taxa were all apparently predominately omnivorous (taenio-
donts and most condylarths) or were adapted to a diet of relatively nonfibrous
browse (other condylarths, pantodonts, and dinoceratans). There was little evidence
of any herbivores subsisting on fibrous vegetation (with the possible exception of
a couple of smaller condylarth taxa in the latest Palaeocene).
2.2.2 The Eocene Emergence of Modern

Ungulates (55 to 34 Ma)
The early Eocene saw a great diversification of artiodactyls and perissodactyls in

the Northern Hemisphere, and the initial appearance of small to mid-sized mam-
mals that appeared to be specialised folivores (primates as well as ungulates). This
mammalian community shift suggests that something had happened to the structure
of the vegetational habitat in the higher latitudes where these mammals were found,
with leaves now available as a broad dietary resource. The early Eocene vegetation
of mid to high latitudes was similar to that of modern tropical forests in terms of
plant diversity (e.g., Collinson et al. 1981; Wolfe 1985). Eocene fossil localities
supported a wide diversity of small-to-medium sized (5–100 kg) terrestrial herbivores,
unlike the situation in present-day equatorial forest habitats, which have a paucity
of small terrestrial herbivores (Hooker 2000; Janis 2000). The initial diversity of
folivorous ungulates was among the perissodactyls (see Fig. 2.1). The main diver-
sity at this time was among the ‘tapiroids’, mostly sheep-sized or smaller. Some of
these were ancestral to modern tapirs, others to rhinos (first appearing in the middle
Eocene), and others were evolutionary dead ends. Most tapiroids had bilophodont
cheek teeth (see Fig. 2.2B), indicative of folivory. Hyracotheriine Equids, and the
Eurasian equid-related palaeotheres were also common, especially in the early
Eocene, most ranging in size from the proverbial wire-haired fox terrier to the size
of a large sheep. In general, their teeth were more bunodont than those of the
tapiroids and rhinos, indicative of a more omnivorous diet, although some European
forms had more lophed teeth, suggestive of a greater degree of folivory. Other fami-
lies of extinct browsing perissodactyls, included chalicotheres (dog-sized at this
time) and brontotheres (see Glossary, Box 2.1).
Artiodactyls were also common in the early Eocene, at this time small (tragulid-
sized) with bunodont cheek teeth (see Fig. 2.2A) indicative of a generalised omniv-
orous diet. Artiodactyls started to diversify into the modern lineages: suines
(pig-related forms), tylopods (camel-related forms), and ruminants (see Fig. 2.1) in
the late Eocene, coincident with the high-latitude climatic deterioration. Dental
changes included more specialised bunodont cheek teeth in many of the suines,
indicative of a more specialised omnivorous diet, and the evolution of more seleno-
dont cheek teeth (see Fig 2.2D) in the tylopods and ruminants, indicative of more
specialised herbivory. The common selenodont artiodactyls of the late Eocene and
Oligocene were various types of now-extinct small to medium-sized traguloid
ruminants and tylopods.
The declining temperatures in the higher latitudes during the late Eocene
(see Fig. 2.3) basically resulted in the extinction of smaller, generalised omnivorous
forms among all ungulate groups, and the rise of more specialised folivores among
surviving artiodactyl and perissodactyl lineages, while the remaining condylarths
and ungulate-like mammals became extinct (Janis 2000). Most higher latitude
primates also disappeared during this time. The megaherbivore brontotheres also
went extinct at the end of the Eocene, possibly unable to sustain their specialised
browsing diet in higher latitudes at this time. Rhinos persisted through this time
period, but tapiroids were badly hit, and equoids (horses and palaeotheres) became
completely extinct in the Old World. The familiar story of horse evolution relates
an unbroken phylogeny through the North American Eocene, but this masks the
fact that the abundance of individual equid fossils decreased sharply following the
early Eocene diversity of the subfamily Hyracotheriinae. Only in the late Eocene
did equids return to the North American fossil record in abundance, with the first
member of the subfamily Anchitheriinae, the sheep-sized Mesohippus. Mesohippus
was a very different beast from the earlier hyracotheres, with more strongly-lophed
teeth, indicative of committed folivory, and more cursorially-adapted limbs.
This pattern of late Eocene perissodactyl decline and artiodactyl diversification
has often been held as indicative of competitive replacement, ascribed to the
supposedly superior foregut system of digestion in ruminating artiodactyls (see
34 C. Janis
discussion in Janis 1976), but this is not supported by the patterns in the fossil
record (Janis 1989). The late Eocene fossil record shows a general shift from
omnivory to folivory (in terms of adaptive dental morphologies) in rodents as well
as ungulates (Collinson and Hooker 1991; Meng and McKenna 1998). Ruminating
artiodactyls may have been fortuitously better-adapted than perissodactyls to cope
with the changing vegetation of the later Eocene, due to their ability to subsist on
smaller amounts of more selectively chosen vegetation. The climatic changes,
including increased patterns of seasonality, of the later Eocene may have resulted
in changes in vegetational abundance, but perhaps more importantly would proba-
bly have resulted in a greater differentiation of fibre content between plant leaf and
stem, allowing for the selective feeding habits that characterise present-day rumi-
nants (see Janis 1989). Foregut fermentation would also have been useful for the
detoxification of plant secondary compounds (Bodmer and Ward 2006). Small-
to-medium-sized ruminants fare better than hindgut fermenters where food is lower
in quantity but higher in quality. The increased body size of many perissodactyls in
the late Eocene (e.g., among the brontotheres) may also reflect an adaptation to
changing patterns of food abundance and quality.
2.2.3 The Lull Before the Storm: Oligocene

and Early Miocene Times (34 to 20 Ma)
The 10 million years of the Oligocene and the first few million years of the Miocene
were a time of relative calm, following the high latitude temperatures of the later
Eocene, the rapid plunge in temperatures just after the Eocene/Oligocene boundary
(see Fig. 2.3), and the extinctions in Europe (the ‘Grande Coupure’) that resulted
not only from climatic change but from an influx of taxa from Asia (Hooker 2000).
The mid to high latitude climate would have been equable, but also temperate and
seasonal, with winter frosts (see Wolfe 1985). The prominent vegetation was
deciduous forest or woodland, with some areas of relative aridity in North America
(Retallack 2001), but no true modern-type grasslands or deserts.
Among North American and Eurasian ungulates, tapiroids were few, and the
modern family Tapiriidae was now well-established. (Note that at this time there
was not yet faunal exchange between Eurasia and Africa, and India was still an
isolated island.) Rhinos were represented by a variety of forms of medium to large
body size, including the huge (up to 15,000 kg) indricotheres (see Glossary,
Box 2.1). Anchitheriine equids underwent a moderate diversification in North
America, and Anchitherium itself migrated to Eurasia in the early Miocene. All of
these perissodactyls had skulls and dentitions suggestive of generalised browsing.
The general artiodactyl diversity was primarily a continuation of the late Eocene
radiation of small or medium-sized browsers and omnivores.
However, it was among the Oligocene artiodactyls that the first incidences of
hypsodonty appeared among ungulates, at least in the Northern Hemisphere. (While
some native South American ungulates had hypsodont teeth from Eocene times, this
feature alone—as noted below—is not necessarily indicative of grazing behavior, and
many hypsodont notoungulates in fact had dental microwear indicative of browsing
or mixed feeding; Townsend and Croft 2005). Oligocene hypsodont ungulates include
several North American taxa: the diminutive hypertragulid Hypisodus, rock-
hyrax-like oreodonts such as Sespia, and gazelle-like camelids such as Stenomylus
(see Glossary, Box 2.1). The appearance of these hypsodont ungulates has been used
in support of the notion of an early spread of grasslands and grazers in North America
at this time (Retallack 1983), but the cranial morphology of these animals does not
support a grazing diet. These ungulates all had very narrow muzzles, typical of spe-
cialised mixed-feeders in open habitats, where grit or abrasive types of browse were
the main cause of high rates of tooth wear. It is likely that they represented some sort
of specialised selective feeders living in open, arid areas. Some grass may have
formed some part of their diet, but they were far from being specialised grazers. All
of these lineages went extinct without issue during the Miocene: they were not
ancestral to any of the later groups of hypsodont mammals.
2.2.4 The Rise of the Grasslands (20 to 10 Ma)
During the late early Miocene, around 20–17 Ma, both floral and faunal change
became prominent, with mid-latitude grasslands diversifying in North America and
eastern Asia. Grasslands may have been established several million years earlier in
southern South America, and a fauna characterised by ungulates with hypsodont
teeth occurred as early as the early Oligocene (Jacobs et al. 1999; see Fig. 2.3). Sea
levels fell, and a land bridge opened up between Africa and Eurasia, allowing the
migration of proboscideans out of Africa and the immigration into Africa of
northern types of ungulates (rhinos and various ruminants such as bovids and
giraffoids; see Turner and Antón 2004). There is little evidence for grasslands in
Africa at this time, although the middle Miocene (14 Ma) site of Fort Ternan in Kenya
contains evidence of fossil grasses and some ungulates that were probably mixed
feeders (Jacobs et al. 1999).
In North America palaeobotanical evidence exists for grasslands during the late
Oligocene and earliest Miocene from both plant phytoliths (Strömberg 2004, 2006)
and fossil soils (Retallack 2001). However, the soil evidence, as well as evidence
from other organisms such as legumes and insects, points to initial arid bunchgrass
shrubland with more savanna-like short sod grasslands developing in the late early
Miocene (Retallack 2001). The advent of grasslands in North America before the
evolution of hypsodont ungulates such as equids has been argued as evidence for
‘adaptive lag’ (Retallack 1983; Strömberg 2006)—that is, that morphological
evolution had not yet caught up to behavioural evolution. This evolutionary scenario
is highly unlikely: as mentioned previously, hypsodonty is essential for an animal
eating an abrasive diet, and hypsodonty also seems to be an easy feature to evolve,
as it has evolved so many times convergently among mammals (Janis and Fortelius
1988; Damuth and Janis 2005).
36 C. Janis
The first pecorans (horned ruminants), including early members of the modern
families Bovidae, Cervidae, and Giraffidae, made their first appearance in Eurasia
at around 18 Ma. The earliest definite bovid, Eotragus, is known from around
18 Ma in Europe and Pakistan, some cervids of a similar age are known from
Europe, and some large giraffoids are known from the early Miocene of Spain
(see Gentry 2000). New types of ruminant artiodactyls appeared in North America,
including antilocaprids (pronghorns) and palaeomerycids (see Glossary, Box 2.1).
Also in North America larger and more derived types of camelids appeared, along
with a great radiation of the first member of the modern equid subfamily, Equinae,
the genus Merychippus. However, at this point none of these ungulates had
craniodental morphologies indicative of grazing, even though hypsodonty was
apparent among the equids, antilocaprids, and camelids.
This initial rise in the numbers of hypsodont ungulates has long been interpreted
as the radiation of a ‘savanna-like’ fauna (e.g., Webb 1977), but it is not at all clear
that these hypsodont taxa were actually specialised grazers: the majority of the
hypsodont Miocene equids have dental microwear indicating mixed feeding (Solounias
and Semprebon 2002). Many lineages of camelids also became more hypsodont at
this time, but again their craniodental morphology does not support the notion of a
grazing habit (Dompierre and Churcher 1996), and all of these taxa display rather
moderate levels of hypsodonty in comparison with modern grazers.
Until recently, the fate of the browsing ungulates during the Miocene was
thought to have been one of gradual extinction, replaced by the ‘better-adapted’
grazers (see discussion in Janis et al. 2004). In the familiar story of horse evolution
there is rarely a mention of the parallel diversification to the hypsodont forms of
specialised browsing anchitheriine equids that survived into the late Miocene in
both North America and Eurasia, ranging from the goat-sized Archaeohippus up to
the horse-sized Hypohippus and Megahippus (Janis et al. 1994; Agustí and Antón
2002). This radiation of browsing horses is merely a portion of the diversity of late
early and middle Miocene browsers. Others browsers included the palaeomerycids,
the bizarrely-clawed chalicotheres, various types of rhinos, and a diversity of mid-
sized proboscideans such as mastodons (Mammutidae) and gomphotheres (see
Glossary, Box 2.1). The evolutionary pattern of these browsers has been best docu-
mented in North America (Janis et al. 2000, 2002, 2004), but similar patterns are
apparent in western Europe (Fortelius et al. 1996). In North America, not only were
browsing ungulates highly diverse taxonomically on a continental scale (see Fig. 2.4),
but they were also exceedingly species rich at individual fossil localities. The num-
bers of browsing ungulates at mid-Miocene fossil sites in North America greatly
exceeds those in any habitat today, and similar patterns appear to hold true for
ungulates at localities in Eurasia and Africa (Janis et al. 2004), and for terrestrial
herbivores in South America (Kay and Madden 1997) and Australia (Myers 2002).
This abundance of browsers disappears from North America by the start of the late
Miocene, around 11 Ma, and an overall decline of ungulates also occurs at this time
in North America, Western Eurasia, and in Pakistan (Barry et al. 1995). These
palaeocommunities clearly represent some sort of woodland savanna habitat, but
with no precise modern analog.
Fig. 2.4 Continent-wide generic diversity of ungulates in North America. H = hypsodont

(i.e., probable grazers or mixed feeders in open habitats); M = mesodont (somewhat hypsodont;
i.e., probable mixed feeders in open or closed habitats); B = brachydont (i.e., probable browsers).
Modified from Janis et al. 2004
It is not clear how these mid Miocene habitats achieved a greater density of
species, especially in terms of the numbers of browsers, than in their modern-day
equivalents. Our (Janis et al. 2004) preferred explanation is for greater levels of
atmospheric carbon dioxide than the preindustrial levels, leading to greater levels
of primary productivity, but we acknowledge that this hypothesis is at odds with the
current geochemical evidence. Additionally, species richness does not correlate in
a simple fashion with productivity in modern ungulate faunas: Prins and Olff
(1996) note that species richness of African grazers is actually highest at intermedi-
ate levels of grass productivity. Whatever the explanation, this great diversity of mid-
Miocene browsers, occurring concurrently with the early diversification of more
hypsodont, open-habitat forms, is an unappreciated evolutionary event.
2.2.5 The Rise of Grazing Ungulates (10 to 2 Ma)
It was not until the start of the late Miocene, around 10 Ma, that ungulates appeared with
morphologies consistent with more specialised grazing adaptations. Note that, in gen-
eral, the browsing lineages of the earlier Miocene went extinct without issue: although
38 C. Janis
all grazers were obviously derived from browsers at some point in evolutionary history,
the lineages that remained as browsers in the mid Miocene did not later transform into
grazers. The remaining browsers in the late Miocene of North America were consider-
ably bigger in body size than previously, indicating a declining quality and abundance
of suitable forage (Janis et al. 1994).
A classic evolutionary pattern of distinct grazing versus browsing lineages can be
seen among the North American horses. In the early Miocene the equid lineage split
into the one leading to the modern Equinae (with the emergence of the genus
Parahippus) and the lineage leading to the specialised large browsing horses of the
later Miocene (derived members of the Anchitheriinae). These browsing horses (e.g.,
Anchitherium, Hypohippus, Megahippus) were a successful radiation for a good ten
million years, but went extinct without issue during the climatic changes of the later
Miocene. They did not change their diet at this time: instead the radiation into the
grazing niche came from the Equinae line that had opted for a more mixed feeding
diet back in the early Miocene. Another example is the North American dromomery-
cids (palaeomerycid ruminants). The final members of this predominantly browsing
lineage showed some craniodental changes indicative of a more mixed-feeding strat-
egy in the latest Miocene, but still declined in numbers and went extinct fairly rapidly,
while at the same time there was an expansion of the antilocaprids that were previ-
ously more adapted for mixed feeding (Semprebon et al. 2004). The general evolu-
tionary pattern in the late Miocene is for browsing lineages to be replaced by ones
adapted for more fibrous diets (e.g., the replacement of tragulids and cervids by
bovids in the Siwalik faunas of Pakistan; see, e.g., Barry 1995 and Barry et al 1995),
rather than the browsers themselves undergoing evolutionary change. Likewise ungu-
late lineages do not transform back into browsing forms once they have evolved the
more derived craniodental apparatus of grazers or mixed feeders. (A possible excep-
tion to this is the evolution of the specialised high-level browsing gerunuk, Litocranius
walleri, that has a brachydont dentition presumably evolved from the more hypsodont
dentition of other gazelles.)
In North America hypsodont horses, members of the extant tribe Equini and
the extinct tribe Hipparionini (see Glossary, Box 2.1), diversified immensely in the
late Miocene, and one lineage of hipparionines (or perhaps a couple) migrated into
Eurasia by 10 Ma and from there into Africa. This time marks the appearance of the
classic mid-latitude savanna faunas, the ‘Clarendonian chronofauna’ in North
America and the ‘Hipparion fauna’ in Eurasia. These later equines were generally
larger than early members of the Equinae (such as Merychippus), approaching the
size of modern equids (although some secondarily dwarfed lineages also existed),
and had more hypsodont cheek teeth and craniodental features indicative of a more
fibrous diet, such as deeper mandibles and broader muzzles. However, patterns of
dental microwear suggest that these equids were still predominantly mixed feeders
(Hayek et al. 1992; Solounias and Sempebron 2002).
The taxonomic diversity of early late Miocene (around 11–8 Ma) North American
equids was very high: continent-wide there were ten sympatric genera, each with a large
diversity of species, and individual fossil localities commonly contained half a dozen
different equid species (Janis et al., 2004), making equids as taxonomically diverse
as grazing bovids are today in Africa. With the exception of the short-legged
rhinoceros Teleoceras, no other ungulates in North America appeared to challenge
equids as grass specialists until the very end of the Miocene, when hypsodont
proboscideans (gomphotheres) first appeared (Lambert and Shoshoni 1998). During
the late Miocene there was a significant radiation of hypsodont camelids, and also of
more derived antilocaprids (the antilocaprines, which include the modern pronghorn,
and which were more hypsodont than the earlier merycodontine antilocaprids); but
the craniodental features of these taxa (e.g., relatively narrow muzzles) are indicative
of mixed feeding rather than grazing.
In the early late Miocene of Eurasia there was a reduction in numbers of
browsers such as suids, tapirs, tragulids, and brachydont rhinos, suggesting a loss
of forest habitat, and a radiation of bovids, equids, and more hypsodont rhinos and
giraffoids, suggesting the spread of more grass-dominated habitats (Barry 1995;
Fortelius et al. 1996; Agustí and Antón 2002; Costeur et al. 2004.). However, these
faunas contained few undoubted grazing species, the hypsodont forms most likely
being mixed feeders, and the habitat of these localities appears to have been wood-
land and shrubland rather than open savanna (Solounias and Dawson-Saunders
1988; Prins 1998). Few highly hypsodont bovids were present in Africa: the late
Miocene fauna contained a large diversity of suids, rhinos, proboscideans, and
giraffids, as well as bovids and hipparionine equids, probably representing an
assemblage of browsers and mixed feeders (Turner and Antón, 2004).
2.2.6 The Late Cenozoic Dawn of the Modern Ungulate Fauna
Global temperatures fell dramatically in the latest Miocene (see Fig. 2.3), as did
mammalian taxonomic diversity, and the mid-latitude vegetational habitats
apparently became more arid (see e.g., Fortelius et al. 2002). Around 8 Ma, there
was a shift in grass photosynthetic biochemistry from the C3 cycle to the C4 cycle
in lower latitudes in North America and Asia (Cerling et al. 1993, 1997). It has been
claimed that this event had a significant impact on the grazing mammals: a greater
diversity of hypsodont artiodactyls, including bovids, appeared in Pakistan at this
time while brachydont taxa such as suids and giraffids declined (Barry et al. 1995),
but there was little overall impact on the hypsodonty levels of ungulates in North
America (Janis et al. 2000).
Various examples of gigantism existed among ungulates at this time. There
were four sympatric genera of oversized browsing or mixed-feeding camelids
(i.e., ∼1,500 kg) in the latest Miocene and the Pliocene of North America. The North
American rhinos (one browsing and one grazing lineage) also increased in size
throughout the Miocene to reach a similar size to present-day African rhinos by the
end of the epoch. In the Old World, elephantids (all grazers) diversified at this time,
as did large browsers such as chalicotheres, deinotheres (see Glossary, Box 2.1),
and modern types of giraffids. There was also a radiation of the sivatheriine giraf-
fids (see Glossary, Box 2.1) of similar size to the giant camelids.
40 C. Janis
In North America the rhinos were gone by the end of the Miocene, and the giant
camelids did not survive past the Pliocene, but the diversity of large browsers
survived in the Old World through the Pleistocene. Again, the situation of North
America as an island continent probably accounts for its greater (and earlier)
pattern of extinctions of large mammals. The only megaherbivores to survive into
the Pleistocene in North America were the mammutid and gompthotheriid
proboscideans, which were joined by the elephantid Mammuthus (mammoths,
including the woolly mammoth and the Imperial mammoth) at this time, along with
a small number of equids, antilocaprids, and camelids, and the newly immigrant
bovids and cervids. The extinction of all the North American endemic ungulates
and proboscideans at the end of the Pleistocene, with the exception of the
pronghorn, Antilocapra americana, remains enigmatic, with numerous arguments
for climatic change, human hunting, or some combination of the two.
African savanna habitats were first prominent in the Pliocene (Jacobs et al. 1999),
with the initial radiation of specialised grazing bovids such as the hippotragines
(e.g., sable) and alcelaphines (e.g., wildebeest). But the establishment of extensive
open savannas of dry, secondary grasslands, versus woodland savanna or seasonally-
flooded edaphic grasslands, was not apparent until the start of the Pleistocene, around
2 Ma, which is also when C4 grasses first appeared in Africa (Cerling 1992). At this time
the first bovids with definitive specialised craniodental grazing adaptations appeared
(Spencer 1997), evolving from earlier mixed feeding forms, and the total numbers of
grazing taxa approached those of today (Reed 1997). The modern equid genus Equus,
first evolving in North America, appeared in the Old World the late Pliocene, around
2.5 Ma; some remaining hipparionine equids (persistently three-toed forms) survived
alongside Equus for a while, but eventually became extinct by the mid Pleistocene
(Turner and Antón 2004). The grazing white rhino (genus Ceratotherium) also made
its first appearance in the African Pliocene.
During the Pleistocene grazing megaherbivores such as the woolly mammoth
(Mammuthus), and several types of hypsodont rhinos, including the woolly
rhino (Coelodonta), were prevalent at mid to high latitudes, as were horses and
large grazing bovids such as bison (Bison) and musk oxen (Ovibos). This array
of large grazing ungulates, along with some smaller mixed-feeders [e.g., saiga
antelope (Saiga) and reindeer (Rangifer)] comprised the fauna of the late
Pleistocene ‘steppe tundra’ biome of northern Eurasia and Alaska, a habitat type
with no modern analogue (Prins 1998; Agustí and Antón, 2002). In both North
America and the Old World today bovids remain the prime grazers, although their
diversity is much less in the Northern Hemisphere than it is in Africa. A few low
latitude Asian cervids are today at least fresh-grass grazers, such as the barasinga
(Cervus duvaucelli) and Père David’s deer (Elaphurus davidianus; there is no
equivalent radiation among cervids to bovids such as alcelaphines and hippotragines).
However, it is not clear why cervids never really expanded into the grazing niche,
and never colonised Africa south of the Sahara.
Ungulates such as deer, tapirs, and camelids (llamas), as well as now-extinct
or extirpated forms such as gomphothere proboscideans and equids, dispersed
into South America in the Pliocene, around 2.5 Ma (Webb 1991). The equids
and gomphotheres diversified into the grazing niche, but no ungulate is a true
grazer there today. Tapirs have remained as specialised browsers; llamas never
diversified much beyond their present day high-altitude, mixed-feeding eco-
logical role; and while cervids underwent the most profound diversification,
they never evolved true grazing forms, although the marsh deer (Blastocerus
dichotomous) is a fresh grass grazer. Bovids never reached South America until
introduced by humans, and it is unclear why no cervid evolved there to be a
specialised grazer.
2.3 Discussion and Conclusions
Grazing is a fairly recent evolutionary specialization among ungulates. Although

extensive grasslands first appeared around 25 Ma—and along with them ungulates
with hypsodont teeth designed to withstand a greater amount of abrasion—
specialised grazers did not appear until around 10 Ma, and most true grazers are of
Plio-Pleistocene age (Fig. 2.3). Many different lineages have evolved intermediate
feeding types, but few have progressed to specialised grazing.
Although the majority of grazers today are bovids—and specialised grazing
evolved a minimum of four times within this clade (within the Bovini, Caprini,
Hippotragini, and Alcelaphini)—no other ruminant, past or present, has evolved a
specialised grazing form (although a few modern cervids, such as Blastocerus,
Elaphurus and species of Cervus and Axis, have a grass-dominated diet). Within
other artiodactyls, grazers have appeared recently (Plio-Pleistocene) among suids
(the warthog) and hippos, but were not in evidence before this time. In contrast,
grazing appears to have been easier to evolve among hindgut fermenters.
Clauss et al. (2003) present an elegant argument for why ruminants, past and
present, do not attain the size of rhinos and elephants. Ruminants have longer
passage times of the digesta than hindgut fermenters of a comparable body size
(around 40 hours for a horse versus around 60 hours for a cow), and the time for
passage of the digesta increases with body size. However, if the passage time
approaches 4 days, an acute problem develops with the growth of methanogenic
bacteria, which convert acetic acid (the main volatile fatty acid product of cellulose
fermentation) to methane and carbon dioxide, with resultant high energy losses.
Elephants solve this problem by adopting relatively shorter and broader guts, thus
speeding up the passage rate of the digesta; but this would not be a possible solution
for a ruminant, where the rumenoreticulum portion of the stomach is specifically
adapted to delay food passage. Thus ruminants are limited to body sizes where their
digesta passage time is less than four days (likely under 1,500 kg).
Within the equids, several different clades in the subfamily Equinae progressed from
mixed feeding to grazing (although only Equus survives). Grazing appeared
independently within several different clades of rhinos, including the white rhino,
and among extinct forms the North American Teleoceras, and the Old World woolly
rhino (Coelodonta) and steppe rhino (Elasmotherium). Grazing also evolved
42 C. Janis
three times within the Proboscidea (within the Elephantidae, Stegodontidae, and
Gomphotheriidae), and once in the Hyracoidea (Procavia).
It is commonly thought that bovids somehow out-competed equids, but this is
erroneous. Equids were predominantly a New World radiation, and their decline in
taxonomic diversity in the late Cenozoic was due to being stranded in an island con-
tinent, with no tropical refuge zone when the more productive savanna grasslands
turned to less productive prairie. North American equids never encountered bovids
until the immigration of sheep and bison in the Pleistocene, by which time only one
genus, Equus, remained. However, this taxon was still a very prominent part of the
Pleistocene faunas, and the fossils are extremely numerous, suggesting individual
abundance in life. North American equids only went extinct in the end-Pleistocene
extinctions that decimated the megafauna. Equids that migrated into the Old World
encountered an existing bovid-dominated faunal community, which may explain
their lack of taxonomic diversification there in comparison with North America.
While modern equids have diversified into a variety of different species (e.g., asses in
Asia, zebras in Africa), and are individually numerous when encountered, there is still
usually only one species of equid in any faunal community (this was also broadly true
of Old World fossil communities), and all extant equid species belong to a single
genus with essentially the same ecology (i.e., herd-living, specialised grazing).
The bias of today’s world suggests a normality of the predominance of ruminant
grazers, and also a diversity of grazers among other ungulates. However, the fossil
record shows us that we need to think more about why grazing is so difficult to
attain in non-bovid ruminants, and why in general it appears to be easy to evolve
an animal that takes some grass in its diet, but not to evolve a specialised grazer.
Acknowledgements This chapter was greatly improved by comments from Herbert Prins and an
anonymous reviewer.
References
Agustí J, Antón M (2002) Mammoths, sabertooths, and hominids: 65 million years of mammalian
evolution in Europe. Columbia University Press, New York
Barry JC (1995) Faunal turnover and diversity in the terrestrial Neogene of Pakistan. In: Vrba ES,
Denton GH, Partridge TC, Buckle LH (eds) Paleoclimate and evolution with emphasis on
human origins. Yale University Press, New Haven, pp 115–134
Barry JC, Morgan ME, Flynn LJ, Pilbeam D, Jacobs LL, Lindsay EH, Raza SM, Solounias N
(1995) Patterns of faunal diversity and turnover in the Neogene Siwaliks of northern Pakistan.
Palaeogeogr. Palaeoclimatol 115:209–226
Benton MJ (1993) The fossil record 2. Chapman and Hall, London
Bodmer R, Ward D (2006) Frugivory in large mammalian herbivores. In: Danell K, Duncan P,
Bergström R, Pastor J (eds) Large herbivore ecosystem dynamics and conservation. Cambridge
University Press, Cambridge, pp 232–260
Cerling TE (1992) Development of grasslands and savannas in East Africa during the Neogene.
Cerling TE, Wang Y, Quade J (1993) Global ecological change in the late Miocene: expansion of
C4 ecosystems. Nature 361:344–345
Cerling TE, Harris JM, MacFadden BJ, Leakey MG, Quade J, Eisenmann V, Ehleringer JR (1997)
Global vegetation change through the Miocene-Pliocene boundary. Nature 389:153–158
Clauss M, Frey R, Kiefer B, Lechner-Doll M, Loehlein W, Polster C, Rössner GE, Streich WJ
(2003) The maximum attainable body size of herbivorous mammals: morphophysiological
constraints on foregut, and adaptations of hindgut fermenters. Oecologia 136:14–27
Collinson ME, Fowler MK, Boulter MC (1981) Floristic change indicates a cooling climate in the
Eocene of southern England. Nature 291:315–317
Collinson ME, Hooker JJ (1991) Fossil evidence of interactions between plants and plant-eating
mammals. Philos T Roy Soc B 333:197–208
Costeur L, Legendre S, Escarguel G (2004) European large mammals palaeobiogeography and
biodiversity during the Neogene. Palaeogeographic and climatic impacts. Rev Paeobiol
Geneve 9:99–109
Damuth JD, Fortelius M, Andrews P, Badgley C, Hadley EA, Hixon S, Janis C, Madden RH, Reed K,
Smith FA, Theodor J, Van Dam JA, Van Valkenburgh B, Werdelin L (2002) Reconstructing
mean annual precipitation based on mammalian dental morphology and local species richness.
J Vertebr Paleontol 22(suppl):48A
Damuth J, Janis CM (2005) Paleoecological inferences using tooth wear rates, hypsodonty and
life history in ungulates. J Vertebr Paleontol 25(suppl):49A
Dompierre H, Churcher CS (1996) Premaxillary shape as an indicator of the diet of seven extinct
late Cenozoic New World camelids. J Vertebr Paleontol 16:141–148
Fortelius M, Werdelin L, Andrews P, Bernor RL, Gentry A, Humphrey L, Mittman H-W, Viranta S
(1996) Provinciality, diversity, turnover, and paleoecology in land mammal faunas of the later
Miocene of western Eurasia. In: Bernor RL, Fahlbusch V, Mittman H-W (eds) The evolution
of western Eurasian Neogene mammal faunas. Columbia University Press, New York,
pp 414–448
Fortelius M, Eronen J, Jernvall J, Liu L, Pushkina D, Rinne J, Tesakov A, Vislobokva I, Zhang Z,
Zhou L (2002) Fossil mammals resolve regional patterns of Eurasian climatic change over 20
million years. Evol Ecol Res 4:1005–1016
Gentry AW (2000) The ruminant radiation. In: Vrba ES, Schaller GB (eds) Antelopes, deer and
relatives. Yale University Press, New Haven, pp 11–25
Gordon IJ (2003) Browsing and grazing ruminants: are they different beasts? Forest Ecol Manag
181:13–21
Gordon IJ, Illius AW (1988) Incisor arcade structure and diet selection in ruminants. Funct Ecol
2:15–22
Hayek L-A, Bernor RL, Solounias N, Steigerwald P (1992) Preliminary studies of hipparionine
horse diet as measured by tooth microwear. Ann Zool Fenn 28:187–200
Hooker JJ (2000) Palaeogene mammals: crises and ecological change. In: Culver SJ, Rawson
PF (eds) Biotic response to global change: the last 145 million years. Cambridge University
Press, Cambridge, pp 333–349
Jacobs BF, Kingston JD, Jacobs LL (1999) The origin of grass-dominated ecosystems. Ann Mo
Bot Gard 86:590–643
Janis C (1976) The evolutionary strategy of the Equidae, and the origin of rumen and cecal diges-
tion. Evolution 30:757–774
Janis CM (1988) An estimation of tooth volume and hypsodonty indices in ungulate mammals,
and the correlation of these factors with dietary preferences. Mem Mus Hist Nat, Paris C
53:367–387
Janis CM (1989) A climatic explanation for patterns of evolutionary diversity in ungulate mam-
mals. Palaeontology 32:463–481
Janis CM (1993) Tertiary mammal evolution in the context of changing climates, vegetation, and
tectonic events. Annu Rev Ecol Syst 24:467–500
Janis CM (1995) Correlation between craniodental morphology and feeding behavior in
ungulates: reciprocal illumination between living and fossil taxa. In: Thomason JJ (ed)
Functional morphology in vertebrate paleontology. Cambridge University Press, Cambridge,
pp 76–98
44 C. Janis
Janis, CM (2000) Patterns in the evolution of herbivory in large terrestrial mammals: the Paleogene
of North America. In: Sues H-D, Labanderia C (eds) Origin and evolution of
herbivory in terrestrial vertebrates. Cambridge University Press, Cambridge, pp 168–221
Janis CM, Fortelius M (1988) On the means whereby mammals achieve increased functional
durability of their dentitions, with special reference to limiting factors. Biol Rev 63:197–230
Janis CM, Gordon I, Illius A (1994) Modelling equid/ruminant competition in the fossil record.
Hist Biol 8:15–29
Janis CM, Scott KM, Jacobs LL (1998) Evolution of Tertiary mammals of North America.
Cambridge University Press, Cambridge
Janis CM, Damuth J, Theodor JM (2000) Miocene ungulates and terrestrial primary productivity:
Where have all the browsers gone? Proc Natl Acad Sci 97:7899–7904
Janis CM, Damuth J, Theodor JM (2002) The origins and evolution of the North American grassland
biome: the story from the hooved mammals. Palaeogeogr. Palaeoclimatol 177:183–198
Janis CM, Damuth J, Theodor JM (2004) The species richness of Miocene browsers, and implica-
tions for habitat type and primary productivity in the North American grassland biome.
Jernvall J, Hunter JP, Fortelius M (1996) Molar tooth diversity, disparity, and ecology in Cenozoic
ungulate radiations. Science 274:1489–1492
Kay RF, Madden RH (1997) Mammals and rainfall: paleoecology of the middle Miocene at La
Venta (Colombia, South America). J Hum Evol 32:161–199
Kowalevsky W (1876) Sur l’Anchitherium aurelianse Cuv. et sur l’histoire paléontologie des
chevaux. Mem Acad Imp Sciences de St. Petersburg, VII, 20:1–73
Lambert WD, Shoshoni J (1998) Proboscidea. In Janis CM, Scott KM, Jacobs LL (eds) Evolution
of Tertiary mammals of North America. Cambridge University Press, Cambridge, pp
606–621
Leopold E, Liu G, Clay-Poole S (1992) Low biomass vegetation in the Oligocene? In: Prothero
DR, Berggren WA (eds) Eocene–Oligocene climatic change and biotic evolution. Princeton
University Press, Princeton, pp 399–420
Matthew WD (1926) The evolution of the horse: its record and interpretation. Quart Rev Biol
1:139–185
Mendoza M, Janis CM, Palmqvist P (2002) Characterizing complex craniodental patterns related
to feeding behaviour in ungulates: a multivariate approach. J Zool Lond 58:223–246
Meng J, McKenna MC (1998) Faunal turnovers of Palaeogene mammals from the Mongolian
Plateau. Nature 394:364–367
Myers TJM (2002) Paleoecology of Oligo-Miocene local faunas from Riversleigh. Unpublished
PhD Thesis. University of New South Wales, Sydney
Pérez-Barbería FJ, Gordon IJ, Nores C (2001) Evolutionary transitions among feeding styles and
habitats in ungulates. Evol Ecol Res 3:221–230
Prasad V, Strömberg CAE, Alimohammadian H, Sahni A (2005) Dinosaur coprolites and the early
evolution of grasses and grazers. Science 310:1177–1180
Prins HHT (1998) The origins of grassland communities in northwestern Europe. In: Wallis
deVries MF, Bakker JP, van Wieren SE (eds) Grazing and conservation management. Kluwer,
Boston, pp 55–105
Prins HHT, Olff H (1996) Species-richness of African grazer assemblages: towards a functional
explanation. In: Newbery DM, Prins HHT, Brown ND (eds) Dynamics of tropical communities.
Blackwell, Oxford, pp 449–490
Prothero, DR, Schoch RM (2002) Horns, tusks and flippers: the evolution of hoofed mammals.
Johns Hopkins Press, Baltimore
Reed, KE (1997) Early hominid evolution and ecological change through the Africa Plio-
Pleistocene. J Hum Evol 32:289–322
Retallack GJ (1983) Late Eocene and Oligocene paleosols from Badlands National Park, South
Dakota. Geol Soc Am Spec Pap 193:1–82
Retallack GJ (2001) Cenozoic expansion of grasslands and climatic cooling. J Geol 109:407–426
Rose KD (2006) The beginning of the age of mammals. Johns Hopkins Press, Baltimore
Semprebon G, Janis CM, Solounias N (2004) The diets of the Dromomerycidae (Mammalia;
Artiodactyla) and their response to Miocene climatic and vegetational change. J Vertebr
Paleontol 24:430–447
Solounias N, Dawson-Saunders B (1988) Dietary adaptations and paleoecology of the late Miocene
ruminants from Pikermi and Samos in Greece. Palaeogeogr. Palaeoclimatol 65:149–172
Solounias N, Semprebon G (2002) Advances in the reconstruction of ungulate ecomorphology
with application to early fossil equids. Am Mus Novit 3366:1–49
Spencer LM (1997) Dietary adaptations of Plio-Pleistocene Bovidae: implications for hominid
habitat use. J Hum Evol 32:201–228
Sponheimer M, Reed KE, Lee-Thorpe JA (1999) Comparing isotopic and ecomorphological data
to refine bovid paleodietary reconstruction. J Hum Evol 36:705–718
Springer MS, Murphy WJ, Eizirik E, O’Brien SJ (2005) Molecular evidence for major placental
clades. In Rose KD, Archibald JD (eds) The rise of placental mammals: origins and relation-
ships of major clades. John Hopkins Press, Baltimore, pp 37–49
Strömberg CAE (2004) Using phytolith assemblages to reconstruct the origin and spread of grass-
dominated habitats in the great plains of North America during the late Eocene to early
Miocene. Palaeogeogr. Palaeoclimatol 207:239–275
Strömberg CAE (2006) Evolution of hypsodonty in equids: testing a hypothesis of adaptation.
Paleobiology 32: 236–258
Townsend KE, Croft DA (2005) Low-magnification microwear analyses of South American
endemic herbivores. J Vertebr Paleontol 25 (suppl):123A
Turner A, Antón M (2004) Evolving Eden: an illustrated guide to the evolution of the African
large-mammal faunas. Columbia University Press, New York
Webb SD (1977) A history of savanna vertebrates in the New World. Part I: North America. Annu
Rev Ecol Syst 8:355–380
Webb SD (1991) Ecogeography and the Great American Interchange. Paleobiology 17:266–280
Wolfe JA (1985) Distribution of major vegetational types during the Tertiary. Geophys Monogr
32:357–375
Zachos J, Pagani M, Sloan L, Thomas E, Billups K (2001) Trends, rhythms, and aberrations in
global climate 65 Ma to present. Science 292:686–693
Chapter 3
The Morphophysiological Adaptations
of Browsing and Grazing Mammals
Marcus Clauss, Thomas Kaiser, and Jürgen Hummel
3.1 Introduction
The behaviour, physiology and morphology of animals are the outcome of

adaptations to particular ecological niches they occupy or once occupied. Studying
the correlation between a given set of characteristics of an ecological niche and the
morphological and physiological adaptations of organisms to these characteristics
is one of the most basic approaches to comparative biology, and has fuelled
scientific interest for generations (Gould 2002). However, current scientific stand-
ards cannot be met by mere descriptions of both the characteristics of the niche and
the organism, and a (hypothetical) intuitive explanation for the adaptive relevance
of the latter; the presence or absence of a characteristic must be demonstrated in
sound statistical terms (Hagen 2003)1 ideally supported by experimental data (from
in vivo, in vitro, or model assays) on its adaptive relevance.
In this chapter, we adopt an approach that first presents the relevant characteristics
of the ecological niche of the ‘grazer’ (GR) and of the ‘browser’ (BR), outlines
hypotheses based on these characteristics, and finally addresses examples where
such hypotheses have been tested. As we consider that the discussion of morpho-
physiological differences between GR and BR is still unresolved, subjects of future
interest, such as the particular adaptations of mixed feeders, or more elaborate clas-
sifications of feeding types (e.g., including frugivores, and differences between
grasses, sedges, forbs, herbs, and woody browse), are not considered here. With
respect to botanical entities, ‘browse’ in this chapter refers to herbs, forbs, and
leaves and twigs of woody plants.
The terms ‘grazer’ and ‘browser’ have been used for a long time to characterise
feeding types; however, it was Hofmann and co-workers (Hofmann and Stewart
1972; Hofmann 1973; 1988; 1989; 1991; 1999) who brought them into common use.
Partly in connection with the original term ‘concentrate selector’, which will not be
adopted in this chapter (Clauss et al. 2003b; 2003c), the term ‘browser’ has become
synonymous with an organism feeding selectively on relatively easily digestible
1
But see recent methodological work in human medicine showing that statistical significance
alone cannot be used as an argument to support or falsify a hypothesis (Ioannidis 2005).
© Springer 2008
48 M. Clauss et al.
material. Hence, even mites (Siepel and de Ruiter-Dijkman 1993) and carnivo-
rous fish (Lechanteur and Griffiths 2003) have been classified as ‘grazers’ and
‘browsers’. In this chapter, these terms are used strictly in relation to their botani-
cal connotation and are not used as indicators of selectivity. Demment and
Longhurst (1987) proposed a classification scheme that demonstrated that there
are both selective and unselective species within the GR and BR classes.
Selectivity generally decreases with body size (Jarman 1974; Owen-Smith 1988),
and differences between feeding type on the one hand and degree of selectivity
on the other have been incorporated into a model to explain niche separation
(Owen-Smith 1985).
Potential adaptations to browse or grass diets have often been compared to
consequences of difference in body mass between species (Hofmann 1989; Gordon
and Illius 1994; Gordon and Illius 1996). In this chapter, therefore, body mass is
only included as an alternative explanation, but the influence of body mass itself on
digestive processes is not reviewed.
3.2 Grass and Browse
Whereas data compilations of animal species have been published in large number
(see Sects. 3.3 and 3.5 for references), there is, as far as we are aware, a surprising
lack of any systematic evaluation of differences between grasses and browses in
terms of their physical and chemical characteristics. In other words, the debate
about differences between grazers and browsers is often based on hearsay, as far as
the assumed differences between grass and browse are concerned; for example, the
often quoted increased amount of grit adhering to grass forage is a conceptual
cornerstone of many investigations on the hypsodont dentition of grazers (Fortelius
1985; Janis 1988; Janis and Fortelius 1988; Williams and Kay 2001), but has never
been demonstrated quantitatively. Here, we only cite works that generated or at
least collated comparative data (even if not statistically testing differences). When
considering the literature, we think there is agreement on the forage characteristics
(Table 3.1) that are of relevance for the topic of this chapter.
Growth Pattern/Location
These have the potential to influence overall body design and the food selection
mechanism.
A1. It is generally assumed that grasses predominate in open landscapes,
whereas browse predominates in forests or spatially more structured landscapes.
A2. It is generally assumed that grasses typically grow close to the ground (with
evident exceptions such as napier grass), whereas browse grows at different heights
(with forbs often at even lower growth levels than surrounding grasses, and woody
browse of shrubs and trees mostly above grass level).
3 The Morphophysiological Adaptations of Browsing and Grazing Mammals 49
Table 3.1 Summary of characteristics of browse and grass used for the generation of predictions
of morphophysiological differences between browsers and grazers. The functional relevance (FR)
code links these predictions to the following tables. See text for more detailed explanations and
references
Subject groups FR Characteristic Browse Grass
Growth A1 landscape forests/spatially open
pattern/location more structured
A2 growth pattern at different heights mostly close
to ground
A3 nutritional homogeneity less more
of a ‘bite’
Chemical B1 protein content higher (including lower
composition nitrogenous
secondary
compounds)
B2 fibre content lower but higher but
more lignified less lignified
B3 pectin content higher lower
B4 secondary compounds more less
Physical C1 abrasive silica less more
characteristics C2 adhering grit less more
C3 resistance to chewing less more
C4 fracture pattern polygonal longish
fibre-like
C5 change in specific less more
gravity during
fermentation
Digestion/ D1 overall digestibility lower higher
fermentation
D2 speed of digestion fast slow
A3. On the scale of single bites, differences in nutritional quality are more pro-
nounced in browse (Van Soest 1996).
Chemical Composition
These have the potential to influence overall metabolism.

B1. Grass generally contains less protein than browse (Dougall et al. 1964;
Owen-Smith 1982; Codron et al. 2007a). The high protein content of browse should
be regarded with some caution, since some part of the nitrogen in browse may stem
from, or may be bound to, plant secondary compounds (Oftedal 1991). For the
calculation of crude protein content, nitrogen content it is generally multiplied by
6.25; in contrast, Milton and Dintzis (1981) suggest that this nitrogen conversion
factor should be as low as 4.4 for tropical browse.
50 M. Clauss et al.
B2. Grass contains more fibre, and a greater proportion of this fibre is cellu-
lose, while browse has less total but more lignified fibre (Short et al. 1974;
Oldemeyer et al. 1977; Owen-Smith 1982; McDowell et al. 1983; Cork and
Foley 1991; Robbins 1993; Van Wieren 1996b; Iason and Van Wieren 1999;
Holechek et al. 2004; Hummel et al. 2006; Codron et al. 2007a). These differ-
ences are more pronounced if C4 grasses are compared to browse (Caswell et al.
1973). The fact that no difference in fibre content between grass and browse
was demonstrated in a comprehensive set of samples of East African forage
plants (Dougall et al. 1964) is explained by the inclusion of twigs in the browse
analysis and the use of the crude fibre method to estimate fibre content, which
can considerably underestimate lignin and hemicellulose content of tropical
forage, especially tropical grasses (Van Soest 1975; McCammon-Feldman
et al. 1981).
B3. Although few data exist, grass and browse contain comparable low levels of
easily digestible carbohydrates, such as sugar and starch (Cork and Foley 1991;
Robbins 1993). This is different for pectins, an easily fermentable part of the cell
wall, which is much more prominent in the browse cell wall at concentrations of
6–12 % of total forage dry matter (Robbins 1993).
B4. Browse leaves contain secondary plant compounds that can act as feeding
deterrents either by poisoning or reducing plant digestibility (Freeland and Janzen
1974; Bryant et al. 1992; Iason and Van Wieren 1999). Common secondary plant
compounds such as tannins occur more often in woody browse (80% of taxa) as
compared to forbs (15% of taxa; Rhoades and Gates 1976) (see Duncan and Poppi
in this book, Chapter 4).
Physical Characteristics
These have the potential to influence adaptations of oral food processing, and might
be important drivers of the differentiation of ruminant forestomach morphology.
C1. Grasses contain abrasive silica (Dougall et al. 1964; McNaughton et al.
1985); silica is harder than tooth enamel and thus wears it down (Baker et al. 1959;
but see Sanson et al. 2007).
C2. As grasses grow close to the ground, it is assumed that grass forage contains
more adhering grit than browse forage, but as stated before, this has not been tested
quantitatively. Herbs, typically included in the ‘browse’ category, should share this
characteristic with grasses.
C3. Differences in the masticatory force required to comminute grass/browse
have been hypothesised (e.g., Solounias and Dawson-Saunders 1988; Mendoza
et al. 2002), but not described. Spalinger et al. (1986) attribute thicker cell walls to
grass leaves than to forb and woody browse leaves (while cell walls of twigs were
thickest). The grinding of C4 grasses needs distinctively more force than that of C3
grasses (Caswell et al. 1973), possibly due to a greater percentage of bundle sheaths
in C4 grasses (Heckathorn et al. 1999).
C4. Differences in fracture patterns of grass and browse have been noted
(Spalinger et al. 1986; Kay 1993; Van Wieren 1996a). Several authors have
reported more polygonal particles from herbaceous forage leaves and more
longish particles from grass leaves (Troelsen and Campbell 1968; Moseley and
Jones 1984; Mtengeti et al. 1995). Although empirical studies are lacking,
browse is thought to be a more heterogeneous material with different levels of
tissue thicknesses and of resistance to breakage, whereas grass is considered
more homogenous in this respect. The fibre bundles in grasses are believed to
be more evenly distributed and at higher density than in most browse species
(Sanson 1989).
C5. Once submitted to fermentation, different forages show different buoyancy
characteristics, due to differences in fibre composition, fracture shape, hydration
capacity, and bacterial attachment (Martz and Belyea 1986; Lirette et al. 1990;
Wattiaux et al. 1992); particles rich in cellulose are expected to change their
functional density at a slower rate. Nocek and Kohn (1987) and Bailoni et al.
(1998) found that the absolute change in functional specific gravity was greater
for grass than for alfalfa hays, suggesting that there may be systematic differ-
ences between forages.
Digestion/Fermentation Characteristics
These have the potential to influence overall digestive physiology.

D1. Grass should yield more energy from fermentation per unit forage
(Codron et al. 2007a), which explains why both in vivo and in vitro overall
digestibility (generally measured as digestibility after fermentation times >24 h)
are often found to be higher in grass than in woody browse (Wofford and
Holechek 1982; Wilman and Riley 1993; Van Wieren 1996b; Hummel et al.
2006). However, the evidence is equivocal, as some other references give higher
digestibilities for woody browse than for grass (Short et al. 1974; Blair et al.
1977; Holechek et al. 2004). All references agree that forbs have a compara-
tively high digestibility.
D2. Browse is fermented at a faster rate than grass during the initial stages of
fermentation (Short et al. 1974; Holechek et al. 2004; Hummel et al. 2006). This
characteristic distinction has been well established for alfalfa (higher fermenta-
tion rate; lower digestibility) and for grass (lower fermentation rate; higher
digestibility; Waldo et al. 1972). The reasons for this might be differences in cell
wall structure, but have not been explored in detail. In vivo digestion of different
forages confirms the pattern found in vitro (Short 1975). In comparison to C3
grasses and legumes, C4 grasses have some characteristics, such as a more
strongly attached epidermis, less loosely arranged mesophyll cells, and a paren-
chymal bundle sheath, which all may prolong the fermentative process (Wilson
1993), and a slower fermentation rate for a C4 than a C3 grass has been demon-
strated by Wilson et al. (1989).
52 M. Clauss et al.
3.3 Predictions
Based on the knowledge of chemical and physical differences between grass

and browse, we can put forward hypotheses regarding the morphological
adaptations of herbivores to these feeding niches (Table 3.2). In this section, we
include visual analyses of data, for example comparative graphs or tables that
have not been assessed statistically; such works are considered exploratory or
preliminary here. We do not infer such contributions to be of minor relevance;
they simply represent an early step in hypothesis testing. Publications that
actually test such predictions statistically are not included in this section but appear
later in section 3.5 (Results).
It must be noted that the alternative hypothesis to most of the predictions
listed here is that there is no influence of feeding type (i. e., browser or grazer)
on the respective parameter, but only an influence of body mass and/or
phylogenetic descent.
Chemical and Physical Ingesta Properties
B1-B4, C4-C5, D2.2 It is to be expected that stomach contents of BR and GR differ

systematically in their content of chemical, physical, and fermentation characteris-
tics. To our knowledge, no systematic evaluation of protein or fibre content, or of
fracture or buoyancy properties, has been performed of the gut contents of
herbivores of different feeding types. Owen-Smith (1988) demonstrated correla-
tions between body size and the protein or non-stem fractions in the gut contents
of herbivores, thus confirming a principal correlation between body size and
selectivity; however, no test for differences between feeding types was performed.
It has been suggested that differences in fermentation patterns should be reflected
in fermentation rates and their products between BR and GR ruminants (Hofmann
1989), or that differences in content of easily digestible carbohydrates should lead
to differences in ruminal pH (Gordon and Illius 1996)—a fact confirmed for a
limited number of species by Jones et al. (2001).
Overall Body Design
A1. It is expected that most GR exhibit adaptations in their limb anatomy to

living in open habitats, enabling them to use the available open space by increased
cursorial activity. Actually, many correlations between postcranial skeleton parameters
2
These codes refer to the characteristics of grass and browse outlined in the previous section.
Table 3.2 Summary of predicted morphophysiological differences between grazers and browsers
based on the plant characteristics summarised in Table 3.1 [use the functional relevance (FR) code
to link plant and animal characteristics]. See text for more detailed explanations and references
Subject groups FR Characteristic Browser Grazer
Ingesta B1, B2 gut contents nutri- higher protein?, lower protein?
propertiesa ent content lower fibre/ higher fibre/
higher lignin lower lignin
B3/D2 gut contents higher? lower?
volatile fatty
acid production
rate
C4 gut contents polygonal longish fibre-like
particle pattern
C5 gut contents homogeneous inhomogeneous
buoancy
characteristics
Overall body A1 limb anatomy shorter longer
design
A2 adaptations to less pronounced: more pronounced:
feeding on the thoracic thoracic verte-
ground vertebrae hump, brae hump, mus-
muscles sup- cles supporting
porting head head and skull in
and skull in grass cropping,
grass cropping, paracondylar
paracondylar and glenoid
and glenoid attachment areas,
attachment face length
areas, face
length
A3 oral anatomy more pronounced less pronounced
for selective incisor incisor
feeding differentiation, a differentiation,
pointed/narrow a square/wide
incisor arcade, incisor arcade,
a short muz- a long muzzle
zle width with width with short
long lips, a long lips, a short
mouth opening, mouth opining,
pronounced lip less flexible lip
muscles muscles
Metabolism B1 protein require- higher? lower?
ments
B4 adaptations against salivary tannin- no salivary tannin-
secondary plant binding binding proteins,
compounds proteins and smaller livers/
larger salivary decreased
glands, larger detoxification
livers/increased capacity
detoxification
capacity
(continued)
54 M. Clauss et al.

Dental/buccal C1, C2 adaptations against no hypsod- hypsodonty; enamel
morphology abrasion onty; enamel adaptations;
adaptations consequences
of hypsodonty
such as deeper
mandibles,
posterior
displacement of
orbita
C3 adaptations to dif- slender mandible, robust mandible,
ferent chewing longer molar shorter molar
forces tooth row, wider tooth row,
palate, smaller narrower palate,
masseter and larger masseter
insertion areas and insertion
areas
C4 adaptations to dif- tooth morphology; tooth morphology;
ferent fracture when consuming when consuming
properties grass forage: grass forage:
larger ingesta smaller ingesta
particles particles
General digestive D1 daily food intake higher lower
physiology
D2 adaptations to shorter ingesta reten- longer ingesta
differences tion; on a compa- retention; on
in digestion rable diet: lower a comparable
kinetics digestibility; diet: higher
lower gut capac- digestibility;
ity; increased greater gut
faecal losses capacity; lower
(sodium); less faecal losses;
cellulolytic activ- more cellulolytic
ity, less diverse activity; diverse
protozoal fauna protozoal fauna
Ruminant C4, C5 adaptations to the weaker rumen pillars, stronger rumen
forestomach tendency of lower reticular pillars, higher
physiology forage to stratify crests, smaller reticular crests,
omasa, more larger omasa, less
viscous saliva, viscous saliva,
frothy contents, stratified contents
complete rumen with fibre mat,
papillation, more fluid layer
acid-producing and gas dome,
abomasal mucosa unpapillated
dorsal rumen
area, less acid-
producing
abomasal mucosa
(continued)

Readiness to C1–C5 problems observed less acceptance of good acceptance of
ingest forage in zoo animals grass hay grass hay
in captivity
a
Predictions about ingesta properties are made more difficult by the fact that gut contents always
consist of a mixture of material recently ingested, of material that has been digested for varying
lengths of time, and of symbiotic gut microbes
and habitat have been demonstrated in bovids (Scott 1985, 1987; Kappelman 1988;
Köhler 1993; Plummer and Bishop 1994; Kappelmann et al. 1997; DeGusta and
Vrba 2003; DeGusta and Vrba 2005a; DeGusta and Vrba 2005b; Mendoza and
Palmqvist 2006b). As these represent morphological correlates of habitat rather
than diet, they are not dealt with explicitly in this chapter.
A2. In accordance with comparisons using two species (Haschick and Kerley
1996; du Plessis et al. 2004), it is expected that the preferred feeding height of GR
is lower than that of BR; consequently, it is expected that GR show adaptations to
feeding close to the ground. A lower angle between braincase and the facial
cranium should be a positional adaptation to ground feeding. The peak of the hump
of the thoracic vertebrae should be correlated with the preferred feeding height,
with a tall hump close to the head being advantageous for ground-level feeding
(increases the moment arm of the nuchal musculature; Guthrie 1990); however,
results from analyses on this hump so far are equivocal (Spencer 1995), which
could be due to the fact that those browsers feeding on herbs/forbs would, by
necessity, also have to feed close to the ground. Muscles supporting head and skull
movements in grass cropping should be more pronounced in GR with accordingly
more pronounced attachment areas (paracondlyar and glenoid). A longer face could
serve to keep the eyes away from the grass, which might protect them (Janis 1995),
help to maintain good visibility for predator detection during feeding (Gentry
1980), and/or enhance moment arms of the head and mandible for more efficient
cropping and mastication. An interesting potential example of a skeletal feeding
height adaptation is the decreasing length of the metatarsus in Sivatheres with
increasing proportion of grass in their diet (Cerling et al. 2005).
A3. BR are expected to show adaptations for a more selective feeding, whereas
GR are expected to show adaptations for a more unselective food intake. Browsers
are expected to have a higher dental incisor index (i.e., a more pronounced size
difference between the individual incisor teeth and the canine teeth; Boué 1970).
BR are expected to have a lower muzzle width, a more pointed and narrower incisor
arcade than GR, which have more square incisor arcades (Boué 1970; Bell 1971;
Owen-Smith 1982; Bunnell and Gillingham 1985; Gordon and Illius 1988;
Solounias et al. 1988; Solounias and Moelleken 1993). GR are assumed to have a
small mouth opening and short lips, whereas BR should have a larger mouth
opening and longer lips (Hofmann 1988). As BR are thought to need more flexible
56 M. Clauss et al.
lips, a larger lip muscle attachment area has been suggested in BR (Solounias and
Dawson-Saunders 1988), and a seemingly larger size of the infraorbital and stylo-
mastoidal foramina in BR ruminants has been interpreted as indicative of more
innervation of the lip muscles compared to GR (Solounias and Moelleken 1999).
Metabolism
B1. In the zoo animal literature, it has been proposed that BR have higher protein
requirements for maintenance than GR. No statistical treatment of this question is
known to us. However, comparisons of experimentally established protein maintenance
requirements between BR and GR species (collated in Robbins 1993; Clauss et al.
2003b) do not suggest any relevant systematic difference between the feeding types.
B2. A particular adaptation of GR to the digestion of cellulose would be
expected. This is subsumed under the prediction D2, as fibre content cannot be
separated from fermentation characteristics.
B4. In order to counteract plant secondary compounds, it has been suggested
that BR produce salivary proteins that bind to these compounds. See Clauss (2003)
and Shimada (2006) for reviews of species in which such proteins have been dem-
onstrated; however, the number of species investigated thus far makes a statistical
comparison between BR and GR unfeasible. Supposedly larger salivary glands of
BR ruminants (Kay et al. 1980; Kay 1987b; Hofmann 1988) have been considered
to be a morphological correlate of a high production of these salivary proteins
(Robbins et al. 1995). The BR that has been demonstrated to deviate from the
general ruminant pattern, the greater kudu (Tragelaphus strepsiceros; Robbins et
al. 1995), has been noted to suffer from die-offs due to tannin poisoning (Van
Hoven 1991), although even kudus can include plants in their diet that are known
to be poisonous for livestock (Brynard and Pienaar 1960). A larger liver, used for
secondary plant compound detoxification, has been postulated in BR ruminants
(Hofmann 1988; Duncan et al. 1998) and in the black rhinoceros (Diceros bicornis),
a browser (Kock and Garnier 1993), as compared to GR ruminants and GR rhinos.
It has been demonstrated, in pair-wise comparisons in rodents, macropods, and
ruminants, that BR are less affected by dietary secondary compounds than GR
(Iason and Palo 1991; Robbins et al. 1991; Hagerman et al. 1992; McArthur and
Sanson 1993). Statistical treatments of these topics for a large range of species is
still required.
Dental and Buccal Morphology
C1, C2. Silica is harder than enamel, and a grass diet should wear down teeth
faster than does browse. Therefore, significant differences between GR and BR in
enamel microwear or molar wear rates (Solounias et al. 1994; Solounias and
Semprebon 2002) are expected. As molar wear is a function of both attrition

(tooth-to-tooth contact, which maintains sharp edges) and abrasion (tooth to food
contact, which produces blunt edges), differences in the ‘mesowear’ pattern [the
macroscopically evaluated shape and reliefs of the cusps of selected teeth first
introduced by Fortelius and Solounias (2000)] between feeding types are also
expected. Differences in mesowear patterns between upper and lower molars in dif-
ferent feeding types indicate different morphological adaptations (Franz-Oftedaal
and Kaiser 2003; Kaiser and Fortelius 2003). Hypsodonty (as an adaptation to
increased tooth wear due to abrasion and maybe increased attrition; C.3) has been
observed in many GR groups such as marsupials, rodents, lagomorphs, and ungu-
lates (Simpson 1953; Fortelius 1985; Janis and Fortelius 1988). Experimental work
on the influence of crown height on survival is summarised in Williams and Kay
(2001). Hypsodonty will have other consequences for cranial morphology, such as
deeper mandibles to accommodate the hypsodont teeth (Vrba 1978). As a second-
ary change due to hypsodonty and the increased space requirement for the higher
maxillary molars (and the masseteric insertion areas; C.3), it is assumed that in GR
the orbita needs to be positioned more posteriorly than in BR (Solounias et al.
1995). This will also lead to an elongation of the whole skull. In a comparison of
three macropod species, enamel hardness was greater in the GR than in the BR
species (Palamara et al. 1984).
C3. The purportedly tougher consistency of grasses and higher chewing pressure
needed by GR to break down grass would require a more robust mandible structure
(that also accommodates hypsodont teeth; C1, C2). A possible mechanism to with-
stand increasing pressure/torsional forces in GR ruminants is the reduction of the
premolar tooth row length (Greaves 1991), and a reduction of palatal width. In
perissodactyls, premolar tooth reduction does not occur, perhaps due to the require-
ment for increased food intake (Janis and Constable 1993). A greater masseter
muscle mass, and accordingly larger masseter insertion surfaces, are assumed for
GR, which is thought to reflect the higher masticatory forces required to grind grass
material (Turnbull 1970; Stöckmann 1979; Axmacher and Hofmann 1988). A
larger masseter, requiring more nerve tissue for innervation, has been hypothesised
to be the reason for seemingly larger foramina ovale (through which the masseter
nerves pass) in GR ruminants (Solounias and Moelleken 1999).
C4. Different fracture properties of browse and grass should be reflected in
differences in tooth occlusal surface morphology (Fortelius 1985; Sanson
2006); and these differences in tooth morphology should be reflected in differ-
ences in ingesta particle reduction between species on a similar food source
(Lentle et al. 2003c).
General Digestive Physiology
D1. Given the higher potential digestibility of grass, one would either expect BR
to have lower basal metabolic rates at similar intake levels, or similar BMRs at
higher intake levels, or GR to have lower BMRs at relatively low intake levels.
58 M. Clauss et al.
The BMR data available for ruminants (Williams et al. 2001) does not suggest a
systematic difference in BMR between the feeding types, and hence it would be
expected that BR have higher intakes.
D2. Optimal digestion theory (Sibly 1981) predicts that animals adapted to
a forage that yields energy and nutrients quickly would have short retention
times, and those that ingest a forage that yields energy and nutrients more
slowly would have longer retention times. This has been postulated for rumi-
nants (Hanley 1982; Kay 1987a; Hofmann 1989; Clauss and Lechner-Doll
2001; Behrend et al. 2004; Hummel et al. 2005a), however, a comprehensive
dataset based on comparable measurements is still lacking.
Differences in retention time would have far-reaching consequences: on
comparable diets, BR should achieve lower digestion coefficients than GR (for
ruminants: Owen-Smith 1982; Prins et al. 1983; Demment and Longhurst
1987). A higher food intake in BR, due to the supposedly shorter passage times,
has been suggested (Owen-Smith 1982; Baker and Hobbs 1987; Prins and
Kreulen 1991). Forage that ferments faster and is retained shorter should also
be ingested in shorter time intervals, and a higher feeding bout frequency has
been suggested in browsing ruminants (Hofmann 1989; Hummel et al. 2006). A
combination of higher food intake and lower digestibility should theoretically
result in a comparatively higher faecal output in BR, which could be expected
to have further consequences. For example, Robbins (1993) states that sodium
losses are a function of faecal bulk, and higher faecal sodium losses have been
observed in the browsing black rhinoceros (Diceros bicornis) than in the
domestic horse (Clauss et al. 2006a).
Cellulolytic activity in the rumen of BR is expected to be lower than that of
GR (Prins et al. 1984; Deutsch et al. 1998). GR have a more diverse protozoal
fauna, whereas BR protozoa are mostly Entodinium sp.; as these are particu-
larly fast-growing ciliates, it has been suggested that other protozoa cannot
establish viable populations in the reticulorumen (RR) of BR (Prins et al. 1984;
Dehority 1995; Dehority et al. 1999; Clauss and Lechner-Doll 2001; Dehority
and Odenyo 2003; Behrend et al. 2004). Other parameters indicative of shorter
ingesta retention times (Clauss and Lechner-Doll 2001; Behrend et al. 2004)
are a lower degree of unsaturated fatty acid hydrogenation in the RR, a greater
number of glucose transporters, and a higher amylase activity in the small intes-
tine of BR,3 as well as larger faecal particles in BR. In order to maintain intake
levels whilst still having prolonged ingesta retention in the rumen, GR should
have more capacious rumens (Prins and Geelen 1971; Giesecke and Van
3
The hypothesis that BR maintain a functional reticular groove throughout their adult lives
(Hofmann 1989) has not been tested directly (Ditchkoff 2000); however, a comparison of fluid
retention data from two different trials on roe deer (Capreolus capreolus) indicates that bypass of
soluble substances from the rumen via the reticular groove is probably not a quantitative factor, at
least in this species (Behrend et al. 2004).
Gylswyk 1975; Drescher-Kaden 1976; Hoppe 1977; Kay et al. 1980; Owen-
Smith 1982; Bunnell and Gillingham 1985; Van Soest 1994). It has been sug-
gested that BR have more capacious hindguts (Hofmann 1988), but this view
has recently been modified (Clauss et al. 2003a; Clauss et al. 2004).
Macropods
C1C3. Macropods have been classified in feeding types based on their dental mor-
phology (Sanson 1989). Macropod teeth have a crushing action over a relatively
large occlusal contact area in BR, with dentine basins making up a large percentage
of this surface area. As macropods have evolved to feed more on grass, the area of
occlusal contact has been decreased by increasing the complexity of the enamel
ridges and increasing the curvature of the tooth row (Janis and Fortelius 1988;
Sanson 1989; Lentle et al. 2003b; Lentle et al. 2003a). The macropod feeding type
classification of Sanson (1989) was tested for two macropod species by Sprent and
McArthur (2002); the results were in accord with Sanson’s prediction. Differences
in ingesta particle size distribution between four macropod species, tested in sets of
both free-ranging and captive animals, support the notion that teeth of GR are more
suited to the fine-grinding of grass material (Lentle et al. 2003c).
C4-C5. It has been postulated that grazing macropods have a lower proportion
of the sacciform relative to the tubiform forestomach (Freudenberger et al. 1989). This
could be due to the fact that macropods do not ruminate, and as a result, ingesta
stratification in a larger sacciform forestomach would not be beneficial to GR. At
the same time, comparisons indicate that, in GR macropods, the length of the large
intestine is greater than in BR, indicating additional fermentation of the slower-
fermenting grasses in this site (Freudenberger et al. 1989).
Ruminant Forestomach Morphophysiology4
C4-C5. Recently, it has been hypothesised that a key to the understanding of

ruminant forestomach physiology is the presence or absence of a stratification of
RR contents (Clauss and Lechner-Doll 2001; Clauss et al. 2001, 2002, 2003c).
4
We feel that a major basis of the discussion on potential differences in digestive morphophysiology
between GR and BR ruminants has been to confirm or refute Hofmann’s original observations and
hypotheses, rather than necessarily to understand the functional relevance of his findings. Here,
we present a new, complex interpretation of ruminant forestomach physiology; this is not a “refu-
tation” of Hofmann’s hypotheses but a refinement and readjustment, based on his anatomical
observations, the validity of which is not drawn into question (but should be submitted to
statistical evaluation).
60 M. Clauss et al.
Physical and chemical characteristics of grass are thought to enhance the

development of this stratification, with particle separation occurring according to
flotation/sedimentation, and a ‘fibre raft’ or ‘fibre mat’ on top of a liquid layer;
to these, the animal has to adapt, for example with stronger rumen pillars (to
work against the tough consistency of the ‘mat’), with deeper reticular honey-
comb cells (traps for the sedimenting particles in the liquid phase), or larger
omasa (for water re-absorption from the liquid outflow out of the RR). As the
stratification enhances particle retention and, hence, fibre digestion (Beaumont
and Deswysen 1991; Lechner-Doll et al. 1991), adaptations encouraging stratifi-
cation would be expected to have evolved in GR. The most controversial part of
the traditional concept of wild ruminant RR physiology is that the larger salivary
glands of BR (Hofmann 1988) are assumed to translate into a higher saliva pro-
duction rate and, hence, a higher fluid throughput through the RR. If this was the
case, then the comparatively smaller omasa of BR (Hofmann 1988; Clauss et al.
206c) would not make sense; the primary function of the omasum is fluid re-
absorption. If fluid throughput was particularly high in BR, then they, not GR,
should have the larger omasa. The concept of higher salivary flow rate in BR was
challenged by Robbins et al. (1995) on the basis of a three-species comparison.
It has been proposed that the difference between particle and fluid retention in the
RR is a distinguishing characteristic of the different feeding types, with the
difference between the phases being large in GR—indicative of ingesta stratifica-
tion—and small in BR (Clauss and Lechner-Doll 2001; Hummel et al. 2005a).
However, the question whether this difference stemmed from longer particle or
shorter fluid retention times in GR, or both, has not been addressed comprehensively
to date. Comparisons between the few species on which data exist indicate that
GR have both a shorter fluid, and a longer particle retention time in the RR than
do BR (Clauss et al. 2006c). We propose that the production of a large amount of
non-viscous saliva is a particular adaptation of GR and supports the development
of RR contents stratification, whereas the more viscous saliva of BR delays any
separation of ingesta particles by flotation/sedimentation. Thus, the size of the
salivary glands is probably not correlated with saliva production but with saliva
protein content (c.f. B4) and hence viscosity (Robbins et al. 1995). Results of
Jones et al. (2001) who found that several browsers had a higher dry matter con-
tent in rumen fluid as compared to grazers supports the concept of a higher fluid
viscosity in browsers. In a low-viscosity medium, the fermentation gases, carbon
dioxide and methane, can easily rise and gather in the dorsal RR. Rumen papillae
development is stimulated by the presence of volatile fatty acids (in particular, of
butyrate) (Warner et al. 1956). The continuous presence of a gas dome of CO2
and methane in the dorsal rumen will prevent any significant concentration of
volatile fatty acids in this region, leading to unpapillated dorsal rumen surfaces
in GR (Hofmann 1973). In a more viscous medium, fermentation gases cannot
dissociate from food particles to rise and gather in the dorsal rumen as easily,
which results in the typical ‘frothy’ appearance of BR RR contents (Clauss et al.
2001) and an even RR papillation (Hofmann 1973). The escape of a more a viscous
ingesta that traps CO2 into the abomasum would also explain why browsers have
a thicker layer of acid-producing abomasal mucosa (Hofmann 1988), inasmuch

as the presence of CO2 would increase the buffering capacity of the ingesta.
Captive Animals: Readiness to Ingest Forage
Based both on dental and, in the case of ruminants, forestomach characteristics, GR

should have less problems in eating browse than BR in eating grass (Clauss et al.
2003c). This hypothesis finds support in reports for captive wild ruminants (Clauss
et al. 2003b), for macropods (Lentle et al. 2003c), suids (Leus and MacDonald 1997),
and from the well-known reluctance of captive tapirs to ingest grass hay (Foose 1982).
Captive BR ruminants ingest less hay and have a higher incidence of rumen acidosis
(due to a lack of fibre) than GR ruminants (Clauss et al. 2003b). Commercially availa-
ble pelleted feeds for captive wild herbivores in general or GR in particular have a
distinctively lower fibre content than diets designed especially for BR, again indicating
that one cannot rely on a sufficient fibre intake via hay consumption by BR (Clauss
and Dierenfeld 2007). Due to the high abrasiveness that characterizes the diets of both
free-ranging grazers and that fed to captive animals in general, the tooth wear pattern
of captive browsers resembles that of free-ranging grazers (Clauss et al. 2007a).
3.4 Testing the Hypotheses
Testing these hypotheses involves problems that have, historically, been

addressed in different ways in different publications. The following discussion
raises issues about the approaches used to test for morphophysiological
adaptations to diet in herbivores.
Body Mass
Body mass (BM) is the single most influential factor on the absolute size of any
anatomical, and of most (but not necessarily all) physiological parameters (Schmidt-
Nielsen 1984; Peters 1986; Calder 1996). Therefore, the inclusion of BM in statisti-
cal evaluations is self-evident. Ideally, datasets should cover similar BM ranges for
all feeding types investigated. On average, grazing ruminants are larger than brows-
ing ruminants (Bell 1971; Case 1979; Bodmer 1990; Van Wieren 1996b; Pérez-
Barbería and Gordon 2001). However, there is either no correlation (Van Wieren
1996b; Clauss et al. 2003c; Sponheimer et al. 2003), or it is very weak (Gagnon and
Chew 2000), between BM and the proportion of grass in the natural diet of rumi-
nants. This is because browsers are found across the body size range (Sponheimer
et al. 2003). The largest extant ruminant, the giraffe (Giraffa camelopardalis), is a
62 M. Clauss et al.
browser, the largest marsupial herbivores were browsers (Johnson and Prideaux
2004), and the largest known terrestrial mammalian herbivore ever, the Indricotherium
(Fortelius and Kappelman 1993), was also a browser; extensive grasslands did not
exist when this species inhabited the earth (Janis 1993). Large BM, therefore, does
not preclude a browsing lifestyle (Hofmann 1989).
The question of whether there are different upper and lower body size thresholds
for the feasibility of grazing or browsing in ruminants has been addressed by
Demment and Van Soest (1985) and by Clauss et al. (2003a). Assuming feeding type
independent general relationships between body mass and forestomach capacity,
and between body mass and ingesta retention, Demment and Van Soest (1985)
demonstrated that, theoretically, grazing by ruminants is feasible at greater body
masses than browsing; using, in contrast, feeding type specific relationships between
body mass and forestomach capacity and ingesta retention, Clauss et al. (2003a)
demonstrated that browsing is theoretically feasible at larger body masses—a result
seemingly in better accord with the extant and fossil ruminant record.
Two examples illustrate the importance of choosing a specific BM value when
evaluating morphological and physiological data: Hofmann (1973) gives data on
the length of the curvature of the omasum (which he claims is larger in GR than in
BR) for the giraffe (52–71 cm) and the African buffalo (Syncerus caffer) (72 cm).
The BM data given in Hofmann (1973) are derived from the literature (giraffe,
750 kg; buffalo, a range of 447–751 kg). Should one chose to compare both meas-
urements on the basis of the maximal BM (750 vs. 752 kg), hardly any difference
between the species would be evident; should one chose to use the averages/
medians of the given BM data (750 kg vs. 599 kg), then the GR buffalo would be
assumed to display a relatively larger omasum. Another example is the greater
kudu in Hofmann’s (1973) dataset. The BM range, again taken from the literature,
is 170–257 kg. However, the handwritten notes in Hofmann’s archive record
estimated BM of the animals investigated to range from 220 to 350 kg (M. Clauss,
pers. obs.). These actual BM data would link the anatomical measurements to a
higher average BM, thus reinforcing potential differences between the feeding
types. The importance of measurements of morphophysiological traits and BM
from the same individuals, therefore, cannot be overemphasised. In studies that use
measurements on museum skeleton specimens (for which live BM data is usually
missing), the use of a parameter that can be measured on the museum specimen and
is known to correlate closely with BM or BM-independent ratios are alternatives
(for example Janis 1988; Spencer 1995; Archer and Sanson 2002).
Definition of Feeding Type
In order to compare adaptations of different species, the species have to be

classified according to niches that are relevant to the adaptations under investiga-
tion. This has mostly been done by allocating feeding type labels, such as ‘BR’ or
‘GR’, as discrete variables. The available information on diet composition on
which such a classification is based differs between species (Gagnon and Chew
2000). A common practice has been to collect published data on the botanical com-
position of a species’ diet, calculate an average value for the different reports, and
then use pre-defined thresholds to allocate a feeding type. These thresholds have
not been used consistently in the literature; in particular, some publications allocate
species with >75 % of the respective forage to the BR or GR category (Pérez-
Barbería and Gordon 1999; Pérez-Barbería et al. 2001a; Mendoza et al. 2002),
whereas other publications reserve these categories only for species consuming >90 %
of the respective forage (Janis 1990; Pérez-Barbería et al. 2001b). The impact of the
choice of allocation of species to feeding types is demonstrated for example by
Gordon and Illius (1994), who showed that results differed depending on the clas-
sification used. A more consistent approach (Janis 1995; Clauss et al. 2003c;
Sponheimer et al. 2003; Pérez-Barbería et al. 2004) does not use a discrete variable,
but uses the percentage of grass and/or browse in the natural diet as a continuous
variable. But while such an approach overcomes the need to make arbitrary ‘thresh-
old decisions’, it should be remembered that the information contained in such a
continuous variable is not perfect since there can be enormous geographical and
seasonal variation in diet composition in some species (Owen-Smith 1997). An
important limitation of the description of ‘natural’ diets is explained by Sprent and
McArthur (2002): in any natural setting, the ‘typical’ forage preference pattern is
evidently modified by the available forage. Ideally, a selected diet should always be
expressed in terms of the available diet.
It should be borne in mind that allocating feeding types on the basis of actual
observations does not provide full information on the nutritional adaptation of
species. Although it is generally viewed that species diversification followed the
sequence of BR/closed habitat, mixed feeder, GR/open habitat (Pérez-Barbería
et al. 2001b), the reverse has been suggested or noted occasionally for extant and
extinct species (Solounias and Dawson-Saunders 1988; Thenius 1992; Cerling
et al. 1999; MacFadden et al. 1999). The morphology of species that are in a
transition/regression state in this respect may not be completely correlated with
dietary behaviour yet. The different evolutionary directions that led species to their
present state can potentially make convergent evolutionary traits more difficult to
discern (Gould 2002).
Phylogenetic Descendence
If values for individual species are used in statistical tests, these values cannot be
viewed as independent because the species are phylogenetically related (Harvey
and Pagel 1991; Martins and Hansen 1996). In recent years, phylogenetic control
in statistical tests has become standard procedure for evaluating differences
between or correlations with feeding types (c.f. the work of Pérez-Barberìa et al.).
Published results can be classified into those that do not remove phylogenetic
effects in the analysis (generally earlier studies) and those that do. This leads to the
64 M. Clauss et al.
dilemma that results from earlier studies cannot be quoted with confidence, but
direct replication of tests are rarely performed on the same datasets.
The method of phylogenetic control has been criticised (Westoby et al. 1995),
but this discussion shall not be reviewed here. The most informative approach is to
conduct two analyses, without and with phylogenetic control. If, for example, a
certain measure shows a difference between feeding types, after controlling for
body mass alone, this indicates that it represents either (1) a case of convergent
evolution between lineages or (2) evolution within a certain lineage that dominates
the dataset. If, in a second step, no difference between feeding types is found, when
phylogeny is controlled for, then the hypothesis of convergent evolution between
lineages can be rejected, but not necessarily the hypothesis of evolution within a
certain lineage, nor the adaptive value of the trait as such. The rejection of the
hypothesis of convergent evolution should not be confused with a rejection of the
hypothesis of adaptive value, which can only be tested experimentally.
Two important choices have to be made when phylogenetic control is applied.
The phylogenetic tree should, ideally, be based on characters unrelated to the
character that is submitted to the test. In this respect, one should, for example,
note that many of the dental characters understood to be adaptations to feeding
niches (Janis 1990) have also been used to establish phylogenetic relations in
ungulates (Janis and Scott 1987). The case of the more recently discovered mam-
malian taxonomic clades of the Afrotheria, Laurasiatheria, and Euarchontoglires
implies a widespread accumulation of homoplasious morphological features in
various placental clades and thus exemplifies the difficulty of basing phyloge-
netic trees on morphological characters (Robinson and Seiffert 2004). The other
choice refers to the spectrum of species included in the analysis, that is, the level
at which convergent evolution is to be assessed. On the one hand, if there is some
trait showing convergent evolution within the ruminants, then this trait will be
more difficult to detect in a dataset comprising only bovids than in a dataset that
comprises both bovids and cervids. On the other hand, an expansion of the species
dataset beyond certain phylogenetic borders may appear unreasonable. When
rumen capacity or rumination activity is compared between feeding types, only
ruminants (and maybe camelids) will be included but not other ungulates.
Similarly, ingesta retention should be analysed separately for ruminants and hindgut
fermenters, due to the differences in digestive physiology (Illius and Gordon
1992). Other examples include systematic differences in cranial morphology. For
example, in perissodactyls, the lower premolar row increases in length in grazing
species whereas it decreases in grazing ruminants and macropods (Janis 1990;
Mendoza et al. 2002), and grazing horses have relatively smaller muzzles than
corresponding grazing ruminants (Janis and Ehrhardt 1988). Therefore, inclusion
of phylogenetically distant groups in one analysis might yield different results
from an analysis within each of these groups; in this respect, the finding that a
parameter shows no convergent evolution in GR or BR ungulates does not falsify
the finding that such convergent evolution occurs within GR or BR ruminants.
This latter question would have to be addressed by an analysis using only ruminant
data. The power of a variable to predict the feeding type correctly, hence, usually
increases as the taxonomic level is narrowed from ungulates to either artiodactyls

or perissodactyls (Janis 1995).
Statistical Procedure
Generally one can distinguish uni- or bivariate statistics (testing one trait, for
example a ratio, or two if control for BM is included, between feeding types), or
a multivariate approach. The advantage of a multivariate approach is that a
number of characteristics are included that will, each and together, contribute to
the adaptation of a feeding type. A potential disadvantage of a multivariate
approach might be the temptation to include as many data as possible without
giving attention to the functional relevance of the different parameters. In this
respect, multivariate analysis can be considered to be exploratory, unless it is
followed or preceded by further detailed investigations (Spencer 1995; Archer
and Sanson 2002). In the selection of data for a multivariate approach, partly
repetitive information—for example, both volumetric and linear measurements of
the same organ, or both the length of the molar tooth row and the length of the indi-
vidual molars—has either been included (Pérez-Barbería et al. 2001a) or excluded
(Mendoza et al. 2002).
Parameter Choice and Measurement Resolution
All previous considerations notwithstanding, the most important prerequisite for a

meaningful analysis is that the parameters investigated have a functional relevance,
and that they are measured with appropriate resolution so that meaningful
differences can be detected. The necessary degree of resolution is intuitively more
evident in anatomical than in physiological studies. No one would conceive of
comparing incisor breadths measured to the closest 0.1 meter. Yet, for example, the
method employed by Foose (1982) for the estimation of ingesta retention (using
one or two pooled faecal samples per day) can be expected to be of sufficient reso-
lution to differentiate between hindgut and foregut fermenters (the aim of Foose’s
study), but hardly so between browsers and grazers of either category. If data
compilations from different publications are used for an analysis, the consistency
of the methods applied, therefore, is of prime importance; an additional solution is
to include data source as a random variation in the analysis (Pérez-Barbería et al.
2004). In many studies, the functional significance of parameters tested has not
only not been proven by experimentation, but is sometimes not even defined in
logical terms. For example, the concept stated by Hofmann (1988) and repeated
many times by others that ‘openings’ in the ruminant forestomach such as the
ostium intraruminale or the ostium ruminoreticulare, the diameters of which are by
magnitudes greater than the particles that leave the rumen, should have any
66 M. Clauss et al.
influence on ingesta retention is not self-evident and would have to be backed by

engineering models (or rapidly dismissed).
3.5 Results
In this section, publications are summarised that generated or collated comparative

data and submitted these data to statistical tests for differences between the feeding
types. The individual morphological and physiological traits are listed in tables
3.3–5, and comprehensive multivariate analyses are explained in the text. In gen-
eral, the lack of phylogenetically controlled studies on different taxonomic groups
does not allow definite conclusions.
Macropods
Results of morphological craniodental comparisons within the macropods are given

in Table 3.3. No phylogenetically controlled approach has been used thus far.
Some, but not all of the predictions regarding craniodental design seem to be met.
As with ruminants, GR macropods also show a shorter premolar row length, but
have smaller muzzles than do BR macropods.
Table 3.3 Statistical tests for craniodental differences between grazing (GR) and browsing (BR)
macropods. From (Janis 1990); n = 52, original data (not given), discrete feeding type allocation.
FR = functional relevance, FT = feeding types (significance without/with phylogenetic control).
BM = body mass
Parameter FR FT Direction
Basicranial angle A2 */- GR BR
Muzzle width A3 */- GR BR
Distance orbita tooth row C1-C2 */- GR > BR
Lower premolar row length C3 */- GR < BR
Lower molar row length C3 ns/-
Depth of mandibular angle C3 ns/-
Maximum width of mandibular angle C3 ns/-
Length of coronoid process C3 ns/-
Length of masseteric fossa C3 ns/-
Palatal width C3 ns/-
Rodents
The only analysis pertaining to rodents indicates that GR have a higher hypsodonty
index (after phylogenetic control) than do BR (Williams and Kay 2001).
Ungulates
Results of uni- and bivariate analyses are given in Table 3.4. Additionally, a
multivariate analysis (without phylogenetic control) of a set of 22 craniodental
variables in 115 species (Mendoza et al. 2002) indicated that differences exist
between the feeding types. As in macropods and rodents, hypsodonty is identified
as a primary distinction between BR and GR, being greater in GR than in BR. For
several parameters (skull length, muzzle width, occlusal surface, tooth row
lengths), opposing trends between artiodactyls and perissodactyls might have led
to nonsignificant results in the ungulate comparison. Similarly, masseter parame-
ters could be significant simply because equids—all GR—have larger masseter
muscles than ruminants (Turnbull 1970). Thus, phylogenetic control leaves only
very few parameters of convergent evolution within the ungulates, either indicat-
ing that no differences between feeding types exist, or that they should be looked
for at lower taxonomic levels. Recently, in an explorative analysis including
phylogenetic control, without a specific hypothesis, Pérez-Barbería and Gordon
(2005) did not find any relevant correlation between feeding type and brain size
in ungulates.
Perissodactyls
The dental anatomy of browsing rhinoceroses (D. bicornis, Dicerorhinus sumatrensis)

differs distinctively from that of the grazing Ceratotherium simum but not
Rhinoceros unicornis (Palmqvist et al. 2003); similarly, production of salivary
tannin-binding proteins differ between C.simum on one hand, and D. bicornis and
R. unicornis on the other (Clauss et al. 2005b). This could indicate that R. unicornis
ingests more browse or fruit than previously reported, or that this species has
switched to a grass-dominated diet recently. There is a significant difference in
tooth sharpness, with particularly blunt tips in the grazing C. simum as compared
to D. bicornis (Popowics and Fortelius 1997). It has been demonstrated that ingesta
retention in the browsing D. bicornis was shorter than expected for its body mass
when compared to grazing rhinoceroses and equids (Clauss et al. 2005a). Digestion
coefficients achieved by the two browsing rhinoceros species D. bicornis and
D. sumatrensis are lower than those of grazing rhinoceroses at comparable levels
of dietary fibre content (Clauss et al. 2006b).
Table 3.4 Statistical tests for morphological differences between grazing (GR) and browsing (BR) ungulates. FR = functional relevance, n = number of species,
68
dg = data given (yes/no), od = original data (yes/no), FTC = feeding-type classification (d = discrete; c = continuous), FT = difference between grazers/browsers
significant without/with phylogenetic control, BM = correlation with body mass significant without/with phylogenetic control. * = significant, ns = not significant,
ds = difference in slope, - = not done
Parameter FR n dg od FTC FT BM Direction Source
Basicranial angle A2 136 n y d */- GR BR (Janis 1990)
Rel. muzzle width (palate/muzzle) A3 95 y y d */- GR > BR (Janis and
Ehrhardt 1988)
Muzzle width A3 136 n y d nsa/- (Janis 1990)
Muzzle width A3 104 y n d */ns */* (Pérez-Barbería
and Gordon 2001)
Width central incisor A3 136 n y d */- GR > BR (Janis 1990)
Width lateral incisor A3 136 n y d */- GR > BR (Janis 1990)
Rel. incisor width (I1/I3) A3 70 y y d */- GR < BR (Janis and
Ehrhardt 1988)
Rel. incisor width (I1/I3) A3 66 y n d ns/ns */* (Pérez-Barbería
and Gordon 2001)
Incisor protrusion A3 25 y n d */ns */* (Pérez-Barbería
and Gordon 2001)
Hypsodonty index C1, C2 128 y y d */- GR > BR (Janis 1988)
Hypsodonty index C1, C2 136 n y d */- GR > BR (Janis 1990)
Hypsodonty index C1, C2 79 n y c */- GR > BR (Janis 1995)
Hypsodonty index C1, C2 57 n n d -/* GR > BR (Williams and Kay
2001)
Hypsodonty index C1, C2 19 n n c */- GR > BR (Codron et al.
2007b)
M3 height C1, C2 121b y y d */- */- GR > BR (Janis 1988)
M3 height C1, C2 113 y n d */* */* GR > BR (Pérez-Barbería
and Gordon 2001)
M. Clauss et al.
M3 volume C1, C2 121 y y d */- */- GR > BR (Janis 1988)
M3 volume C1, C2 113 y n d (ds) */* GR > BR (Pérez-Barbería
and Gordon 2001)
Distance orbita tooth row C1-C2 136 n y d */- GR > BR (Janis 1990)
Postcanine tooth row volume C1, C3 121 y y d */- */- GR > BR (Janis 1988)
Molar row volume C1, C3 113 y n d */* */* GR > BR (Pérez-Barbería
and Gordon 2001)
Occlusal surface C3 92 y n d nsa/ns */* (Pérez-Barbería
and Gordon 2001)
Lower M2 area C3 136 n y d */- GR BR (Janis 1990)
Max. width mandibular angle C3 136 n y d */- GR > BR (Janis 1990)
Length coronoid process C3 136 n y d */- GR > BR (Janis 1990)
Length masseteric fossa C3 136 n y d */- GR > BR (Janis 1990)
Palatal width C3 136 n y d ns/- (Janis 1990)
L. premol. tooth row length C3 136 n y d nsa/- (Janis 1990)
L. molar tooth row length C3 136 n y d nsa/- (Janis 1990)
17 jaw and 4 skull traits, including 94 n y d */ns (Pérez-Barbería
most of the parameters and Gordon 1999)
listed above [except those
used in Pérez-Barbería
and Gordon (2001)]
3 The Morphophysiological Adaptations of Browsing and Grazing Mammals
a
opposing trends in artiodactyls and equids
b
excluding equids from the dataset
69
70 M. Clauss et al.
While extant equids are uniformly classified as GR, there is a large number of
BR equids in the fossil record (MacFadden 1992). The feeding type of fossil
horses is generally determined by a combination of isotope, hypsodonty, and
microwear data (MacFadden et al. 1999). In theory, it would be feasible to
compare feeding types classified in this manner for other osteological measure-
ments. MacFadden (1992, pp 241–242) gives the example of an equid with a low
hypsodonty index and a pointed muzzle shape as would be expected of a BR, and
another one of an equid with a high hypsodonty index and a broad muzzle as
would be expected of a GR. However, a quantitative approach to such correlations
in the fossil record is lacking.
Proboscids
Extant elephants are intermediate feeders with a preference for browse, but iso-
topic investigations show that both lineages were once grazers and are in a transi-
tion back to browsing (Cerling et al. 1999). Isotopic evidence suggests that the
Asian elephant (Elephas maximus) might ingest a higher proportion of grass than
the African elephant (Loxodonta africana). This difference is confirmed by
microwear results (Solounias and Semprebon 2002). The differences in molar
structure between the two species (with L. africana having less enamel ridges
than E. maximus; Maglio 1973) could be interpreted as a higher degree of
adaptation for grass forage in E. maximus. Elephants differ in their digestive
physiology from other ungulates due to their very short retention times and low
digestion coefficients (Clauss et al. 2003d; Loehlein et al. 2003). Hackenberger
(1987) found significantly longer ingesta retention times in E. maximus compared
to L.africana and correspondingly higher digestion coefficients for E. maximus
than L.africana when both species were fed hay diets. Data from Foose (1982)
confirms this pattern. Anatomical data compilations suggest that E. maximus has
a longer gastrointestinal tract, a larger masseteric insertion area, and smaller
parotid glands than L.africana (Clauss et al. 2007b) —seemingly in parallel to
similar differences between GR and BR ruminants.
Hyraxes
The GR Procavia capensis and the more browsing Heterohyrax brucei and
Dendrohyrax dorsalis have similar differences in microwear pattern as found in
other herbivore taxons (Walker et al. 1978), but they have the same hypsodonty
index (Janis 1990). No differences in tooth sharpness were observed between GR
and BR hyraxes (Popowics and Fortelius 1997).
Suids
The suids comprise the GR warthog (Phacochoerus aethipicus) or the browsing/

omnivorous babyrousa (Babyrousa babyrussa) and bushpig (Potamochoerus por-
cus), and the BR forest hog (Hylochoerus meinertzhageni; Harris and Cerling 2002;
Mendoza et al. 2002; Cerling and Viehl 2004). Differences in hypsodonty (Harris
and Cerling 2002; Mendoza et al. 2002) accord with observations in other taxo-
nomic groups. Preliminary results indicate that B. babyrussa digests grass fibre less
efficiently than domestic pigs (Leus and MacDonald 1997).
Ruminants
Results of uni- and bivariate analyses are given in Table 3.5. A multivariate analysis
by Solounias and Dawson-Saunders (1988) using 13 traits mainly related to masseter
muscle insertion in 27 species, and another multivariate analysis by Sponheimer et al.
(1999) using four craniodental traits in 23 species, both found significant differences
between the feeding types. In a multivariate, stepwise discriminant analysis of data
from 72 bovid species, Mendoza and Palmqvist (2006a) demonstrated systematic
differences in craniodental morphology between feeding types. With four exceptions,
none of the studies listed in Table 3.5 were performed with phylogenetic control;
therefore, although many characters do differ according to the predictions, it cannot be
determined whether this represents a case of true convergent evolution. For
physiological measurements such as digestibility (Robbins et al. 1995; Iason and Van
Wieren 1999) or particle retention time (Gordon and Illius 1994; Hummel et al. 2006),
larger datasets yielded different results than previous studies on more limited datasets.
The fact that GR digest fibre more efficiently than BR (Pérez-Barbería et al. 2004)
supports the general concept of feeding type differentiation. With respect to anatomi-
cal measurements of the forestomach, very few have been submitted to tests, and the
basic dataset (Hofmann 1973) has hardly been expanded. In the multivariate analysis
of Pérez-Barbería et al. (2001a), no specific functionality of the traits analysed was
addressed; instead, the study tested whether Hofmann’s conclusions could be sup-
ported or derived from the majority of the data given in Hofmann (1973). It was found
that, after controlling for both body mass and phylogeny, the forestomach structures
of BR and GR are similar, whereas those of mixed feeders differ; this finding is diffi-
cult to reconcile with the result that mixed feeders represent an intermediate
evolutionary state between BR and GR (Pérez-Barbería et al. 2001b). Recently, it has
been shown that GR have larger omasal laminal surface areas than BR (Clauss et al.
2006c), supporting Hofmann’s (1968) observation that GR have larger omasa, and
that BR have larger salivary glands (Hofmann et al., in press), confirming the indica-
tions of earlier studies. Even though the respective forages differ in fermentation rate
(D2), it is difficult to predict in what way RR ingesta samples, which do not represent
fresh forage but forage in varying states of fermentation (that is, proportions of which
Table 3.5 Statistical tests for morphological differences between grazing (GR) and browsing (BR) ruminants. FR = functional relevance, n = number of
72
species, dg = data given (yes/no), od = original data (yes/no), FTC = feeding-type classification (d = discrete; c = continuous), FT = difference between
grazers/browsers significant without/with phylogenetic control, BM = correlation with body mass significant without/with phylogenetic control. * = significant,
ns = not significant, ds = difference in slope, - = not done
Braincase angle A2 33 y y d */- GR BR (Spencer 1995)
Paracondylar process A2 33 y y d */- GR > BR (Spencer 1995)
Diastema length A2, C3 33 y y d */- GR > BR (Spencer 1995)
Length of skull A2, C3 33 y y d */- GR > BR (Spencer 1995)
Predental length A2, C3 33 y y d */- GR > BR (Spencer 1995)
Incisor arcade breadth A3 88 y y d */- */- GR > BR (Gordon and Illius 1988)
Incisor arcade breadth A3 33 y y d */- GR > BR (Spencer 1995)
Incisor arcade breadth A3 79 y n d */ns */* (Pérez-Barbería and Gordon 2001)
Incisor arcade shape A3 72 y n d (ds) ns/ns (Pérez-Barbería and Gordon 2001)
Muzzle height A3 27 ya y d */- GR < BR (Solounias and Dawson-
(lip muscle attachment) Saunders 1988)
Parotid gland size B4 22b n yn d */- */- GR < BRc (Robbins et al. 1995)
Parotid gland size B4 20 y n d */(-) GR < BR (Jiang and Takatsuki 1999)
Parotid gland size B4 62 y y c */* */- GR < BR (Hofmann et al., in press)
Mandibular gland size B4 61 y y c */* */- GR < BR (Hofmann et al., in press)
Size of the ventral buccal glands B4 44 y y c */* */- GR BR (Sponheimer et al. 2003)
Hypsodonty index C1-C2 37 y n c */- GR > BR (Cerling et al. 2003)
Hypsodonty index C1-C2 13 n n c */- GR > BR (Codron et al. 2007b)
Molar wear rates C1-C2 9 y n d */- */- GR > BR (Solounias et al. 1994)
Distance orbita tooth row C1-C2 22 y y d */- GR > BR (Solounias et al. 1995)
Mandible depth C1-C3 33 y y d */- GR > BR (Spencer 1995)
Mandible depth C1-C3 27 n n c */- GR > BR (Sponheimer et al. 2003)
M. Clauss et al.
Mandible width C1-C3 33 y y d */- GR > BR (Spencer 1995)

Tooth blade sharpness C1-C2, C4 14 n y d ns/- (Popowics and Fortelius 1997)
Length of premolar tooth row C3 33 y y d */- GR BR (Solounias et al. 1995)
Palatal width C3 33 y y d */- GR BR (Solounias and Dawson-
Saunders 1988)
Molar enamel ridge C4 26 y y d */- GR > BR (Archer and Sanson 2002)
pattern/occlusal
surface complexity
Faecal particle size C4, D2 81 y y d *d/- */- GR BR (Demment and Longhurst 1987)
RR capacity (water fill volume) C4, C5, D2 25 n n d ns/- */- (Van Wieren 1996b)
RR capacity (water fill volume) C4, C5, D2 36 y n d */(-) GR > BR (Jiang and Takatsuki 1999)
RR contents (wet weight) C4, C5, D2 21 y n d nse/- */- (Gordon and Illius 1994)
RR contents (wet weight) C4, C5, D2 21 y n d *f/- */- GR > BR (Gordon and Illius 1994)
RR contents (wet weight) C4, C5, D2 47 y n d */(-) GR > BR (Jiang and Takatsuki 1999)
RR contents (wet weight) C4, C5, D2 29 y n d/c */- */- GR > BR (Clauss et al. 2003c)
RR contents (dry) ? 21 y n d nse/- */- (Gordon and Illius 1994)
Rumen pillar thickness C4, C5, D2 27 y n d/c */- */- GR > BR (Clauss et al. 2003c)
Rumen total surface area C4, C5 45 y y d *(ds)g/- GR < BR (Demment and Longhurst 1987)
Density papillation ventr. rumen ? 25 n n d *h/- */- GR < BR (Van Wieren 1996b)
Maximum SEF (of any RR loca- ? 25 n n d ns/- */- (Van Wieren 1996b)
tion)
Ostium rumino-reticularei ? 25 n n d -/- */- (Demment and Longhurst 1987)
Omasal laminar surface area C4, C5 34 y y c */* */* GR > BR (Clauss et al. 2006c)
RR liquid flow rate D2 8 nj yn d ns/- */- (Robbins et al. 1995)
RR liquid retention time D2 8 n yn d ?k (Robbins et al. 1995)
RR liquid retention time D2, C5 14 n n d nsl/- nsl/- (Clauss et al. 2006c)
73
RR particle retention time D2, C5 10 n n d nsm/- nsm/- (Clauss et al. 2006c)

(continued)
n
74
RR VFA concentrations B2, B3 16 y y d * /- */- GR < BR (Clemens and Maloiy 1983)
RR contents crude fibre content B2 16 y y d ns/- (Woodall 1992)
RR fermentation rates (mols per B2, B3, D2 21 y n d ns/- */- (Gordon and Illius 1994)
DM and d)
Energy by VFA in RR per day D2 21 y n d *o/- */- GR > BR (Gordon and Illius 1994)
Energy by VFA in RR per day D2 21 y n d nsp/- */- (Gordon and Illius 1994)
RR acetate:propionate ratio B2-B3 16 y y d ns/- */- (Clemens et al. 1983)
Energy content of RR VFA B2, B3q 21 y n d ns/- ns/- (Gordon and Illius 1994)
(J per mol)
RR acetate proportions B2-B3 16 y y d ns/- ns/- (Clemens et al. 1983)
RR propionate propotion B2-B3 16 y y d */- */- GR < BR (Clemens et al. 1983)
RR butyrate proportion B2-B3 16 n y d ns/- ns/- (Clemens et al. 1983)
Dry matter RR (% of wet weight) ? 16 y y d ns/- ns/- (Clemens and Maloiy 1983)
Volume omasum-abomasum C4, C5 25 n n d ns/- */- (Van Wieren 1996b)
Ratio Vol. RR:Omas-Abomas ? 25 n n d ns/- ns/- (Van Wieren 1996b)
DM abomasum (%) C4, C5, D2 16 y y d */- ns/- GR > BR (Clemens and Maloiy 1983)
DM small intestine (%) C4, C5, D2 16 y y d */- */- GR > BR (Clemens and Maloiy 1983)
Hindgut contents (wet weight) D2 15 y n d *r/- */- GR < BR (Gordon and Illius 1994)
Caecum VFA concentration B2, B3, D2 18 n y d ns/- -/- (Maloiy and Clemens 1991)
Colonic VFA concentration B2, B3, D2 16 y y d */- ns/- GR < BR (Clemens and Maloiy 1983)
Hindgut VFA reabsorption B2, B3, D2 16 y y d */- ns/- GR < BR (Clemens and Maloiy 1984)
Hindgut fermentation rates B2, B3, D2 15 y n d ns/- */- (Gordon and Illius 1994)
Energy by VFA in hindgut D2 15 y n d ns/- */- (Gordon and Illius 1994)
per day
Hindgut acetate:propionate ratio B2, B3, D2 16 y y d ns (*)s/- ns/- GR > BR (Clemens et al. 1983)
Hindgut acetate proportions B2, B3, D2 16 y y d */- */- GR > BR (Clemens et al. 1983)
Hindgut propionate propotion B2, B3, D2 16 y y d ns/- ns/- (Clemens et al. 1983)
Hindgut butyrate proportion B2, B3, D2 16 n y d */- GR < BR (Clemens et al. 1983)
Lactic acid in caecum and colon B2, B3, D2 16 y y d */- */- GR BR (Van Wieren 1996b)
Fiber digestion D2 20 n n d */- *t/- GR > BR (Iason and Van Wieren 1999)
Fibre digestion D2 24 n n c -/* -/ns GR > BR (Pérez-Barbería et al. 2004)
Total GIT particle mean D2 26 y n d nsu/- */- (Gordon and Illius 1994)
retention time
Total GIT particle mean D2 33 n n d *u/- */- GR > BR (Hummel et al. 2006)
retention time
Colonic fluid absorption ? 16 y y d ns/- ns/- (Clemens and Maloiy 1984)
Dry matter distal colon (% of wet ? 16 y y d */- GR < BR (Clemens and Maloiy 1983)
weight)
Dry matter faeces (% of wet ? 81 y y d nsv/- *w/- (Clauss et al. 2004)
weight)
a
data not measured but given as discrete categorical variables
b
not all species stated in the methods can be found on the graph in the results
c
the tragelaphinae, especially the greater kudu, were outliers to this pattern
d
difference in slopes, GR 80 kg; only captive animals on zoo winter diets used
e
see Clauss et al. (2003c); data compared to gut-content-free BM, using the authors’ own feeding type classification
f
using Hofmann’s classification; not mentioned in results, only in discussion
g
as observed (Pérez-Barbería et al. 2001a), the claim of statistical significance is not justified due to the difference in slopes
h
significance disappeared when three small species were excluded; note that according tests (exclusion of certain species) were not performed in other analyses
i
misquoted as the ostium reticulo-omasale (Illius and Gordon 1999; Gordon 2003); in contrast to the claim by these authors, no test for difference between
feeding types is found in the original work
j
not all data from the publications cited in methods section visible in the graph
k
no evident pattern, unclear whether statistical test was performed, not all data from the publications cited in methods section visible in the graph
l
number of species too small to allow confirmation or refutation of hypotheses; note that in pair-wise comparison, cattle had shorter RR fluid retention than
large browsers
m
number of species too small to allow confirmation or refutation of hypotheses
75
76

n
statistical significance could not be confirmed (Streich, pers. comm.)
o
using their own feeding type classification; note that the individual factors used to calculate this factor (RR DM load, VFA production rate, VFA energy
content) did not differ significantly
p
using Hofmann’s classification
q
due to chemical composition of forages, no difference is to be expected
r
using Hofmann’s classification
s
p < 0.045 is stated in the paper
t
correlation with BM weak but significant
u
see comments in text and in Clauss and Lechner-Doll (2001)
v
wider ranges in faecal DM observed in GR than in BR
w
decreasing faecal DM with BM
M. Clauss et al.
do not yield further volatile fatty acids), will reflect this. In particular, a large set of
truly comparable ingesta passage measurements is missing.
3.6 Conclusion and Outlook
Although many of the morphophysiological adaptations expected in BR and GR

(Table 3.2) seem to be met in tests using conventional statistical methods, the number
of such characteristics tested with a phylogenetic control remains low to date (Tables
3.3-3.5). Most of the studies including phylogenetic controls have addressed ungulates
as a phylogenetic group, whereas the according parameters have hardly been investi-
gated within lower taxonomic units, such as the ruminants, or, also for lack of extant
species, the hindgut-fermenting ungulates. In this respect, the findings of Pérez-
Barberia et al. (2004), Clauss et al. (2006b), and Hackenberger (1987) on differences
in digestive efficiency between feeding types in ruminants, rhinoceroses, and elephants
should incite more comparative investigations in mechanisms involved in digestive
physiology. Differences in the correlation between food intake and ingesta retention
(Clauss et al. 2007c) could be particularly revealing for the differentiation of BR and GR.
Controlled studies that measure retention time and digestibility, on both ad libitum
intake of natural forages (which is hypothesised to result in different intake levels) and
on similar intake levels of the same forages, could be performed in a multitude of
species. Effects of forages should be tested by feeding the same amounts of different
forages to a species. The variety of macropod, rodent (Kay and Madden 1997), suid
and ruminant species offers an ideal research area for such future studies.
The craniodental patterns found in ruminants with conventional statistical
methods should be evaluated within the ruminant guild by phylogenetically
controlled studies. More experimental (engineering or biomechanical) approaches
are warranted to corroborate the proposed adaptive relevance of craniodental
parameters. Amongst others, the pioneering studies of Witzel and Preuschoft (1999,
2002, 2005) could serve as an example here. In order to test for potential differences
in soft tissue morphology, new data on both macropods and ruminants is desper-
ately needed, and until more cervid species are included in a phylogenetically con-
trolled study, tests for convergent evolution within the ruminants must remain
tentative. In terms of functional relevance, we think that soft tissue anatomy poses
the most challenging questions that have to be addressed by engineering models,
such as the one created by Langer and Takács (2004) addressing the functional
relevance of taeniae and haustra in mammalian intestines, or by invasive experi-
ments such as the one fashioned by Kaske and Midasch (1997) who impeded the
motility of the reticulum in sheep in order to elucidate the role of this forestomach
compartment for the digestive process in ruminants.
While a comparison of measurements of physiological processes, such as
digestive efficiency, ingesta retention time, particle size reduction, amount of food
ingested, etc., represent valuable bases for the discussion of different feeding types
and their ecological impact on their feeding niches, we believe that only progress
78 M. Clauss et al.
in the quantitative understanding of both the function of hard and soft tissue
anatomical features and functionally relevant forage properties will allow meaning-
ful interpretations of potential morphological adaptations. It is only by such knowl-
edge that questions like whether different sets of adaptations, with an overexpression
of one morphological feature compensating for the underexpression of another fea-
ture or vice versa, will facilitate the exploitation of the same niche; or whether the
evolution of a particular anatomical feature exclusively allows the use of a new
niche or broadens the range of niches available to the species, leaving open the path
to a back-switching to formerly used niches. We want to conclude this chapter with
the puzzling example of the ruminant reticulum, well aware that differences in this
organ have not been statistically demonstrated between the feeding types. The
reticular honeycomb cells of many grazing ruminants are particularly pronounced
and deep (Hofmann 1988). In domestic ruminants, their function as sedimentation
traps for small particles which are subsequently transported into the next forestom-
ach compartment has been determined experimentally (Kaske and Midasch 1997).
In contrast, many browsing ruminants, deemed representatives of evolutionary
older ruminants, have extremely shallow reticular crests (Neuville and Derscheid
1929), the mechanical function of which is beyond imagination so far. Their shape
could be explained if they could be considered ‘atavisms’, an interpretation ruled
out by the common understanding of the evolutionary sequence of feeding types to
date. Therefore, the shallow crests of roe deer, moose, and giraffe remain a chal-
lenging example for the fact that the evolution of morphological characters can
only be understood by their functional relevance.
Acknowledgements MC thanks R.R. Hofmann for years of support and hospitality. We thank
B. Schneider for tireless help with literature acquisition. This contribution is dedicated to all those
who relish the beauty of molar enamel ridges and reticular honeycomb cells, and whose hearts
beat faster at the smell of acetate.
References
Archer D, Sanson G (2002) Form and function of the selenodont molar in southern African rumi-
nants in relation to their feeding habits. J Zool 257:13–26
Axmacher H, Hofmann RR (1988) Morphological characteristics of the masseter muscle of 22
ruminant species. J Zool 215:463–473
Bailoni L, Ramanzin M, Simonetto A, Obalakov N, Schiavon S, Bittan G (1998) The effect of in
vitro fermentation on specific gravity and sedimentation measurements of forage particles.
J Anim Sci 76:3095–3103
Baker DL, Hobbs NT (1987) Strategies of digestion: digestive efficiency and retention times of
forage diets in montane ungulates. Can J Zool 65:1978–1984
Baker G, Jones LHP, Wardrop ID (1959) Cause of wear in sheep’s teeth. Nature 184:1583–1584
Beaumont R, Deswysen AG (1991) Mélange et propulsion du contenu du réticulo-rumen. Reprod
Nutr Dev 31:335–359
Behrend A, Lechner-Doll M, Streich WJ, Clauss M (2004) Seasonal faecal excretion, gut fill,
liquid and particle marker retention in mouflon (Ovis ammon musimon), and a comparison
with roe deer (Capreolus capreolus). Acta Theriol 49:503–515
Bell RHV (1971) A grazing ecosystem in the Serengeti. Sci Am 225:86–93

Blair RM, Short HL, Epps EA (1977) Seasonal nutrient yield and digestibility of deer forage from
a young pine plantation. J Wildl Manage 41:667–676
Bodmer RE (1990) Ungulate frugivores and the browser–grazer continuum. Oikos 57:319–325
Boué C (1970) Morphologie fonctionelle des dents labiales chez les ruminants. Mammalia
34:696–711
Bryant JP, Reichardt PB, Clausen TP, Provenza FD, Kuropat PJ (1992) Woody plant – mammal
interactions. In: Rosenthal GA, Berenbaum MR (eds) Herbivores: their interaction with sec-
ondary plant compounds, vol 2. Academic Press, San Diego, pp 343–370
Brynard A, Pienaar VdV (1960) Annual report of the biologist, 1958/1959, Koedoe. 3:1–205
Bunnell FL, Gillingham MP (1985) Foraging behavior: the dynamics of dining out. In: Hudson
RJ, White RG (eds) Bioenergetics of wild herbivores. CRC Press, Boca Raton, FL, pp 53–59
Calder WA (1996) Size, function and life history. Harvard University Press, Cambridge, MA
Case TJ (1979) Optimal body size and an animal’s diet. Acta Biotheor 28:54–69
Caswell H, Reed F, Stephenson SN, Werner PA (1973) Photosynthetic pathways and selective
herbivory: a hypothesis. Am Nat 107:465–480
Cerling TE, Viehl K (2004) Seasonal diet changes of the forest hog (Hylochoerus meinertzhageni)
based on the carbon isotopic composition of hair. Afr J Ecol 42:88–92
Cerling TE, Harris JM, Leakey MG (1999) Browsing and grazing in elephants: the isotope record
of modern and fossil proboscideans. Oecologia 120:364–374
Cerling TE, Harris JM, Passey BH (2003) Diets of East African bovidae based on stable isotope
analysis. J Mammal 84:456–470
Cerling TE, Harris JM, Leakey MG (2005) Environmentally driven dietary adaptations in African
mammals. In: Ehleringer JR, Cerling TE, Dearing MD (eds) A history of atmospheric CO2 and
its effects on plants, animals and ecosystems. Springer, New York, pp 258–272
Clauss M (2003) Tannins in the nutrition of captive wild animals. In: Fidgett A, Clauss M, Ganslosser
U, Hatt JM, Nijboer J (eds) Zoo animal nutrition, vol 2. Filander, Fuerth, Germany, pp 53–89
Clauss M, Lechner-Doll M (2001) Differences in selective reticulo-ruminal particle retention as a
key factor in ruminant diversification. Oecologia 129:321–327
Clauss M, Dierenfeld ES (2007) The nutrition of browsers. In: Fowler ME, Miller RE (eds) Zoo
and wild animal medicine, vol VI. Saunders, Philadelphia (in press)
Clauss M, Lechner-Doll M, Behrend A, Lason K, Lang D, Streich WJ (2001) Particle retention in
the forestomach of a browsing ruminant, the roe deer (Capreolus capreolus). Acta Theriol
46:103–107
Clauss M, Lechner-Doll M, Streich WJ (2002) Faecal particle size distribution in captive wild
ruminants: an approach to the browser/grazer-dichotomy from the other end. Oecologia
131:343–349
Clauss M, Frey R, Kiefer B, Lechner-Doll M, Loehlein W, Polster C, Rössner GE, Streich WJ
(2003a) The maximum attainable body size of herbivorous mammals: morphophysiological
constraints on foregut, and adaptations of hindgut fermenters. Oecologia 136:14–27
Clauss M, Kienzle E, Hatt JM (2003b) Feeding practice in captive wild ruminants: peculiarities in
the nutrition of browsers/concentrate selectors and intermediate feeders. A review. In: Fidgett
A, Clauss M, Ganslosser U, Hatt JM, Nijboer J (eds) Zoo animal nutrition, vol 2. Filander,
Fuerth, Germany, pp 27–52
Clauss M, Lechner-Doll M, Streich WJ (2003c) Ruminant diversification as an adaptation to the
physicomechanical characteristics of forage. A reevaluation of an old debate and a new
hypothesis. Oikos 102:253–262
Clauss M, Löhlein W, Kienzle E, Wiesner H (2003d) Studies on feed digestibilities in captive
Asian elephants (Elephas maximus). J Anim Physiol Anim Nutr 87:160–173
Clauss M, Lechner-Doll M, Streich WJ (2004) Differences in the range of faecal dry matter con-
tent between feeding types of captive wild ruminants. Acta Theriol 49:259–267
Clauss M, Froeschle T, Castell J, Hummel J, Hatt JM, Ortmann S, Streich WJ (2005a) Fluid and
particle retention times in the black rhinoceros (Diceros bicornis), a large hindgut-fermenting
browser. Acta Theriol 50:367–376
80 M. Clauss et al.
Clauss M, Gehrke J, Hatt JM, Dierenfeld ES, Flach EJ, Hermes R, Castell J, Streich WJ, Fickel J
(2005b) Tannin-binding salivary proteins in three captive rhinoceros species. Comp Biochem
Physiol A 140:67–72
Clauss M, Castell J, Kienzle E, Dierenfeld ES, Flach EJ, Behlert O, Ortmann S, Hatt JM, Streich
WJ, Hummel J (2006a) Macromineral absorption in the black rhinoceros (Diceros bicornis) as
compared to the domestic horse. J Nutr 136:2017S–2020S
Clauss M, Castell JC, Kienzle E, Dierenfeld ES, Flach EJ, Behlert O, Ortmann S, Streich WJ,
Hummel J, Hatt JM (2006b) Digestion coefficients achieved by the black rhinoceros (Diceros
bicornis), a large browsing hindgut fermenter. J Anim Physiol Anim Nutr 90:325–334
Clauss M, Hofmann RR, Hummel J, Adamczewski J, Nygren K, Pitra C, Reese S (2006c) The
macroscopic anatomy of the omasum of free-ranging moose (Alces alces) and muskoxen
(Ovibos moschatus) and a comparison of the omasal laminal surface area in 34 ruminant
species. J Zool 270:346–358
Clauss M, Hummel J, Streich WJ (2006d) The dissociation of the fluid and particle phase in the
forestomach as a physiological characteristic of large grazing ruminants: an evaluation of
available, comparable ruminant passage data. Eur J Wildl Res 52:88–98
Clauss M, Franz-Odendaal TA, Brasch J, Castell JC, Kaiser TM (2007a) Tooth wear in captive
giraffes (Giraffa camelopardalis): mesowear analysis classifies free-ranging specimens as
browsers but captive ones as grazers. J Zoo Wildl Med (in press)
Clauss M, Steinmetz H, Eulenberger U, Ossent P, Zingg R, Hummel J, Hatt JM (2007b)
Observations on the length of the intestinal tract of African (Loxodonta africana) and Asian
elephants (Elephas maximus). Eur J Wildl Res 53:68–72
Clauss M, Streich WJ, Schwarm A, Ortmann S, Hummel J (2007c) The relationship of food intake
and ingesta passage predicts feeding ecology in two different megaherbivore groups. Oikos
116:209–216
Clemens ET, Maloiy GMO (1983) Digestive physiology of East African wild ruminants. Comp
Biochem Physiol A 76:319–333
Clemens ET, Maloiy GMO (1984) Colonic absorption and secretion of fluids, electrolytes and
organic acids in East African wild ruminants. Comp Biochem Physiol A 77:51–56
Clemens ET, Maloiy GMO, Sutton JD (1983) Molar proportions of volatile fatty acids in the
gastrointestinal tract of East African wild ruminants. Comp Biochem Physiol A 76:217–224
Codron J, Lee-Thorp JA, Sponheimer M, Codron D, Grant RC, De Ruiter DJ (2006) Elephant
(Loxodonta africana) diets in Kruger National Park, South Africa: spatial and landscape dif-
ferences. J Mammal 87:27–34
Codron D, Lee-Thorp JA, Sponheimer M, Codron J (2007a) Nutritional content of savanna plant
foods: implications for browse/grazer models of ungulate diversification. Eur J Wildl Res
53:100–111
Codron D, Lee-Thorp JA, Sponheimer M, Codron J, de Ruiter D, Brink JS (2007b) Significance
of diet type and diet quality for ecological diversity of African ungulates. J Anim Ecol
76:526–537
Cork SJ, Foley WJ (1991) Digestive and metabolic strategies of arboreal mammalian folivores in
relation to chemical defenses in temperate and tropical forests. In: Palo RT, Robbins CT (eds)
Plant defenses against mammalian herbivores. CRC Press, Boca Raton, pp 133–166
DeGusta D, Vrba E (2003) A method for inferring paleohabitats from the functional morphology
of bovid astagali. J Archaeol Sci 30:1009–1022
DeGusta D, Vrba E (2005a) Methods for inferring paleohabitats from discrete traits of the bovid
postcranial skeleton. J Archaeol Sci 32:1115–1123
DeGusta D, Vrba E (2005b) Methods for inferring paleohabitats from the functional morphology
of bovid phalanges. J Archaeol Sci 32:1099–1113
Dehority BA (1995) Rumen ciliates of the pronghorn antelope (Antilocapra americana), mule
deer (Odocoileus hemionus), white-tailed deer (Odocoileus virginianus) and elk (Cervus
canadensis) in the Northwestern United States. Arch Protistenkd 146:29–36
Dehority BA, Odenyo AA (2003) Influence of diet on the rumen protozoal fauna of indigenous
African wild ruminants. J Eukaryot Microbiol 50:220–223
Dehority BA, Demarais S, Osborn DA (1999) Rumen ciliates of white-tailed deer (Odocoileus
virginianus), axis deer (Axis axis), sika deer (Cervus nippon) and fallow deer (Dama dama)
from Texas. J Eukaryot Microbiol 46:125–131
Demment MW, Van Soest PJ (1985) A nutritional explanation for body-size patterns of ruminant
and nonruminant herbivores. Am Nat 125:641–672
Demment MW, Longhurst WH (1987) Browsers and grazers: constraints on feeding ecology
imposed by gut morphology and body size. Proceedings of the IVth International Conference
on Goats, Brazilia, Brazil, 989–1004
Deutsch A, Lechner-Doll M, Wolf AG (1998) Activity of cellulolytic enzymes in the contents of
reticulorumen and caecocolon of roe deer (Capreolus capreolus). Comp Biochem Physiol A
119:925–930
Ditchkoff SS (2000) A decade since “diversification of ruminants”: has our knowledge improved?
Oecologia 125:82–84
Dougall HW, Drysdale VM, Glover PE (1964) The chemical composition of Kenya browse and
pasture herbage. E Afr Wildl J 2:86–121
Drescher-Kaden U (1976) Tests on the digestive system of roe deer, fallow deer and mouflon.
Report 1: Weight statistics and capacity measurements on the digestive system, particularly of
the rumenreticulum. Z Jagdwiss 22:184–190
du Plessis I, van der Waal C, Webb EC (2004) A comparison of plant form and browsing
height selection of four small stock breeds - preliminary results. S Afr J Anim Sci
34(Suppl. 1):31–34
Duncan P, Tixier H, Hofmann RR, Lechner-Doll M (1998) Feeding strategies and the physiology
of digestion in roe deer. In: Andersen R, Duncan P, Linell JDC (eds) The European roe deer:
the biology of success. Scandinavian University Press, Oslo, pp 91–116
Foose TJ (1982) Trophic strategies of ruminant versus nonruminant ungulates. Dissertation,
University of Chicago, Chicago
Fortelius M (1985) Ungulate cheek teeth: developmental, functional, and evolutionary interrelations.
Acta Zool Fenn 180:1–76
Fortelius M, Kappelman J (1993) The largest land mammal ever imagined. Zool J Linn Soc–Lond
107:85–101
Fortelius M, Solounias N (2000) Functional characterization of ungulate molars using the abra-
sion–attrition wear gradient: a new method for reconstructing paleodiets. Am Mus Novit
3301:1–36
Franz-Oftedaal TA, Kaiser TM (2003) Differential mesowear in the maxillary and mandibular
cheek dentition of some ruminants (Artiodactyla). Ann Zool Fenn 40:395–410
Freeland WJ, Janzen DH (1974) Strategies in herbivory by mammals: The role of secondary
compounds. Am Nat 108:269–289
Freudenberger DO, Wallis IR, Hume ID (1989) Digestive adaptations of kangaroos, wallabies and
rat-kangaroos. In: Grigg G, Jarman P, Hume I (eds) Kangaroos, wallabies and rat-kangaroos.
Surrey-Beatty, Sydney, pp 151–168
Gagnon M, Chew AE (2000) Dietary preferences in extant African bovidae. J Mammal
81:490–511
Gentry AW (1980) Fossil bovidae (Mammalia) from Langebaanweg. Ann S Afr Mus
79:213–337
Giesecke D, Van Gylswyk NO (1975) A study of feeding types and certain rumen functions in six
species of South African wild ruminants. J Agr Sci 85:75–83
Gordon IJ (2003) Browsing and grazing ruminants: Are they different beasts? Forest Ecol Manag
181:13–21
2:15–22
Gordon IJ, Illius AW (1994) The functional significance of the browser–grazer dichotomy in
African ruminants. Oecologia 98:167–175
Gordon IJ, Illius AW (1996) The nutritional ecology of African ruminants: a reinterpretation. J Anim
Ecol 65:18–28
82 M. Clauss et al.
Gould SJ (2002) The structure of evolutionary theory. Harvard University Press, Cambridge, MA
Greaves W (1991) A relationship between premolar loss and jaw elongation in selenodont artio-
dactyls. Zool J Linn Soc–Lond 101:121–129
Guthrie RD (1990) Frozen fauna of mammoth steppe: the story of blue babe. University of
Chicago Press, Chicago
Hackenberger MK (1987) Diet digestibilities and ingesta transit times of captive Asian and
African elephants. University of Guelph, Guelph
Hagen J (2003) The statistical frame of mind in systematic biology from Quantitative zoology to
Biometry. J Hist Biol 36:353–384
Hagerman AE, Robbins CT, Weerasuriya Y, Wilson TC, McArthur C (1992) Tannin chemistry in
relation to digestion. J Range Manage 45:57–62
Hanley TA (1982) The nutritional basis for food selection by ungulates. J Range Manage
35:146–151
Harris JM, Cerling TE (2002) Dietary adaptations of extant and Neogene African suids. J Zool
256:45–54
Harvey PH, Pagel MD (1991) The comparative method in evolutionary biology. Oxford University
Press, Oxford
Haschick SL, Kerley GIH (1996) Experimentally determined foraging heights of buchbuck
(Tragelaphus scriptus) and Boer goats (Capra hircus). S Afr J Wildl Res 26:64–65
Heckathorn SA, McNaughton SJ, Coleman JS (1999) C4 plants and herbivory. In: Sage RF,
Monson RK (eds) C4 plant biology. Academic Press, San Diego, pp 285–312
Hofmann RR (1968) Comparison of the rumen and omasum structure in East African game rumi-
nants in relation to their feeding habits. Sym Zool Soc Lond 21:179–194
Hofmann RR (1973) The ruminant stomach. East African Literature Bureau, Nairobi
Hofmann RR (1988) Morphophysiological evolutionary adaptations of the ruminant digestive
system. In: Dobson A, Dobson MJ (eds) Aspects of digestive physiology in ruminants. Cornell
University Press, Ithaca, NY, pp 1–20
Hofmann RR (1989) Evolutionary steps of ecophysiological adaptation and diversification of
ruminants: a comparative view of their digestive system. Oecologia 78:443–457
Hofmann RR (1991) Endangered tropical herbivores - their nutritional requirements and habitat
demands. In: Ho YW, Wong HK, Abdullah N, Tajuddin ZA (eds) Recent advances on the
nutrition of herbivores. Malaysia Society of Animal Production, UPM Serdang, pp 27–34
Hofmann RR (1999) Functional and comparative digestive system anatomy of Arctic ungulates.
Rangifer 20:71–81
Hofmann RR, Stewart DRM (1972) Grazer or browser: a classification based on the stomach-
structure and feeding habit of East African ruminants. Mammalia 36:226–240
Hofmann RR, Streich WJ, Fickel J, Hummel J, Clauss M. Convergent evolution in feeding types:
salivary gland mass differences in wild ruminant species. J Morphal (in press)
Holechek JL, Pieper RD, Herbel CH (2004) Range management. Principles and practices, 5th edn.
Pearson/Prentice Hall, Upper Saddle River, NJ
Hoppe PP (1977) Rumen fermentation and body weight in African ruminants. In: Peterle TJ (ed)
13th Congress of Game Biology, vol 13. The Wildlife Society, Washington, DC, pp 141–150
Hummel J, Clauss M, Zimmermann W, Johanson K, Norgaard C, Pfeffer E (2005) Fluid and article
retention in captive okapi (Okapia johnstoni). Comp Biochem Physiol A 140:436–444
Hummel J, Südekum KH, Streich WJ, Clauss M (2006) Forage fermentation patterns and their
implications for herbivore ingesta retention times. Funct Ecol 20:989–1002
Iason G, Palo RT (1991) The effects of birch phenolics on a grazing and a browsing mammal. A
comparison of hares. J Chem Ecol 17:1733–1743
Iason GR, Van Wieren SE (1999) Digestive and ingestive adaptations of mammalian herbivores
to low-quality forage. In: Olff H, Brown VK, Drent RH (eds) Herbivores: between plants and
predators. 38th Symp Brit Ecol Soc Blackwell, Oxford, pp 337–369
Illius AW, Gordon IJ (1992) Modelling the nutritional ecology of ungulate herbivores: evolution
of body size and competitive interactions. Oecologia 89:428–434
Illius AW, Gordon IJ (1999) The physiological ecology of mammalian herbivory. In: Jung HJG,
Fahey GC (eds) Nutritional ecology of herbivores. The American Society of Animal Science,
Savoy, IL, pp 71–96
Ioannidis JPA (2005) Why most published research findings are false. PLos Med 2:e124
(696–701)
Janis CH (1988) An estimation of tooth volume and hypsodonty indices in ungulate mammals and
the correlation of these factors with dietary preference. Teeth revisited. Proceedings of the VIIth
International Symposium on Dental Morphology. Mem Mus Hist Naturelle Paris C
53:367–387
Janis CM (1990) Correlation of cranial and dental variables with dietary preferences in mammals:
a comparison of macropodoids and ungulates. Mem Queensland Mus 28:349–366
Janis CM (1993) Tertiary mammal evolution in the context of changing climates, vegetation, and
tectonic events. Annu Rev Ecol Syst 24:467–500
Janis CM (1995) Correlations between craniodental morphology and feeding behavior in ungu-
lates: reciprocal illumination between living and fossil taxa. In: Thomason JJ (ed) Functional
morphology in vertebrate paleontology. Cambridge Univ. Press, New York, pp 76–98
Janis CM, Scott KM (1987) The interrelationships of higher ruminant families with special
emphasis on the members of the cervoidea. Am Mus Novit 2893:1–85
Janis CM, Ehrhardt D (1988) Correlation of the relative muzzle width and relative incisor width
with dietary preferences in ungulates. Zool J Linn Soc–Lond 92:267–284
Janis CM, Fortelius M (1988) The means whereby mammals achieve increased functional durabil-
ity of their dentitions, with special reference to limiting factors. Biol Rev 63:197–230
Janis CM, Constable E (1993) Can ungulate craniodental features determine digestive physiology?
J Vertebr Paleontol 13:abstract
Jarman PJ (1974) The social organization of antelope in relation to their ecology. Behaviour
48:215–266
Jiang Z, Takatsuki S (1999) Constraints on feeding type in ruminants: a case for morphology over
phylogeny. Mammal Study 24:79–89
Johnson CN, Prideaux GJ (2004) Extinctions of herbivorous mammals in the late Pleistocene of
Australia in relation to their feeding ecology: no evidence for environmental change as cause
of extinction. Austr Ecol 29:553–557
Jones RJ, Meyer JHF, Bechaz FM, Stolzt MA, Palmer B, van der Merwe G (2001) Comparison of
rumen fluid from South African game species and from sheep to digest tanniniferous browse.
Aust J Agr Res 52:453–460
Kaiser TM, Fortelius M (2003) Differential mesowear in occluding upper and lower molars: opening
mesowear analysis for lower molars and premolars in hypsodont horses. J Morphol 258:67–83
Kappelman J (1988) Morphology and locomotor adaptations of the bovid femur in relation to
habitat. J Morphol 198:119–130
Kappelmann J, Plummer T, Bishop L, Duncan A, Appleton S (1997) Bovids as indicators of plio-
pleistocene paleoenvironments in East Africa. J Hum Evol 32:229–256
Kaske M, Midasch A (1997) Effects of experimentally impaired reticular contractions on digesta
passage in sheep. Brit J Nutr 78:97–110
Kay RF, Madden RH (1997) Mammals and rainfall: paleoecology of the middle Miocene at La
Venta (Columbia, South America). J Hum Evol 32:161–199
Kay RNB (1987a) Comparative studies of food propulsion in ruminants. In: Ooms LAA, Degtyse
AD, Van Miert ASJ (eds) Physiological and pharmacological aspects of the reticulo-rumen.
Marinus Nijhoff, Boston, pp 155–170
Kay RNB (1987b) Weights of salivary glands in some ruminant animals. J Zool 211:431–436
Kay RNB (1993) Digestion in ruminants at pasture. World Conference on Animal Production
Edmonton, Canada, pp 461–474
Kay RNB, Engelhardt Wv, White RG (1980) The digestive physiology of wild ruminants. In:
Ruckebush Y, Thivend P (eds) Digestive physiology and metabolism in ruminants. MTP Press,
Lancaster, pp 743–761
84 M. Clauss et al.
Kock RA, Garnier J (1993) Veterinary management of three species of rhinoceros in zoological
collections. In: Ryder OA (ed) Rhinoceros biology and conservation. Zoological Society of
San Diego, San Diego, pp 325–338
Köhler M (1993) Skeleton and habitat of recent and fossil ruminants. Münchner Geowiss Abh A:
Geol Paläontol 25:1–88
Langer P, Takács A (2004) Why are taeniae, haustra, and semilunar folds differentiated in the
gastrointestinal tract of mammals, including man? J Morphol 259:308–315
Lechanteur YARG, Griffiths CL (2003) Diets of common suprabenthic reef fish in False Bay,
South Africa. Afr Zool 38:213–227
Lechner-Doll M, Kaske M, Engelhardt Wv (1991) Factors affecting the mean retention time of
particles in the forestomach of ruminants and camelids. Proc Internat Sym Ruminant Physiol
7:455–482
Lentle RG, Hume ID, Stafford KJ, Kennedy M, Haslett S, Springett BP (2003a) Comparisons of
indices of molar progression and dental function of brush-tailed rock-wallabies (Petrogale
penicillata) with tammar (Macropus eugenii) and parma (Macropus parma) wallabies. Aust J
Zool 51:259–269
Lentle RG, Hume ID, Stafford KJ, Kennedy M, Haslett S, Springett BP (2003b) Molar progression
and tooth wear in tammar (Macropus eugenii) and parma (Macropus parma) wallabies. Aust
J Zool 51:137–151
Lentle RG, Hume ID, Stafford KJ, Kennedy M, Springett BP, Haslett S (2003c) Observations on
fresh forage intake, ingesta particle size and nutrient digestibility in four species of macropod.
Aust J Zool 51:627–636
Leus K, MacDonald A (1997) From barbirusa (Babyrousa babyrussa) to domestic pig: the nutri-
tion of swine. Proc Nutr Soc 56:1001–1012
Lirette A, Milligan LP, Cyr N, Elofson RM (1990) Buoyancy separation of particles of forages,
feces, and ruminal contents and nuclear magnetic resonance examination. Can J Anim Sci
70:1099–1108
Loehlein W, Kienzle E, Wiesner H, Clauss M (2003) Investigations on the use of chromium oxide
as an inert, external marker in captive Asian elephants (Elephas maximus): passage and
recovery rates. In: Fidgett A, Clauss M, Ganslosser U, Hatt JM, Nijboer J (eds) Zoo animal
nutrition, vol 2. Filander, Fuerth, Germany, pp 223–232
MacFadden BJ (1992) Fossil horses. Cambridge University Press, Cambridge
MacFadden BJ, Solounias N, Cerling TE (1999) Ancient diets, ecology, and extinction of 5-
million-year-old horses from Florida. Science 283:824–827
Maglio V (1973) Origin and evolution of the elephantidae. Trans Am Philos Soc 63:1–149
Maloiy GMO, Clemens ET (1991) Aspects of digestion and in vitro fermentation in the caecum
of some East African herbivores. J Zool 224:293–300
Martins EP, Hansen TF (1996) The statistical analysis of interspecific data: a review and
evaluation of phylogenetic comparative methods. In: Martins EP (ed) Phylogenies and the
comparative method in animal behavior. Oxford University Press, Oxford, pp 22–75
Martz FA, Belyea RL (1986) Role of particle size and forage quality in digestion and passage by
cattle and sheep. J Dairy Sci 69:1996–2008
McArthur C, Sanson GD (1993) Nutritional effects and costs of a tannin in a grazing and a
browsing macropodid marsupial herbivore. Funct Ecol 7:690–969
McCammon-Feldman B, van Soest PJ, Horvath P, McDowell RE (1981) Feeding strategy of the
goat. Cornell University, Ithaca, NY
McDowell RE, Sisler DG, Schermerhorn EC, Reed JD, Bauer RP (1983) Game or cattle for meat
production on Kenya rangelands? Cornell International Agriculture Monograph, Ithaca, NY
McNaughton SJ, Tarrants JL, MacNaughton MM, Davis RH (1985) Silica as a defense against
herbivory and a growth promotor in African grasses. Ecology 66:528–535
Mendoza M, Palmqvist P (2006a) Characterizing adaptive morphological patterns related to diet
in bovidae. Acta Zool Sin 52:988–1008
Mendoza M, Palmqvist P (2006b) Characterizing adaptive morphological patterns related to
habitat use and body mass in bovidae. Acta Zool Sin 52:971–987
Mendoza M, Janis CM, Palmqvist P (2002) Characterizing complex craniodental patterns related
to feeding behaviour in ungulates: a multivariate approach. J Zool 258:223–246
Milton K, Dintzis FR (1981) Nitrogen-to-protein conversion factors for tropical plant samples.
Biotropica 13:177–181
Moseley G, Jones JR (1984) The physical digestion of perennial ryegrass (Lolium perenne) and
white clover (Trifolium repens) in the foregut of sheep. Brit J Nutr 52:381–390
Mtengeti EJ, Wilman D, Moseley G (1995) Physical structure of white clover, rape, spurrey and
perennial ryegrass in relation to rate of intake by sheep, chewing activity and particle break-
down. J Agr Sci 125:43–50
Neuville H, Derscheid JM (1929) Recherches anatomiques sur l’okapi. IV. L’estomac. Rev Zool
Afr 16:373–419
Nocek JE, Kohn RA (1987) Initial particle form and size on change in functional specific gravity
of alfalfa and timothy hay. J Dairy Sci 70:1850–1863
Oftedal O (1991) The nutritional consequences of foraging in primates: the relationship of nutrient
intakes to nutrient requirements. Philos Trans R Soc B 334:161–170
Oldemeyer JL, Franzmann AW, Brundage AL, Arneson PD, Flynn A (1977) Browse quality and
the Kenai moose population. J Wildlife Manage 41:533–542
Owen-Smith N (1982) Factors influencing the consumption of plant products by large herbivores.
In: Huntley BJ, Walker BH (eds) Ecology of tropical savannas. Springer, Berlin Heidelberg
New York, pp 359–404
Owen-Smith N (1985) Niche separation among African ungulates. In: Vrba ES (ed) Species and
Speciation, vol 4. Transvaal Museum, Pretoria, pp 167–171
Owen-Smith N (1988) Megaherbivores - the influence of very large body size on ecology.
Owen-Smith N (1997) Distinctive features of the nutritional ecology of browsing versus grazing
ruminants. Z Säugetierkd 62 (Suppl. 2):176–191
Palamara J, Phakey PP, Rachinger WA, Sanson GD, Orams HJ (1984) On the nature of the opaque
and translucent enamel regions of some macropodinae (Macropus giganteus, Wallabia bicolor
and Peradorcas concinna). Cell Tissue Res 238:329–337
Palmqvist P, Groecke DR, Arribas A, Farina RA (2003) Paleoecological reconstruction of a lower
Pleistocene large mammal community using biogeochemical and ecomorphological approaches.
Paleobiology 29:205–229
Pérez-Barbería FJ, Gordon IJ (1999) The functional relationship between feeding type and jaw
and cranial morphology in ungulates. Oecologia 118:157–165
Pérez-Barbería FJ, Gordon IJ (2001) Relationships between oral morphology and feeding style
in the Ungulata: a phylogenetically controlled evaluation. Proc R Soc Lond B Bio
268:1023–1032
Pérez-Barbería FJ, Gordon IJ (2005) Gregariousness increases brain size in ungulates. Oecologia
145:41–52
Pérez-Barbería FJ, Gordon IJ, Illius A (2001a) Phylogenetic analysis of stomach adaptation in
digestive strategies in African ruminants. Oecologia 129:498–508
Pérez-Barbería FJ, Gordon IJ, Nores C (2001b) Evolutionary transitions among feeding styles and
habitats in ungulates. Evol Ecol Res 3:221–230
Pérez-Barbería FJ, Elston DA, Gordon IJ, Illius AW (2004) The evolution of phylogenetic
differences in the efficiency of digestion in ruminants. Proc Roy Soc Lond B Bio
271:1081–1090
Peters RH (1986) The ecological implications of body size. Cambridge University Press,
Cambridge
Plummer TW, Bishop LC (1994) Hominid paleoecology at Olduvai Gorge, Tanzania as indicated
by antelope remains. J Hum Evol 27:47–75
Popowics TE, Fortelius M (1997) On the cutting edge: tooth blade sharpness in herbivorous and
faunivorous mammals. Ann Zool Fenn 34:73–88
Prins RA, Geelen MJH (1971) Rumen characteristics of red deer, fallow deer and roe deer. J
Wildlife Manage 35:673–680
86 M. Clauss et al.
Prins RA, Kreulen DA (1991) Comparative aspects of plant cell wall digestion in mammals. In:
Hoshino S, Onodera R, Minoto H, Itabashi H (eds) The rumen ecosystem. Japan Scientific
Society Press, Tokyo, pp 109–120
Prins RA, Rooymans TP, Veldhuizen M, Domhof MA, Cliné-Theil W (1983) Extent of plant cell wall
digestion in several species of wild ruminants kept in the zoo. Zool Garten NF 53:393–403
Prins RA, Lankhorst A, Van Hoven W (1984) Gastro-intestinal fermentation in herbivores and the
extent of plant cell wall digestion. In: Gilchrist FMC, Mackie RI (eds) Herbivore nutrition in
the subtropics and tropics. Science Press, Craighall, South Africa, pp 408–434
Rhoades DF, Gates RG (1976) Towards a general theory of plant antiherbivore chemistry. Recent
Adv Phytochem 10:168–213
Robbins C, Hagerman A, Austin P, McArthur C, Hanley T (1991) Variation in mammalian physio-
logical responses to a condensed tannin and its ecological implications. J Mammal 72:480–486
Robbins CT (1993) Wildlife feeding and nutrition. Academic Press, San Diego
Robbins CT, Spalinger DE, Van Hoven W (1995) Adaptations of ruminants to browse and grass
diets: are anatomical-based browser–grazer interpretations valid? Oecologia 103:208–213
Sanson GD (1989) Morphological adaptations of teeth to diets and feeding in the macropodoidea.
In: Grigg G, Jarman P, Hume I (eds) Kangaroos, wallabies and rat-kangaroos. Surrey-Beatty,
Sydney, pp 151–168
Sanson GD (2006) The biomechanics of browsing and grazing. Am J Bot 93:1531–1545
Sanson GD, Kerr SA, Gross KA (2007) Do silica phytoliths really wear mammalian teeth? J
Archaeol Sci 34:526–531
Schmidt-Nielsen K (1984) Scaling: Why is animal size so important? Cambridge University Press,
Cambridge
Scott KM (1985) Allometric trends and locomotor adaptations in the Bovidae. Bull Am Mus Nat
Hist 179:197–288
Scott KM (1987) Allometry and habitat-related adaptations in the postcranial skeleton of cervidae.
In: Wemmer CM (ed) Biology and management of the cervidae. Smithsonian Press,
Washington, DC, pp 65–79
Short HL (1975) Nutrition of southern deer in different seasons. J Wildlife Manage 39:321–329
Short HL, Blair RM, Segelquist CA (1974) Fiber composition and forage digestibility by small
ruminants. J Wildl Manage 38:197–209
Sibly RM (1981) Strategies of digestion and defecation. In: Townsend C, Calow P (eds) Physiological
ecology: an evolutionary approach to resource utilization. Blackwell, Oxford, pp 109–139
Siepel H, de Ruiter-Dijkman EM (1993) Feeding guilds of oribatid mites based on their carbohy-
drase activities. Soil Biol Biochem 25:1491–1497
Simpson GG (1953) The major features of evolution. Columbia University Press, New York
Solounias N, Dawson-Saunders B (1988) Dietary adaptations and palaecology of the late Miocene
ruminants from Pikermi and Samos in Greece. Palaeogeogr Palaeoecl 65:149–172
Solounias N, Moelleken S (1993) Dietary adaptations of some extinct ruminants determined by
premaxillary shape. J Mammal 74:1059–1071
Solounias N, Moelleken SMC (1999) Dietary determination of extinct bovids through cranial
foraminal analysis, with radiographic applications. Ann Mus Goulandris 10:267–290
Solounias N, Semprebon G (2002) Advances in the reconstruction of ungulate ecomporphology
with application to early fossil equids. Am Mus Novit 3366:1–49
Solounias N, Teaford M, Walker A (1988) Interpreting the diet of extinct ruminants: the case of a
non-browsing giraffid. Paleobiology 14:287–300
Solounias N, Fortelius M, Freeman P (1994) Molar wear rates in ruminants: a new approach. Ann
Zool Fenn 31:219–227
Solounias N, Moelleken S, Plavcan J (1995) Predicting the diet of extinct bovids using masseteric
morphology. J Vertebr Paleontol 15:795–805
Spalinger DE, Robbins CT, Hanley TA (1986) The assessment of handling time in ruminants: the
effect of plant chemical and physical structure on the rate of breakdown of plant particles in
the rumen of mule deer and elk. Can J Zool 64:312–321
Spencer LM (1995) Morphological correlates of dietary resource partitioning in the African

bovidae. J Mammal 76:448–471
Sponheimer M, Reed KE, Lee-Thorp JA (1999) Combining isotopic and ecomorphological data
to refine bovid paleodietary reconstruction: a case study from the Makapansgat Limeworks
hominin locality. J Hum Evol 36:705–718
Sponheimer M, Lee-Thorp JA, DeRuiter D, Smith JM, Van der Merwe NJ, Reed K, Grant CC,
Ayliffe LK, Robinson TF, Heidelberger C, Marcus W (2003) Diets of Southern African bovi-
dae: stable isotope evidence. J Mammal 84:471–479
Sprent JA, McArthur C (2002) Diet and diet selection of two species in the macropodid browser–
grazer continuum: do they eat what they ‘should’? Aust J Zool 50:183–192
Stöckmann W (1979) Differences in the shape of the mandibles of African bovidae in relation to
food composition. Zool Jahrb Syst 106:344–373
Thenius E (1992) The okapi from Zaire - a “living fossil” or a secondary rainforest inhabitant? Z
Zool Syst Evol 30:163–179
Troelsen JE, Campbell JB (1968) Voluntary consumption of forage by sheep and its relation to
the size and shape of particles in the digestive tract. Anim Prod 10:289–296
Turnbull WD (1970) Mammalian masticatory apparatus. Fieldiana Geol 18:147–356
Van Hoven W (1991) Mortalities in kudu populations related to chemical defence in trees. J Afr
Zool 105:141–145
Van Soest PJ (1975) Physico-chemical aspects of fiber digestion. In: McDonald IW, Warner ACI
(eds) Digestion and metabolism of the ruminant. University of New England Publishing,
Armidale, NSW, pp 352–365
Van Soest PJ (1994) Nutritional ecology of the ruminant, 2nd edn. Cornell Univ Press, Ithaca,
New York
Van Soest PJ (1996) Allometry and ecology of feeding behavior and digestive capacity in
herbivores: a review. Zoo Biol 15:455–479
Van Wieren SE (1996a) Nutrient extraction from mixed grass-browse diets by goats and sheep.
In: Van Wieren SE (ed) Digestive strategies in ruminants and nonruminants. Thesis University
of Wageningen, pp 67–79
Van Wieren SE (1996b) Browsers and grazers: foraging strategies in ruminants. In: Van Wieren
SE (ed) Digestive strategies in ruminants and nonruminants. Thesis University of Wageningen,
pp 119–146
Vrba ES (1978) The significance of bovid remains as indicators of environment and predation
patterns. University of Chicago Press, Chicago
Waldo DR, Smith LW, Cox EL (1972) Model of cellulose disappearance from the rumen. J Dairy
Sci 55:125–129
Walker A, Hoeck HN, Perez L (1978) Microwear of mammalian teeth as indicators of diet.
Science 201:908–910
Warner RG, Flatt WP, Loosli JK (1956) Dietary factors influencing the development of the
ruminant stomach. Agr Food Chem 4:788–792
Wattiaux MA, Satter LD, Mertens DR (1992) Effect of microbial fermentation on functional
specific gravity of small forage particles. J Anim Sci 70:1262–1270
Westoby M, Leishman MR, Lord JM (1995) On misinterpreting the “phylogenetic correction”. J Ecol
83:531–534
Williams SH, Kay RF (2001) A comparative test of adaptive explanations for hypsodonty in
ungulates and rodents. J Mammal Evol 8:207–229
Williams J, Ostrowski S, Bedin E, Ismail K (2001) Seasonal variation in energy expenditure, water
flux and food consumption of Arabian oryx (Oryx leucoryx). J Exp Biol 204:2301–2311
Wilman D, Riley JA (1993) Potential nutritive value of a wide range of grassland species. J Agr
Sci 120:43–49
Wilson JR (1993) Organization of forage plant tissues. In: Jung HG, Buxton DR, Hatfield RD,
Ralph J (eds) Forage cell wall structure and digestibility. American Society of Agronomy,
Madison, pp 1–32
88 M. Clauss et al.
Wilson JR, McLeod NM, Minson DJ (1989) Particle-size reduction of the leaves of a tropical and
a temperate grass by cattle. I. Effect of chewing during eating and varying times of digestion.
Grass Forage Sci 44:55–63
Witzel U, Preuschoft H (1999) The bony roof of the nose in humans and other primates. Zool Anz
238:103–115
Witzel U, Preuschoft H (2002) The functional shape of the human skull, as documented by three-
dimensional FEM studies. Anthropol Anz 60:113–135
Witzel U, Preuschoft H (2005) Finite-element model construction for the virtual synthesis of the
skulls in vertebrates: case study of Diplodocus. Anat Rec Part A 283:391–401
Wofford H, Holechek JL (1982) Influence of grind size on four- and forty-eight hour in vitro
digestibility. Proc W Sect Am Soc Anim Sci 33:261–263
Woodall PF (1992) An evaluation of a rapid method for estimating digestibility. Afr J Ecol
30:181–185
Chapter 4
Nutritional Ecology of Grazing and Browsing
Ruminants
Alan J. Duncan and Dennis P. Poppi
4.1 Introduction
Ruminants are, without exception, obligate herbivores subsisting as they do on

a diet composed entirely of plant material. However, plant material is a diverse
resource and within the Ruminantia there is a range of feeding niches with
different herbivore classes focussing their foraging effort on different vegetation
types (Hofmann 1989). The plant material available to herbivores comes in a
range of morphological types with the major types being grasses, forbs, and
browse. Grasses (including the morphologically similar sedges) are
monocotyledonous plants characterised by a basal meristem, a low growth form
and a relative lack of lignified support structures (except for some of the tall
tropical grasses which could be considered to be morphologically classed as
browse). Browse plants, comprising shrubs and trees are largely dicotyledo-
nous plants characterised by an apical meristem, a low or high growth form and
a well developed system of lignified support structures. Forbs are generally
intermediate in character being dicotyledonous, with apical meristems but gen-
erally showing low growth forms and a relative lack of lignified growth struc-
tures. This variation in growth form, degree of lignification and location of the
meristem has implications for resource allocation within the plant, e.g., relative
investment in structural tissue and defensive chemicals. This in turn affects the
nutritional value of plants for potential herbivores. The purpose of this chapter
is to consider the way in which plants are digested by ruminants and how this
varies with feeding habit. The chapter will first provide some simple background
on digestion of plants by ruminants and how this is influenced by the presence
of plant secondary metabolites in plant material. The discussion will then turn
to differences between browsers and grazers and review the evidence for
differences in the way in which they process the food they consume. Finally,
the extent to which browsers and grazers deal differently with plant secondary
metabolites will be explored.
© Springer 2008
90 A.J. Duncan and D.P. Poppi
4.2 Nutritive Value of Plant Material for Ruminant Herbivores
4.2.1 Digestion of plant material by Herbivores
Plant material can be considered to be composed of cell wall material and cell
contents. The structure of plants is provided by cell walls composed of cellulose
and hemicellulose with varying amounts of lignin to increase the strength of the
cell walls. Cellulose cannot be digested by enzymes of mammalian origin so
herbivores rely on symbiotic micro-organisms within their digestive tracts capable
of cellulolytic activity (Hungate 1966). Bacteria are presumed to exist in the digestive
tract of all herbivores because cellulase is not in the suite of mammalian enzymes.
In ruminants, as well as bacteria, protozoa and fungi are present in the digestive
tract. They are contained largely within a pre-gastric chamber known as the rumen
but also in hind-gut organs such as the caecum and colon. Cellulolytic micro-
organisms convert cellulose to volatile fatty acids (VFAs), predominantly acetate,
butyrate and propionate, mainly by anaerobic metabolic pathways. It is these VFAs
which are used as an energy source by ruminants. The degree to which cell walls
are digested by micro-organisms depends on a number of factors, but notably on
the extent to which the cell wall is lignified and thus protected from colonisation
and utilisation by microbes, on the length of time the plant material resides in the
rumen prior to further passage via the omasum to the abomasum, and on the
availability of nitrogen for microbial growth within the rumen. The length of time
required for degradation of cell walls by micro-organisms has led to the evolution
of a mechanism for selective retention of cell-wall particles while allowing soluble
components of plant material and small particles to pass quickly on (Van Soest
1994). In the ruminant this selective retention occurs at the outflow of the reticulo-
rumen into the omasum. Cell contents consisting mainly of sugars, starch, and
protein are all rapidly and completely digestible by ruminant animals.
4.2.2 How do Browse and Grass Differ in Nutritive Value?
Grasses and browse represent very different food resources for ruminant herbivores.
These differences in food characteristics led to the ideas of niche separation
amongst ungulates (Bell/Jarman principle; Bell 1971; Jarman 1974) which were
adopted in Hoffman’s classic papers that classified the spectrum of ruminant
herbivores into the grazers, intermediate feeders, and browsers (‘concentrate
selectors’, to use Hoffman’s terminology; Hofmann and Stewart 1972; Hofmann
1973, 1989). Hofmann presents a view of the nutritive characteristics of grass and
browse in which grasses are rich in relatively unlignified cell wall material and
hence available for cellulolysis by rumen micro-organisms. Browse, on the other
hand, has a lower proportion of cell-wall material but is rich in cell contents.
Furthermore, the cell-wall component of browse in heavily lignified and hence
4 Nutritional Ecology of Grazing and Browsing Ruminants 91
refractory to microbial cellulolysis (Hofmann 1989). These characteristics,

according to Hofmann, led to distinct feeding styles among herbivores specialising
on either grass or browse. Grazers would tend to retain material in their rumen for
long periods to allow maximum opportunity for micro-organisms to digest cell-wall
material. Browsers, on the other hand, with most of their nutrition derived from
rapidly fermentable cell contents, should be expected to adopt a strategy of passing
material rapidly through the rumen.
Does Hofmann’s view of the nutritive characteristics of grasses and browse still
hold with the advent of more biologically realistic methods of assessing nutritive
value? It has been suggested that Hofmann’s view of nutritive value came from an
era during which chemical assessments of nutritive value predominated and the role
of plant secondary metabolites in limiting the nutritive value of browse plants was
not fully realised (Robbins et al. 1995). Grasses certainly show higher neutral
detergent fibre (NDF) percentages than the leaves of browse and forbs. Typical
values for grasses are in the region of 50–70% NDF while browse values tend to be
in the region of 30–50% (Van Soest 1982). However, numerous recent papers point
to the inadequacy of detergent extraction methods for characterising the nutritive
value of browse (reviewed in Mould 2003). Polyethylene glycol (PEG) interacts
with tannins, effectively neutralising their protein-binding properties (Jones and
Mangan 1977). Addition of PEG to in vitro systems for assessing nutritive value
has repeatedly demonstrated the often large depressive effects of tannins on fer-
mentation rates in the rumen. As an example, in a study of the nutritive value of a
range of plant species found on the Tibetan Plateau, Long et al. (1999) found that
grasses and sedges had higher NDF values and, by implication, lower digestibility
than browse and forbs. These differences in digestibility were less marked when in
sacco techniques were used to assess digestibility, and when gas production meth-
ods were used the differences disappeared altogether. Addition of PEG to gas
production syringes demonstrated the large effect of tannins on rates of fermenta-
tion in forb and browse samples (Fig 4.1).
Poppi et al. (1999) reviewed the chemical composition of a large number of
grasses and dicotyledonous plants including tropical and temperate species. They
concluded that whilst crude protein (CP) content varied widely among plant species
and stages of maturity, the CP/neutral detergent-soluble ratio was relatively
constant, indicating that within the cell content and metabolic fraction of the plant
there was a constancy in the mix of photosynthetic enzymes and enzymes involved
in other metabolic pathways. The proportion of plant cell wall was the feature
which separated plants in terms of composition per unit of dry matter.
As well as the chemical differences in the properties of grass and browse, large
structural differences exist. The grass material consumed by ruminants tends to
be long and fibrous in structure whereas the leaves of browse plants break down
into small polygonous particles during mastication (Spalinger et al. 1993). It has
recently been suggested that the physical characteristics of grass and browse are
a major determinant of the nutritional strategies of grazers and browsers. Clauss
et al. (2003) argue that the tendency of grass-based diets to stratify in the rumen
into a fibrous raft floating on a liquid phase has led to morphological adaptation
100
90
80
70
60 Forb
% or ml
Grass
50
Sedge
40 Shrub
30
20
10
0
NDF (%) In sacco D Gas Increase in
(%) production gas
(ml) production
with PEG (%)
Fig. 4.1 Mean values of NDF, in sacco digestibility, gas production, and increase in gas produced
in presence of PEG for a range of plant species of different morphological types found on the
Tibetan Plateau (based on the data of Long et al. 1999) Error bars represent SEM. (Only one shrub
representative appeared in the original dataset)
of the rumen among grass specialists. The stratification of rumen contents, they
argue, facilitates longer retention of solids in the rumen and hence more thorough
fermentation. Browse material, on the other hand, is distributed more uniformly
within the rumen leaving less opportunity for selective retention of large parti-
cles, and browsers therefore have to rely on alternative strategies for deriving
nutrients from their food (see below). Thornton and Minson (1973) observed a
higher packing density in the rumen of sheep of legume forage compared to grass
forage and used that concept to explain the higher intake of legumes. This sug-
gests that it is the different morphology of the herbage types per se that is impor-
tant and that the way in which the herbage material behaves in the rumen is not
controlled by the animal.
4.2.3 Plant Secondary Metabolites
As well as cell walls and cell contents, plant material contains the so-called plant
secondary metabolites (PSM), a diverse array of biologically active compounds
some of which may have an adaptive role in protecting plant material from
herbivory (Rosenthal and Janzen 1979). Plant secondary metabolites are largely
absent from graminaceous species but are generally abundant in forbs and
especially browse species (Harborne 1988). As well as performing a range of
other ecological functions, some plant secondary metabolites limit nutritive value
of plant material in a number of ways. Condensed tannins, which are polymers of
flavonoid sub-units, form complexes with proteins and, to some extent carbohy-
drates, (Zucker 1983). In this way they limit the extent of microbial degradation
of cell wall material in the rumen. A number of other classes of PSMs act at the
level of the rumen by inhibiting microbial fermentation directly. These include
phenolic acids (Theodorou et al. 1987) and terpenes (Oh et al. 1967) as well as a
range of other compounds (Wallace 2004). Still other PSMs exert their effects
following absorption and distribution to the tissues. These include the alkaloids
such as the pyrrolizidine alkaloids found in Senecio species which form stable
pyrroles in the liver and thus act as cumulative hepatotoxins (Stegelmeier et al.
1999). The range of PSMs is large and their physiological action is known in only
sketchy detail for a few well-studied plants, usually those of agronomic
importance.
4.3 Do Browsers and Grazers Differ in the Way

They Process Their Food?
It was Hofmann who laid the groundwork for the debate over the different
digestive strategies of browsers vs. grazers. Hofmann’s arguments are summa-
rised in Hofmann (1989). In summary, Hofmann argues that the different feed-
ing niches of browsers and grazers have led to morpho-physiological adaptations
connected to the chemical characteristics of their diets. Thus, browsers, which
consume a diet rich in cell contents but whose cell walls are refractory to diges-
tion adopt a strategy involving rapid passage rates, small reticulo-rumen vol-
ume, copious saliva production to hasten passage and buffer high fermentation
rates in the rumen, rapid absorption of VFAs, less selective particle retention in
the rumen, and greater reliance on hind-gut fermentation. Grazers on the other
hand have adopted a different strategy involving slow passage rate, large retic-
ulo-rumen volumes to allow sufficient time for cellulose digestion, slower
saliva flow and less need to absorb VFAs quickly. Grazers also retain small
particles more effectively to allow extensive energy extraction from cellulose in
the rumen. These ideas have sparked a long-running debate in the literature; the
main sceptics of the Hofmann hypothesis have argued that variation in strate-
gies for dealing with different diets can be explained wholly by body size
effects without the need to invoke classifications based on morpho-physiological
adaptation (Gordon and Illius 1994, 1996; Robbins et al. 1995). More recent
contributions to the debate have suggested that the polarity of positions is
unnecessary with both body size and morpho-physiological specialisation con-
tributing to observed variation in digestive strategy (Iason and Van Wieren
1999; Perez-Barberia et al. 2004). Some of the evidence for these varying posi-
tions is reviewed in the following sections (summarised in Table 1).
94
Table 4.1 Summary of evidence for and against physiological adaptations of browsers (BR) and grazers (GR) to their feeding habit
Supporting Equivocal Contrary
Adaptation Browsers Grazers evidence evidence evidence Reference
Reticulo-rumen Low High Meta-analysis of wet RR Clauss et al. 2003
(RR) volume contents indicated that
relative to GR had higher values
body size than BR
Van Wieren conducted statistical Van Wieren 1996
analysis on Hofmann’s data
and found that RR volume
did not differ between
BR and GR
Passage rate High Low Indirect evidence of faster
passage rate in BR
comes from their body
fat composition which
is richer in unsaturated
lipids, pointing to less
biohydrogenation in the
rumen
Meta-analysis shows that for a given Gordon and Illius
body mass, MRT of BR and GR do 1994
not differ However, GR and BR body
masses did not overlap to any great
extent in the dataset analysed. Also,
the quality of the MRT data may have
been limited since Foose data relied on
bulked daily faecal samples. It was
not clear what foods were used in the
MRT trials; if BR were fed on grass
hay, one would not expect shorter
retention times
A.J. Duncan and D.P. Poppi
Mean particle High Low Larger faecal particle size Clauss 2002
size escaping in BR
rumen
Meta-analysis indicates that Clauss 2001
GR retain particles more
efficiently in the rumen
than BR
Efficiency of Low High Meta-analysis of digest- Iason and van
digestion of ibility trials shows that Wieren 1999
fibre in the digestibility (D) is nega-
rumen tively related to lignin
and positively related
to body mass but that
GR show more efficient
NDF digestion than BR
NDF (cell wall) digestibility values from Robbins et al.
published studies were compared 1995
across a range of
herbivores and found not to
4 Nutritional Ecology of Grazing and Browsing Ruminants
differ between BR and GR

Salivary gland High Low BR have higher parotid sali- Robbins et al.
size to buffer vary gland weights than 1995
highly fer- GR for a given body
mentable weight
feeds
Resting saliva flow does not differ Robbins et al.
between GR and BR, but, saliva 1995
flow during browsing was not
measured. This is the parameter
that would be expected to differ
if an oesophageal groove
mechanism operated in BR
95
(continued)
96
Absorptive sur- High Low Van Wieren (1996) reana- Van Wieren 1996
face area of lysed Hofmann’s 1973
rumen to rap- data and found that den-
idly absorb sity of rumen papillae
VFAs to pre- on the ventral wall was
vent acidosis higher in BR than GR
Measurements of absorptive Demment and
area of rumen between Longhurst
BR and GR show that, 1987
for a given body size,
BR have higher surface
area than GR
Rumen muscu- poorly well Meta-analysis of published Clauss et al. 2003
lature to deal devel- devel- data shows that RR pil-
with fibrous oped oped lar thickness greater in
diets GR than BR
Fermentation High Low Robbins argues that the nutritive Robbins et al.
rate in rumen value and hence rumen fermentation 1995
rates of browse diets have been
over-emphasised without considering
the suppressing effects of PSMs/
tannins. However, this does not
tally with the dataset of Gordon
and Illius (1994) which found
fermentation rates in small
ruminants (mainly BR) to be
higher than in large ruminants
Meta-analysis of fermentation rates Gordon and Illius
shows that the same relationship 1994
applies to GR and BR when
body size effects are taken
into account
Tannin-binding Occurs Does not Salivary tannin-binding pro- Austin et al. 1989
salivary occur teins found in saliva of
proteins mule deer (Odocoileus
hemionus hemionus), a
browser, but not in the
saliva of sheep and cattle
When sheep and deer were Robbins 1991
fed quebracho tannin-
containing diets, sheep
showed a reduction in
digestibility while deer
did not
STBPs not found in tannin- Makkar and
adapted cattle (grazer) Becker 1998
Rhinoceros saliva Clauss et al. 1995

shows tannin-
binding capacity
that conforms
to feeding niche
except for inter-
mediate feeders
(continued)
97
Table 4.1 (continued) 98
Plant secondary High Low Bacterium (Streptococcus Brooker et al.
metabolite caprinus) capable of 1994
tolerance or degrading tannin-protein
detoxify- complexes described;
ing ability found in feral goats
of rumen grazing Acacia but
microbes not domestic goats.
Presumably its pres-
ence in the rumen was
a function of the diet
consumed
Found strains of tannin-tolerant bacteria Odenyo and Osuji
in the rumen of sheep, goats and ante- 1998
lope adapted to Acacia diets. Bacteria
were not cellulolytic. Since they were
found in the rumen of sheep (a grazer/
intermediate feeder), this suggests that
the presence of tannin-tolerant bacteria
in the rumen is a function of diet
Gas production (GP) Odenyo et al.1999
from rumen fluid
from a range
of ruminants
incubated with a
range of browse
species was
tested. Dik-dik
showed particu-
larly high GP but
other results were
less clear-cut. As
argued by Jones
et al. (2001), the high
dik dik GP could be
related to the high
DM of the rumen
fluid which was
not corrected for in
this study. Difficult
to separate out the
effects of diet and
animal species
Rumen fluid from African game species Jones et al.

and sheep was screened for its ability 2001
to degrade forages rich in tannins.
Game-species-derived rumen fluid
was found to be no better at degrad-
ing tannin-rich forages than sheep

rumen fluid. Addition of PEG consist-
ently improved protein degradation.
Authors argue that if browser rumen
fluid contains tannin-resistant
bacteria, then one would expect to see
similar in vitro nitrogen digestibility
(IVND) values in the presence and
absence of PEG. The fact that this
was not observed is taken as evidence
that tannin-tolerant bacteria are not
present in BR rumen fluid
99
4.3.1 Reticulo-Rumen Size
According to Hofmann, browsers would be expected to have a relatively smaller

reticulo-rumen than grazers since their strategy is to retain food in the rumen for a
relatively short period of time, relying less as they do on the energy derived from
cellulolysis of cell wall material, focussing instead on deriving nutrients from cell
contents. Van Wieren (1996) conducted a statistical analysis of Hofmann’s original
data (Hofmann 1973) and concluded that there was no statistically significant dif-
ference in reticulo-rumen size between grazers and browsers as classified by
Hofmann. Perez-Barberia et al. (2001) also carried out a multivariate analysis of
stomach morphology traits of Hofmann, controlling for the effects of body size and
phylogeny. Their study also supported the conclusion that browsers and grazers
cannot be separated morphologically once the effects of body size and phylogeny
have been taken into account. Interestingly, in a follow-up study Perez-Barberia
et al. (2004) found that even after accounting for phylogeny and body size, differences
in NDF digestibility between browsers and grazers were evident, suggesting that
the anatomical basis of digestive function has not yet been properly understood.
Clauss et al. (2003) conducted a meta-analysis of wet rumen contents using data
from a range of published sources. Rumen volume was regressed against body
weight, and a significantly larger intercept was found for grazers than browsers.
These authors assumed that gut volume was linearly related to body weight (Parra
1978), but, interestingly, their regression of wet rumen contents as a percentage of
body weight against body weight showed a positive slope, thus raising questions
about the quality of the data used to investigate these relationships.
4.3.2 Retention Time and Passage Rate
Browsers would be expected to pass material through their rumens relatively

quickly since soluble cell components are given up rapidly and cell wall digestion
is not a priority for them. There is no good quality data on passage rate in the exist-
ing literature. Gordon and Illius (1994) used published data to show that variation
in mean retention time of a range of ruminants could be wholly explained by body
size effects; values for browsers and grazers lay along a common regression line.
However, much of the base data was derived from a study in which retention times
were estimated using daily bulked samples (Foose 1982), which would yield rather
crude estimates of mean retention time. Furthermore, the diet types used in the
published studies were unclear. As pointed out by Clauss and Lechner-Doll (2001),
indirect anecdotal evidence of shorter retention time in browsing ruminants comes
from the composition of their body fat. The unsaturated lipid found in plant mate-
rial is saturated during passage through the rumen. Meat from browsing animals has
a higher unsaturated lipid content and some would speculate that this indicates
shorter rumen residence times.
Most wildlife studies are compromised because they do not strictly compare
animals on the same diets under the same feeding conditions, and there is an
inevitable difference in the feed type and physiological state of the animals.
The most controlled conditions arise in comparison of domestic ruminants such
as sheep, cattle, goats, and deer. The largest difference is between cattle and sheep;
under controlled conditions cattle have a longer retention time and slower pas-
sage rate of material out of the rumen, with which is associated a higher digest-
ibility (Hendricksen et al. 1981; Poppi et al. 1981). Despite the large difference
in live weight between the two species, the rumen fill of dry matter (DM),
expressed relative to metabolic live weight, was similar and could not explain
these differences. These studies showed that the difference in digestibility was
not related to any inherent difference in rate of digestion but rather to the rate
of passage from the rumen, which affected the time available for digestion and
hence the final extent of digestion. Most of the rumen particles were of a size
that had little resistance to escape, and yet rumen conditions, presumably aspects
of rumen volume and raft characteristics, affected the rate of passage. deVega and
Poppi (1997) showed by inserting common particles (either digested or undi-
gested of one size <1mm) in a range of rumens that the rumen conditions, as set
by the diet, was the most important factor affecting rate of passage rather than
the extent of digestion of the particle. It seems likely that for retention time,
passage rate, and possibly other digestion parameters, apparent grazer/browser
differences may be less to do with morpho-physiological differences than with
the nature of the diet consumed.
4.3.3 Mean Particle Size Escaping Rumen
According to the Hofmann view, browsers would be expected to be less efficient

at retaining material in their reticulo-rumen since they rely less on cellulolytic
digestion of fibre in the rumen. They would therefore be expected to allow larger
particles to escape the rumen than the grazers. This hypothesis is supported by
the work of Clauss and co-authors who propose selective retention of large parti-
cles in the rumen of grazers as being a driving force (Clauss et al. 2003) which
allowed the exploitation of graminaceous vegetation and radiation of ruminant
forms following the appearance of grasslands during the late Miocene. A meta-
analysis of ratios of solid to liquid phase markers in a range of ruminants indi-
cated browsers to have a lower ‘selectivity index’ than grazers, i.e., they were
less selective in their retention of solid material (Clauss and Lechner-Doll 2001).
In further work, comparison of mean particle size in the faeces of around 90 cap-
tive ruminants showed higher mean particle size in the faeces of browsers than
grazers after allowing for the effects of body size (Clauss et al. 2002). The effects
of diet were not controlled for and, as acknowledged by the authors, most of the
samples analysed would have been derived from animals offered alfalfa hay diets.
Clauss et al. (2003) go on to argue that the physico-mechanical characteristics

of browse and grass herbage have been key to the diversification of ruminants.
As previously described, grass material tends to form long, fibrous particles in
the rumen which form a fibrous raft floating on a liquid phase, while browse
material is more friable and breaks down into smaller, polygonal particles which
do not stratify to the same extent (see section 4.2.2, above). According to this
view, an important adaptation among grazing ruminants is therefore a stronger
rumen musculature to deal with the stratified rumen contents typical of animals
consuming grass-based diets. Browsing ruminants’ inability to subsist on pre-
dominantly grass diets is therefore attributed to their weak rumen musculature
which cannot deal with the rumen load resulting from grass diets (Clauss et al.
2002). In support of this idea Clauss et al. (2003) included data on rumen muscle
dimensions in their meta-analysis and showed that grazers have thicker ruminal
pillars for a given body weight than browsers.
4.3.4 Absorptive Surface Area of Rumen
Faster passage rates and smaller rumen volumes in browsers compared to

grazers would reduce the time available for absorption of nutrients and VFAs.
Hofmann asserts that browsing ruminants have adapted to this constraint by
increasing the surface area of the rumen (surface enlargement in Hofmann’s
terminology) to facilitate the faster absorption of VFAs that would be necessary
given the rapid fermentation rates found in browsers (Hofmann 1989). Van
Wieren (1996) reanalysed Hofmann’s data and showed that browsers do appear
to show more extensive papillation of the rumen wall than grazers of similar
body size. A similar conclusion was arrived at by Demment and Longhurst
(1987) using their own data derived from samples of East African ruminants
collected from shot animals. Extent of papillation varies within a species
depending on metabolisable energy intake or VFA production (Van Soest 1994)
and so experiments need to be carefully designed to avoid this confounding
factor. The biggest difference is seen between cattle on concentrate diets com-
pared to forage-based diets where the intakes are quite different and the papillae
density and size markedly different. The weight of rumen contents (as distinct
from surface area) appears to be directly proportional to liveweight (W1.0; Van
Soest 1994). Rumen volume varies with physiological state and as energy
demand is increased the rumen can increase in size up to a point to overcome
the physico-chemical limitations to increasing intake to meet this increased
energy demand. For example, in an experiment where rumen volume was com-
pared in barren cows and lactating cows suckling twins and fed the same for-
age, the volume of digesta and size of the rumen were both found to be higher
in the lactating cows (Tulloh 1966a, 1966b).
4.3.5 Saliva Flow
Saliva flow would be expected to be more rapid in browsers than grazers for two
reasons. Firstly, the strategy of passing material rapidly through the rumen to
maximise utilisation of nutrients derived from cell contents would be facilitated by
high rates of liquid flow through the rumen. Secondly, an important function of
saliva in the ruminant is the buffering of rumen contents. The rapid fermentation
rates envisaged for browsers would require rapid rates of saliva flow to buffer the
rapid production rates of VFAs. The larger salivary glands found in browsers than
in grazers by Robbins et al. (1995) provided indirect evidence for more copious
production of saliva in browsing compared with grazing ruminants. However,
Robbins et al. (1995) also used a meta-analysis of resting saliva flow rates in a
range of ruminants to show that saliva flow rates did not differ between grazers and
browsers once body size effects had been accounted for. The study was subse-
quently criticized for only measuring resting saliva flow rates when saliva flow
rates during food ingestion would have been more appropriate (Ditchkoff 2000).
Ditchkoff (2000) also suggested that critics of Hofmann had misunderstood his
hypotheses related to salivary flow. Hofmann suggests that browsers have a mecha-
nism for passing food past the rumen, directly into the abomasum via the ventricu-
lar groove thus avoiding the inefficiencies of rumen fermentation for readily
available substrates (Hofmann 1989). A similar mechanism is used by juvenile
ruminants to allow ruminal bypass of milk (Orskov et al. 1970), although in juve-
nile ruminants it is a conditioned reflex. Ditchkoff contends that conventional
measurements of passage rates do not take account of this mechanism and can-
not therefore be used to refute Hofmann’s ideas. However, since the passage
rates used in meta-analyses were whole tract values, the basis for Ditchkoff’s
criticism is unclear.
4.3.6 Fermentation Rate and Fibre Digestibility in the Rumen
The supposed browser strategy of focussing on cell contents (neutral detergent

solubles; NDS) to derive nutrients from plants would be accompanied by faster
fermentation rates since cell contents are more-or-less immediately available
sources of energy and protein, and would not require the slow cellulolysis for
extraction of energy from cell wall material. On the other hand, grazers would
be expected to be more efficient at extracting nutrients from cell wall material
and hence would be expected to show higher rates and extents of NDF diges-
tion. The meta-analysis of Gordon & Illius (1994) indicates that once body size
effects are taken into account, fermentation rates in the rumen of browsers and
grazers do not differ. Similarly, Robbins et al. (1995) did not detect differences
in digestibility of NDF between browsers and grazers. However, the opposite
conclusion was reached in a similar meta-analysis but using a different data set
by Iason Van Wieren (1999), and a similar study which accounted for the
effects of phylogeny reached a similar conclusion (Perez-Barberia et al. 2004).
There is, however, no strong empirical evidence to suggest substantial differ-
ences in extent of NDF digestion or digestion of cell contents among species.
Thus, all rumens irrespective of size or classification of animal (concentrate
feeder, grazer, etc) essentially digest NDF at the same rate; any differences in
digestion can be explained simply by differences in rate of passage and retention
time allowing greater or lesser time for digestion when species are compared.
Van Soest (1994) highlighted this by stating that the claims for differences in
digestibility are not supported by ‘rational biochemistry and kinetics’ and that
‘there are no magic enzymes and no magic rumen bacteria that can exceed
physico-chemical limitations’. The differences observed in rate or extent of DM
digestion are largely influenced by the diet selected by the animal. If the propor-
tion of NDF (and hence cell contents) varies because of diet selection, it follows
that rate and extent of digestion will also differ. Strict comparisons on the
same diet are not common and Van Soest’s conclusions would appear most
appropriate.
There is one circumstance relating to low CP diets in which differences
between species may appear. In these diets, rumen ammonia levels are low and
rate of fibre digestion may be reduced. This is the classical explanation for the
response of ruminants to urea (or non-protein N supplements) where rate of fibre
digestion and microbial protein production are increased in response to the limit-
ing nutrient, N. Several species comparisons have been done on the same low CP
diets whereby differences between species have been explained by differences in
N recycling to the rumen (Norton et al. 1979; Watson & Norton 1982; Alam et al.
1984). Species which have a higher salivary flow rate or maintain higher plasma
urea levels place themselves at a physiological advantage to enhance N recycling
to the rumen. Increasing N recycling to the rumen will be of advantage to the
species in times of low CP diets. The evidence is variable but it does indicate that
in times of acute shortage of N for the microbes in the rumen that any species
which can increase N recycling to the rumen will have an advantage (but still
within the classical physico-chemical response of N requirement by microbes for
digestion and growth).
For example, goats have been compared with sheep in many studies and they are
categorised differently by the Hofmann criteria. With high digestibility and CP
diets, there were no differences between the species in site of digestion, digestibility,
and use of nutrients (Alam et al. 1983, 1987a, 1987b). With low CP diets there was
an advantage to goats in many digestion parameters, one of which was maintenance
of a higher rumen ammonia N level (Watson and Norton 1982; Alam et al. 1984).
This appeared to occur in goats because of their lower water intake, higher plasma
urea concentration, and hence higher recycling of urea to the rumen (Alam et al.
1984, 1987b). The water conservation mechanism of goats enhanced the N conser-
vation mechanism leading to the maintenance of better N conditions within the
rumens of N deficient animals.
4.3.7 Rumen Microbial Genetic Profiles

and Digestion End Products
The end products of digestion in the rumen are volatile fatty acids (VFA), carbon
dioxide, methane, and microbial cells. VFA and microbial cells provide the bulk of
the nutrients absorbed by the host. In general, there are three main VFAs produced
by all anaerobic rumen fermentation: acetic acid, propionic acid, and butyric acid,
with acetic acid predominant under most forage feeding regimes. Smaller amounts
of short chain branch chain fatty acids are also found when there are large amounts of
protein degraded within the rumen. The changes in VFA proportions have been
extensively studied in ruminants, with propionic acid increasing in concentration as
starch is fermented and when legume forages are digested (Annison et al. 2002).
Assuming that browsers and concentrate feeders select diets high in temperate plant
species and/or high in cell content, then it may be implied that there would also
be a shift in VFA proportion from acetate to propionate. Clemens et al. (1983)
examined a wide range of East African wild ruminants ranging in weight from 5 or
6 kg (dik-dik) to 850 kg (African buffalo) and found that acetate as a proportion of
total VFA did not vary greatly; what variation occurred was related more to body
size than diet type. Domestic ruminants show a much greater range in VFA propor-
tion than occurred in this study, probably because the dietary intake of starch and
other soluble carbohydrates vary more widely for domestic ruminants than can be
achieved by animals that only graze. Clemens and Maloiy (1984) showed with the
same wide grouping of animals a wide variation in colonic function (absorption of
fluids, VFA concentration, etc.) but did not report VFA proportions. It is unlikely
that VFA proportions in the large intestine would vary unless large amounts of
starch escaped digestion, which is unlikely in animals consuming herbage.
Ishaque et al. (1971) in a little known study showed that a group of sheep on
a common diet could be separated into two groups based on level of propionate
and microbial protein production, with both being positively associated. The high
propionate, high microbial protein production group was suggested to have a dif-
ferent rumen microbial population. The methods to determine this were not avail-
able at that time but now the use of the 16S rDNA sequence has allowed the
microbial diversity within the rumen to be explored (Mackie et al. 2002). Some
studies have already shown that only a small percentage of the diversity of micro-
bial species has been described or cultured to date (Mackie et al. 2002). Larue
(2005) have outlined novel lines of Clostridium species associated with the
microbial sub-population tightly adhered to the fibre, and Tolosa et al. ( 2004)
have identified species which increase under high molasses-based diets. This
work is rapidly expanding and we will see more information on microbial ecol-
ogy under different species and diets, especially with respect to wild ruminants
in natural situations. For example, McEwan et al. (2005) have shown in, Soay sheep a
day-length-sensitive sheep species that, despite being fed the same diet, the
rumen microbial diversity varied with the day length under which the sheep were
housed. Mackie et al. (2003) and Sundset et al. (2004), studying reindeer in
Norway, showed that there was greater variation in the diversity of Oscillospira
sp. in wild reindeer than in cattle and sheep. They identified novel rumen bacteria
and showed differences in rumen microbial diversity between reindeer consum-
ing natural diets and formulated diets. The extent to which these differences
relate to diet, to genotype of the host animal, or simply to the interaction of indi-
viduals with sources of microbial inoculums has yet to be determined.
The same approach has been used to study kangaroos (herbivores with a rumen-
like forestomach) in Australia. Ouwerkerk et al. (2005a) identified a range of new
bacterial species in kangaroos using denaturing gradient gel electrophoresis
(DGGE), PCR, and the phylogenetic-tree-construction approach. Most interesting
is that they were able to explain the lack of methane production under anaerobic
fermentation in kangaroos by the discovery of a group of reductive acetogens in
these animals (Ouwerkerk et al. 2005b). Kangaroos had reductive acetogens and no
methanogens, while the reverse was true for sheep. Thus two different microbial
populations had evolved under anaerobic foregut fermentation in two distinct spe-
cies despite co-existence for around 200 years within Australia; no cross transfer of
these microbes appears to have occurred. It is, therefore, possible that browsers and
grazers may differ in their microbial populations, and this could have implications
for food digestion.
4.4 Problems with the Meta-Analysis Approach
From the foregoing it is clear that consensus has yet to be reached on whether
browsers and grazers show morpho-physiological adaptation to diet type or
whether variation in digestive parameters can be explained by body size effects.
The reality is probably somewhere between these two extremes (Iason and Van
Wieren 1999). Given the very different physical and chemical characteristics of
vegetation consumed by browsers and grazers, it would be surprising if different
ruminant classes had not developed physiological adaptations to their diet type.
Unfortunately, most of the evidence used in arguments for one or other position
is derived from meta-analyses which often rely on confounded data of questiona-
ble quality. Browsers tend to be smaller than grazers, and so defining relation-
ships between body size and digestive parameters can be problematic. Browsers
tend to eat browse while grazers tend to eat grass. It is, therefore, difficult to sepa-
rate the effects of diet type from the effects of morpho-physiological variation in
influencing the way in which ruminants handle their food. For example, faecal
particle size was found to be greater in the faeces of browsing compared with
grazing ruminants (Clauss et al. 2002) but these samples came from captive zoo
animals during winter when all animals would be consuming a predominantly
alfalfa hay diet. Furthermore, it is unclear what influence diet type has on the
fermentation characteristics used in the various meta-analyses used to support or
refute the Hofmann classification.
4.5 Detoxification of Plant Secondary Metabolites:

Do Browsers and Grazers Differ?
One of the major differences between the chemical characteristics of grass- and
browse-dominated diets is the presence of plant secondary compounds which are
prominent in the diets of browsers but largely absent from the diet of grazers. This
marked distinction in chemistry is thought to result from the different growth forms
of monocotylenous and dicotyledonous plants: monocots have a basal meristem so
that growth originates low in the plant. This makes them more resilient to offtake of
material by foraging herbivores. Dicots have an apical meristem and loss of tissue
to herbivory has more serious resource implications than for monocots. This leads to
strong selection pressure to defend tissue by chemical means and has led to the
diverse array of plant secondary compounds found in forbs, shrubs and trees
(Harborne 1988). Plant secondary compounds are diverse in both chemical structure
and physiological action on herbivores. Many attempts have been made to classify
plant secondary compounds according to their mode of action, and a useful and broad
division is into those compounds which reduce digestive efficiency in the gut and
those which act at the tissue level following absorption and circulation in the blood.
The most important digestibility reducers are the tannins which bind dietary and
endogenous protein leading to reduced protein digestibility via a direct effect on
dietary protein and an indirect effect through suppression of digestive enzyme activity
(Robbins et al. 1987a, 1987b). A range of other secondary compounds exert negative
effects on digestion through anti-microbial effects in the rumen and hind-gut (Wallace
2004). Plant secondary compounds which exert their effects post-absorptively can act
as either acute or chronic toxins. For example, cyanogenic glycosides yield free
cyanide following ingestion and this is an acute mammalian toxin (Majak 1992).
Pyrrolizidine alkaloids, on the other hand, cause the deposition of toxic pyrroles in
the liver leading to chronic toxicity (Mattocks 1986).
A range of physiological adaptations to the presence of plant secondary
metabolites in the diet have been proposed and some research has reported on
possible differences between browsers and grazers. This will now be reviewed.
4.5.1 Salivary Tannin-Binding Proteins
Salivary tannin-binding proteins were originally demonstrated in rats fed tannin-

containing sorghum diets (Mehansho et al. 1983). Sorghum-fed rats showed a
dramatic increase in salivary gland weight which was accompanied by increased
synthesis of a group of salivary glycoproteins with a strong affinity for tannins.
The high proportion of the amino acid proline in these salivary glycoproteins led
to them being described as ‘proline-rich proteins’ but the more functionally
descriptive term ‘salivary tannin-binding protein’ (STBP) is now widely used. Their
hypothesised mode of action is as sacrifice proteins to limit the digestion-reducing
effects of plant tannins. Because of their high affinity for tannin, STBPs are
hypothesised to spare dietary protein for utilisation further down the gut. The
presence of similar proteins in ruminant saliva has subsequently been reported
although the proportion of proline in the salivary mucoproteins of ruminants is
much lower than in rats (Austin et al. 1989). Furthermore, in ruminant herbivores,
STBPs appear to be constitutive and not to be inducible following exposure to
dietary protein.
The role of salivary tannin-binding proteins has been reviewed by McArthur
et al. (1995) who suggest that their ancestral role was in the maintenance of oral
homeostasis but that in mammals with a high concentration of dietary tannins they
have evolved to protect animals against the detrimental effects of plant tannin.
There is certainly evidence for the occurrence of STBPs in the saliva of browsers
and their absence in grazing herbivores; in one study salivary tannin-binding
proteins were found in saliva of mule deer (Odocoileus hemionus hemionus), a
browser, but not in the saliva of sheep and cattle, grazers (Austin et al. 1989).
Further indirect evidence for a browser/grazer difference came from a study in
which sheep and deer were fed quebracho tannin-containing diets; sheep showed a
reduction in digestibility while deer did not (Robbins et al. 1991). STBPs were not
found in the saliva of cattle even following a period of adaptation to a tannin-rich
diet (Makkar and Becker 1998). Salivary tannin-binding proteins are also found in
non-ruminant herbivores such as rhinoceroses. Salivary tannin-binding capacity in the
saliva of captive rhinoceros species appears to conform to dietary feeding niche, at
least to some extent; Black rhinoceroses (Diceros bicornis, browser) have higher
salivary tannin binding capacity than White rhinoceroses (Ceratotherum simum,
grazer) although Indian rhinoceroses (Rhinoceros unicornis, intermediate feeder)
appear to have even higher tannin binding capacity than the Black species in saliva
(Clauss et al. 2005). Although evidence supports the idea that STBPs are an impor-
tant physiological adaptation to browse diets, tannins are only one group of plant
secondary compounds and there is no evidence that the saliva of browsing animals
can neutralise the many other secondary compounds present in browse diets.
Histatins have also been shown to precipitate tannins, and to a greater extent than
STBPs. They are a group of peptides found in human saliva but not studied in
ruminants (Naurato et al. 1999).
The binding of tannins with protein can have both detrimental and beneficial
effects. At low levels of condensed tannin, protein from plant material and saliva is
bound most probably during mastication but also within the rumen contents. This
protein is protected from degradation and increases the supply of protein to the
small intestine where it is released and absorbed. However, high levels of condensed
tannin may lead to over-protection and detrimental effects on fibre and protein
digestion within the animal. In these cases the infusion of PEG experimentally is
able to determine the effect of condensed tannins on the host by binding to and
reversing the effects of condensed tannin. The role of STBPs in these circumstances
of high condensed tannin will most likely be beneficial. These issues have been
extensively reviewed in Foley et al. (1999). They identified a range of herbivore
species where STBPs have a role in binding the condensed tannin, enabling
species to deal with variable and high levels of condensed tannin in their diet.
Perez-Maldonado and Norton (1996a, 1996b) clearly showed that no difference
exists between goats and sheep in binding and metabolising condensed tannins in
tropical legumes with high levels of condensed tannin. They also could find no
evidence for STBPs in goats (B.W. Norton, pers. comm.).
4.5.2 Ruminal Detoxification
The anterior position of the microbial fermentation chamber in the ruminant

digestive tract increases the potential for microbial biotransformation of plant
secondary compounds in ruminants—as compared to herbivores occupying similar
feeding niches but relying on hind-gut fermentation. The rumen is a rich source of
enzymatic activity and many of the biotransformations that occur under the action
of mammalian detoxification enzymes can occur readily within the rumen
environment. The role of the rumen in microbial detoxification of plant secondary
metabolites has been well reviewed (Carlson and Breeze 1984). Classic examples
of the role of the rumen micro-flora in reducing toxicity of plants consumed by
ruminant herbivores include the degradation of oxalates found in Halogeton
glomeratus (Allison and Reddy 1984), mimosine found in Leucaena leucocephala
(Jones 1981), glucosinolate breakdown products found in Brassica species (Duncan
and Milne 1992) and pyrrolizidine alkaloids found in Senecio species (Lanigan
1970). Mammalian metabolism of plant secondary compounds can result in activa-
tion rather than detoxification and a similar phenomenon can occur in the rumen.
For example, S-methyl cysteine sulphoxide found in cruciferous species is trans-
formed by rumen bacteria into the toxic principle dimethyl disulphide, which is
responsible for the haemolytic anaemia observed in farm livestock fed on brassica
material (Smith 1980).
Although the role of the rumen in metabolism of plant secondary compounds is
not in doubt, there is some question over whether the rumen microbial populations of
different ruminant species have markedly different capacities to degrade plant sec-
ondary compounds. This uncertainty also applies to the question of whether browsers
and grazers differ at the rumen level in their capacity to degrade plant secondary
compounds. The extent to which the host animal can manipulate the composition of
the rumen microbial population is unclear and this would be a pre-requisite for
browser/grazer differences. The classic experiments of Jones and colleagues with
mimosine-degrading bacteria demonstrated that transfer of bacteria from one rumi-
nant population to another could also transfer tolerance to an otherwise detrimental
plant toxin (Jones 1981). This work was influential and suggested many possibilities
for using rumen microbial communities for protection of host animals from toxicity
in the agricultural context. However, manipulating rumen microbial communities for
particular ends has proved more challenging than first supposed, and attempts to
transfer tolerance to plant toxins using rumen inoculae have not been convincing
(Brooker et al. 1994; Miller et al. 1995, 1996). Since rumen microbial populations are
largely controlled by substrate availability, mechanisms by which host animals could

vary their resident rumen microbial population, other than by altering the diet
selected, are unclear. Genotypic differences in digestive tract function in ruminants
have been recently reviewed (Hegarty 2004) with the conclusion that the main diges-
tive function trait under genetic control is mean retention time. Factors such as rate
of saliva flow, the morphology of the reticulo-omasal orifice, and the degree of urea
recycling to the rumen (Hungate et al. 1960) could all contribute to variation in mean
retention time, and are all factors which could vary genetically.
There has been particular interest, in the last decade, in isolating bacterial strains
which are capable of degrading tannin, or are at least tolerant of tannin. Such strains had
been isolated (Brooker et al. 1994; Odenyo and Osuji 1998) but they have not been
found to be necessarily host-animal specific. In a study by Odenyo et al. (1999) gas
production from rumen fluid from a range of ruminants incubated with a range of
browse species was measured. Rumen fluid from dik-dik showed particularly high gas
production values but results for other species were less clear-cut. The study design
made it difficult to distinguish the effects of diet and animal species. Jones et al. (2001)
screened rumen fluid from various African game species for its ability to degrade test
forages rich in tannins. Results were compared with values found with sheep. It was
shown that game-species-derived rumen fluid was no better at degrading tannin-rich
forages than sheep rumen fluid. Addition of PEG consistently improved protein degra-
dation. The authors argue that if browser rumen fluid contains tannin-resistant bacteria
then one would expect to see similar in vitro nitrogen digestibility values in the presence
and absence of PEG. The fact that this was not observed was taken as evidence that
tannin-tolerant bacteria are not present in browser rumen fluid.
Mackie et al. (2002) collated a list of known secondary compounds which were
metabolised by specific microbial species. The compounds were non-protein
amino acids, phenolics, trihydrobenzenoids, ferulic and p-coumaric acid, con-
densed and hydrolysable tannins, phyto-oestrogens, oxalate, pyrrolizidine alka-
loids, and mycotoxins. Whilst these compounds may be metabolised, there is no
guarantee that the toxic effects are completely removed as the action depends on level
and extent of exposure. Whether the populations of these secondary compound-
degrading bacteria differ between grazers and browsers has not been investigated.
In conclusion, the question of whether the rumen microbial populations of browsers
and grazers show different capacities to degrade plant secondary compounds, thus
conferring differential tolerance to toxicity on their hosts, has not yet been resolved.
Intuitively, one would expect such genotypic differences to be more readily expressed
at the level of the detoxification mechanisms of the host itself, rather than indirectly
through the rumen microbial population.
4.5.3 Post Absorptive Metabolism
Following ingestion and digestion of plant material, nutrients and toxins are
absorbed from the digestive tract. Biotransformation in the rumen may lead to
detoxification of plant secondary metabolites but this is not always the case; some
PSMs may escape detoxification while others may increase in toxicity as a result of
microbial action as described above. PSMs absorbed from the digestive tract may
be detoxified in the tissues, primarily in the liver. Most of our understanding of
post-absorptive metabolism of xenobiotics comes from the discipline of pharmacology,
and most research in the area is focussed on the metabolism of artificial drugs in
the context of human and veterinary medicine. As a result, most of the animal
models that have been used in this area of research are either laboratory rodents or
domestic livestock. Very little information on tissue-level detoxification is available
for wildlife species. Watkins and Klaassen (1986) used probe substrates to quantify
cytochrome P450 activity in a range of species including fish and birds as well as
domestic livestock. Vast differences in apparent enzyme activity were reported
which did not follow obvious phylogenetic lines. The study highlighted the com-
plexity of hepatic biotransformation of xenobiotics, the multiple enzymes involved,
the overlap in substrate specificity, and the difficulties in using probe substrates
developed in laboratory rats for quantifying activity of enzymes in other species.
Our understanding of the P450 superfamily of enzymes has increased considerably
since the Watkins and Klaassen (1986) study with the development of molecular
and immunological methods of quantifying hepatic enzyme activity. There is a need
to use some of this information for cross-species comparisons in the context of
browser-grazer differences, but this has not yet been reported.
Comparisons to date have been at the level of gross anatomy. A general observa-
tion is that concentrate feeders have a larger liver than grazers (Van Soest 1994;
Foley et al. 1999) and that this confers an advantage to them in the further detoxifi-
cation of plant secondary compounds or their derivatives after gastro-intestinal tract
degradation. They are also more likely to be exposed to such compounds given their
feeding habit. Goats metabolise anthelmintic drugs faster than sheep, indicating
some differences in metabolic activity (Hennessy et al. 1993). In the case of the
latter study there was no difference between goats and sheep in metabolism of con-
densed tannins and their degradation products and appearance in urine.
4.6 Conclusions
Plant material is variable in structural characteristics and hence in nutritive value.

Plant secondary compounds also show marked diversity across the plant kingdom.
Browsers and grazers thus subsist on very different diets. Despite this variation, our
discussion has demonstrated that there is still much that we do not understand about
how herbivores have adapted nutritionally and physiologically to this variation in
diet. Much of our understanding of ruminant nutrition derives from the agricultural
context and has focussed on a very few species which do not adequately represent
the browser-grazer continuum. What is now needed is to apply some of the
methodology derived from several decades of agricultural research to conduct con-
trolled experiments aimed at understanding the ways in which a range of wild her-
bivore species deal with the varied plant sources that form their diets in natural situations.
Such research will allow a better understanding of the impact of wild herbivores on
vegetation in natural habitats and will facilitate the more effective management of
vegetation using herbivores as tools for land management.
In this chapter we have considered the various levels at which food processing
by browsers and grazers might differ. As we have seen, the research community
has, so far, given most attention to possible differences in digestive anatomy and
physiology and how this might differ between grazers and browsers. There has been
much debate in the literature but no clear consensus has emerged about whether
grazers and browsers show marked morpho-physiological differences which might
influence their digestive efficiency. Considerable recent attention has also been
given to the question of whether rumen microbial degradation of plant secondary
metabolites is more effective in browsers than grazers. The balance of evidence
suggests no great differences at the rumen level. In contrast, very scant attention has
been paid to possible browser-grazer differences in post-absorptive metabolism of
plant secondary metabolites. It is perhaps at this level that we might expect physio-
logical differences between browsers and grazers to be most strongly expressed.
since species-level genetic variation in Phase 1 and Phase 2 enzyme activity has
been shown to be extensive in biomedical studies. Despite the difficulties inherent
in making measurements of tissue enzyme activity in wildlife, we suggest that this
would be an interesting avenue for future work.
References
Alam MR, Poppi DP, Sykes AR (1983) Intake, digestibility and retention time of two forages by
kids and lambs. Proc New Zeal Soc An 43:119–121
Alam MR, Borens F, Poppi DP, Sykes AR (1984) Comparative digestion in sheep and goats. In:
Barker SK, Gawthorne JB, Mackintosh JB, Purser DB (eds) Ruminant physiology – concepts
and consequences. Proceedings Symposium, University of Western Australia, Perth, p 184
Alam MR, Lawson GD, Poppi DP, Sykes AR (1987a) Comparison of the site and extent of diges-
tion of nutrients of a forage in kids and lambs. J Agr Sci 109:583–589
Alam MR, Poppi DP, Sykes AR (1987b) Comparative aspects of water-intake and its flow through
the gastrointestinal-tract of kids and lambs. J Agr Sci 108:253–256
Allison MJ, Reddy CA (1984) Adaptations of gastrointestinal bacteria in response to changes in
dietary oxalate and nitrate. In: Klug MJ, Reddy CA (eds) Current perspectives in microbial
ecology. American Society of Microbiology, Washington, DC, pp 248–256
Annison, EF, Lindsay, DB and Nolan, JV (2002). Digestion and metabolism, In:Freer M, Dove H
(eds) Sheep nutrition. CABI/CSIRO, Wallingford New York, pp 95–118
Austin PJ, Suchar LA, Robbins CT, Hagerman AE (1989) Tannin-binding proteins in saliva of
deer and their absence in saliva of sheep and cattle. J Chem Ecol 15:1335–1347
Bell RHV (1971) Grazing ecosystem in Serengeti. Sci Am 225:86–93
Brooker JD, O’Donovan LA, Skene I, Clarke K, Blackall L, Muslera P (1994) Streptococcus caprinus
sp.nov., a tannin-resistant ruminal bacterium from feral goats. Lett Appl Microbiol 18:313–318
Carlson JR, Breeze RG (1984) Ruminal metabolism of plant toxins with emphasis on indolic
compounds. J Anim Sci 58:1040–1049
Clauss M, Lechner-Doll M (2001) Differences in selective reticulo-ruminal particle retention as a
key factor in ruminant diversification. Oecologia 129:321–327
Clauss M, Lechner-Doll M, Streich WJ (2002) Faecal particle size distribution in captive wild
ruminants: an approach to the browser/grazer dichotomy from the other end. Oecologia
131:343–349
Clauss M, Lechner-Doll M, Streich WJ (2003) Ruminant diversification as an adaptation to the
physicomechanical characteristics of forage. A reevaluation of an old debate and a new
hypothesis. Oikos 102:253–262
Clauss M, Gehrke J, Hatt JM, Dierenfeld ES, Flach EJ, Hermes R, Castell J, Streich WJ, Fickel J
(2005) Tannin-binding salivary proteins in three captive rhinoceros species. Comp Biochem
Phys A 140:67–72
Clemens ET, Maloiy GMO (1984) Colonic absorption and secretion of fluids, electrolytes and
organic-acids in East-African wild ruminants. Comp Biochem Phys A 77:51–56
Clemens ET, Maloiy GMO, Sutton JD (1983) Molar proportions of volatile fatty-acids in the
gastrointestinal-tract of East-African wild ruminants. Comp Biochem Phys A 76:217–224
Demment MW, Longhurst WM (1987) Browsers and grazers: constraints on feeding ecology
imposed by gut morphology and body size. In: Santana OP, Da Silva AG, Foote WC (eds)
Proceedings of the IVth International Conference on Goats, Symposia, vol 2, March 8–13,
Brasilia, Brazil, pp 989–1004
deVega A, Poppi DP (1997) Extent of digestion and rumen condition as factors affecting passage
of liquid and digesta particles in sheep. J Agr Sci 128:207–215
Ditchkoff SS (2000) A decade since “diversification of ruminants”: has our knowledge improved?
Duncan AJ, Milne JA (1992) Rumen microbial degradation of allyl cyanide as a possible expla-
nation for the tolerance of sheep to brassica-derived glucosinolates. J Sci Food Agr
58:15–19
Foley WJ, Iason GR, McArthur C (1999) Role of plant secondary metabolites in the nutritional
ecology of mammalian herbivores: how far have we come in 25 years? Pages In: Jung H-JG,
Fahey GC (eds) Nutritional ecology of herbivores. Proceedings Vth International Symposium
Nutrition of Herbivores, Am Soc Anim Sci, Savoy, IL, pp 130–209
Foose TM (1982) Trophic strategies of ruminant versus non-ruminant herbivores. Thesis.
University of Chicago
Gordon IJ, Illius AW (1994) The functional significance of the browser-grazer dichotomy in
Gordon IJ, Illius AW (1996) The nutritional ecology of African ruminants - a reinterpretation.
J Anim Ecol 65:18–28
Harborne JB (1988) Introduction to ecological biochemistry. Academic Press, London
Hegarty RS (2004) Genotype differences and their impact on digestive tract function of ruminants:
a review. Aust J Exp Agr 44:458–467
Hennessy, DR (1993) Pharmokinetic disposition of benzimidazole drugs in the ruminant gastroin-
testinal tract. Parasitol Today 9:329–333
Hendricksen RE, Poppi DP, Minson DJ (1981) The voluntary intake, digestibility and retention
time by cattle and sheep of stem and leaf fractions of a tropical legume (Lablab purpureus).
Hofmann RR (1973) The ruminant stomach: stomach structure and feeding habits of East African
game ruminants. East African Literature Bureau, Nairobi
ruminants: a comparative view of their digestive system. Oecologia 78:443
Hofmann RR, Stewart DRM (1972) Grazer or browser: a classification based on the stomach
structure and feeding habits of East African ruminants. Mammalia 36:226–240
Hungate RE (1966) The rumen and its microbes. Academic Press, New York
Hungate RE, Phillips GD, Hungate DP, MacGregor A (1960) A comparison of rumen fermentation
in European and Zebu cattle. J Agr Sci 54:196–201
Iason GR, Van Wieren SE (1999) Digestive and ingestive adaptations of mammalian herbivores to
low-quality forage. In: Olff H, Brown VK, Drent RH (eds) Herbivores: between plants and
predators. Blackwell, Oxford, pp 337–369
Ishaque M, Thomas PC, Rook JAF (1971) Consequences to host of changes in rumen microbial
activity. Nature–New Biol 231:253–256
Jarman PJ (1974) The social organisation of antelope in relation to their ecology. Behaviour
48:215–266
Jones RJ (1981) Does ruminal metabolism of mimosine explain the absence of Leucaena toxicity
in Hawaii? Aust Vet J 57:55–56
Jones RJ, Meyer JHF, Bechaz FM, Stoltz MA, Palmer B, Van der Merwe G (2001) Comparison
of rumen fluid from South African game species and from sheep to digest tanniniferous
browse. Aust J Agr Res 52:453–460
Jones WT, Mangan JL (1977) Complexes of condensed tannins of sainfoin (Onobrychis viccifolia
Scop.) with fraction 1 leaf protein and submaxillary mucoprotein, and their reversal by
polyethylene glycol and pH. J Sci Food Agr 28:126–136
Lanigan GW (1970) Metabolism of pyrrolizidine alkaloids in the ovine rumen II. Some factors
affecting rate of alkaloid breakdown by rumen fluid in vitro. Aust J Agr Res 21:633–639
Larue R, Yu ZT, Parisi VA, Egan AR, Morrison M (2005) Novel microbial diversity adherent to
plant biomass in the herbivore gastrointestinal tract, as revealed by ribosomal intergenic spacer
analysis and rrs gene sequencing. Environ Microbiol 7:530–543
Long RJ, Apori SO, Castro FB, Orskov ER (1999) Feed value of native forages of the Tibetan
Plateau of China. Anim Feed Sci Tech 80:101–113
Mackie RI, McSweeney CS, Klieve AV (2002) Microbial ecology of the ovine rumen. In: Freer
M, Dove H (eds) Sheep nutrition. CABI/CSIRO, Wallingford/New York, pp 71–94
Mackie RI, Aminov RI, Hu WP, Klieve AV, Ouwerkerk D, Sundset MA, Kamagata Y (2003)
Ecology of uncultivated Oscillospira species in the rumen of cattle, sheep, and reindeer as
assessed by microscopy and molecular approaches. Appl Environ Microbiol 69:6808–6815
Majak W (1992) Metabolism and absorption of toxic glycosides by ruminants J Range Manage
45:67–71
Makkar HPS, Becker K (1998) Adaptation of cattle to tannins: role of proline-rich proteins in
oak-fed cattle. Anim Sci 67:277–281
Mattocks AR (1986) Chemistry and toxicology of pyrrolizidine alkaloids. Academic Press, San
Diego
McArthur C, Sanson GD, Beal AM (1995) Salivary proline-rich proteins in mammals: roles in
oral homeostasis and counteracting dietary tannin. J Chem Ecol 21:663–691
McEwan NR, Abecia L, Regensbogenova M, Adam CL, Findlay PA, Newbold CJ (2005) Rumen
microbial population dynamics in response to photoperiod. Lett Appl Microbiol 41:97–101
Mehansho H, Hagerman A, Clements S, Butler L, Rogler J, Carlson DM (1983) Modulation of
proline-rich protein biosynthesis in rat parotid glands by sorghums with high tannin levels.
Proc Nat Acad Sci 80:3948–3952
Miller SM, Brooker JD, Blackall LL (1995) A feral goat rumen fluid inoculum improves nitrogen-
retention in sheep consuming a mulga (Acacia aneura) diet. Aust J Agr Res 46:1545–1553
Miller SM, Brooker JD, Phillips A, Blackall LL (1996) Streptococcus caprinus is ineffective as a
rumen inoculum to improve digestion of mulga (Acacia aneura) by sheep. Aust J Agr Res
47:1323–1331
Mould FL (2003) Predicting feed quality - chemical analysis and in vitro evaluation. Field Crop
Res 84:31–44
Naurato N, Wong P, Lu Y, Wroblewski K, Bennick A (1999) Interaction of tannin with human
salivary histatins. J Agr Food Chem 47:2229–2234
Norton BW, Moran JB, Nolan JV (1979) Nitrogen metabolism in Brahman cross, buffalo, Banteng
and shorthorn steers fed on low-quality roughage. Aust J Agr Res 30:341–351
Odenyo AA, Osuji PO (1998) Tannin-tolerant ruminal bacteria from East African ruminants. Can
J Microbiol 44:905–909
Odenyo AA, McSweeney CS, Palmer B, Negassa D, Osuji PO (1999) In vitro screening of rumen
fluid samples from indigenous African ruminants provides evidence for rumen fluid with
superior capacities to digest tannin-rich fodders Aust J Agr Res 50:1147–1157
Oh HK, Sakai T, Jones MB, Longhurst WM (1967) Effect of various essential oils isolated from
Douglas Fir needles upon sheep and deer rumen microbial activity. Appl Microbiol
15:777–784
Orskov ER, Benzie D, Kay RNB (1970) The effect of feeding procedure on closure of the
oesophageal groove in sheep. Brit J Nutr 24:785–795
Ouwerkerk D, Klieve AV, Forster RJ, Templeton JM, Maguire AJ (2005a Characterization of cul-
turable anaerobic bacteria from the forestomach of an eastern grey kangaroo, Macropus gigan-
teus. Lett Appl Microbiol 41:327–333
Ouwerkerk D, Maguire AJ, Klieve AV (2005b) Reductive acetogenesis in the foregut of macropod
marsupials in Australia. In: Soliva CR, Takahashi J, Kreuzer M (eds) Publication Series, vol
27. Institute of Animal Science, EFTH, Zurich, pp 98–101
Parra, R (1978) Comparison of foregut and hindgut fermentation in herbivores. In: Montgomery
GG (ed) The ecology of arboreal folivores. Smithsonian Institute, Washington, DC, pp
205–230
Perez-Barberia FJ, Gordon IJ, Illius AW (2001) Phylogenetic analysis of stomach adaptation in
digestive strategies in African ruminants. Oecologia 129:498–508
Perez-Barberia FJ, Elston DA, Gordon IJ, Illius AW (2004) The evolution of phylogenetic differ-
ences in the efficiency of digestion in ruminants. Proc R Soc Lond B Bio 271:1081–1090
Perez-Maldonado RA, Norton BW (1996a) The effects of condensed tannins from Desmodium
intortum and Calliandra calothyrsus on protein and carbohydrate digestion in sheep and goats.
Brit J Nutr 76:515–533
Perez-Maldonado RA, Norton BW (1996b) Digestion of c-14-labeled condensed tannins from
Desmodium intortum in sheep and goats. Brit J Nutr 76:501–513
Poppi DP, Minson DJ, Ternouth JH (1981) Studies of cattle and sheep eating leaf and stem frac-
tions of grasses. 1. The voluntary intake, digestibility and retention time in the reticulo-rumen.
Robbins CT, Hagerman AE, Hjelijord O, Baker DL (1987a) Role of tannins in defending plants
against ruminants: reduction in protein availability. Ecology 68:98–107
Robbins CT, Mole S, Hagerman AE, Hanley TA (1987b) Role of tannins in defending plants
against ruminants: reduction in dry matter digestion. Ecology 68:1606–1615
Robbins CT, Hagerman AE, Austin PJ, McArthur C, Hanley TA (1991) Variation in mammalian
physiological responses to a condensed tannin and its ecological implications J Mammal
72:480–486
Robbins CT, Spalinger DE, Vanhoven W (1995) Adaptation of ruminants to browse and grass
diets - are anatomical-based browser-grazer interpretations valid? Oecologia 103:208–213
Rosenthal GA, Janzen DH (1979) Herbivores: their interaction with secondary plant metabolites.
Academic Press, New York
Smith RH (1980) Kale poisoning: the brassica anaemia factor. Vet Rec 107:12–15
Spalinger DE, Robbins CT, Hanley TA (1993) Adaptive rumen function in elk (Cervus elaphus
nelsoni) and mule deer (Odocoileus hemionus hemionus). Can J Zool 71:601–610
Stegelmeier BL, Edgar JA, Colegate SM, Gardner DR, Schoch TK, Coulombe RA, Molyneux RJ
(1999) Pyrrolizidine alkaloid plants, metabolism and toxicity. J Nat Toxins 8:95–116
Sundset MA, Cann IOK, Mathiesen SD, Mackie RI (2004) Rumen microbial ecology in reindeer
- adaptations to a unique diet. J Anim Feed Sci 13:717–720
Theodorou MK, Gascoyne DJ, Akin DE, Hartley RD (1987) Effect of phenolic acids and pheno-
lics from plant cell walls on rumen-like fermentation in consecutive batch culture. Appl
Environ Microbiol 53:1046–1050
Thornton RF, Minson DJ (1973) Relationship between apparent retention time in rumen, volun-
tary intake, and apparent digestibility of legume and grass diets in sheep. Aust J Agr Res
24:889–898
Tolosa MX, Dinh TV, Klieve AV, Ouwerkerk D, Poppi DP, McLennan SR (2004) Molecular char-
acterisation of rumen bacterial populations in cattle fed molasses diets. Anim Prod Aust
25:328
Tulloh NM (1966a) Physical studies of the alimentary tract of grazing cattle. IV. Dimensions of
the tract in lactating and non-lactating cows. New Zeal J Agr Res 9:999–1008
Tulloh NM (1966b) Physical studies of the alimentary tract of grazing cattle. III. Seasonal changes
in capacity of the reticulo-rumen of dairy cattle. New Zeal J Agr Res 9:252–260
Van Soest PJ (1982) Nutritional ecology of the ruminant. O&B Books, Corvallis, OR
Van Soest PJ (1994) Nutritional ecology of the ruminant. Cornell University Press, Ithaca, NY
Van Wieren SE (1996) Digestive strategies in ruminants and non-ruminants. University of
Wageningen, Netherlands
Wallace RJ (2004) Antimicrobial properties of plant secondary metabolites. P Nutr Soc
63:621–629
Watkins JBI, Klaassen CD (1986) Xenobiotic biotransformation in livestock: comparison to other
species commonly used in toxicity testing. J Anim Sci 63:933–942
Watson C, Norton BW (1982) The utilisation of pangola grass hay by sheep and Angora goats.
P Aust Soc Anim Prod 41:467–470
Zucker WV (1983) Tannins: does structure determine function? An ecological perspective. Am
Nat 121:335–365
Chapter 5
The Comparative Feeding Behaviour of Large
Browsing and Grazing Herbivores
Kate R. Searle and Lisa A. Shipley
5.1 Introduction
Herbivores exploit a food resource that is fundamentally different from that of most
other trophic levels. Their food exists in an apparent surplus, food items are rarely
eaten in entirety, and the concepts of pursuit and catchability are irrelevant (Owen-
Smith and Novellie 1982; Spalinger & Hobbs 1992a). Plant foliage is generally of
low nutritive value so foraging time and digestive time are important constraints.
Ungulate herbivores can be broadly classified into two groups—those that feed
mainly on grass, and those than feed mainly on browse. This division is important
in relation to the both the way in which an animal forages, and the definition of
what constitutes a patch, because grass and browse material present themselves in
very different ways to the foraging herbivore, and because grasses differ from
browses in architecture and physical structure (Table 5.1).
5.2 The Functional Response
An herbivore’s foraging behaviour derives from the interaction between

characteristics of the herbivore and characteristics of the plants on which it
feeds. Intake rate is the currency through which an herbivore measures the out-
comes of different foraging strategies (Stephens and Krebs 1986), with an aim
to maximising energy intake rate over short time scales (i.e., minutes, hours).
At longer time scales (days) other currencies and constraints come to influence
the foraging process, such as digestive constraints, gut capacity (Provenza
1995; Hirakawa 1997; Iason et al. 1999; Villalba and Provenza 1999, 2000, 2005;
Launchbaugh et al. 2001), and predation risk (Brown 1988, 1999; Beecham and
Farnsworth 1999; Abrams and Schmitz 1999; Frair et al. 2005). Hence, over the
course of a day an herbivore may simultaneously choose short-term strategies
that maximise instantaneous intake rate (e.g., selection of bites within a feeding
© Springer 2008
118 K.R. Searle and L.A. Shipley
Table 5.1 A comparison of general chemical and structural characteristics of grasses (monocots)
and browses (herbaceous and woody dicots)
Characteristic Grasses Browses
Cell wall Thick Thin
Greater proportion Greater proportion lignin
cellulose/hemicellulose
Plant defences Silica Thorns and spines,
secondary chemicals
Plant architecture Fine-scaled heterogeneity Coarse-scaled heterogeneity
within plant within plant
Meristem at base Meristem at tips
Low growth form Low to high growth from
Compact Complex, diffuse,
branching architecture
Dispersion Uniform Dispersed
station), whilst employing more long-term strategies that seek to minimise time
spent foraging (e.g., choosing patches within a habitat; Bergman et al. 2001).
These decisions occur very frequently and form the grain of the foraging
hierarchy, representing an important underlying base to higher level behaviours.
First, we will examine important components that control the relationship
between plant characteristics and intake rate, and examine the short-term decisions
herbivores make in relation to maximising instantaneous intake rate. Because her-
bivores can be classified into two general groups, browsers and grazers, by the main
components of their diets, we will examine how these foraging behaviours vary
between these groups.
An animal’s intake rate is typically expressed using the functional response,
which describes the relationship between intake rate and some measure of plant
abundance (e.g., biomass, sward height, density, plant mass, bite mass). The
concept of the functional response was first developed for predators (disk equation;
Holling 1959, 1965), though it has since been adapted for herbivores (Spalinger
Hobbs 1992a, 1992b; Gross et al. 1993; Beckerman 2005). As such, it is a key
concept in herbivore foraging behaviour, linking individual behaviour to higher
level processes such as plant–herbivore interactions and population dynamics.
Dynamics of the functional response provide the impetus for herbivore movement—
changes to intake rate motivate movements in search of new food sources. The
functional response is therefore a key component in the reciprocal interactions of
herbivores and the spatio-temporal patterns of their food resources, which ultimately
determine important ecosystem functions such as nutrient cycling, fire regimes, and
sediment and water transfer (Fig. 5.1).
The seminal paper of Spalinger and Hobbs (1992a) provided a mechanistic
interpretation of the way several key processes shape and control the functional
response of herbivores foraging within patches (Eq. 1). For herbivores feeding on
spatially concentrated foods in which foods are apparent and spaced such that
animals can reach the next bite before completely chewing the previous (Process
5 The Comparative Feeding Behaviour of Large Browsing and Grazing Herbivores 119
Nutrient dynamics Fire Water and

sediment transfer
Plant-herbivore Population Spatial

interactions dynamics heterogeneity of
Numerical landscape
response
Movement Individual
patterns foraging
success
Intake
rate
Functional
response
Bite mass Bite interval
Bite volume Bite depth Cropping Chew rate

time
Plant size, architecture and fibrousness Animal size, morphology and dentition
Fig. 5.1 Conceptual model demonstrating linkages between plant characteristics, intake rate and
behaviour, and higher order population dynamics and ecosystem function
III; Spalinger and Hobbs 1992a), such as a grass sward or browse patch,
instantaneous intake rate (I, g min−1) is represented as an asymptotic function of bite
size (S, g):
Rmax S
I= (Eq. 1)
Rmax h + S
where Rmax (g min−1) is the maximum rate of processing of plant tissue in the
mouth that would occur in the absence of cropping, and h (min) is the average
time required to crop a single bite in the absence of chewing. This model rests
upon the assumption that intake rate is controlled by the competition between
cropping and chewing leading to a Type II functional response, just as competi-
tion between searching for and handling prey items creates a Type II functional
response in Holling’s (1959) original disk equation.
When foods are apparent and dispersed in space such that previous bites are
fully consumed before encountering the next (Process II; Spalinger and Hobbs
1992a), variables describing handling time (h and Rmax) drop out of the model and
variables describing rate of encounter with plants (Vmax, m/min, travel velocity in
the absence of cropping, D, bites/m2, bite density, and δ, m/bite, decrease in velocity
caused by cropping bites) are included in the model:
Vmax D
I= ⋅S
1+ d D (Eq. 2)
Finally, when bites are dispersed, but are cryptic, animals must search to encounter
bites (Process I; Spalinger and Hobbs 1992a), thus W (m), the width of the search
path or the detection limits imposed by vision or smell are included in the model:
DVmaxW
I= ⋅S
1 + dWD (Eq. 3)
5.2.1 Components of the Functional Response
Because herbivores experience selective pressures for obtaining food rapidly while
foraging (Stephens and Krebs 1986), coevolutionary processes between plants and
herbivores have shaped different mouth morphology for grazing and browsing
herbivores (Table 5.2; Hofmann and Stewart 1972; Jarman 1974; Janis and Ehrhardt
1988; Pérez-Barbería and Gordon 2001). Therefore, plant architecture and animal
anatomy influence each of the components of functional response, including
finding, cropping, and chewing bites of vegetation.
5.2.1.1 Cropping Time
The time necessary to crop bites of vegetation (h, min, Eq. 1) is influenced by fibre
composition, dispersion of plant parts, and structural defences of plants, all of
which differ between grasses and browses. First, grasses tend to have thicker
Table 5.2 A comparison of mouth and tooth anatomy of grazing and browsing herbivores
Characteristic Grazer Browser
Mouth Wide muzzle Narrow muzzle
Smaller mouth opening Larger mouth opening
Stiffer lips Flexible, often muscular
lips, long tongue
Teeth Lower incisors of similar Central incisors broader than
size, wider contact area outside ones, less contact
area for manipulation
Wide incisor row Narrow incisor row
Incisors project forward Incisors more upright
High-crowned teeth Low-crowned teeth
Less surface area, more Large surface area of contact
shearing action, more ridges between lower and upper
teeth for crushing action
cell walls than do browses. Within the cell wall, grasses contain a larger proportion
of cellulose, whereas browses, especially the woody parts, contain more lignin. Many
grasses contain silica, causing leaf blades to have sharp and rigid edges (Demment
and Van Soest 1985; Bodmer 1990; Gordon and Illius 1994; Owen-Smith 1997).
Cellulose, lignin, and silica increase the tensile strength, or ‘toughness’ of the plant,
thus requiring an herbivore to exert greater force and expend more time to sever the
bite (Reid 2000). Tensile strength, or ‘toughness’, increases as a function of age and
diameter of tillers and stems cropped (Van Soest 1981; Nelson and Moser 1994;
Wright and Illius 1995; Shipley et al. 1999). Tensile strength can also limit the size
of bites cropped by both grazers and browsers, thus further reducing intake rates
(Vivås et al. 1991). For example, large bites taken by grazers necessarily require
cropping more tillers and a greater proportion of mature tillers, both of which increase
tensile strength. When grazing swards of high tensile strength, grazers sometimes fail
to sever the bites of grasses apprehended, and must do so in order to release some of
the forage before completing the biting motion. In grazing macropods, for example,
this requires longer occlusal contact between upper and lower incisors (Hume 1999).
When browsers take larger bites, they often must include more woody material and
crop larger-diameter or older stems. As twig diameter increases, browsers must
switch from using incisors to molars to crop bites, which slows intake (Cooper and
Owen-Smith 1986). Eventually, stem diameters become too large to crop at all.
The dispersion of plant parts and physical plant defences can also influence
cropping time. When leaves and tillers are small and dispersed, herbivores must
spend more time rounding up multiple leaves and stems with the tongue and
lips before cropping the bite, than they would when consuming more uniform,
compact bites (Illius and Gordon 1987). Because browses are typically more
spatially diffuse than grasses, cropping bites is often more time-consuming for
browsers. Furthermore, structural defences of woody browses, such as spines and
thorns, also slow cropping by impeding stripping motions and by separating
leaves (Dunham 1980; Cooper and Owen-Smith 1986; Belovsky et al. 1991;
Gowda 1996; Illius et al. 2002). Animals must manipulate thorny plants more
slowly and carefully in their mouths to avoid pain and injury (Dunham 1980;
Cooper & Owen-Smith 1986; Belovsky et al. 1991). Grazers and browsers typically
have evolved different mouth anatomy that may allow them to reduce cropping time
on the plants for which they are adapted. For example, the wide incisor row of
grazers may allow them to spread out the force of severing many tillers in one
bite (Janis and Ehrhardt 1988). In contrast the pointed muzzles and mobile lips
of browsers are likely adaptations for quickly obtaining small bites separated in
space or defended by thorns, and for gathering or stripping leaves from branches
in one motion (Myers and Bazeley 1992). In browsing macropods, the first upper
incisor is larger than the second and third incisor, minimising the amount of
upper and lower incisor contact, which allows manipulation of food items by the
incisors (Sanson 1989). Unfortunately, few studies have directly compared the
ability of browsers and grazers to crop bites in order to evaluate quantitatively
whether their different mouth anatomy influences cropping efficiency on similar
plants. When comparing a group of 13 herbivores cropping bites of fresh alfalfa,
however, Shipley et al. (1994) found that relative to their size, grazers generally
had a shorter cropping time than the intermediate and browsing species, especially
those with long muzzles.
5.2.1.2 Bite Size
Bite size cropped (S, g, Eq. 1–3) strongly influences intake and the functional response
of browsing and grazing herbivores. In many herbivores, intake rate varies as much as
10-fold with increasing bite size (Black and Kenney 1984; Spalinger and Hobbs 1992b;
Gross et al. 1993; Shipley et al. 1994). Animals achieve this greater intake rate because
large bites require relatively fewer interruptions in chewing and swallowing caused by
cropping new bites than do small bites. The size of bites animals obtain depends on
both the architecture of the plant and the size and morphology of its mouth.
Relative to many browses, grasses are often uniformly distributed in a 3-dimensional
layer of food. The bite mass a grazer can obtain from this layer depends on the height
and bulk density of grass, and the width of the grazer’s incisor row (Black and Kenney
1984; Gordon and Illius 1988; Ungar et al. 1991; Illius and Gordon 1992). Therefore,
the wider the grazer’s incisor width, the larger the bite it can obtain from a grass sward
(Illius and Gordon 1987; Janis and Ehrhardt 1988). When grasses are particularly short,
intake rates can become so low, that feeding on them becomes unprofitable and they
are avoided (Arnold 1987; Laca et al. 1993). Even within grazers, Murray and Illius
(2000) found that when feeding on short swards the wildebeest (Connochaetes tauri-
nus) obtained twice the bite size and intake rate of narrower-muzzled topi (Damaliscus
lunatus). On mid-length, differentiated grass swards, however, topi were able to select
the higher quality leaves and thus have a higher intake than wildebeest.
In contrast to grasses, browses often grow with a fractal, branching geometry
which separates plant tissue in space and makes predicting bite mass cropped by
browsers difficult to model and predict from plant volume (Shipley et al. 1994).
For example, browsers often avoid cropping small thin stems branching at wide
angles from which they can obtain little mass (Vivås et al. 1991; Myers &
Bazeley 1992). Instead, large browsers tend to prefer woody plants that provide
larger leaves, thicker or longer annual growth stems, and non-thorny plants like
willows that allow them to strip many leaves in one bite (Danell et al. 1994;
Shipley et al. 1998; Stapley 1998). Browses also tend to vary in nutritional quality
on a small scale, thus browsers must often select specific plant parts in each bite
(Jarman 1974; de Reffye and Houllier 1997). Furthermore, structural defences,
such as thorns, separate leaves and reduce the size of bite a browser can take
(Pellew 1984; Milewski et al. 1991; Stapley 1998; Illius et al. 2002). The browser’s
narrower muzzle, prehensile lips (e.g., elephant’s trunk) and long tongue allow
greater selection of smaller, more nutritious plant parts, and allow it to obtain
bites interspersed with thorns (Janis and Ehrhardt 1988; Hofmann 1989; Belovsky
et al. 1991). These features also permit large browsers, like moose, kudus, and
giraffes, to obtain large bites by stripping many leaves from one stem in a sideways
motion (Pellew 1984; Stapley 1998).
Although few direct comparisons of bite size of grazing and browsing

herbivores have been made, Shipley et al. (1994) compiled data on maximum
bite sizes reported in the literature for 16 grazing and 17 browsing herbivores
feeding on their natural forages. Although maximum bite mass increased with
body mass with the same slope, browsers of the same size took larger maximum
bites than grazers. This result suggests that grazers may be more limited by
their mouth size when cropping bites of grasses than are browsers when
cropping browses. This result has implications for the form of gain functions
observed in grazers and browsers. Linear gain functions have often been
reported for browsing herbivores (Åström et al. 1990; Rominger 1995;
Rominger et al. 1996; Illius et al. 2002). Illius et al. (2002) observed linear gain
functions in roe deer feeding in browse patches, and demonstrated that deceler-
ating gain functions were not observed in browse patches because increased
bite rate compensated for declines in bite mass as patches were depleted. In
contrast, grazing herbivores are often observed to exhibit decelerating gain
functions, especially on tall grasses (Laca et al. 1994; Ginnett et al. 1999).
Searle et al. (2005) showed that decelerating gain curves in herbaceous patches
are fundamentally a function of declining bite size. However, this relationship
is mediated through changes in the time required to form and process those
bites. Their results showed that bite mass declined during the course of patch
depletion, but that cropping time and processing time increased due to accumu-
lation of plant material in the mouth interfering with bite formation (increased
cropping time) and bite processing (increased chewing rate).
5.2.1.3 Chewing Investment and Chewing Rates
How efficiently food can be processed in the mouth (Rmax, g/min, Eq. 1) depends
on both the rate of chewing (chews/min) and how many chews must be invested
per g of food comminuted (chews/g) The fibre content of plants and the size
and quality of the teeth play a role in these processes (Balch 1971; McLeod et al.
1990; Shipley and Spalinger 1992; Shipley et al. 1994). Because cell wall thickness
and fibre content is often relatively greater in grasses, herbivores must spend more
time chewing a unit mass of grass than of browse, either while ingesting it, or later
during rumination (Robbins 1983; Choong et al. 1992; Wright and Illius 1995). In
addition, grazers are expected to have a smaller reticulo-omasal orifice (Hofmann
& Stewart 1972; Clauss & Lechner-Doll 2001, but see also Pérez-Barbería and
Gordon 2001), thus grazers may need to comminute food to a smaller particle size
to escape the rumen than do browsers, further increasing their chewing investment.
Fibre content, and thus chewing time, increases as both grasses and browses mature
(Hacker and Minson 1981). Spines on leaves of dicots may also reduce chewing
efficiency for browsers (Illius et al. 2002).
The efficiency with which an herbivore can chew, and thus ingest, its food
also depends on the size and wear of its molars. Silica in grasses can promote
rapid tooth wear and thus reduce chewing efficiency (McNaughton and Georgiadis
1986; Riet-Correa et al. 1986). As a response, grazers of some taxonomic groups

have developed high-crowned, or hypsodont, molars (Vaughan et al. 2000).
These high-crowned molars allow for longer wear when chewing higher fibre and
silica. For example, molars and premolars of rock hyraxes (Procavia johnsoni),
which feed preferentially on grasses, are hypsodont, having high crowns and
short roots (Hoeck 1975, 1989). In contrast, yellow-spotted hyraxes (Heterohyrax
brucei), a browsing hyrax, have brachydont tooth structure with shorter crowns
and longer roots. In herbivorous marsupials, grazing macropod marsupials have
extended the wear of molars through sequential molar progression anteriorly
along a curved tooth row (Sanson 1980, 1989). This morphology allows only two
molars to occlude at a time, concentrating shearing force in a smaller area of
occlusion to maximise the effectiveness of mastication. As the anterior molars
wear out and are lost, they are replaced by posterior molars. In contrast, browsing
macropod marsupials have a flat tooth row in which all molars occlude at once.
The larger occlusal surface area allows a crushing action and precise molar
occlusion. Furthermore, the tooth enamel of the grazing red kangaroo (Macropus
giganteus) is harder than that of the browsing, swamp wallaby (Wallabia bicolour;
Palamara et al. 1984). However, similar patterns in tooth structure have not been
found in ruminants. Larger molars provide more area for breaking down plant
tissue while chewing, and thus Rmax increases with body mass at the same rate as
does the surface area of the molars (Shipley et al. 1994). However, relative to
body size, grazers and browsers had the same molar surface area, chewing investment,
and chewing rates (Shipley et al. 1994).
5.2.1.4 Encountering
The rate of encounter with bites of vegetation (λ, bites/min) may also differ in
habitats occupied by grazing and browsing herbivores. However, modelling
encounter rate is complex because the animal’s mechanics of travel, perceptive
abilities, spatial memory, and volition play a strong and interconnected role.
Navigating to food items requires an animal to detect, travel to, and choose to
eat food items at more than one spatial scale. Detection depends on character-
istics of the animal (e.g., sight and olfaction, spatial memory, and travel speed
(Bailey et al. 1989; Roese et al. 1991; Speakman and Bryant 1993), and
characteristics of the plant (e.g., crypticity, size and contrast, visual and olfactory
cues, predictability, (Blough 1989; Howery et al. 2000), and occasionally
characteristics of the animal’s conspecifics (e.g., social facilitation; Howery et al.
1998; Bailey et al. 2000). Efficient encounter rate requires that an animal avoid
unfilled or low quality patches and move directly among food items. Although
we have no reason to believe that grazers and browsers differ fundamentally in
their perceptual abilities or memory, grasses and browses tend to differ in their
spatial distribution and contrast.
Homogeneous food sources that are relatively fixed and predictable in time and
space, as may be true of grasses, provide fewer visual cues for easily locating foods,
but animals are more able to depend on their spatial memory to find food resources
in these habitats (Howery et al. 2000). Spatial memory allows herbivores to revisit
nutrient rich sites and avoid low quality sites, but often requires animals to store an
enormous amount of information (Bailey et al. 1989; Bailey and Rittenhouse 1989;
Laca 1998). On the other hand, heterogeneous food sources, as may often be true
of browses, may provide more visual cues that herbivores can use to predict and
locate forage resources from a distance. Visual cues allow animals to speed up and
travel directly between food items or patches. Because patches are easier to detect
than individual food items, herbivores tend to have lower intake when food is dis-
tributed randomly or uniformly than when it is distributed in patches (Laca and
Ortega 1996).
Animals that forage within a social group may more easily detect patches of
forages through social facilitation (Howery et al. 1998; Bailey et al. 2000). In
general, grazers that occupy open habitat are more likely to live and forage in
groups than are browsers that more typically occupy areas with taller and denser
vegetation, possibly because of the added advantage in detecting and avoiding
predators (Krebs and Davies 1993). In feral cattle, for example, the integrity of
the social group is high (Lazo 1994). Therefore, social facilitation beyond that
provided by the mother (Howery et al. 1998) may enhance encounter rates at
larger scale for grazers.
After food items have been detected, encounter rate depends on the dis-
tance between food items and how fast an animal can travel while foraging.
Maximum foraging speed is likely similar between browsers and grazers,
even if they differ in body mass. Shipley et al. (1996) found no relationship
in maximum travel velocity while foraging for animals ranging from 5 g to
500 kg. However, average travel speed was slower for large animals when
plants were spaced closer together because it took larger animals longer to
accelerate and decelerate between bites. If food items are difficult to detect,
animals must slow their travel rate while foraging to increase detection ability
(Getty and Pulliam 1991; Speakman and Bryant 1993; Shipley et al. 1996).
Animals may also purposely choose to increase travel speed when plants are
apparent, but spaced widely apart (Shipley and Spalinger 1995). Therefore,
herbivores often show rapid coarse-scaled movements among patches within
a matrix and slower, fine-scaled movements when searching for bites within
a patch.
Finally, diet selectivity can also influence encounter rate. When food is more
heterogeneous in quality and dispersion, animals must make decisions about
maximizing nutrient intake more frequently than do animals encountering more
homogenous resources. The more selective an animal is, the slower it may have
to travel to detect and make a decision about a plant item, and likely the further
apart and more cryptic the items may be (Murray 1991; Laca & Demment
1996). Browsers are expected to be more selective when foraging because
browses tend to vary greatly in nutritional content and browsers may not be able
to digest plant fibre as well as grazers. Therefore, encounter rate may be more
influenced by diet selection in browsers than grazers.
5.3 Foraging in Patches
Spatially subdividing natural environments into units that have functional relevance
for an ecological process remains a fundamental challenge in many areas of
ecological research. For foraging herbivores, the question of what constitutes a
patch has been at the core of behavioural research for several decades. The patch
concept forms a framework on which most research into foraging behaviour for
herbivores rests, and it underpins many of the widely used models in foraging
behaviour research. However, it is a concept that continues to defy rigorous
definition, and applications that rely on it often fail to produce predictions that
match with observed behaviour. One of the greatest challenges in defining patches
for foraging herbivores arises because herbivores exploit food resources that are
generally continuous, making it very difficult to break up the foraging environment
into discrete units that are robust and repeatable.
Early attempts to define patches centred on reference to their appearance or dif-
ferences from their surroundings (Wiens 1976; Kotliar and Wiens 1990). However,
a more functional definition has since been introduced where patches are delineated
by reference to a change in the rate of a process or behavior that relates specifically
to the foraging animal (Sih 1980; Senft et al. 1987; Bailey et al. 1996). Hierarchy
theory (O’Neill et al. 1986) has been used to provide a framework for the delinea-
tion of heterogeneous foraging landscapes for large herbivores (Senft et al. 1987;
Kotliar and Wiens 1990; Milne 1991; Laca and Ortega 1995; Bailey et al. 1996). This
framework captures heterogeneity by subdividing a landscape into a set of nested
spatial scales, or hierarchical levels, each having functional relevance to the forag-
ing animal.
Use of this hierarchical framework requires us to define some terms; for a large
herbivore these hierarchical levels may consist of individual plants, nested into
feeding stations (an aggregation of bites, where all bites can be removed without
movement of the animal’s forelegs, Goddard 1968), nested into small patches, then
plant communities, ultimately aggregating up to home ranges. We can treat each of
these hierarchical levels as a patch in the traditional sense (sensu MacArthur and
Pianka 1966; Charnov 1976) because at each level the foraging herbivore must
make a decision as to which individual units it chooses to eat from, and how much
effort to devote to exploiting an individual unit before moving on. For instance, at
lower hierarchical levels an herbivore encounters feeding stations as it moves
through its foraging environment, and it must first decide which individual feeding
stations to forage within, and second, how much forage to consume from each
feeding station before moving on.
Natural landscapes present the foraging herbivore with considerable variation
within all these hierarchical levels (i.e., different plant species co-occurring within
a feeding station). Any attempt to define patches for large herbivores must, therefore,
acknowledge this spatial heterogeneity, and act to incorporate the multi-layered
decision process that foraging herbivores follow in complex landscapes. This
section examines the selection of forage patches by ungulates over a range of
spatial scales; specifically asking whether patches are functionally different for
grazers versus browsers, and what consequences these differences have for the way
these animals perceive and interact with their foraging environment.
5.3.1 Definition of Patches
The growth form of grasses differs from that of browse, and these differences have
important consequences for the ways in which herbivores interact with these two
food resources. Jarman (1974) initiated much of our current understanding of these
interactions in his seminal paper discussing the ecology of antelope in Africa.
Through describing significant differences in the growth forms and dispersion of
grasses and browse (differences in meristem position, proportion of the plant
composed of actively growing tissue, chemical and physical defences), Jarman
(1974) was able to relate behavioural aspects of antelope ecology to specific differ-
ences in forage resources. Of particular relevance to any discussion of patch defini-
tion for herbivores is Jarman’s (1974) conclusion that, on average, a grass plant is
likely to be more homogeneous in food quality than a browse plant, and that grasses
typically present a more contiguous distribution than browse plants. Woody browse
plants tend to be composed of ‘an assembly of parts of heterogeneous value’
(Jarman 1974); younger shoots and leaves will offer relatively high nutritional
content, whilst more mature leaves and stems or trunks will offer only negligible
food value. This underlying heterogeneity derived from differences in plant growth
form and plant dispersion interacts with the selective capabilities of the herbivore
to determine the ultimate variation in food quality experienced by grazers and
browsers. It is this variation that ultimately drives the perception of patches of
functional significance to the forager.
The distribution of grasses and browse across a landscape can differ substantially.
Grasses often occur in extensive contiguous swards that may be dominated by
relatively few grazing-tolerant species under intermediate to high levels of grazing
(Jarman 1974; McNaughton 1979; Walker 1987; Stuart-Hill and Tainton 1989),
whereas browse species often occur in more isolated clusters within a landscape
(Jarman 1974). Both grasses and browse employ physical and chemical defences
against herbivory, but these are generally more effective in browse species (Cooper
and Owen-Smith 1986; Milewski et al. 1991; Cooper and Ginnett 1998). Browse
species tend to offer the forager a greater opportunity for higher nutritional value
because on average grasses have a higher dietary fibre content, though for browsing
animals ultimate fibre intake is determined by how much woody stem is eaten (Van
Soest 1981). Intra-plant variation in quality is a general feature of both grasses and
browse; individual plants may contain considerable morphological and phenological
differences in quality. These differences vary widely through a number of
mechanisms, and have profound consequences for the pattern and scale of spatial
variation in the quality of grass and browse forage. It is this spatial variation that
browsers and grazers are able to exploit to increase foraging efficiency and diet
quality. We will now examine the patterns and mechanisms that underlie spatial
variation in the quality of browse and grasses for foraging herbivores.
5.4 Spatial Variation in the Quality of Grass
Grasslands can appear at first glance as monotonous stretches of physiologically

and structurally similar grasses, uniformly distributed within the landscape. This
perception has encouraged the view that the foraging behaviour of large grazers is
relatively simple, with animals using large, unselective mouths to uniformly mow
their way through grass swards. In reality, grasslands are a ‘highly dynamic,
spatially and temporally heterogeneous complex of ecological relationships’
(McNaughton 1989), and grazing herbivores interact in a reciprocal manner with
this variation, both responding to, and creating heterogeneity.
For grazing herbivores the quality of their food resources is determined not so
much by plant defence mechanisms (secondary compounds and physical struc-
tures), but more by the concentrations of nutrients and minerals—namely nitrogen,
phosphorous, potassium, sodium, and calcium. Whilst grasses can contain
secondary compounds, they are generally lower in defensive chemicals than are
other forages (Cates and Rhoades 1977; McNaughton 1983; Malachek and Balph
1987). Physical defences, such as morphological adaptations and silicification
appear to be the predominant evolutionary response of grasses to grazers
(McNaughton 1989). However, it is the spatial variation in the concentrations of
nutrients and minerals in natural grasslands that is the key driver underlying the
diets and distributions of grazing mammals (Owen-Smith and Novellie 1982;
McNaughton 1988; McNaughton 1990; Owen-Smith 1994; Olff and Ritchie 1998).
Explaining the causes of this spatial variation in grass quality is difficult because
the patterns evident in natural grasslands are the outcome of many interacting biotic
and abiotic processes.
Additional complexity is added because, as with most ecological patterns, the
relative importance of these constituent processes changes with spatial scale
(Roberts 1987; Mutanga et al. 2004). However, spatial patterns in grass quality
have been quantified in several grassland systems. The presence of coherent
plant communities has been demonstrated in the Serengeti and Masai Mara,
arguably the most extensive mammalian grazing systems in the world.
McNaughton (1983) demonstrated that almost a hundred stands of grasses
within these systems could be classified into just seventeen objectively identifi-
able plant communities, each having a characteristic, coherent species composi-
tion. Spatial variation in grass quality has been documented at most of the
functionally relevant spatial scales for grazing herbivores—from individual
plants and stands of plants, up to landscape and regional scale variation.
McNaughton (1989) presented a foraging hierarchy for the Serengeti ecosystem
and used forage nutrient analyses to indicate that grazing herbivores confront
substantial heterogeneity in forage quality during the growing season at each
hierarchical level. Clearly then, there is ample opportunity for grazers to increase
their foraging efficiency by choosing to forage in those communities that best

suit their dietary requirements.
The majority of research into spatial variation in grass quality has focused on
regional and landscape scale variation, and has tended to examine the influence
of individual factors in isolation. Traditional approaches have linked patterns in
grass quality at regional scales with temperature and rainfall (Robbins 1983;
McNaughton 1983; Roberts 1987; Skarpe 1992; Kumar et al. 2002). At landscape
scales, topography becomes an important determinant of grass quality through
mechanisms expressed differentially with slope, aspect, and altitude (McNaughton
1983; Roberts 1987; Seagle and McNaughton 1992; Chaneton et al. 2005), and
soil characteristics greatly influence grass quality through spatial variation in
soil nutrients, structure, and texture (Anderson and Talbot 1965; Bell 1982;
McNaughton 1983; Bakker et al. 1983; Kumar et al. 2002; Mutanga et al. 2004).
For instance, McNaughton (1989) found evidence for considerable variation in
mineral contents within a grass species distributed between local patches of forage
in the Serengeti; phosphorous content varied by as much as 52% between swards,
and sodium by as much as 486% among swards.
At a finer scale, spatial variation in grass quality is influenced by biotic fac-
tors. Studies have emphasized the importance of species type and genotype,
phenology, aboveground biomass, and tree/shrub canopy cover (Bakker et al.
1983; Wilson 1984; McNaughton 1988; McNaughton 1990; Olff and Ritchie
1998; Ludwig et al. 2001, 2004). Edaphic factors play a significant role in
manipulating grass quality on local scales. A study in the Kruger National Park
in South Africa demonstrated that grass nutrient distribution was significantly
correlated with slope, aspect, aboveground biomass, and percentage grass cover
(Mutanga et al. 2004). A strong general relationship was found between grass
quality and micro-variations in slope, altitude and aspect, factors which influ-
ence grass quality through their effects on soil temperature and water run-off.
The interplay between grass quality, topography, and hydrology is important in
many semi-arid grazing systems. Fine-scale redistribution of sediments and
nutrients through run-off and run-on is well documented, particularly in systems
dominated by tussock-forming grasses (Greene et al. 1994; Mcivor et al. 1995,
2005; Turner 1998; Eldridge 1998; Northup et al. 1999, 2005; Ludwig et al.
1999, 2000, 2005; HilleRisLambers et al. 2001; Rietkerk et al. 2002, 2004;
Augustine 2003). The nitrogen concentration of individual grasses within a spe-
cies was found to vary with slope in Kruger National Park: some species had
greater nitrogen concentrations on steep slopes, whereas others had higher con-
centrations on flat to gentle slopes (Mutanga et al. 2004). The same study found
an inverse relationship between percentage cover and biomass of grass, and
grass quality. It is thought that this can be explained by the dilution of nutrients
in high biomass regions (Wilson 1984; Mutanga et al. 2004).
Trees in savanna ecosystems can cause significant changes in the quality of
grasses under, and nearby, their canopies. Trees alter nutrient availability for
grasses, generally increasing soil nutrient availability and plant productivity in the
understorey (Kellman 1979; Bernhard-Reversat 1982; Belsky et al. 1989, 1993;
Belsky 1994), though some negative net effects of savanna trees on understorey
productivity have been documented (Stuart-Hill and Tainton 1989; Mordelet and
Menaut 1995). A study in East Africa found that soil nutrient availability around
Acacia tortilis trees increased with tree age and size, being highest under dead trees
and lowest in open grassland patches (Ludwig et al. 2004). The presence of trees
seemed to cause a shift from nitrogen limitation in open grassland to phosphorous
limitation under tree canopies; soil moisture was lower under tree canopies, and the
species composition was very different under tree canopies in comparison to open
areas (Ludwig et al. 2004). Whilst this study found no difference in aboveground
biomass under versus outside live tree canopies, likely due to increased competition
for soil moisture, the quality of grasses was different. A previous study reported
similar results, finding greater mid-wet season nutrient concentrations in grasses
under than outside tree canopies (Ludwig et al. 2001). This variation represents an
important opportunity for herbivores to select for better quality forage beneath tree
canopies in this system.
Within-species level variation is also present in grass species. Lignification is
under genetic control and considerable differences in lignin concentration and
composition have been found among individuals growing in different temperatures,
soil moisture and fertility, and light levels, and even among genotypes within a
species (see Moore and Jung 2001 for a review). Within-plant variation in quality
increases as grass plants mature and the proportion of tissues with large amounts of
structural components increases, but some plant parts such as leaves remain of
relatively high quality. McNaughton (1989) demonstrated significant within-plant
variation in minerals such as aluminium, calcium, iron, potassium, and magnesium
for two grass species within two swards in the Serengeti during the wet season. This
presents grazers with a great potential for exercising selectivity (Fryxell 1991), and
provides opportunities for selective exploitation of plant parts of high nutrient
values within the sward. Grass blades can, therefore, present the grazing animal
with considerable nutrient and mineral heterogeneity.
5.5 Spatial Variation in the Quality of Browse
Species of browse plants vary dramatically in their digestibility to both mammalian

and insect herbivores (Kamupingene et al. 2004; Scogings et al. 2004). This
variation is of undoubted importance to diet selection of browsing herbivores, but
perhaps more relevant to their perception of patches is the considerable variation
among individuals of the same browse species in a suite of compounds known to
influence herbivore digestion and energy gain.
Several mechanisms can cause variation in plant defences within a population.
First, variance in defence against browsing can have a strong genetic component
(Dimock et al. 1976; Silen et al. 1986; Laitinen et al. 2002), and genetic variation in
plant resistance to browsing has been documented in several browse species (Danell
et al. 1990; Rousi et al. 1997; Roche and Fritz 1997; Mutikainen et al. 2000;
Vourc’h et al. 2002; Laitinen et al. 2002). Lignification is also under genetic
control, and considerable differences in lignin concentration and composition,

and hence digestibility, can be found amongst browse species, and even amongst
genotypes within species (Moore and Jung 2001). Second, palatability of plants can
be altered as environmental variation in plant resource availability affects defence
chemical expression (Coley et al. 1985; Moore and Jung 2001; Vourc’h et al. 2002;
Vila et al. 2002; Laitinen et al. 2002). Third, ontogeny causes variations in plant
defences as stage-specific constraints and selections influence a plant at different
stages in its life cycle (Kearsley and Whitham 1989; Karban and Thaler 1999;
Laitinen et al. 2000; Swihart and Bryant 2001; Moore and Jung 2001; Scogings et al.
2004). Finally, the browsing history of a plant can influence its defence levels; both
increases (Chapin et al. 1985; Stolter et al. 2005), decreases (Karban and Baldwin
1997; Agrawal and Karban 1999; Vourc’h et al. 2002), and no difference (Vila et al.
2002) in nutritional quality have been documented in species subjected to browsing.
All of these mechanisms contribute to variation in browse quality for herbivores
across multiple hierarchical levels. Species-level variation in important compounds
has been well documented (Ricklefs and Matthew 1982; Hatcher 1990; Kamupingene
et al. 2004; Scogings et al. 2004), but additional important variation exists at lower
hierarchical levels. Individual trees within a species can vary greatly in concentrations
of nutrients and secondary metabolites related to palatability and deterrency of
grazing (Suomela and Ayres 1994). Strong variability in concentrations of terpenes
and phenolic compounds within individual eucalyptus trees has been demonstrated
(Lawler et al. 1998), and variation in leaf quality (total phenolics) was best
explained at the individual tree level for mountain birch (Betula pubescens tortuosa;
Suomela et al. 1995). Individuals of two tropical deciduous forest tree species
(Spondias mombin and Bursera simaruba) varied significantly in palatability and
leaf characteristics (Howard 1990). Variation in phenolic and monoterpenoid
compounds amongst individuals of several Artemesia spp was found to be large
(Welch and McArthur 1981; White et al. 1982; Cluff et al. 1982; Zhang and States
1991; Gershenzon and Croteau 1991; Wilt et al. 1992).
At still lower levels, extensive within-tree variation is thought to arise from the mod-
ular organisation of trees into hierarchical levels that are more or less physiologically
autonomous and can even differ genetically due to somatic mutations (Suomela and
Ayres 1994). In mountain birch up to 44% of the variance in water content, specific
weight, and toughness was explained at the ramet, branch, and shoot level (Suomela and
Ayres 1994). Leaf quality (secondary chemistry, nutrient availability, and moisture con-
tent) within tropical tree species has been found to vary within individuals (Howard
1990). Leaf nitrogen content has been found to be relatively invariant within individual
trees, suggesting that the processes of nitrogen allocation are relatively well integrated
on a whole plant basis (Suomela and Ayres 1994). Woody shrubs show similar trends
of within-plant variation. Significant variation at the sub-individual level is not uncom-
mon in browse species; variation in leaf quality, measured as the content of carbohy-
drates and amino acids, was generally larger amongst branches within a tree, than
variation between trees in mountain birch (Suomela et al. 1995). Indeed, different geno-
types have highly individualistic responses to environmental variation in terms of
defence against herbivores (Rousi et al. 1993; Mutikainen et al. 2000; Laitinen et al.
2000, 2002). Up to 70% greater resistance to mammalian herbivores was found at the
intra-clonal level of B pendula from a restricted area in south-central Finland

(Mutikainen et al. 2000). As a result, the resistance to herbivory of an individual tree in
relation to other trees in the population may be dependent on the specific conditions of
a particular year (Laitinen et al. 2000); this presents the browsing herbivore with an
incredible array of variation in its feeding choices. However, at least one study has
shown that the degree of resistance to mammalian herbivores in one tree species (Betula
pubescens) was influenced more by genotype than environmental variation (artificial
fertilisation) (Mutikainen et al. 2000). Not only does the chemical composition vary
dramatically within individuals of a browse species, leaf biomass can also vary signifi-
cantly at different heights within a tree canopy (Woolnough and du Toit 2001;
Nordengren et al. 2003). Indeed the vertical distribution of nitrogen, fibre, and defensive
compounds varied substantially within trees of birch (B pubescens) and willow (Salix
spp; Nordengren et al. 2003).
Several studies have suggested that the temporal component of variation in quality
within browse plants may be less than the spatial component. Studies on mountain birch
in Finland have found qualitative variation within trees, ramets, and branches to be quite
constant from year to year (Suomela and Ayres 1994; Suomela et al. 1995). This con-
stancy may reflect the presence of sources (e.g., mature leaves) and sinks (e.g., growing
meristems and reproductive structures) that remain relatively stationary throughout the
tree, or persistent microclimatic effects such as shading by neighbouring trees. This
temporal qualitative consistency may provide the browsing herbivore with an opportu-
nity to learn about its foraging environment, and exploit variable resources more effi-
ciently by returning to previously browsed patches in successive years. It has been
suggested that browsing can increase the quality of trees and shrubs in subsequent years
for browsing herbivores. Repetitive browsing from year to year has been demonstrated
in several studies; moose foraging on Scots Pine in Sweden and Finland (Löyttyniemi
1985; Faber and Lavsund 1999), moose browsing on birch in Sweden (Danell et al.
1985), moose browsing on willow in Alaska (Bowyer and Bowyer 1997), and red deer
and roe deer browsing on Sitka spruce trees in Scotland (Welch et al. 1991). These stud-
ies suggest that initial browsing resulted in browse of a higher quality making them
more palatable increasing the likelihood for re-browsing. Few studies have examined
quantitative changes in browse quality after browsing, though one study on willow
leaves found an increase in leaf nitrogen concentrations in those leaves severely
browsed by moose in the previous winter (Stolter et al. 2005). This change in quality
may well promote the perception of individual trees, or portions of individual trees as
patches, representing a functionally relevant unit that can be manipulated through
browsing to maximise foraging efficiency.
5.6 Perception of Patches
Hierarchical models of patchiness in environments provide us with a framework

for functionally linking the physiology and behaviour of foraging animals to the
spatial patterns of their food resources (Senft et al. 1987; Kotliar and Wiens 1990;
Bailey et al. 1996; Hobbs 1999). The smallest scale at which an animal can
respond to patchiness is called the grain (Kotliar and Wiens 1990). This effectively
represents the lower limit at which selection by the foraging animal can occur;
below this limit the animal is simply not able to respond to any patchiness in the
environment. The differing growth forms and spatial distributions of grasses and
browse promote important differences in the most effective ways a foraging
animal can exploit these resources. This is because the grain for an animal graz-
ing a grass sward will not be the same as that of an animal browsing a clump of
willow. The lower spatial limit at which selection by the animal remains profitable
will differ, and as a consequence the perception of patches by grazers and
browsers will likely to be quite different.
Because the architecture and dispersion of forage resources of grazers
exhibits a more coarse-grained pattern in natural landscapes, grazers are more
likely to make food choices on a larger spatial scale than do browsers. In
grasslands, bite mass, and thus intake rate, is predominantly constrained by the
height and bulk density of the grass canopy. The same is true for browsing
herbivores, where the arrangement of leaves on a branch determines the poten-
tial range of bite sizes available to the animal. Because grasses tend to occur in
contiguous swards, it may not pay a grazing herbivore to invest in behaviours
that allow it to discriminate on very fine scales, such as the bite. However, for
the browsing herbivore, the structural arrangement of leaves within a single tree
can present considerable complexity. For instance, some species such as willow
allow animals to strip many leaves in one bite, but in species where stems
branch at wide angles bites are effectively separated in space, and so individual
groups of leaves effectively represent separate decisions for the foraging ani-
mal. As such it may pay the browsing herbivore to invest in behaviours that
allow for discrimination on fine scales that are more appropriate for its fine-grained
foraging environment.
In addition, grazers focus more on nutrient and mineral concentrations as the
main currency for forage selection (McNaughton 1988; Seagle and McNaughton
1992; Papachristou and Nastis 1993), which vary most among plants and swards,
whereas browsers focus more on secondary chemicals (Provenza and Malecheck
1984;; Provenza et al. 1990; Palo et al. 1992; Papachristou and Nastis 1993; Jia
et al. 1995; Lawler et al. 1998; Stolter et al. 2005; but see Shipley et al. 1998),
which vary most within plants. Thus browsers and grazers may respond to funda-
mentally different currencies while foraging, each of which has a different spatial
scale of variation in the environment. As a consequence, we expect that grazers
perceive their environment on a larger spatial scale than browsers, and that this
will be reflected in the evidence for their patch perceptions. Defining patches for
grazers and browsers requires considering these differences in the spatial hetero-
geneity of food resources; a practice rendered difficult because there is no innate
way to know the scale at which an animal perceives its environment. However,
several approaches have been developed whereby likely patch perceptions can be
identified using analysis of observed intake rates and movement patterns of
foraging herbivores.
5.6.1 Evidence for Patch Perceptions of Grazers and Browsers
Attempts to functionally define patches for herbivores consider how changes in plant
characteristics and spatial distribution interact with intake rate. Competing models
describing how intake rate might be influenced by plant characteristics, such as
biomass and density, can be confronted with data to identify the mechanics most
likely responsible for observed behaviour. Hobbs et al. (2003) used this approach to
analyse the short-term foraging behaviour of a range of herbivores feeding in artificial
patches of alfalfa. They found that intake rates were best described using a threshold
model that utilises a distance d*, which identifies the plant spacing at which intake
rate switches from being regulated by food processing (bite mass) to regulation by
food encounter (plant density). This distance d* thus represents a threshold for the
functional expression of heterogeneity (Hobbs et al. 2003). At distances below d*
the foraging herbivore will be insensitive to changes to heterogeneity in the spatial
arrangement of plants because its intake rate is regulated by bite mass, not encounter
rate of plants. Hobbs et al. (2003) predicted that d* would scale positively with
increasing body size in herbivores. However, Fortin (2006) examined the allometry
of distance threshold d* in several species of herbivores foraging in artificial patches
of alfalfa, and found overall greater evidence for a negative relationship between d*
and body size. This suggests that larger herbivores should generally experience a
decrease in intake rate at closer spacing of plants than smaller herbivores. The
consequences of this finding are that small herbivores should be able to maintain
close to maximal intake rates when foraging in environments where their food
resource is patchily distributed (Fortin 2006). This outcome has considerable
significance for patch perceptions of the two feeding guilds, given that grazing
herbivores are generally of larger body size than browsing herbivores.
5.6.2 Patch Perceptions by Grazers
Grazing can create areas of relative homogeneity within swards, or larger grazing
lawns of grasses of similar age, structure and physiognomy (Bakker et al. 1983;
McNaughton 1984). Studies have demonstrated that grazing sustains grasses in an
actively growing, highly nutritious state (Seagle et al. 1992). Grassland systems
with a long evolutionary history of mammalian grazing tend to contain grasses
with a high capacity for compensatory growth after defoliation (McNaughton
1979, 1983; McNaughton and Chapin 1985; Milchunas et al. 1988). Consequently,
grazing herbivores can maintain small, highly productive areas within which grass
is kept in a state of active growth with higher overall nutrient content (particularly
nitrogen), digestibility, and biomass than surrounding areas (Misleavy et al. 1982;
McNaughton 1983; 1984, 1988; Bakker et al. 1983). However, grazing is very
rarely uniform and as such both natural and commercial grasslands are often a
mosaic of short, heavily grazed patches and tall, lightly grazed patches (McNaughton
1984). For instance, ten years of sheep grazing in an initially uniform Holcus
lanatus-dominated sward created a micro-pattern of heavily grazed patches

alternating with taller, lightly grazed patches on a scale of 5–30 m (Bakker et al.
1983). This micro-pattern was more or less stable over time, probably maintained
by a higher protein content (greater proportion of young leaves) in the patches
subjected to repetitive sheep grazing (Bakker et al. 1983). The creation of patches
within grasslands, and observations on the movement patterns of herbivores can
be used to quantify the scale at which grazing herbivores perceive their
environment.
Foraging patterns of elk (Cervus elaphus) have been examined using first
passage times that describe the time an animal spends in an area of a given size.
This approach identified movement scales of approximately 270 m while elk were
foraging; elk tended to forage in areas having intermediate biomass and low
movement costs (Frair et al. 2005). Similarly, elk in Yellowstone National Park
were found to allocate more time to foraging in areas of greater biomass of
grasses and forbs, and searched more intensively (creating larger craters in the
snow) where food biomass was higher, consistent with patch use models (Fortin
et al. 2005). Elk readily use all areas within small patches but tend to restrict their
movements to smaller areas within larger patches when foraging (McIntyre and
Wiens 1999). Cattle have been shown to respond to patches of forage. In one of
the rare occasions of quantitative support for the Marginal Value Theorem
(Charnov 1976), Laca et al. (1993) demonstrated that cattle grazing from grassland
patches adjusted their patch residence times in accordance with predictions for
different patch types and inter-patch distances. On a smaller spatial scale, cattle
have further been shown to exhibit selectivity within patches for feeding stations
(WallisDeVries et al. 1999). However, the number of bites per feeding station was
not affected by patch size, suggesting that selection between and within feeding
stations are essentially different processes (WallisDeVries et al. 1999). Once a
feeding station had been chosen, no further selectivity was shown; the number of
bites per feeding station was similar in different patch types (WallisDeVries et al.
1999). Selection therefore operated mainly in directing the foraging path and in
selecting feeding stations (WallisDeVries et al. 1999).
More detailed studies suggest that the behaviour of large grazers at the level
of feeding stations appears to be consistent with patch use models: an alternating
pattern of a series of steps between feeding stations followed by grazing (a series
of bites) within each feeding station (Laca et al. 1994). Sheep have been shown
to exhibit no discrimination between individual plant species within feeding
stations unless there were very strong preference differences between the com-
posite plant species (Gordon et al., unpubl). Sheep simply consumed the different
grass species in proportion to their abundance within each feeding station. Cattle
have been shown to graze grass swards systematically, removing grass in upper
horizons before biting from the lower horizon (Ungar et al. 1992; Laca et al.
1994; Ginnett et al. 1999). Both cattle (Wade et al. 1989; Ungar et al. 1991; Laca
et al. 1992) and sheep (Milne et al. 1982; Burlison et al. 1991) have been shown
to remove a constant proportion of the available sward height when grazing.
However, unlike cattle and sheep, reindeer—a mixed feeder (Rangifer tarandus
tarandus)—took constant bite sizes irrespective of vascular plant biomass on

tundra (Trudell and White 1981), and wood bison (Bison bison athabascae) took
constant bite sizes irrespective of sward height in an experimentally manipulated
Carex grassland (Bergman et al. 2000). The authors suggest this is because,
unlike other Bovinae, bison rarely use their tongues to increase bite area on tall
swards, so bite area tends to remain constant. This evidence suggests that large
grazing herbivores are relatively unselective within feeding stations in grass
swards, and that perception of heterogeneity in their environment likely occurs at
the feeding station scale.
Selectivity at the sub-feeding station level seems, for grazers, to be limited to
manipulating the level of depletion before moving on, especially in large-bodied
herbivores that have a limited capacity to benefit from exploitation of fine-scale
heterogeneity (WallisDeVries et al. 1999). WallisDeVries et al. (1999) demon-
strated that the selectivity of cattle was significantly impaired in a fine-grained
environment. They suggest that cattle were unable to adjust their movement
patterns to a sufficient degree as to match the scale of patchiness in the fine-
grained environment once grain size becomes sufficiently small—in this study
approximating the body length of the animals—the cost of frequently turning to
seek out the most profitable food becomes too high (Jiang and Hudson 1993;
WallisDeVries et al. 1999). Elk grazing grassland patches have been shown to tend
to leave feeding stations when their lateral neck angle reaches a critical point, sug-
gesting a biokinetic mechanism for feeding station departure rules (Jiang and
Hudson 1993).
5.6.3 Patch Perceptions by Browsers
Browsers may treat a portion of an individual tree or shrub as a patch, the entire
tree or shrub as a patch, or alternatively, whole stands of trees could be considered
as a patch. The clustering of leaves on browse species may prompt browsing
herbivores to perceive individual trees as patches of food (Gordon 2003). Several
studies have indicated that moose (Alces alces) perceive individual trees as
patches. Åström et al. (1990) demonstrated that handling time per tree increased
with increasing tree size for moose feeding on deciduous trees. Danell et al. (1991)
found that moose did not consume from stands of trees in proportion to their avail-
ability, rather they disproportionately directed foraging towards more profitable
tree types, indicating that selection again occurred on the scale of the tree.
Furthermore, Edenius et al. (2002) were able to show that moose did not perceive
aspen stands as discrete patches; random sites and aspen stands were utilized
equally by moose in terms of overall use of forage. Instead, moose seemed to use
individual aspen ramets in accordance with diet theory. Owen-Smith and Novellie
(1982) found that kudus (Tragelaphus strepsiceros) foraging in South Africa
tended to end step sequences in less than four steps, which approximately
corresponds to the size of individual bushes; they observed that kudus commonly
took 2 or 3 paces without leaving the bush upon which they had been feeding. This
suggests that kudus were exploiting their foraging environment treating individual
browse bushes as a patch unit. A further patch size of approximately 18 paces was
identified, corresponding to a patch radius of about 75m (Owen-Smith and
Novellie 1982). The authors suggest that this area may correspond to the ‘field of
view’ over which a kudo can see ahead and choose its direction of movement to
the next feeding station. Movements over greater distances probably involved
search for the next group of woody plants or cluster of forbs, rather than for
individual food items (Owen-Smith and Novellie 1982).
Browsers also make diet choices at a smaller scale, the bite or the grain of
the patch hierarchy. Browsers respond to both food quality and availability by
eating a higher quality diet and taking less from each plant at high food densities
(Vivås and Sæther 1987; Andersen and Sæther 1992; Shipley and Spalinger
1995). This behaviour seems to be primarily controlled at the bite scale.
Browsing roe deer (Capreolus capreolus) have been shown to exhibit a high
degree of selectivity at the bite scale whilst feeding from patches of browse
species (Illius et al. 2002). Animals took larger bites from larger patches
(branches)—similar to grazers on taller swards—but bite mass declined
continuously as patch exploitation progressed, implying that animals were
selecting larger items to eat first, prompting the authors to reject optimal patch
use as the mechanism for exploitation of food in favour of diet optimization—a
trade-off between diet quality and quantity (Illius et al. 2002). Moose (Alces
alces) have also been shown to select larger bites on their first visit to Scots
pine (Pinus sylvestris) and aspen (Populus tremula; Edenius 1991). Other
studies have shown that moose tend to consume birch twigs of greater diameter
as food abundance declines, with a higher proportion of the intake taken from
nutritious ‘top twigs’ in high density plots (Vivås and Sæther 1987; Shipley and
Spalinger 1995). If browsers were responding to variation in their food
resources at the patch scale, the near linearity of gain functions reported for
several browsers (Åström et al. 1990; Shipley and Spalinger 1995; Illius et al.
2002) would predict that optimal patch use should involve nearly complete
defoliation of browse patches. Yet many studies have proven otherwise,
demonstrating low amounts of biomass removal by browsing ungulates. In
contrast great success in predicting diet optimization at the bite scale has been
demonstrated. Shipley et al. (1999) developed a general model that predicted
optimal bite size for browsing herbivores, defined as the bite size that results in the
greatest daily net energy intake, based on constraints in harvesting and digest-
ing foods. Tests of this model with moose, red deer (Cervus elaphus) and roe
deer provided good evidence that browsing herbivores selected bite sizes in
relation to the chemistry and morphology of plants, animal body size, and
digestive strategy (Shipley et al. 1999). An additional model predicting optimal
bite size for moose browsing on birch in Norway based on maximization of net
energetic gain, incorporating intake rate and digestibility, also resulted in accu-
rate predictions (Vivås et al. 1991). This suggests that browsers exploit variation
in their food resource, primarily through selectivity at the bite scale.
5.7 Summary
All herbivores are faced with meeting their energy requirements from a food source
that provides relatively little digestible energy and protein, requires extensive time
to harvest and digest, and is patchily-distributed over landscapes. As a conse-
quence, the foraging behaviours of grazers and browsers have many elements in
common. The shape of the functional response for both grazers and browsers
responds to plant and animal characteristics, and determines how much time
animals have to spend to acquire nutrients, and how much time remains for other
life-sustaining and fitness-enhancing activities, such as predator evasion and
rearing of young. There are commonalities in some of the mechanistic components
of the functional response for grazers and browsers; cropping time and chewing
investment increase with bite size for both grazers and browsers because of the
increase in structural tissues associated with larger bites. Grasses and browses each
have structural components that reduce bite size and chewing rates, and grazers and
browsers both must search for and travel to food resources. The different architecture
of grasses and browses, especially the more spatially diffuse organisation of browse
plants, and how that architecture influences bite size forms the most important
difference in functional response between grazers and browsers.
Herbivores interact reciprocally with the distribution of their forage resources,
perceiving and responding to variation in plant quality and quantity across multiple
spatial scales. However, structural and chemical differences in grasses and browses
have required grazers and browsers to adapt morphologies and behaviours that
allow them to best acquire nutrients from these different plant types. In response,
grasses and browses have co-evolved mechanisms for reducing the effects of
herbivory on their fitness. These differences can have striking consequences for the
organisation of herbivores across landscapes. For instance, most species of herbiv-
ores that aggregate into large herds are grazers rather than browsers (Fryxell and
Sinclair 1988), and this may be because grasses tend to be of particularly poor qual-
ity at maturation (Owen-Smith 1982) so the benefits of aggregating and maintaining
swards of forage in an early maturational stage are high (Fryxell 1991). The
combination of a relatively contiguous distribution of grasses, and a tendency for
variation in quality to occur between swards of grasses, encourages grazing herbiv-
ores to interact with their forage resource on a comparable spatial scale, utilising
their environment to make use of small patches and feeding stations as the basic
unit of exploitation. In contrast, the more diffuse distribution of browse plants, and
the tendency for variation in quality to occur within individual trees and shrubs,
promotes a more fine-scale interaction between browsers and their forage resources
such that browsers utilise their environment to make use of portions of individual
plants and bites as the basic unit of exploitation. There is mounting evidence that
the application of patch optimisation models to browsers may be misplaced. Gain
functions for browsing herbivores are often linear or piecewise-linear, and the
majority of studies have failed to support predictions based on patch optimisation
models. Standard applications of patch models for browsing herbivores are in need
of revision; energy intake, time constraints, predation avoidance, social behaviours,

and diet optimisation incorporating digestive constraints should all be included in
future research directions.
The differences in foraging strategies we have outlined here are manifested
across the life histories of these two ungulate guilds, having important conse-
quences for social and mating systems, production and care of young, and distribu-
tion and migration patterns. As such, seemingly minor differences in forage and
foraging tactics between grazers and browsers can lead to profound contrasts at
higher ecological levels.
References
Abrams PA, Schmitz OJ (1999) The effect of risk of mortality on the foraging behaviour of ani-
mals faced with time and digestive capacity constraints. Evol Ecol Res 1:285–301
Agrawal A, Karban R (1999) Why induced defenses may be favoured over constitutive strategies
in plants. In: Tollrian R, Harvell CD (eds) The ecology and evolution of inducible defenses.
Princeton University Press, Princeton, NJ, pp 45–61
Andersen R, Sæther BE (1992) Functional response during winter of a herbivore, the moose, in
relation to age and size. Ecology 73:542–550
Anderson GD, Talbot LM (1965) Soil factors affecting the distribution of the grassland types and
their utilization by wild animals on the Serengeti plains, Tanganyika. J Ecol 53:33–56
Arnold GW (1987) Influence of the biomass, botanical composition and sward height of annual
pastures on foraging behaviour by sheep. J Appl Ecol 24:759–772
Åström M, Lundberg P, Danell K (1990) Partial prey consumption by browsers - trees as patches.
Augustine DJ (2003) Spatial heterogeneity in the herbaceous layer of a semi-arid savanna ecosys-
tem. Plant Ecol 167:319–332
Bailey DW, Rittenhouse LR, (1989) Management of cattle distribution. Rangelands 11:159–161
Bailey DW, Rittenhouse LR, Hart RH, Richards RW (1989) Characteristics of spatial memory in
cattle. Appl Anim Behav Sci 23:331–340
Bailey DW, Gross JE, Laca EA, Rittenhouse LR, Coughenour MB, Swift DM, Sims PL (1996)
Mechanisms that result in large herbivore grazing distribution patterns. J Range Manage
49:386–400
Bailey DW, Howery LD, Boss DL (2000) Effects of social facilitation for locating feeding sites
by cattle in an eight-arm radial maze. Appl Anim Behav Sci 68:93–105
Bakker JP, de Leeuw J, van Wieren S (1983) Micropatterns in grassland vegetation created and
sustained by sheep grazing. Vegetatio 55:153–161
Balch CC (1971) Proposal to use time spent chewing as an index of the extent to which diets for
ruminants possess the physical property of fibrousness characteristic of roughages. Brit J Nutr
26:383–392
Beckerman AP (2005) The shape of things eaten: the functional response of herbivores foraging
adaptively. Oikos 110:591–601
Beecham JA, Farnsworth KD (1999) Animal group forces resulting from predator avoidance and
competition minimization. J Theor Biol 198:533–548
Bell RHV (1982) The effect of soil nutrient availability on community structure in African eco-
systems. In: Huntley BJ, Bakker BH (eds) Ecology of tropical savannas. Springer, Berlin
Heidelberg New York, pp 193–216
Belovsky GE, Schmitz OJ, Slade JB, Dawson TJ (1991) Effects of spines and thorns on Australian
arid zone herbivores of different body masses. Oecologia 88:521–528
Belsky AJ (1994) Influences of trees on savanna productivity - tests of shade, nutrients, and
tree–grass competition. Ecology 75:922–932
Belsky AJ, Amundson RG, Duxbury JM, Riha SJ, Ali AR, Mwonga SM (1989) The effects of
trees on their physical, chemical, and biological environments in a semi-arid savanna in Kenya.
J Appl Ecol 26:1005–1024
Belsky AJ, Mwonga SM, Amundson RG, Duxbury JM, Ali AR (1993) Comparative effects of
isolated trees on their undercanopy environments in high-rainfall and low-rainfall savannas.
J Appl Ecol 30:143–155
Bergman CM, Fryxell JM, Gates CG (2000) The effect of tissue complexity and sward height on
the functional response of Wood Bison. Funct Ecol 14:61–69
Bergman CM, Fryxell JM, Gates CG, Fortin D (2001) Ungulate foraging strategies: energy maxi-
mizing or time minimizing? J Anim Ecol 70:289–300
Bernhard-Reversat F (1982) Biogeochemical cycle of nitrogen in a semi-arid savanna Oikos
38:321–332
Black JL, Kenney PA (1984) Factors affecting diet selection by sheep: II. Height and density of
pasture. Aust J Agr Res 35:551–563
Blough DS (1989) Contrast as seen in visual search reaction times. J Exp Anal Behav
52:199–211
Bodmer RE (1990) Ungulate frugivores and the browser–grazer continuum. Oikos 57:319–325
Bowyer JW, Bowyer RT (1997) Effects of previous browsing on the selection of willow stems by
Alaskan moose. Alces 33:11–18
Brown JS (1988) Patch use as an indicator of habitat preference, predation risk, and competition.
Behav Ecol Sociobiol 22:37–47
Brown JS (1999) Vigilance, patch use and habitat selection: foraging under predation risk Evol
Ecol Res 1:49–71
Burlison AJ, Hodgson J, Illius AW (1991) Sward canopy structure and the bite dimensions and
bite weight of grazing sheep. Grass Forage Sci 46:29–38
Cates RG, Rhoades CJ (1977) Patterns in the production of anti-herbivore chemical defences in
plant communities. Biochem Syst Ecol 5:185–194
Chaneton EJ, Perelman SB, León RJC (2005) Floristic heterogeneity of flooding pampas grass-
lands: a multi-scale analysis. Plant Biosyst 139:245–254
Chapin III FS, Bryant JP, Fox JF (1985) Lack of induced chemical defense in juvenile Alaskan
woody plants in response to simulated browsing. Oecologia 67:457–459
Charnov EL (1976) Optimal foraging, the Marginal Value Theorem. Theor Popul Biol
9:129–136
Choong MF, Lucas PW, Ong JSY, Pereira B, Tan HTW, Turner IM (1992) Leaf fracture toughness
and sclerophyll: their correlations and ecological implications. New Phytol 121:597–610
Clauss M, Lechner-Doll M (2001) Difference in selective reticulo-omasal particle retention as a
key factor in ruminant diversification Oecologia 129:321–327
Cluff LK, Welch BL, Pederson JC, Brotherson JD (1982) Concentration of monoterpenoids in the
rumen ingesta of wild mule deer. J Range Manage 35:192–194
Coley PD, Bryant JP, Chapin III FS (1985) Resource availability and plant antiherbivore defense.
Science 230:895–899
Cooper SM, Owen-Smith N (1986) Effects of plant spinescence on large mammalian herbivores.
Cooper SM, Ginnett TF (1998) Spines protect plants against browsing by small climbing mam-
mals. Oecologia 113:219–221
Danell K, Huss-Danell K, Bergström R (1985) Interactions between browsing moose and two
species of birch in Sweden. Ecology 66:1867–1878
Danell K, Gref R, Yazdani R (1990) Effects of mono- and diterpenes in Scots Pine needles on
moose browsing. Scand J Forest Res 5:535–539
Danell K, Edenius L, Lundberg P (1991) Herbivory and tree stand composition - moose patch use
in winter. Ecology 72:1350–1357
Danell K, Bergstrom R, Iedenius L (1994) Effects of large mammalian browsers on architecture,
biomass, and nutrients of woody-plants. J Mammal 75:833–844
de Reffye P, Houllier F (1997) Modelling plant growth and architecture: some recent advances and
applications to agronomy and forestry. Curr Sci India 73:984–992
Demment MW, Van Soest PJ (1985) A nutritional explanation for body-size patterns of ruminant
Dimock II EJ, Silen RR, Allen VE (1976) Genetic resistance to Douglas-fir damage by snowshoe
hare and black-tailed deer. Forest Sci 22:106–121
Dunham KM (1980) The feeding behaviour of a tame impala Aepyros melampus. Afr J Ecol
18:253–257
Edenius L (1991) The effect of resource depletion on the feeding behavior of a browser winter
foraging by moose on Scots pine. J Appl Ecol 28:318–328
Edenius L, Ericsson G, Naslund P (2002) Selectivity by moose versus the spatial distribution of
aspen: a natural experiment. Ecography 25:289–294
Eldridge DJ (1998) Trampling of microphytic crusts on calcareous soils, and its impact on erosion
under rain-impacted flow. Catena 33:221–239
Faber WE, Lavsund S (1999) Summer foraging on Scots pine Pinus sylvestris by moose Alces
alces in Sweden - patterns and mechanisms. Wildl Biol 5:93–106
Fortin D, Morales JM, Boyce MS (2005) Elk winter foraging at fine scale in Yellowstone National
Park. Oecologia 145:335–343
Fortin, D (2006) The allometry of plant spacing that regulates food intake rate in mammalian
herbivores. Ecology 87:1861–1866
Frair JL, Merrill EH, Visscher DR, Fortin E, Beyer HL, Morales JM (2005) Scales of movement
by elk (Cervus elaphus) in response to heterogeneity in forage resources and predation risk.
Landscape Ecol 20:273–287
Fryxell JM (1991) Forage quality and aggregation by large herbivores Am Nat 138:478–498
Fryxell J M, Sinclair ARE (1988) Causes and consequences of migration by large herbivores.
Trends Ecol Evol 3:237–241
Gershenzon J, Croteau R (1991) Herbivores: their interactions with secondary plant metabolites.
Academic Press, San Diego
Getty T, Pulliam HR (1991) Random prey detection with pause-travel search. Am Nat 138:1459–1477
Ginnett TF, Dankosky JA, Deo G, Demment MW (1999) Patch depression in grazers: the roles of
biomass distribution and residual stems. Funct Ecol 13:37–44
Goddard J (1968) Food preferences of two black rhinoceros populations. E Afr Wildl J 6:1–18
Gordon IJ (2003) Browsing and grazing ruminants: are they different beasts? Forest Ecol Manage
181:13–21
2:15–22
Gowda JH (1996) Spines of Acacia tortilis: what do they defend and how? Oikos 77:279–284
Greene RSB, Kinnell PIA, Wood JT (1994) Role of plant cover and stock trampling on runoff and
soil-erosion from semiarid wooded rangelands. Aust J Soil Res 32:953–973
Gross JE, Shipley LA, Hobbs NT, Spalinger DE, Wunder BA (1993) Functional response of
herbivores in food-concentrated patches: tests of a mechanistic model. Ecology 74:778–791
Hacker JB, Minson DJ (1981) The digestibility of plant parts. Herbage Abstracts 51:459–482
Hatcher PE (1990) Seasonal and age-related variation in the needle quality of five conifer species.
HilleRisLambers R, Rietkerk M, van den Bosch F, Prins HHT, de Kroon H (2001) Vegetation
pattern formation in semi-arid grazing systems. Ecology 82:50–61
Hirakawa H (1997) Digestion-constrained optimal foraging in generalist mammalian herbivores.
Oikos 78:37–47
Hobbs NT (1999) Responses of large herbivores to spatial heterogeneity in ecosystems. In: Jung
HG , Fahey GC (eds) Nutritional ecology of herbivores: Proceedings of the Vth International
Symposium on the nutrition of herbivores. Am Soc Anim Sci, Savoy IL, pp 97–129
Hobbs NT, Gross JE, Shipley LA, Spalinger DE, Wunder BA (2003) Herbivore functional
response in heterogeneous environments: a contest among models. Ecology 84:666–681
Hoeck HN (1975) Differential feeding behaviour of sympatric Hyrax procavia-johnstoni and

Heterohyraz brucei. Oecologia 22:15–47
Hoeck, H N (1989) Demography and competition in Hyrax - a 17 year study Oecologia
79:353–360
Hofmann RA (1989) Evolutionary steps of ecophysiological adaptation and diversification of
Hofmann RA, Stewart DRM (1972) Grazer or browser: a classification based on the stomach-
structure and feeding habits of East African ruminants. Mammalia 36:226–240
Holling CS (1959) Some characteristics of simple types of predation and parasitism. Can Entomol
41:385–398
Holling CS (1965) The functional response of predators to prey density and its role in mimicry
and population regulation. Mem Entomol Soc Can 48:1–46
Howard JJ (1990) Infidelity of leaf-cutting ants to host plants: resource heterogeneity or defense
induction? Oecologia 82:394–401
Howery LD, Provenza FD, Banner PE, Scott CB (1998) Social and environmental factors
influence cattle distribution on rangeland. Appl Anim Behav Sci 55:231–244
Howery LD, Bailey DW, Ryle CB, Renken WJ (2000) Cattle use visual cues to track food loca-
tions. Appl Anim Behav Sci 67:1–14
Hume ID (1999) Marsupial nutrition. Cambridge University Press, Cambridge
Iason GR, Mantecon AR, Sim DA, Gonzalez J, Foreman E, Bermudez FF, Elston DA (1999)
Can grazing sheep compensate for a daily foraging time constraint? J Anim Ecol
68:87–93
Illius AW, Gordon IJ (1987) The allometry of food intake in grazing ruminants. J Anim Ecol
56:989–1000
Illius AW, Gordon IJ (1992) Modelling the nutritional ecology of ungulate herbivores evolution of
body size and competitive interactions. Oecologia 89:428–434
Illius AW, Duncan P, Richard C, Mesochina P (2002) Mechanisms of functional response and
resource exploitation in browsing roe deer. J Anim Ecol 71:723–734
Janis CM, Ehrhardt D (1988) Correlation of relative muzzle width and relative incisor width with
dietary preference in ungulates. Zool J Linn Soc–Lond 92:267–284
Jarman PJ (1974) The social organization of antelope in relation to their ecology. Behaviour
48:215–266
Jia J, Niemela P, Danell K (1995) Moose Alces alces bite diameter selection in relation to twig
quality on four phenotypes of Scots pine Pinus sylvestris. Wildlife Biol 1:47–55
Jiang Z, Hudson RJ (1993) Optimal grazing of wapiti cervus elaphus on grassland patch and
feeding station departure rules. Evol Ecol 7:488–498
Kamupingene GT, Abate AL, Kimambo AE (2004) Crude protein degradability, fibre and tannin
levels of browse forages in an extensive farming system. J Anim Feed Sci 13:111–114
Karban R, Baldwin IT (1997) Induced responses to herbivory. University of Chicago Press,
Chicago
Karban R, Thaler JS (1999) Plant phase change and resistance to herbivory. Ecology 80:510–517
Kearsley MJC, Whitham TG (1989) Developmental changes in resistance to herbivory:
implications for individuals and populations. Ecology 70:442–444
Kellman M (1979) Soil enrichment by neotropical savanna trees. J Ecol 67:565–577
Kotliar NB, Wiens JA (1990) Multiple scales of patchiness and patch structure - a hierarchical
framework for the study of heterogeneity. Oikos 59:253–260
Krebs JR, Davies MJ (1993) An introduction to behavioural ecology. Blackwell, Oxford
Kumar L, Rietkerk M, van Langevelde F, van de Koppel J, van Andel J, Hearne J, de Ridder N,
Stroosnijder L, Skidmore AK, Prins HHT (2002) Relationship between vegetation growth
rates at the onset of the wet season and soil type in the Sahel of Burkina Faso. Ecol Model
149:143–152
Laca EA (1998) Spatial memory and food searching mechanisms of cattle. J Range Manage
51:370–378
Laca EA, Ortega IM (1995) Integrating foraging mechanisms across spatial and temporal scales.
In: West NE (ed) Rangelands in a sustainable biosphere. Society for Range Management,
Denver, pp 129–132
Laca EA, Demment MW (1996) Foraging strategies of grazing animals. In: Hodgson J, Illius AW
(eds) The ecology and management of grazing systems. CABI, Wallingford, UK, pp
137–158
Laca EA, Ortega IM (1996) Integrating foraging mechanisms across spatial and temporal scales.
In: West NE (ed) Vth International Rangeland Congress Society of Range Management, Salt
Lake City, pp 129–132
Laca EA, Ungar ED, Seligman N, Demment MW (1992) Effects of sward height and bulk density
on bite dimensions of cattle grazing homogeneous swards. Grass Forage Sci 47:91–102
Laca EA, Distel RA, Griggs TC, Deo GP, Demment MW (1993) Field test of optimal foraging
with cattle: the marginal value theorem successfully predicts patch selection and utilisation. In:
Proceedings of XVII International Grassland Congress, New Zealand and Queensland,
February, pp 709–701
Laca EA, Distel RA, Griggs TC, Demment MW (1994) Effects of canopy structure on patch
depletion by grazers. Ecology 75:706–716
Laitinen M, Julkunen-Tiitto R, Rousi M (2000) Variation in phenolic compounds within a birch
(Betula pendula) population. J Chem Ecol 26:1609–1622
Laitinen J, Rousi M, Tahvanainen J (2002) Growth and hare, Lepus timidus, resistance of white
birch, Betula pendula, clones grown in different soil types. Oikos 99:37–46
Launchbaugh KL, Provenza FD, Pfister JA (2001) Herbivore response to anti-quality factors in
forages. J Range Manage 54:431–440
Lawler IR, Foley WJ, Eschler BM, Pass DM, Handasyde K (1998) Intraspecific variation in
Eucalyptus secondary metabolites determines food intake by folivorous marsupials. Oecologia
116:160–169
Lazo A (1994) Social segregation and the maintenance of social stability in a feral cattle popula-
tion. Anim Behav 48:1133–1141
Löyttyniemi K (1985) On repeated browsing of Scots Pine saplings by moose (Alces alces). Silva
Fenn 19:387–391
Ludwig JA, Eager RW, Williams RJ, Lowe LM (1999) Declines in vegetation patches, plant diver-
sity, and grasshopper diversity near cattle watering-points in Victoria River District, northern
Australia. Rangeland J 21:135–149
Ludwig JA, Wiens JA, Tongway DJ (2000) A scaling rule for landscape patches and how it applies
to conserving soil resources in savannas. Ecosystems 3:84–97
Ludwig F, de Kroon H, Prins HHT, Berendse F (2001) Effects of nutrients and shade on tree–grass
interactions in an East African savanna. J Veg Sci 12:579–588
Ludwig F, de Kroon H, Berendse F, Prins HHT (2004) The influence of savanna trees on nutrient,
water and light availability and the understorey vegetation. Plant Ecol 170:93–105
Ludwig JA, Wilcox BP, Breshears DD, Tongway D, Imeson AC (2005) Vegetation patches and
runoff–erosion as interacting ecohydrological processes in semiarid landscapes. Ecology
86:288–297
MacArthur RH, Pianka ER (1966) On optimal use of a patchy environment. Am Nat
100:603–609
Malachek JC, Balph DF (1987) Diet selection by grazing and browsing livestock. In: Hacker JB,
Ternouth JH (eds) The nutrition of herbivores. Academic Press, Sydney, pp 121–132
McIntyre NE, Wiens JA (1999) Interactions between landscape structure and animal behavior: the
roles of heterogeneously distributed resources and food deprivation on movement patterns.
Landscape Ecol 14:437–447
Mcivor JG, Williams J, Gardener CJ (1995) Pasture management influences runoff and soil
movement in the semi-arid tropics. Aust J Exp Agr 35:55–65
Mcivor JG, McIntyre S, Saeli I, Hodgkinson JJ (2005) Patch dynamics in grazed subtropical native
pastures in south-east Queensland. Austral Ecol 30:445–464
McLeod MN, Kennedy PM, Minson DJ (1990) Resistance of leaf and stem fractions of tropical
forage to chewing and passage in cattle. Brit J Nutr 63:105–119
McNaughton SJ (1979) Grazing as an optimization process: grass–ungulate relationships in the
Serengeti. Am Nat 113:691–703
McNaughton SJ (1983) Serengeti grassland ecology: the role of composite environmental factors
and contingency in community structure. Ecol Monogr 53:291–320
McNaughton SJ (1984) Grazing lawns: animals in herds, plant form, and coevolution. Am Nat
124:863–866
McNaughton SJ (1988) Mineral nutrition and spatial concentrations of African ungulates. Nature
334:343–345
McNaughton SJ (1989) Interactions of plants of the field layer with large herbivores. Symposium
Zool Soc Lond 61:15–29
McNaughton SJ (1990) Mineral nutrition and seasonal movements of African migratory ungu-
lates. Nature 345:613–615
McNaughton SJ, Chapin IFS (1985) Effects of phosphorus nutrition and defoliation on C4 grami-
noids from the Serengeti Plains. Ecology 66:1617–1629
McNaughton SJ, Georgiadis NJ (1986) Ecology of African grazing and browsing mammals. Annu
Rev Ecol Syst 17:39–66
Milchunas DG, Sala OE, Lauenroth WK (1988) A generalized model of the effects of grazing by
large herbivores on grassland community structure. Am Nat 132:87–106
Milewski AV, Young TP, Madden D (1991) Thorns as induced defenses - experimental evidence.
Milne BT (1991) Heterogeneity as a multiscale characteristic of landscapes. In: Kolasa J, Pickett
STA (eds) Ecological heterogeneity. Springer, Berlin Heidelberg New York , pp 68–84
Milne JA, Hodgson J, Thompson R, Souter WG (1982) The diet ingested by sheep grazing swards
differing in white clover and perennial ryegrass content. Grass Forage Sci 37:209–218
Misleavy P, Mott GO, Martin FG (1982) Effect of grazing frequency on forage quality and stolon
characteristics of tropical perennial grasses. Soil Crop Sci Soc Fl 41:77–83
Moore KJ, Jung HG (2001) Lignin and fiber digestion. J Range Manage 54:420–430
Mordelet P, Menaut JC (1995) Influence of trees on aboveground production dynamics of grasses
in a humid savanna. J Veg Sci 6:223–228
Murray MG (1991) Maximizing energy retention in grazing ruminants. J Anim Ecol 60:1029–1045
Murray MG, Illius AW (2000) Vegetation modification and resource competition in grazing ungu-
lates. Oikos 89:501–508
Mutanga O, Prins HHT, Skidmore AK, van Wieren S, Huizing H, Grant R, Peel M, Biggs H
(2004) Explaining grass–nutrient patterns in a savanna rangeland of southern Africa.
J Biogeogr 31:819–829
Mutikainen P, Walls M, Ovaska J, Keinänen M, Julkunen-Tiitto R, Vapaavuori E (2000) Herbivore
resistance in Betula pendula: effect of fertilization, defoliation and plant genotype. Ecology
81:49–65
Myers JH, Bazeley DR (1992) Thorns, spines, prickles and hairs: are they stimulated by herbivory
and do they deter herbivores? In: Tallamy DW, Raupp NJ (eds) Phytochemical induction by
herbivores. Wiley, New York, pp 325–344
Nelson CJ, Moser LE (1994) Plant factors affecting forage quality. in Fahey Jr GC (ed) Forage
quality, evaluation and utilization. American Society of Agronomy, Crop Science Society of
America, and Soil Science Society, Madison, WI, pp 115–154
Nordengren C, Hofgaard A, Ball JP (2003) Availability and quality of herbivore winter browse in
relation to tree height and snow depth. Ann Zool Fenn 40:305–314
Northup BK, Brown JR, Holt JA (1999) Grazing impacts on the spatial distribution of soil micro-
bial biomass around tussock grasses in a tropical grassland. Appl Soil Ecol 13:259–270
Northup BK, Dias CD, Brown JR, Skelly WC (2005) Micro-patch and community scale spatial
distribution of herbaceous cover in a grazed eucalypt woodland. J Arid Environ 60:509–530
O’Neill RV, Deangelis DL, Waide JB, Allen TFH (1986) A hierarchical concept of ecosystems.
Princeton University Press, Princeton, NJ
Olff H, Ritchie ME (1998) Effects of herbivores on grassland plant diversity Trends Ecol Evol
13:261–265
Owen-Smith N (1982) Factors influencing the consumption of plant products by large herbivores.
New York, pp 359–404
Owen-Smith N (1994) Foraging responses of kudus to seasonal changes in food resources:
elasticity in constraints. Ecology 75:1050–1062
Owen-Smith N (1997) Control of energy balance by a wild ungulate, the kudu (Tragelopahus
strepsiceros) through adaptive foraging behaviour. P Nutr Soc 56:15–24
Owen-Smith N, Novellie P (1982) What should a clever ungulate eat? Am Nat 119:151–178
Palamara J, Phakey PP, Rachinger WA, Sanson GD, Oranus HJ (1984) On the nature of the opaque
and translucent enamel regions of some Macropodinae (Macropus giganteus, Wallabia bicolor
and Peradorcus concinna). Cell Tissue Res 238:329–337
Palo RT, Bergstrom R, Danell K (1992) Digestibility distribution of phenols and fiber at different
twig diameters of birch in winter: implications for browsers Oikos 65:450–454
Papachristou TG, Nastis AS (1993) Factors affecting forage preference by goats grazing kermes
oak shrubland in northern Greece. In: Papanastasis V, Nikolaidis A (eds) Management of
Mediterranean shrublands and related forage resources. Seventh meeting, FAO European sub-
network on Mediterranean pastures and fodder crops. Chania, Greece, pp 167–170
Pellew RA (1984) Food consumption and energy budgets of the giraffe. J Appl Ecol 21:141–159
Pérez-Barbería FJ, Gordon IJ (2001) Relationships between oral morphology and feeding style in
the Ungulata: a phylogenetically controlled evaluation. P Roy Soc Lond B Bio
268:1023–1032
Provenza FD (1995) Postingestive feedback as an elementary determinant of food preference and
intake in ruminants. J Range Manage 48:2–17
Provenza FD, Malecheck JC (1984) Diet selection by domestic goats in relation to blackbrush
twig chemistry. J Appl Ecol 21:831–841
Provenza FD, Burrit EA, Clausen TP, Bryant JP, Reichardt PB, Distel RA (1990) Conditioned fla-
vour aversion: a mechanism for goats to avoid tannins in blackbrush. Am Nat 136:810–828
Reid ED (2000) Differential expression of herbivore resistance in late- and mid-seral grasses of
the tall grass prairie. Dissertation, University of Idaho, Moscow, ID
Ricklefs RE, Matthew KK (1982) Chemical characteristics of the foliage of some deciduous trees
in southeastern Ontario. Can J Botany 60:2037–2045
Riet-Correa F, Mendez MC, Schild AL, Oliveira JA, Zenebon O(1986) Dental lesions in cattle and
sheep due to industrial pollution caused by coal combustion. Pesquisa Vet Brasil 6:23–31
Rietkerk M, Ouedraogo T, Kumar L, Sanou S, van Langevelde F, Kiema A, van de Koppel J,
van Andel J, Hearne J, Skidmore AK, de Ridder N, Stroosnijder L, Prins HHT (2002) Fine-
scale spatial distribution of plants and resources on a sandy soil in the Sahel. Plant Soil
239:69–77
Rietkerk M, Dekker SC, de Ruiter PC, van de Koppel J (2004) Self-organized patchiness and cata-
strophic shifts in ecosystems. Science 305:1926–1929
Robbins CT (1983) Wildlife feeding and nutrition. Academic Press, London
Roberts BR (1987) The availability of herbage. In: Hacker JB, Ternouth JH (eds) The nutrition of
herbivores. Academic Press, London, pp 47–63
Roche MS, Fritz RS (1997) Genetics of resistance of Salix sericea to a diverse community of her-
bivores. Evolution 51:1490–1498
Roese JH, Risenhoover KL, Folse LF (1991) Habitat heterogeneity and foraging efficiency: an
individual-based foraging model. Ecol Monogr 57:133–143
Rominger EM (1995) Late winter foraging ecology of woodland caribou. Dissertation, Washington
State University, Pullman
Rominger EM, Robbins CT, Evans MA (1996) Winter foraging ecology of woodland caribou in
northeastern Washington. J Wildlife Manage 60:719–728
Rousi M, Tahvanainen J, Henttonen H, Uotila I (1993) Effects of shading and fertilization on
resistance of winter-dormant birch (Betula pendula) to voles. Ecology 74:30–38
Rousi M, Tahvanainen J, Henttonen H (1997) Clonal variation in susceptibility of white birches

(Betula spp.) to mammalian and insect herbivores. Forest Sci 43:396–402
Sanson GD (1980) The morphology and occlusion of the malariform cheek teeth in some
Macropodinae (Marsupialia:Macropodidae). Aust J Ecol 28:341–365
Sanson GD (1989) Morphological adaptations of teeth to diets and feeding in the Macropodoidea.
In: Grigg G, Jarman PJ, Hume ID (eds) Kangaroos, wallabies and rat-kangaroos. Surrey
Beatty, Sydney, pp 151–168
Scogings PF, Dziba LE, Gordon IJ (2004) Leaf chemistry of woody plants in relation to season, canopy
retention and goat browsing in a semiarid subtropical savanna. Austral Ecol 29:278–286
Seagle SW, McNaughton SJ (1992) Spatial variation in forage nutrient concentrations and the
distribution of Serengeti grazing ungulates. Landscape Ecol 7:229–241
Seagle SW, McNaughton SJ, Ruess RW (1992) Simulated effects of grazing on soil nitrogen and
mineralization in contrasting Serengeti grassslands. Ecology 73:1105–1123
Searle KR, Vandervelde T, Hobbs NT, Shipley LA (2005) Gain functions for large herbivores:
tests of alternative models. J Anim Ecol 74:181–189
Senft RL, Coughenour MB, Bailey DW, Rittenhouse LR, Sala OE, Swift DM (1987) Large her-
bivore foraging and ecological hierarchies. Bioscience 37:789–799
Shipley LA, Spalinger DE (1992) Mechanics of browsing in dense food patches effects of plant
and animal morphology on intake rate. Can J Zool 70:1743–1752
Shipley LA, Spalinger DE (1995) Influence of size and density of browse patches on intake rates
and foraging decisions of young moose and white-tailed deer. Oecologia 104:112–121
Shipley LA, Gross JE, Spalinger DE, Hobbs NT, Wunder BA (1994) The scaling of intake rate in
mammalian herbivores. Am Nat 143:1055–1082
Shipley LA, Blomquist S, Danell K (1998) Diet choices made by free-ranging moose in northern
Sweden in relation to plant distribution, chemistry, and morphology. Can J Zool
76:1722–1733
Shipley LA, Illius AW, Danell K, Hobbs NT, Spalinger DE (1999) Predicting bite size selection
of mammalian herbivores: a test of a general model of diet optimization Oikos 84:55–68
Shipley LA, Spalinger DE, Gross JE, Hobbs NT, Wunder BA (1996) The dynamics and scaling of
foraging velocity and encounter rate in mammalian herbivores. Funct Ecol 10:234–244
Sih A (1980) Optimal foraging: partial consumption of prey. Am Nat 116:281–290
Silen RR, Randall WL, Mandel NL (1986) Estimates of genetic parameters for deer browsing of
Douglas fir. Forest Sci 32:178–184
Skarpe C (1992) Dynamics of savanna ecosystems. J Veg Sci 3:293–300
Spalinger DE, Hobbs NT (1992a) Mechanisms of foraging in mammalian herbivores: new models
of functional response. Am Nat 140:325–348
Spalinger DE, Hobbs NT (1992b) Herbivore functional response: a mechanistic model of food
intake rate in patchy environments. B Ecol Soc Am 73:209
Speakman JR, Bryant DM (1993) The searching speeds of foraging shorebirds: redshank (Tinga
tetanus) and oystercatcher (Haematopus ostralegus). Am Nat 142:296–319
Stapley L (1998) The interaction of thorns and symbiotic ants as an effective defense mechanism
of swollen-thorn acacias. Oecologia 115:401–405
Stephens DW, Krebs JR (1986) Foraging theory. Princeton University Press, Princeton, NJ
Stolter C, Ball JP, Julkunen-Tiitto R, Lieberei R, Ganzhorn JU (2005) Winter browsing of moose
on two different willow species: food selection in relation to plant chemistry and plant
response. Can J Zool 83:807–819
Stuart-Hill GC, Tainton NM (1989) The competitive interaction between Acacia karroo and the
herbaceous layer and how this is influenced by defoliation. J Appl Ecol 26:285–298
Suomela J, Ayres MP (1994) Within-tree and among-tree variation in leaf characteristics of
mountain birch and its implications for herbivory. Oikos 70:212–222
Suomela J, Ossipov J, Haukioja E (1995) Variation among and within mountain birch trees in
foliage phenols, carbohydrates, and amino acids, and in growth of Epirrata autumnata larvae.
J Chem Ecol 21:1421–1446
Swihart RK, Bryant JP (2001) Importance of biogeography and ontogeny of woody plants in win-
ter herbivory by mammals. J Mammal 81:1–21
Trudell J, White RG (1981) The effect of forage structure and availability on food intake, biting
rate, bite size and daily eating time of reindeer. J Appl Ecol 18:63–81
Turner MD (1998) Long-term effects of daily grazing orbits on nutrient availability in Sahelian
West Africa: I. Gradients in the chemical composition of rangeland soils and vegetation.
J Biogeogr 25:669–682
Ungar ED, Genizi A, Demment MW (1991) Bite dimensions and herbage intake by cattle grazing
short hand-constructed swards. Agron J 83:973–978
Ungar ED, Seligman NG, Demment MW (1992) Graphical analysis of sward depletion by graz-
ing. J Appl Ecol 29:427–435
Van Soest PJ (1981) Nutritional ecology of the ruminant. Cornell University Press, Ithaca, NY
Vaughan TA, Ryan JM, Czaplewski NJ (2000) Mammalogy, 4th edn. Brooks Cole, Toronto
Vila B, Vourc’h G, Gillon D, Martin J, Guibal F (2002) Is escaping deer browse just a matter of
time in Picea sitchensis? A chemical and dendroecological approach. Trees 16:488–496
Villalba JJ, Provenza FD (1999) Effects of food structure and nutritional quality and animal nutri-
tional state on intake behaviour and food preferences of sheep. Appl Anim Behav Sci
63:145–163
Villalba JJ, Provenza FD (2000) Roles of novelty, generalization, and postingestive feedback in
the recognition of foods by lambs. J Anim Sci 78:3060–3069
Villalba JJ, Provenza FD (2005) Foraging in chemically diverse environments: energy, protein,
and alternative foods influence ingestion of plant secondary metabolites by lambs. J Chem
Ecol 31:123–138
Vivås HJ, Sæther BE (1987) Interactions between a generalist herbivore, the moose Alces alces,
and its food resources: an experimental study of winter foraging behaviour in relation to
browse availability. J Anim Ecol 56:509–520
Vivås HJ, Sæther BE, Andersen R (1991) Optimal twig-size selection of a generalist herbivore,
the moose Alces alces: implications for plant–herbivore interactions. J Anim Ecol
60:395–408
Vourc’h G, Vila B, Gillon D, Escarré J, Guibal F, Fritz H, Clausen TP, Martin J (2002)
Disentangling the causes of damage variation by deer browsing on young Thuja plicata. Oikos
98:271–283
Wade MH, Peyraud JL, Lemaire G, Cameron EA (1989) The dynamics of daily area and depth of
grazing of cows in a five day paddock system. In: Proceedings of the 16th International
Grasslands Congress, French Grassland Society, Nice, pp 1111–1112
Walker BH (1987) A general model of savanna structure and function. In: Walker BH (ed)
Determinants of tropical savannas. ICSU Press, Miami, pp 1–12
WallisDeVries MF, Laca EA, Demment MW (1999) The importance of scale of patchiness for
selectivity in grazing herbivores. Oecologia 121:353–63
Welch BL, McArthur ED (1981) Variation of monoterpenoid content among subspecies and
accessions of Artemisia tridentata grown in a uniform garden. J Range Manage 34:380–384
Welch D, Staines BW, Scott D, French DD, Catt DC (1991) Leader browsing by red and roe deer
on young Sitka spruce trees in western Scotland. I. Damage rates and the influence of habitat
factors. Forestry 64:61–82
White SM, Flinders JT, Welch BT (1982) Preference of pygmy rabbits (Brachylagus idahoensis)
for various populations of big sagebrush (Artemisia tridentata). J Range Manage 35:724–726
Wiens JA (1976) Population responses to patchy environments. Annu Rev Ecol Syst 7:81–120.
Wilson JR (1984) Environmental and nutritional factors affecting herbage quality. In: Hacker JB
(ed) Nutritional limits to animal production from pastures. Commonwealth Agricultural
Bureau, Slough, UK, pp 111–131
Wilt FM, Geddes JD, Tamma RV, Miller GC, Everett RL (1992) Interspecific variation of phenolic
concentrations in persistent leaves among six taxa from subgenus Tridentatae of Artemesia
(Asteraceae). Biochem Syst Ecol 20:41–52
Woolnough AP, du Toit JT (2001) Vertical zonation of browse quality in tree canopies exposed
to a size-structured guild of African browsing ungulates. Oecologia 129:585–590
Wright W, Illius AW (1995) A comparative study of the fracture properties of five grasses. Funct
Ecol 9:269–278
Zhang X, States JS (1991) Selective herbivory of ponderosa pine by Abert squirrels: are-examination
of the role of terpenes. Biochem Syst Ecol 19:111–115
Chapter 6
The Comparative Population Dynamics
of Browsing and Grazing Ungulates
Norman Owen-Smith
6.1 Introduction
Among African ruminants a clear dietary distinction exists between grazers,

feeding predominantly on grasses and sedges (graminoids) year-round, and browsers
consuming mostly the leaves and shoots of woody plants as well as herbaceous
dicots (or forbs) and fruits (Owen-Smith 1992, 1997). Even so-called mixed
feeders can be divided between a subset obtaining the greater proportion of their
diet from grasses year-round (mainly grazers, e.g., Thomson’s gazelle (Gazella
thomsoni), and mainly browsers dependent largely on woody and herbaceous
dicots, especially during the dry season (e.g., Grant’s gazelle Gazella granti). Just
two species seem firmly intermediate, switching between consuming mostly
grass in the wet season, and a predominance of browse during the dry season:
impala (Aepyceros melampus) and nyala (Tragelaphus angasi). The grazer-
browser distinction is replicated among the African rhinos, with the white rhino
(Ceratotherium simum) being exclusively a grazer and the black rhino (Diceros
bicornis) almost entirely a browser (Owen-Smith 1988). All extant zebras (Equus spp.)
are exclusively grazers.
The grazer-browser separation is less clear among temperate zone ungulates.
Although primarily grazers, bison (Bison bison and B. bonasus), feral cattle
(Bos primigenius), and even feral horses consume a higher proportion of dicots
than is typical of grazers in African ecosystems (Owen-Smith 1997; Putman
1986). Sheep and goats (Caprinae) consume a variable mix of grass and low
browse (shrubs and herbs). Most deer (Cervidae) are mainly or entirely browsers,
but red deer (Cervus elaphus) and North American elk (or wapiti; C. e. canadensis)
as well as fallow deer (Dama dama) may obtain the greater portion of their summer
diet from grass (Gebert and Verheyden-Tixier 2001). The lack of strict grazers
occurs also in tropical Asia and South America, as well as among Australian
macropods. Hence the contrast I will draw is between species that are exclusively
or mainly grazers, and those that are largely or entirely browsers. Ungulates occu-
pying tropical forests, deserts or tundra, mountainous regions, and tropical Asia
plus South America, will largely be ignored, because of the absence of grazers for
comparison in these environments. Accordingly, this review is focussed mostly on
© Springer 2008
150 N. Owen-Smith
a comparison between African ungulates inhabiting savannas, grasslands and

shrublands, and their temperate zone counterparts from North American or
European woodlands, grasslands, and steppe.
The topics of relevance to this chapter include the spatio-temporal dynamics of
the food resource, population abundance levels and their temporal variability, spatial
and demographic structure of populations, density dependent or independent
processes affecting population dynamics, and regional synchrony in population
fluctuations. Special attention will be paid to the reputed propensity of ungulate
populations to show eruptive dynamics, i.e., rapid growth to densities severely
impacting vegetation resources, followed by population crashes (Riney 1964;
Caughley 1970, 1976a; McShea et al. 1997; Forsyth and Caley 2006). This
phenomenon is often used as justification for culling or other management intervention
to curtail population growth or abundance.
The specific questions to be addressed are these:
1. What distinctions in food resource dynamics in quantity and quality underlie
potential differences in population features between grazers and browsers?
2. How do the abundance levels attained by grazers and browsers differ, independently
of influences from body mass?
3. Does the spatial, social, and demographic structure of populations differ
substantially between these feeding categories?
4. Are the feedback mechanisms regulating population dynamics different in any
consistent way between these feeding types?
5. Which feeding category is most susceptible to eruptive or oscillatory
dynamics?
6. Do these feeding types differ in their susceptibility to broad- or fine-scale
weather influences on population dynamics?
6.2 Spatial and Temporal Dynamics of Grass and Browse
In African savanna ecosystems, grass growth is determined largely by rainfall,

occurring mostly within a restricted wet season. During the dry season, grass leaves
and stems become converted progressively into dead tissues or ‘necromass’. Figure 6.1
compares the seasonal phytomass dynamics of the grass layer for a subtropical
South African savanna, with a single six-month rainy season, with that for equatorial
East African grasslands having one major and one minor rainy period. All of the sites
were subject to grazing, albeit at somewhat different levels. The grass layer showed
a general pattern of green biomass accumulation during the main rainy season,
followed by its progressive conversion to dead leaves and stems. Under light grazing,
the total standing crop of potential food for grazers declines little between the growing
season and the end of the dry season, as shown by the South African site (Fig. 6.1A).
However, the amount of green leaf remaining in this material by the late dry season
is quite small, and diluted among the standing dead biomass. In very dry years, the
6 The Comparative Population Dynamics of Browsing and Grazing Ungulates 151
green fraction may fade to zero, except in bottomlands where the soil retains some
moisture (Prins 1988). Dry season fires alter this pattern by removing most of the
above-ground material, thereby promoting a flush of leafy regrowth if there is sufficient
soil moisture (van de Vijver et al. 1999). Without burning or heavy grazing, the
Fig. 6.1A–D Monthly biomass dynamics of green (live) and brown (dead) grass forage in
representative regions. A Lightly grazed grass layer in Burkea savanna in Nylsvley Nature
Reserve, South Africa; annual rainfall during study period 600 mm (from Grunow et al. 1980,
averaging across years). B Grazed grassland in Nairobi National Park, Kenya; annual rainfall
during study period 800 mm (from Boutton et al. 1988a, averaging across sites).
152 N. Owen-Smith
Fig 6.1 (continued) C Grassland in Masai Mara Reserve, Kenya, heavily grazed especially by
migratory ungulates between July and December; annual rainfall during study period 1200 mm (from
Boutton et al 1988a, averaging across sites). D Grassland on Isle of Hirta, Scotland, heavily grazed by
sheep; mean annual rainfall 1200 mm (from Milner and Gwynne 1974, averaging across sites)
accumulated dead plant parts tend to be carried over into the next growing season, as
is evident in Fig. 6.1A and B. In temperate regions, moisture from winter snow or rain
generally enables grass growth to commence as soon as temperatures rise sufficiently
in spring, leading to a peak in standing phytomass by mid-late summer (Fig. 6.1D).
Total phytomass production seems to be less than that of the African savanna sites,
perhaps because of slower growth, but a much larger fraction may remain green,
especially under wet maritime conditions. Nevertheless, extensive fires in late sum-
mer were formerly a feature of temperate grasslands, as well as many woodland
types, and remain so where not controlled by humans.
Annual variability in rainfall leads to quite wide fluctuations between years in
herbaceous production in African savanna and grassland ecosystems (Le Houerou
et al. 1988; O’Connor et al. 2001). For the semi-arid grassland depicted in Fig. 6.2,
the coefficient of variation (CV) in peak phytomass was nearly double that in
rainfall, and grassland judged to be in poor condition, from its cover and species
composition, produced less forage for the same rainfall (O’Connor et al. 2001).
Forage production seems somewhat less variable annually in mesic grasslands of
the Northern hemisphere than in these tropical or subtropical grasslands (Knapp
and Smith 2001).
The nutritional quality of grass forage in African savanna and grassland ecosystems,
as indexed by crude protein (nitrogen content × 6.25), drops substantially after the
end of the rainy season even in remaining green leaves (Fig. 6.3). There are periods
when the crude protein content of the grass falls below 5 or 6%, the assumed mini-
mum maintenance requirement for a large ruminant (Robbins 1993). East African
sites exhibited a higher grass quality during the short rains than during the main
rainy season (Fig. 6.3A), probably as a result of the nitrogen flux supporting initial
regrowth (Prins 1988). Crude protein levels in temperate (predominantly C3)
grasses are generally higher than in tropical (mostly C4) grasses, although the
sheep-fertilised meadows on Hirta (Fig. 6.3B) are perhaps an extreme example
300
Standing biomass (g/m2)
200
100
0
200 300 400 500 600 700 800 900
Annual rainfall (mm)
Fig. 6.2 Between-year variability in grass standing biomass at the end of the growing season for
a Themeda grassland in good condition in the Free State in relation to annual rainfall; mean annual
rainfall 560 mm (from O’Connor et al. 2001)
154 N. Owen-Smith
(but see Ydenberg and Prins 1981). Individual grass species can differ quite widely
in their nutritional value, with better quality grasses remaining above the 5% crude
protein level for most of the year and poor quality species falling below this level
except for a few months during the wet season (Fig 6.3C; see also Prins and
A
12 Live, Nairobi Dead, Nairobi Live, Mara Dead, Mara Maint Req
Crude Protein (% of DM)
10
2
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb
Month
B
18 80
Crude Protein Digestibility
16 70
Crude protein (% DM)
Digestibility (%OM)
14 60
12 50
10 40
8 30
Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar
Month
Fig. 6.3A–E Seasonal variability in the nutritional quality of grass forage. A Changes in crude
protein contents in live (green) and dead (brown) grass at two sites in Kenya (from Boutton et al.
1988b). B Seasonal changes in crude protein content and digestibility of grass (‘pinch samples’)
on Isle of Hirta, Scotland (from Milner and Gwynne 1974)
C
D decumbens P maximum E rigidior Mainten level Rainfall
16
14
Crude Protein (% of DM)
200
12
Rainfall (mm)
10
6 100
0 0
Aug Sep Oct Nov Dec Jan Feb Mar Apr May June Jul
Month
D
90 D decumbens P maximum E rigidio Mainten leve Rainfall
Digestibility (% of DM)
80 200
Rainfall (mm)
70
60 100
50
40 0
Aug Sep Oct Nov Dec Jan Feb Mar Apr May June Jul
Month
Fig. 6.3 (continued) C Seasonal changes in crude protein contents for individual grass species in
Botswana, comparing good (Panicum maximum), intermediate (Digitaria decumbens), and poor
(Eragrostis rigidior) species (Mosienyane 1979). D Seasonal changes in digestibility comparing
individual grass species in Botswana (Mosienyane 1979).
Beekman 1989). Changes in digestibility largely reflect the changes in crude protein
(Figs. 6.3B, C, and D), because the build-up of indigestible fibre affects both.
Through selective foraging for grass species and green leaves over dry leaves
and stems, grazers may maintain the nutritional value of the forage they consume
above the minimum need throughout the year, even in predominantly poor quality
grassland (Fig. 6.3E). The nutritional value of the prevalent grass species is affected
156 N. Owen-Smith
E
14
Eaten Available
12
Crude protein (% DM)
10
0
Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep
Month
Fig. 6.3 (continued) E Comparison between crude protein contents in the grass consumed by
cattle (from fistula samples) and that in the available grassland in the Nylsvley Nature
Reserve (from Zimmerman 1980 cited by Scholes and Walker 1993)
by soil fertility, dependent largely on clay minerals retaining cations against the
forces of leaching. Nevertheless, even in regions with predominantly poor soils
there are local sites where better quality grasses prevail, such as under tree canopies
and in drainage sumps where nutrients accumulate (Ludwig et al. 2004).
In African savannas the majority of woody plants are deciduous, shedding their
leaves during the dry season. The evergreen or semi-evergreen species that retain
leaves through the dry season tend to have sclerophyllous foliage, defended chemi-
cally as well as high in fibre. Accordingly, an acute bottleneck in the leafy browse
remaining on woody plants can develop by the end of the dry season (Fig. 6.4A and
B). Forbs may contribute some herbaceous browse, but mostly wither away during
the dry season. Leaves shed from higher layers of trees can provide additional for-
age, especially for small to medium-sized browsers, but soon become scattered and
degraded (Owen-Smith and Cooper 1985). Browse availability can vary widely
across the landscape, as well as between ecosystems, depending on the tree canopy
cover. In Serengeti, the peak browse biomass in open woodland of the midslope
region was about one tenth of that for upland regeneration thicket (Fig. 6.4B).
Drainage line thickets produced almost three times as much browse as the upland
woodland, and retained more forage through the dry season due to a greater evergreen
component (Pellew 1983).
Because most shoot and leaf growth on woody plants takes place prior to or
early in the wet season, browse production depends more on the rainfall of the
previous season than on the current season’s rain, and appears more constant over
Fig. 6.4A, B Seasonal biomass dynamics of available browse. A Burkea savanna in Nylsvley
Nature Reserve, South Africa, amount available below 2.5 m to kudu, distinguishing palatable
deciduous, unpalatable deciduous and evergreen tree species plus forbs (from Owen-Smith and
Cooper 1989). B Regenerating acacia thicket on ridge crest in Serengeti National Park, Tanzania,
available below 5.75 m to giraffe (from Pellew 1983)
the years than grass production (Rutherford 1984). Leaf retention through the dry
season is influenced by rain falling in the latter part of the wet season. In drought
years, leaf abscission occurs early and the leaf flush at the start of the next season
is delayed, prolonging the period of food stress for browsers. Herbaceous dicots
respond more directly to rainfall, and some species initiate growth ahead of the
158 N. Owen-Smith
rains (personal observations). Fires can promote a flush of forbs, although these
later become shaded out by grass regrowth.
In temperate regions, deciduous trees and shrubs are of variable nutritional value
(Robbins and Moen 1975), while evergreen species are generally less nutritious due
to high fibre or secondary chemical contents (Renecker and Hudson 1988; Klein
1990). During winter, browsers depend largely on dormant buds of deciduous species,
terpene-rich needle-like leaves of conifers, or evergreen shrubs such as heather
(Belovsky 1981; Saether and Anderson 1990; Tixier and Duncan 1996; Lothan et al.
1999). Highest browse availability occurs at the margins of woodland patches, and in
woodlands regenerating after fire. The production of summer browse within reach
of moose (Alces alces) in the form of leaves of deciduous shrubs amounted to
10–30 gDM/m2 in southwestern Quebec, similar to values reported for other forests
at comparable latitudes in North America and Europe (Crete and Jordan 1982).
Browse remaining available during winter in the form of twigs of deciduous species,
plus needles and twigs of balsam fir, amounted to about 20% of the summer estimates.
Mediterranean-type vegetation shows a different seasonal pattern, with most
woody plant growth occurring during the wet winter. Evergreen trees and shrubs
constituted most of the browse consumed by red deer during the dry summer
(Pappageourgiou 1978; Bugalho and Milne 2003).
Crude protein levels in tree leaves tend to be higher and more consistent than
those in grasses, especially for the thorny Acacia species that occur widely in
African savannas (Fig. 6.5). Unpalatable deciduous species prevalent on infertile
PalSpin PalDec UnpDec Evergr Rainfall

24 150
Crude Protein (% DM)
20
100
16
Rainfall (mm)
12
50
4 0
Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep
Month
Fig. 6.5 Seasonal changes in the nutritional quality of tree foliage in Nylsvley Nature Reserve,
distinguishing (1) palatable spinescent species, mainly Mimosaceae; (2) palatable deciduous species,
including Combretaceae and Tiliaceae; (3) unpalatable deciduous species, Caesalpiniaceae and other
families; and (4) evergreen species, various families (from Owen-Smith 1994)
soils offer protein levels not much different from those in evergreen foliage.
Although crude protein contents in leaves rarely drop below 5% of dry mass, the
digestibility of tree leaves tends to be somewhat less than that of grasses. The fibre
component is more lignified, while tannins and other secondary chemicals may
additionally reduce digestibility (Bryant et al. 1991; Owen-Smith 1993a).
Herbaceous plants with weak supporting stems, especially annuals and creepers,
offer some of the highest nutritional yields, but may also contain toxic chemicals
(Levin 1976). The range in crude protein levels in the leaves of northern trees in
late summer seems basically similar to that recorded for African savannas, e.g.,
12–16% for mountain maple (Acer spicatum) and 10% for balsam fir (Abies balsa-
mea) in Quebec (Crete and Jordan 1982). Protein levels in twigs browsed by moose
in winter are much lower, around 5–6.5%.
6.3 Population Density Levels
The triangular allometric relationship documented for birds by Brown and Maurer
(1987), with density declining with changing body mass both below and above some
pivotal size, is shown also by ungulates (Fig. 6.6). Above a mean body mass of around
A
100 Grazers
Browsers
Mixed
Cervids
Northern grazers
Allometric constraint
Density (/km )
2
10
1
1 10 100 1000 10000
Body mass (kg)

Fig. 6.6A–D Allometric scaling of abundance levels for various ungulate species, assessed via
log–log plots relative to body-mass (M) distinguishing feeding categories. Points represent
averages for each species across localities where data available. A Local numerical densities
160 N. Owen-Smith
35 kg (standardised as three-quarters of adult female mass at population level), the

upper bound to the local or “ecological” population density declines with increasing
body mass, following the allometric trend in energy metabolism as suggested by
B
100000
10000
Biomass (kg/km2)
1000
Grazers
Browsers
100
Mixed
Cervids
Northern grazers
10 Constraint - grazers
Constraint - browsers
1
1 10 100 1000 10000
Body mass (kg)
C
100
Grazers
Browsers
Mixed
Cervids
Northern grazers
Density (/km2)
10 Allometric line
0
1 10 100 1000 10000
Body mass (kg)
Fig. 6.6 (continued) D within the extent of a home range. Line represents allometric scaling
assuming each species uses a constant proportion of the food produced: D = aM−0.75. B Local
biomass (B) densities; lines represent the equivalent allometric scaling, i.e., B = aM0.25, for
grazers and browsers separately. C Regional numerical densities across the extent of the park
or other survey area, including unoccupied regions; line as in A.
D
10000
1000
Biomass (kg/km2)
100
Grazers
Browsers
Mixed
Cervids
10 Northern grazers
Constraint - grazers
Constraint - browsers
1
1 10 100 1000 10000
Body mass (kg)
Fig. 6.6 (continued) D Regional biomass densities, lines as in B
Damuth (1981). Below this mass, numerical abundance also decreases as body mass is
reduced, even when assessed at a local scale. Dikdik (Madoqua kirki) are a notable
exception: the local density within pair territories covering less than a hectare amounts
to over 100 animals per km2. Furthermore, many species do not approach the abun-
dance levels attained by other species of similar size, as shown by the numerous points
falling well below the outer bounds to the triangular distribution. Highest numerical
densities are manifested by medium-sized grazers occupying floodplain or otherwise
locally productive habitats in Africa, as well as by similar-sized deer in North America.
Wildebeest (Connochaetes taurinus) achieve this density within the Serengeti
ecosystem, but not consistently elsewhere. The highest ecological density of around
200 animals per km2 is for feral Soay sheep (Ovis aries) on Hirta (in the Scottish
St. Kilda archipelago).
In the lower body size range, grazers and browsers appear not to differ in the
density levels that they show at either local or regional scale (Fig. 6.6). However,
the maximum abundance levels attained by larger browsers seem to be an order of
magnitude lower than those exhibited by similar-sized grazers in African ecosys-
tems. This is most clearly evident when local abundance is plotted as biomass
rather than numerical density (Fig. 6.6B). Biomass density rises initially with
increasing body mass before reaching the upper asymptote imposed by the scaling
of energy requirements with body size. Impala and nyala, the only clearly interme-
diate feeders, seem intermediate also in their biomass density. Among temperate-
zone species, mixed feeders like red deer and wapiti, as well as browsers like
white-tailed deer (Odocoileus virginianus) and mule deer (O. hemionus), show
162 N. Owen-Smith
local biomass densities matching those of African grazers such as wildebeest,

buffalo (Syncerus caffer) and white rhino, taking into account metabolic scaling.
However, the largest cervid, moose, falls closely along the browser line.
Crude densities assessed at a regional scale incorporate unoccupied regions,
especially for populations with restricted distributions. Accordingly, regional density
or biomass levels are generally lower than local ecological densities. Grazer-browser
distinctions remain little altered, but the density of some of the smallest species is
reduced more at the regional scale than is the case for larger species (Figs. 6.6B and D).
Regional biomass densities generally increase with annual rainfall for most African
ungulate species, both grazers and browsers (East 1984).
6.4 Demographic Patterns
Large grazers may aggregate in vast herds, greatly outnumbering those formed by
similar-sized browsers. African buffalo herds can exceed 2,000 (Sinclair 1977;
Prins 1996), while the largest eland (Taurotragus oryx) herd was just over 400
(Hillman 1987). Browsing ungulates under 35 kg in mean body mass are mostly
solitary, apart from female–offspring associations, while grazing oribi (Oerebia
oerebi) and mountain reedbuck (Redunca fulvorufula) form small herds of 5–12,
and Thomson’s gazelle aggregates of 50 or more. Only among certain small brows-
ers do females maintain exclusive territories, sometimes shared with a male partner
(e.g., dikdik, klipspringer Oreotragus oreotragus). Territories defended by males
for mating purposes are a common feature of grazing ungulates, but not among
browsers, probably because females of the larger browsers do not attain sufficient
local densities to support such a strategy (Jarman 1974, Owen-Smith 1977).
Contrasting patterns have been recorded in the extent of the home range covered
seasonally by African grazers and browsers. Grazers including buffalo (Funston et
al. 1994, Ryan et al. 2006), white rhino (Owen-Smith 1975) and plains zebra
(Equus burchelli; Klingel 1969) move over a wider area in the dry season than in
the wet season, even excluding excursions to water. In contrast, browsers like eland
(Hillman 1988), kudu (Tragelaphus strepsiceros; personal observations) and black
rhino (Goddard 1967) contract their home ranges during the dry season. This could
reflect spatial differences in seasonal predictability of food resources. During the
dry season, grazers may opportunistically seek out areas where localised rain-
showers have produced areas of green regrowth (Talbot and Talbot 1963). In con-
trast, trees do not respond to dry season rainfall, while the localities where evergreen
foliage persists through the dry season are fixed in the landscape. Among temperate
zone ungulates, moose (Cedarlund and Okarma 1988; Mysterud et al. 2001) and
European red deer (Clutton-Brock et al. 1992; Mysterud et al. 2001) generally fol-
lowed the browser pattern, with a contraction in home range during winter.
However, roe deer (Capreolus capreolus; Kjellander et al. 2004; Mysterud et al.
2001), mule deer (Mysterud et al. 2001) and North American wapiti (Anderson et
al. 2005), as well as bison (McHugh 1958), showed an expansion in home range
during the winter months, while white-tailed deer showed no consistent pattern
(Mysterud et al. 2001).
Large-scale migrations are a feature of certain grazer populations, notably wilde-
beest, zebra, topi (Damaliscus lunatus), and Thomson’s gazelle (Gazella thomsoni) in
the Serengeti (Maddock 1979), and elsewhere (Jewell 1972; Williamson et al. 1988)
white-eared kob (Kobus kob leucotis) in Sudan (Fryxell and Sinclair 1988), and in
former times springbok (Antidorcas marsupialis) in South Africa (Skinner 1993).
Among temperate zone ungulates, extensive migrations are also shown by saiga ante-
lope (Saiga tatarica; Bekonov et al. 1998) and Mongolian gazelles (Procapra guttur-
osa; Jiang et al. 2002), which feed partly on grasses as well as herbs and shrubs.
Migratory wildebeest in the Serengeti vastly outnumber the resident wildebeest sub-
population (Fryxell et al. 1988). Where migratory routes have been blocked by fences
or settlements, substantial reductions in population have resulted, e.g., for wildebeest
in the southern Kalahari (Williamson and Mbano 1988), Etosha (Gasaway et al. 1996),
and Kruger Park (Whyte and Joubert 1988). Mainly browsing eland roam widely, but
without any regular migratory pattern (Hillman 1988). Most African browsers are
fairly sedentary, without seasonally distinct home ranges, but so are many grazer popu-
lations. Temperate zone browsers and mixed feeders commonly shift their home
ranges seasonally along altitudinal gradients in mountainous regions, to take advantage
of differences in plant phenology (Craighead et al. 1972; Schoen and Kirchoff 1985;
Albon and Langvatn 1992; Histol and Hjeljord 1994; Mysterud 1999).
Population composition in terms of age and sex classes is largely an outcome of
the recruitment rate, governed by litter size and phase of population growth. All
African bovids produce just a single offspring annually, irrespective of whether they
are grazers or browsers (although springbok can reproduce twice during an annual
cycle; Skinner and Louw 1996). Among cervids, species in the New World sub-
family Odocoilinae commonly produce twins or triplets, while Old World deer in the
Cervinae mostly give birth to single offspring. Sheep and goats are mostly polyto-
cous, although the extent of twinning differs among populations. Differences in litter
size, and hence in the juvenile proportion in the population, determine the maximum
population growth rate that can be sustained. This is around 25% per year for species
producing a single young, but up to 50% per year for roe deer, white-tailed deer and
moose (Anderson and Linnell 2000; McCullough 1997). Most ungulates show a
strongly female-biased sex ratio in the adult segment, indicating higher male than
female mortality, but no grazer-browser distinction is evident (Owen-Smith 1993b;
Berger and Gompper 1999; Owen-Smith and Mason 2005).
6.5 Regulation of Abundance
For small antelope, territories maintained by both sexes may limit abundance rela-
tive to localised food resources, to the extent that these territories cannot be com-
pressed. Among larger ungulates, direct contests for food are rare, because of the
wide distribution and generally low quality of vegetation resources (Prins 2000).
164 N. Owen-Smith
Intraspecific competition arises largely indirectly through the effects of exploitation

for subsequent food availability during the dormant season. There may also be
longer-term consequences for vegetation structure and composition affecting the
productive potential of plants. Because interference with feeding may only be mani-
fested after crowding has reached high levels, and indirect consequences via vegeta-
tion impacts are delayed, a convex or plateau-and-ramp pattern of reduction in the
population growth rate with increasing density is generally expected (Fowler 1981,
1987; McCullough 1992, 1999). However, the trend in population growth rate may
appear approximately linear around the zero growth level, depending on how the
density feedback affects different population segments (Owen-Smith 2006).
Density influences on population growth can be obscured by errors in the
abundance estimates, climatic fluctuations and variable age structure, as well as
by non-linear or delayed responses (Bonenfant et al., unpublished manuscript).
The density effect may be revealed only after controlling for the influence of vari-
able rainfall on food availability (Owen-Smith 1990). Density dependence is
most clearly evident in body mass (Solberg et al. 2004) and the consequences of
growth for rates of reproduction, in particular for the age at first parturition
(Bonenfant et al. unpublished manuscript). Mortality losses among juveniles and
old adults also generally rise with increasing population density, while the sur-
vival of prime-aged females changes little until high abundance levels are attained
(Gaillard et al. 2000). The high sensitivity of juvenile survival to food shortfalls
as well as impacts of predation may obscure the density dependence in popula-
tion growth (Owen-Smith et al. 2005). Most of the change in population growth
rate towards the zero growth level may be due to the reduction in adult survival,
but with a shift in the age structure of the population towards older animals with
lowered survival chances also contributing (Festa-Bianchet et al. 2003; Owen-
Smith et al. 2005; Owen-Smith 2006).
No consistent grazer-browser distinction in the sensitivity of population growth to
either density or rainfall was evident among ungulate species in the Kruger Park
(Mills et al. 1995; Ogutu and Owen-Smith 2003, Owen-Smith and Mills 2006).
Browsing kudus responded positively to rainfall, along with buffalo, waterbuck
(Kobus ellipsiprymnus) and other less common grazers. Giraffe (Giraffa
camelopardalis) seemed as little affected by rainfall variability as grazing wildebeest
and zebra, while impala appeared intermediate.
Predation may either mask the influence of resource availability on population
dynamics, or accentuate it, depending on the extent to which the mortality loss
imposed is additive and how it depends on nutritional status (Chapter 10 in
Owen-Smith 2002a; Owen-Smith and Mills, submitted). Predation pressure
(Klein 1965; Prins and Iason 1989), and in some situations disease impacts (Prins
and Weyerhauser 1987), may hold ungulate populations below the food ceiling,
or cause lagged population responses to changing density levels (Owen-Smith
and Mills, submitted). The risk of predation can amplify the effects of nutritional
deficits on population growth by affecting foraging behaviour (Sinclair and
Arcese 1995). Migratory populations may escape predator control, while resident
populations can be held in a “predator pit” below the abundance level that could
be sustained by resources (Fryxell et al.1988; Gasaway et al. 1992), especially if

human hunting is imposed in addition to predation (Messier and Crete 1985). An
alternative prey species may be reduced to low abundance by a predator population
sustained by a more abundant species, e.g., for non-migratory caribou (Rangifer
tarandus) and moose, with wolves (Canis lupus) as the predator (Seip 1992;
McLoughlin et al. 2003), and for less common ungulate species in the Kruger Park,
including both grazers and browsers (Owen-Smith and Mills 2006).
6.6 Population Dynamics
Large herbivores have the potential to “irrupt” to population levels where their
consequent feeding impact on vegetation resources depresses food availability,
precipitating a population crash (Caughley 1976a). A book addressing “the science
of overabundance” focussed on white-tailed deer and mule deer in North America
as the exemplars (McShea et al. 1997). Most of the examples of population irrup-
tions and crashes come from island situations where dispersal was precluded, and
usually involve animals introduced in the absence of predators (McCullough 1997;
Kaji et al 2004; Forsyth and Caley 2006). The moose population on Isle Royale
(Lake Superior) has shown repeated oscillations in abundance despite the presence
of wolves (Peterson 1999), but the effectiveness of the wolves as predators is lim-
ited by other factors (Peterson et al. 1998). Although theoretical models suggest
that extreme peak densities could result in persistent vegetation degradation,
observations indicate no change in subsequent peak abundance levels attained by
oscillating deer populations (McCullough 1997).
While the above deer are mainly browsers, feral Soay sheep which are
largely grazers, have shown persistent oscillations in abundance over a greater-
than-twofold range in the St Kilda islands (Clutton-Brock et al. 1991). In con-
trast, mixed-feeding red deer inhabiting a similar island environment on Rum
(Inner Hebrides, Scotland) have remained relatively stable in abundance
(Clutton-Brock et al. 1997; Clutton-Brock and Coulson 2002). Contributing to
the susceptibility to irruptions of both the deer and the sheep is a high reproduc-
tive potential. White-tailed deer and mule deer, as well as moose, commonly
produce twin offspring as well as undergoing first parturition at a young age.
Soay sheep also first reproduce at an early age, but twin offspring are infre-
quent (Clutton-Brock et al. 1997). Nevertheless, the last two months of gesta-
tion precede the spring regrowth of grasses, so that in favourable years the high
recruitment potential coupled with the absence of any density feedback during
summer elevates the population well above the level supported by the food
resources by late winter, precipitating a die-off. Supporting the high growth
potential of the sheep is the high quality of the grasses growing in the lush
meadows fertilised by sheep manure. Furthermore, heather may be less effec-
tive as a buffer resource for the sheep than for red deer (Chapter 13 in Owen-
Smith 2002a).
166 N. Owen-Smith
Since browsers appear to be limited mainly by the amount of food remaining

during the adverse season (Sect. 6.2 above), they should be more prone to fluctua-
tions in abundance than grazers, limited more by seasonal changes in food quality
(Chapter 13 in Owen-Smith 2002a). Nevertheless, feral cattle and ponies, which are
mainly grazers, as well as mixed-feeding red deer, showed unusually high (15–30%)
winter mortality in the Oostvadersplassen Reserve in Netherlands, following popula-
tion growth towards high density levels (S. van Wieren, personal communication).
These deaths occurred during exceptionally cold weather, with ice crusts restricting
access to forage. Severe cold weather was also directly responsible for precipitating
population crashes by Soay sheep (Coulson et al. 2001) and white-tailed deer
(Patterson and Power 2002).
For African ungulates, the extreme weather conditions leading to major die-offs
entail a lack of rainfall, restricting food availability. During the 1982/1983 El Nino-
related drought, populations of six grazing ungulates crashed to 20% or less of their
former abundance in the Klaserie Private Nature Reserve, adjoining the Kruger
Park (Walker et al. 1987). In this case the two large browsers (kudu and giraffe)
were less affected, dropping by 30–50% in abundance, while mixed-feeding impala
showed a 75% decline. Contributing to the severe mortality was the close spacing
of waterpoints on this private land, enabling animals to access almost all remaining
grass, until virtually no forage remained. Trees started producing new foliage ahead
of the rains, alleviating food stress for the browsers in this crucial period. Ungulate
populations within the vast Kruger Park, where waterpoint spacing was somewhat
wider, decreased much less during this same drought (Walker et al. 1987).
Wildebeest and zebra were almost unaffected, with the greatest declines by 30–40%
shown by browsing kudu and intermediate-feeding impala, as well as grazing
waterbuck. Over the 13-year period spanned by park-wide surveys of Kruger Park,
which included this and a subsequent extreme drought, the coefficient of variation
(CV) in census totals was consistently larger for the three browsers than for grazers
of similar body size (Fig. 6.7). Impala showed a relatively low CV, probably partly
because sampling error was lowered as a consequence of their large population
total, but notably they were also subject to predation by numerous predator species
(Owen-Smith and Mills, submitted).
During an earlier drought in 1965, starvation-related mortality reduced populations
of browsing giraffe and kudu by about a third in the Timbavati Private Nature
Reserve, bordering Kruger Park, while no starvation occurred among grazing
wildebeest, zebra, and impala (Hirst 1969). In the southern Kalahari, a drought during
1985 was associated with 35% mortality among mainly browsing eland, compared
with 12–18% among grazing wildebeest and hartebeest (Alcelaphus buselaphus),
and little mortality among springbok (Knight 1995).
Kudu are notoriously susceptible to die-offs when cold and wet weather occurs
near the end of the dry season (Wilson 1970, Owen-Smith 2000). For this sub-tropical
species, extreme cold is a maximum daily temperature under 15°C, coupled with
wind and rain accentuating the loss in body heat at a stage when body reserves are
low. Kudu need to forage for extended periods during this generally hot time of the
year when food availability is at a minimum, and thus must tolerate high temperatures
CV in abundance 0.4 Grazers Browsers Mixed Log. (Grazers) Log. (Browsers)
0.3
0.2
0.1
0
10 100 1000
Body mass
Fig. 6.7 Coefficient of variation in annual census totals for ten ungulate species in the Kruger Park
over 1980–93 in relation to body mass, distinguishing feeding types. Semi-log trend-lines are
indicated
at the expense of insulation from cold (Owen-Smith 1998). Die-offs following

cold weather have been recorded for browsing giraffe (Walker et al. 1987) and
intermediate-feeding nyala (Keep 1973). Among grazers, a population crash associ-
ated with cold weather has been documented for reedbuck (Redunca arundimum;
Ferrar and Kerr 1971), but among African buffalo cold-related mortality was
restricted to calves (Jolles personal communication).
Ungulate population die-offs have been recorded even within large, unfenced
protected areas in East Africa. A drought in 2000 resulted in the death of at least
1,500 buffalo, representing 30% of the prior population, in the Ngorongoro Crater
in Tanzania (Estes et al. 2006). However, declines by similar proportions in wilde-
beest and zebra appeared to be due more to movements of animals out of the crater
rather than to mortality. The severe drought of 1993 led to a drop by 75% in the
buffalo population within Maasai-Mara and adjoining group ranches in Kenya
(Ottichilo et al. 2000), and brought about a 22% decline in the migratory wildebeest
population in the Serengeti (Mduma et al. 1999).
Overall, these observations suggest that African browsers are susceptible to fre-
quent but relatively minor die-offs, keeping populations below levels where severe
crashes might be precipitated. In contrast, grazers may be subject to less frequent
but potentially more substantial population declines during major droughts.
Observations on domestic ungulates suggest that the vulnerability of grazers to
large-scale mortality during droughts may be dependent on soil fertility, controlling
the general palatability of grasses (Chapter 13 in Owen-Smith 2002a). On fertile
soils where grasses are mostly palatable, almost all of the grass layer may be con-
sumed in a drought year, so that cattle die through lack of food (Scoones 1993;
Hatch and Stafford Smith 1997). On infertile soils, the less palatable grasses that
168 N. Owen-Smith
are prevalent tend to remain uneaten, and thus serve as a buffer slowing starvation
during drought periods. Bottomlands where grasses retain green leaves through the
dry season may serve as key resource areas buffering herbivore populations against
severe declines (Prins 1988; Scoones 1995; Illius and O’Connor 1999, 2000).
Evergreen woody plants, also predominant especially along river margins, may
serve similarly to buffer population fluctuations in browsers like kudu and giraffe
(Owen-Smith and Cooper 1989). Notably, the population crash of moose on Isle
Royale was associated with conditions under which their main winter food, bal-
sam fir (Abies balsamea), had been severely depressed in abundance by previous
browsing pressure (Risenhoover and Maass 1987). However, the contribution of
different vegetation components can be somewhat more complex, with different
plant types serving as reserve, bridging or buffer resources under different condi-
tions (Owen-Smith and Cooper 1989; Chapter 11 in Owen-Smith 2002a; Knoop
and Owen-Smith 2006).
An additional problem with domestic livestock, mostly grazers, in savannas is a
shift in vegetation composition towards woody plants when high stocking levels
and consequent heavy grazing suppress fires—the problem of bush encroachment
(Walker et al. 1981). High abundance levels reached by wild grazers in the absence
of predation in the Hluhluwe Reserve in South Africa seem to have been the pri-
mary cause of the transformation of large areas of open grassland into thicket,
leading to greatly increased abundance of browsers and mixed feeders at the
expense of grazers (Brooks and Macdonald 1983).
Megaherbivores like rhinos and elephants are largely invulnerable to predation
as adults, and can have a corresponding ‘mega impact’ on vegetation. This raises
concerns that their population dynamics could show cyclic oscillations (Caughley
1976b; Owen-Smith 1981, 1988). The outcome depends largely on lags in the
recovery of plant populations to these impacts, and on how vegetation impacts are
distributed in space and over time. Modelling suggests that sufficient functional
heterogeneity in vegetation quality and forage retention during the adverse season,
assisted by opportunities for dispersal, could suppress the risk of oscillations
(Owen-Smith 2002a, 2002b, 2004).
6.7 Weather Patterns and Population Fluctuations
Large-scale weather patterns related to shifts in atmospheric pressure belts as a

result of changes in oceanic circulation patterns affect environmental condi-
tions, and hence potentially the population dynamics of herbivores (Saether
1997). The El Nino - Southern Oscillation (ENSO) linked to sea surface tem-
perature in the eastern Pacific Ocean generates either high rainfall or extreme
droughts in different regions when it intensifies at irregular intervals. Within
Africa, there is additionally a quasi-decadal alternation between wet and dry
phases in rainfall affecting the summer rainfall region of South Africa, with
conditions expressed in tropical East Africa being opposite to those in southern
Africa (Tyson and Gatebe 2001). In north temperate latitudes, the North
Atlantic Oscillation (NAO) and counterpart North Pacific Oscillation (NPO)
generate an alternation in weather patterns between years of relatively warm
conditions with high winter precipitation and hence deeper snow cover, and
colder winters with less precipitation and hence less snow, although the
regional consequences vary with altitude as well as latitude (Stenseth et al.
2002a, 2003).
These weather patterns can act on populations either by promoting mortality
through the physiological stress imposed by temperature or wind extremes, or by
affecting resource availability. High rainfall generally increases vegetation growth
and hence food availability, while deep snow as a result of high precipitation during
winter makes food less accessible. In African savannas, a lack of rainfall during the
normally dry season can greatly reduce the amount of green leaf persisting, and
hence food quality during this critical period (Mduma et al. 1999; Ogutu and
Owen-Smith 2003). Temperature changes can also affect the timing of plant growth
in spring in temperate latitudes, with potential effects on ungulate populations (Post
and Stenseth 1999).
Such broad-scale weather patterns can induce regional synchrony in population
trends, as documented for mixed-feeding red deer (Forschhammer et al. 1998) and
browsing roe deer in Norway (Grotan et al. 2005), as well as browsing moose and
white-tailed deer in North America (Post and Stenseth 1998). Deep snow also
increases susceptibility to predation by wolves (Post and Stenseth 1998; Hebblewhite
2005). For Soay sheep in the St Kilda islands, where snow is not a factor, warm
winters were adverse through being wet and windy, thereby increasing mortality
(Milner et al 1999) and at the same time synchronising populations on nearby
islands (Grenfell et al. 1998).
In African savanna ecosystems, high rainfall is associated with increased popu-
lations of most ungulates (Mills et al. 1995; Owen-Smith and Ogutu 2003; Ogutu
and Owen-Smith 2003, 2005). Hence individual species within the diverse assem-
blage of grazers and browsers in the Kruger Park fluctuated largely in parallel in
response to rainfall variability. However, wildebeest and zebra performed better
during the dry phase of the rainfall oscillation, due to changing susceptibility to
predation (Mills et al. 1995; Ogutu and Owen-Smith 2005; Owen-Smith and Mills,
submitted). In years of low rainfall there is less grass cover for stalking lions, and
herds can aggregate in more extensive short grassland, thereby diluting predation
pressure (Smuts 1978). This contrasts with the positive response to rainfall of a
zebra population in northern Kenya, where predation pressure is somewhat lower
(Georgiadis et al. 2003).
Ellis and Swift (1988; see also Ellis et al. 1993) suggested that in regions where
the coefficient of variation in rainfall is 30% or greater, herbivore populations do
not reach any equilibrium with the vegetation, and fluctuate widely in response to
droughts while rarely approaching carrying capacity. This standpoint was based
largely on observations on domestic livestock managed by nomadic pastoralists in
northern Kenya. Illius and O’Connor (2000) pointed out that key resource areas
during the dry season may largely maintain herbivore populations and incur the
170 N. Owen-Smith
brunt of grazing or browsing impacts, although populations may be largely decoupled

from resource limitations during the wet season. Towards high rainfall conditions
where the coefficient of variation in annual rainfall drops below 25%, herbivore
dynamics might be more closely coupled to vegetation dynamics, but forage quality
then becomes a potential limitation, especially for grazers.
6.8 Summary and Conclusions
Grazing and browsing ungulates depend largely on different components of the

vegetation, which show contrasting dynamics in quantity and quality seasonally
and between years. For grazers, the nutritional value of the dry grass forming the
bulk of the potential food persisting through the winter or dry season is low. For
browsers, the small amount of green foliage remaining towards the end of the dor-
mant season shows relatively less decline in nutrient contents than grasses. These
resource differences might be expected to produce distinctions in population
dynamics between grazers and browsers. However, the situation is not quite so
simple. Some green grass persists in moister regions of the landscape, while in fer-
tile ecosystems a large fraction of the dry grass may be of adequate nutritional
value. Much of the tree foliage remaining by the end of the dormant season is ever-
green and defended by digestibility-reducing chemicals. Furthermore, fires during
the dry season can greatly alter food availability for grazers.
Distinctions exist between grazers and browsers in maximum population density
levels, but only among larger species, and some grazers attain abundance levels no
greater than those of equivalent-sized browsers. Browsers and mixed feeders in
temperate woodlands can reach biomass levels comparable to those of abundant
grazers in African savannas. Grazers commonly expand their home ranges during
the dry season or winter months, and some populations migrate seasonally in
response to changing food and surface water availability. Browsers are generally
somewhat more sedentary and may contract their ranges when food availability is
restricted.
Density feedbacks largely arise indirectly via seasonal or annual depression of
food resources, and can potentially be delayed. No fundamental distinctions
between grazers and browsers are apparent here. Extreme or persistent oscillations
in abundance seem to be a feature of northern ungulates with high reproductive
potentials, or island situations where constrictions in spatial extent limit the ability
of herbivores to exploit resource heterogeneity. Quite severe die-offs in years when
adverse weather conditions restrict food availability, and perhaps also induce ther-
mal stress, occur even among populations inhabiting large protected areas, but may
be exacerbated when movements are restricted. Long term consequences for the
capacity of the vegetation to support these populations have not been documented,
except where the suppression of fire by heavy grazing has resulted in the transfor-
mation of savanna into thicket. African browsers appear to show more frequent but
relatively minor die-offs, while grazers may show greater population crashes at
longer intervals when severe droughts occur, especially in fertile ecosystems where
food quality is not much of a limitation. African browsers seem more susceptible
to mortality through hypothermia than syntopic grazers.
Large-scale climatic patterns can induce variability in populations either through
physiological stress or by affecting food availability, and be responsible for
regional synchrony in population fluctuations. Synchrony may be disrupted by
local variability in weather effects and relative vulnerability to predation under dif-
ferent conditions. Hence the consequences of the distinctions in quality, quantity
and temporal dynamics of grass and browse resources are complex, and do not
affect abundance levels of the ungulate species feeding on these resource types or
their population fluctuations and distribution in any simple way.
Acknowledgements I am indebted to Herbert Prins and Iain Gordon for helpful suggestions that
greatly improved this chapter.
References
Albon SD, Langvatn R (1992) Plant phenology and the benefits of migration in a temperate
ungulate. Oikos 65:502–613
Anderson DP, Forester JD, Turner MG, Frair JL, Merrill EH, Fortin D, Mao JS, Boyce MS (2005)
Factors influencing female home range sizes in elk in North American landscapes. Landscape
Ecol 20:257–271
Anderson R, Linnell JDC (2000) Irruptive potential in roe deer: density-dependent effects on body
mass and fertility. J Wildl Manage 64:698–706
Bekonov AB, Grachevand IuA, Milner-Gulland EJ (1998) The ecology and management of the
saiga antelope in Kazakhstan. Mammal Rev 28:1–52
Belovsky GE (1981) Food plant selection by a generalist herbivore: the moose. Ecology
62:1020–1030
Berger J, Gompper ME (1999) Sex ratios in extant ungulates: products of contemporary predation
or past life histories? J Mammal 80:1084–1113
Bonenfant C, Gaillard J-M, Loe LE, Loison A, Blanchard P, Garel M, Pettorelli N, Owen-Smith N,
du Toit J, Duncan P (unpublished manuscript, submitted) Empirical evidence of density
dependence in ungulates.
Boutton TW, Tieszen LL, Imbamba SK (1988a) Biomass dynamics of grassland vegetation in
Kenya. Afr J Ecol 26:89–101
Boutton TW, Tieszen LL, Imbamba SK (1988b) Seasonal changes in the nutrient content of East
African grassland vegetation. Afr J Ecol 26:103–116
Brooks PM, Macdonald IAW (1983) In: Owen-Smith RN (ed) The Hluhluwe-Umfolozi Reserve:
an ecological case history. Haum, Pretoria, pp.51–77
Brown JH, Maurer BA (1987) Evolution of species assemblages: effects of energetic constraints
and species dynamics on the diversification of the North American avifauna. Am Nat
130:1–17
Bryant JP, Provenza FD, Pastor J, Reichardt PB, Clausen TP, du Toit JT (1991) Interactions
between woody plants and browsing mammals mediated by secondary metabolites. Annu Rev
Ecol Syst 22:431–446
Bugalho MN, Milne JA (2003) The composition of the diet of red deer in a Mediterranean
environment: a case of summer nutritional constraint. Forest Ecol Manage 181:23–29
Caughley G (1970) Eruption of ungulate populations with emphasis on Himalayan tahr in New
Zealand. Ecology 51:53–72
172 N. Owen-Smith
Caughley G (1976a) Plant-herbivore systems. In: May RM. (ed) Theoretical ecology. Blackwell,
Oxford, pp 94–113
Caughley G (1976b) The elephant problem - an alternative hypothesis. E Afr Wildl J
14:265–283
Cedarlund GN, Okarma H (1988) Home range and habitat use of adult female moose. J Wildl
Manage 52:336–343
Clutton-Brock TH, Price OF, Albon SD, Jewell PA (1991) Persistent instability and population
regulation in Soay sheep. J Anim Ecol 60:593–608
Clutton-Brock TH, Guinness FE, Albon SD (1992) Red deer. behaviour and ecology of two sexes,
2nd edn. Edinburgh University Press, Edinburgh
Clutton-Brock TH, Illius AW, Wilson K, Grenfell BT, MacColl A, Albon SD (1997) Stability and
instability in ungulate populations: an empirical analysis. Am Nat 149:196–219
Clutton-Brock TH, Coulson T (2002) Comparative ungulate dynamics: the devil is in the detail.
Phil Trans R Soc Lond B 357:1285–1298
Coulson T, Catchpole EA, Albon SD, Morgan BJT, Pemberton JM, Clutton-Brock TH, Crawley
MJ, Grenfell BT (2001) Age, sex, density, winter weather, and population crashes in Soay
sheep. Science 292:1528–1531
Craighead JJ, Atwell G, O’Gara BW (1972) Elk migrations in and near Yellowstone National
Park. Wildl Monogr No. 29
Crete M, Jordan PA (1982) Production and quality of forage available to moose in southwestern
Quebec. Can J Forest Res 12:151–159
Damuth J (1981) Home range, home range overlap and energy use among animals. Biol J Linn
Soc 15:185–19
East R (1984) Rainfall, soil nutrient status and biomass of large African savanna mammals.
Afr J Ecol 22:245–270
Ellis JE, Swift DM (1988) Stability of African pastoral ecosystems: alternative paradigms and
implications for development. J Range Manage 41:450–459
Ellis JE, Coughenour MB, Swift DM (1993) Climatic variability, ecosystem stability and the
implications for range and livestock development. In: Behnke RH, Scoones I, Kerven C (eds)
Range ecology at disequilibrium. Overseas Development Institute, London, pp 31–41
Estes RD, Atwood JL, Estes AB (2006) Downward trends in Ngorongoro Crater ungulate populations
1986–2005: conservation concerns and the need for ecological research. Biol Cons 131:106–120
Ferrar AA, Kerr MA (1971) A population crash of the reedbuck in Kyle National Park, Rhodesia.
Arnoldia (Rhodesia) 5:1–9
Festa-Bianchet M, Gaillard J-M, Cote SD (2003) Variable age structure and apparent density
dependence in survival of adult ungulates. J Anim Ecol 72:640–649
Forschhammer M, Stenseth NC, Post E, Langvatn R (1998) Population dynamics of Norwegian
red deer: density dependence and climatic variation. Proc R Soc Lond B 265:341–350
Forsyth DM, Caley P (2006) Testing the irruptive paradigm of large-herbivore dynamics. Ecology
87:297–303
Fowler CW (1981) Density dependence as related to life history strategy. Ecology 62:602–610
Fowler CW (1987) A review of density dependence in populations of large mammals. In: Genoways
HH (ed) Current Mammalogy Vol. 1. Plenum, New. York, pp 401–441
Fryxell JM, Greever J, Sinclair ARE (1988) Why are migratory ungulates so abundant? Am Nat
131:781–798
Fryxell JM, Sinclair ARE (1988) Seasonal migration by white-eared kob in relation to resources.
Afr J Ecol 26:17–31
Funston PJ, Skinner JD, Dott HM (1994) Seasonal variation in movement patterns, home range
and habitat selection of buffaloes in a semi-arid habitat. Afr J Ecol 32:100–114
Gaillard J-M, Festa-Bianchet M, Yoccoz NG, Loison A, Toigo C (2000) Temporal variation in
fitness components and dynamics of large herbivores. Annu Rev Ecol Syst 31:367–393
Gasaway WC, Boertje RD, Grangaard DV, Kelleyhouse DG, Stephenson RO, Larsen DG (1992)
Predation limiting moose at low densities in Alaska and Yukon and implications for conservation.
Wildl Monogr No. 120
Gebert C, Verheyden-Tixier H (2001) Variations of diet composition of red deer in Europe.

Mammal Rev 31:189–201
Georgiadis N, Hack M, Turpin K (2003) The influence of rainfall on zebra population dynamics:
implications for management. J Appl Ecol 40:125–136
Goddard J (1967) Home range behaviour and recruitment rates of two black rhinoceros populations.
E Afr Wildl J 5:133–150
Grenfell BT, Wilson K, Finkenstadt BF et al (1998) Noise and determinism in synchronised sheep
dynamics. Nature 394:674–677
Grotan V, Saether B-E, Engen S, Solberg EJ, Linnelll JDC, Anderson R, Broseth H, Lund E
(2005) Climate causes large-scale spatial synchrony in population fluctuations of a temperate
herbivore. Ecology 86:1472–1482
Grunow JO, Groeneveld HT, du Toit HC (1980) Above ground dry matter dynamics of the grass
layer in the Nylsvley tree savanna. J Ecol 68:877–889
Hatch GP, Stafford Smith DM (1997) The bioeconomic implications of various drought management
strategies for a communal cattle herd in a semi-arid savanna of KwaZulu-Natal. Afr J Range Forage
Sci 14:17–25
Hebblewhite M (2005) Predation by wolves interacts with North Pacific Oscillation (NPO) on a
western North American elk population. J Anim Ecol 74:226–233
Hillman JC (1987) Group size and association patterns of the common eland. J Zool Lond
213:641–663
Hillman JC (1988) Home range and movement of the common eland in Kenya. Afr J Ecol 26:135–148
Hirst SM (1969) Populations in a Transvaal lowveld nature reserve. Zool Afr 4:199–230
Histol T, Hjeljord O (1994) Winter feeding strategies of migrating and non-migrating moose.
Can J Zool 71:1421–1428
Illius AW, O’Connor TG (1999) On the relevance of nonequilibrium concepts to arid and
semi-arid grazing systems. Ecol Applic 9:798–813
Illius AW, O’Connor TG (2000) Resource heterogeneity and ungulate population dynamics. Oikos
89:283–294
Jarman PJ (1974) The social organization of antelope in relation to their ecology. Behaviour 48: 15–267
Jewell PA (1972) Social organisation and movements of topi during the rut at Ishasha, Queen
Elizabeth Park, Uganda. Zool Afr 7:233–255
Jiang Z, Takatsuki S, Li J, Wang W, Gao Z, Ma J (2002) Seasonal variation in foods and digestion
of Mongolian gazelles in China. J Wildl Manage 66:40–58
Kaji K, Okada H, Yamanaka M, Matsuda H, Yabe T (2004) Irruption of a colonizing sika deer
population. J Wildl Manage 68:889–899
Keep, ME (1973) Factors contributing to a population crash of nyala in Ndumu Game Reserve.
Lammergeyer (Natal) 19:16–23
Kjellander P, Hewison AJM, Liberg O, Angibault J-M, Bideau E, Cargnelutti B (2004)
Experimental evidence of density dependence of home-range size in roe deer: a comparison of
two long-term studies. Oecologia 139:478–485
Klein DR (1965) Ecology of deer range in Alaska. Ecol Monogr 35:259–284
Klein DR (1990) Variation in quality of caribou and reindeer forage plants associated with season,
plant part, and phenology. Rangifer, Special Issue No. 3, pp.123–129
Klingel H (1969) The social organisation and population ecology of the plains zebra. Zool Afr
4:249–263
Knapp AK, Smith MD (2001) Variation among biomes in temporal dynamics of aboveground pri-
mary production. Science 291:481–485
Knight MH (1995) Drought-related mortality of wildlife in southern Kalahari and the role of man.
Afr J Ecol 33:377–394
Knoop M-C, Owen-Smith N (2006) Foraging ecology of roan antelope: key resources during
crucial periods. Afr J Ecol 44:228–236
Le Houerou HN, Bingham RL, Skerbek W (1988) Relationship between the variability of primary
production and the variability of annual precipitation in world arid lands. J Arid Envir
15:1–18
174 N. Owen-Smith
Levin DA (1976) Alkaloid-bearing plants: an ecogeographic perspective. Am Nat 110:261–284

Lothan J, Staines BW, Gorman ML (1999) Comparative feeding ecology of red and roe deer in a
Scottish plantation forest. J Zool Lond 247:409–418
Ludwig F, De Kroon H, Berendse F, Prins HHT (2004) The influence of savanna trees on nutrient,
water and light availability and the understorey vegetation. Plant Ecol 170:93–105
Maddock L (1979) The “migration” and the grazing succession. In: Sinclair ARE, Norton-
Griffiths M (eds) Serengeti: dynamics of an ecosystem. Univ Chicago Press, Chicago,
pp 104–129
McCullough DR (1992) Concepts of large herbivore population dynamics. In: McCullough DR,
Barrett RH (eds) Wildlife 2001: populations. Elsevier, Amsterdam, pp 967–984
McCullough DR (1997) Irruptive behavior in ungulates. In: McShea WJ, Underwood HB,
Rappole JH (eds) The science of overabundance. Smithsonian, Washington DC, pp 69–99
McCullough DR (1999) Density dependence and life-history strategies of ungulates. J Mammal
80:1130–1146
McHugh T (1958) Social behavior of American buffalo. Zoologica–New York 43:1–40
McLoughlin PD, Dzus E, Wynes B, Boutin S (2003) Declines in populations of woodland
caribou. J Wildl Manage 67:755–761
McShea WJ, Underwood HB, Rappole JH (1997) The science of overabundance. Smithsonian,
Washington DC
Mduma SAR, Sinclair ARE, Hilborn R 1999 Food regulates the Serengeti wildebeest: a 40-year
record. J Anim Ecol 68:1101–1122
Messier F, Crete M (1985) Moose–wolf dynamics and the natural regulation of moose popula-
tions. Oecologia 65:503–512
Mills MGL, Biggs HC, Whyte IJ (1995) The relationship between rainfall, lion predation and
population trends in African herbivores. Wildl Res 22:75–88
Milner C, Gwynne D (1974) The Soay sheep and their food supply. In: Jewell PA, Milner C,
Morton Boyd J (eds) Island survivors. The ecology of the Soay sheep of St. Kilda.
Athlone Press, London, pp 273–325
Milner JM, Elston DA, Albon SD (1999) Estimating the contributions of population density and
climatic fluctuations to interannual variation in survival of Soay sheep. J Anim Ecol
68:1235–1247
Mosienyane B (1979) Nutrient evaluation of Botswana range grasses for beef production. Thesis,
Cornell University, Ithaca, NY
Mysterud A (1999) Seasonal migration patterns and home range of roe deer along an altitudinal
gradient in southern Norway. J Zool Lond 247:479–486
Mysterud A, Perez-Barbieria FJ, Gordon IJ (2001) The effect of season, sex and feeding style on
home range area versus body mass scaling in temperate ungulates. Oecologia 127:30–39
O’Connor TG, Haines LM, Snyman HA (2001) Influence of precipitation and species composition
on phytomass of a semi-arid grassland. J Ecol 89:850–860
Ogutu J, Owen-Smith N (2005) Oscillations of large herbivore populations: are they due to
predation or rainfall? Afr J Ecol 43:332–339
Ottichilo WK, De Leeuw J, Skidmore AK, Prins HHT, Said MY (2000) Population trends of large
non-migratory wild herbivores and livestock in the Masai Mara Ecosystem, Kenya, between
1977 and 1997. Afr J Ecol 38:202–216
Owen-Smith N (1975) The social ethology of the white rhinoceros. Z Tierpsychol 38:337-384
Owen-Smith N (1977) On territoriality in ungulates and an evolutionary model. Q Rev Biol
52:1–38
Owen-Smith N (1981) The white rhinoceros overpopulation problem, and a proposed solution. In:
Jewell PA, Holt S, Hart D (eds) Problems in management of locally abundant wild mammals.
Academic Press, New York, pp 129–150
Owen-Smith N (1988) Megaherbivores. The influence of very large body size on ecology.
Owen-Smith N (1990) Demography of a large herbivore, the greater kudu, in relation to rainfall.
Owen-Smith N (1992) Grazers and browsers: ecological and social contrasts among African
ruminants. In: Spitz F, Janeau G, Aulagnier S (eds) Ungulates / Ongules 91. SFEPM-IRGM,
Toulouse, pp 175–181
Owen-Smith N (1993a) Woody plants, browsers and tannins in southern African savannas. S Afr
J Sci 89:505–510
Owen-Smith N (1993b) Comparative mortality rates of male and female kudus: the costs of sexual
size dimorphism. J Anim Ecol 62:428–440
Owen-Smith N (1994) Foraging responses of kudus to seasonal changes in food resources: elasti-
city in constraints. Ecology 75:1050–1062
Owen-Smith N (1997) Distinctive features of the nutritional ecology of browsing versus grazing
ruminants. Z Saugetierkd 62 Suppl.II:176–191
Owen-Smith N (1998) How high ambient temperature affects the daily activity and foraging time
of a subtropical ungulate, the greater kudu. J Zool Lond 246:183–192
Owen-Smith N (2000) Modeling the population dynamics of a subtropical ungulate in a variable
environment: rain, cold and predators. Natur Res Mod 13:57–87
Owen-Smith N (2002a) Adaptive herbivore ecology. From resources to populations in variable
environments. Cambridge University Press, Cambridge
Owen-Smith N (2002b) Credible models for herbivore - vegetation systems: towards an ecology
of equations. S Afr J Sci 98:445–449
Owen-Smith N (2004) Functional heterogeneity within landscapes and herbivore population
dynamics. Landscape Ecol 19:761–771
Owen-Smith N (2006) Demographic determination of the shape of density dependence for three
African ungulate populations. Ecol Monogr 76:93–109
Owen-Smith N, Cooper SM (1985) Comparative consumption of vegetation components by
kudus, impalas and goats in relation to their commercial potential as browsers in savanna
regions. S Afr J Sci 81:72–76
Owen-Smith N, Cooper SM (1989) Nutritional ecology of a browsing ruminant, the kudu, through
the seasonal cycle. J Zool Lond 219:29–43
Owen-Smith N, Mason DR (2005) Comparative changes in adult versus juvenile survival affect-
ing population trends of African ungulates. J Anim Ecol 74:762–773
Owen-Smith N, Mills MGL (2006) Manifold interactive influences on the population dynamics of
a multi-species ungulate assemblage. Ecol Monogr 76:73–92
Owen-Smith N, Mills MGL (submitted) Shifting prey selection generates contrasting herbivore
dynamics with a large-mammal predator-prey web. Ecol Monogr
Owen-Smith N, Ogutu JO (2003) Rainfall influences on ungulate population dynamics in the
Kruger National Park. In: du Toit JT , Rogers KH, Biggs HC (eds) The Kruger experience:
ecology and management of savanna heterogeneity. Island Press, Washington, DC,
pp 310–331
Owen-Smith N, Mason DR, Ogutu JO (2005) Correlates of survival rates for ten African ungulate
populations: density, rainfall and predation. J Anim Ecol 74:774–788
Pappageorgiou NK (1978) Food preference, feed intake and protein requirements of red deer in
central Greece. J Wildl Manage 42:940–943
Patterson BR, Power VA (2002) Contributions of forage competition, harvest, and climate fluctu-
ations to changes in population growth of northern white-tailed deer. Oecologia 130:62–71
Pellew RA (1983) The giraffe and its food resource in the Serengeti. I. Composition, biomass and
production of available browse. Afr J Ecol 21:241–267
Peterson RO (1999) Wolf–moose interactions on Isle Royale: the end of natural regulation? Ecol
Applic 9:10–16
Peterson, RO, Thomas NJ, Thurber JM, Vucetich JM, Waite TA (1998) Population limitation and
the wolves of Isle Royale. J Mammal 79:828–841
Post E, Stenseth NC (1998) Large-scale climatic fluctuation and population dynamics of moose
and white-tailed deer. J Anim Ecol 67:537–543
Post E, Stenseth NC (1999) Climatic variability, plant phenology, and northern ungulates.
Ecology 80:1322–1339
176 N. Owen-Smith
Prins HHT (1988) Plant phenology patterns in Lake Manyara National Park, Tanzania. J Biogeogr
15:465–480
Prins HHT (1996) Ecology and behaviour of the African buffalo. Chapman & Hall, London
Prins HHT (2000) Competition between wildlife and livestock in Africa. In: Prins HHT,
Grooterhuis JG, Dolan TT (eds) Wildlife conservation by sustainable use. Kluwer, Dordrecht,
pp.51–80
Prins HHT, Beekman JH (1989) A balanced diet as a goal for grazing: the food of the Manyara
buffalo. Afr J Ecol 27:241–259
Prins HHT, Iason GR (1989) Dangerous lions and nonchalant buffalo. Behaviour 108:262–297
Prins HHT, Weyerhaus FJ (1987) Epidemics in populations of wild ruminants: anthrax and
impala, rinderpest and buffalo in Lake Manyara National park, Tanzania. Oikos 49:28–38
Putman RJ (1986) Grazing in temperate ecosystems: large herbivores and their ecology in the
New Forest. Croon Helm, London
Renecker LA, Hudson RJ (1988) Seasonal quality of forages used by moose in the aspen-
dominated boreal forest, Alberta. Holarctic Ecol 11:111–118
Riney T (1964) The impact of introductions of large herbivores on the tropical environment.
IUCN public, new series no. 4, pp 261–273
Risenhoover KL, Maass SA (1987) The influence of moose on the composition of Isle Royale
forests. Can J For Res 17:357–364
Robbins CT (1993) Wildlife feeding and nutrition. Academic Press, New York
Robbins CT, Moen AN (1975) Composition and digestibility of several deciduous browses in the
north-east. J Wildl Manage 39:337–341
Rutherford MC (1984) Relative allocation and seasonal phasing of growth of woody plant com-
ponents in a South African savanna. Progr Biometeorol 3:200–221
Ryan SJ, Knechel CU, Getz WM (submitted) Seasonal and interannual variation in home range
and habitat selection of African buffalo: a long-term study in the Klaserie Private Nature
Reserve, South Africa. J Wildl Manage 70:764–776.
Saether B-E (1997) Environmental stochasticity and population dynamics of large herbivore:
a search for mechanisms. Trends Ecol Evol 12:143–149
Saether B-E, Andersen R (1990) Resource limitation in a generalist herbivore, the moose: ecologi-
cal constraints on behavioural decisions. Can J Zool 68:993–999
Schoen JW, Kirchoff MD (1985) Seasonal distribution and home-range patterns of Sitka black-
tailed deer on Admiralty Island, Southeast Alaska. J Wildl Manage 49:96–103
Scholes RJ, Walker BH (1993) An African savanna. Synthesis of the Nylsvley study. Cambridge
University Press, Cambridge
Scoones I (1993) Why are there so many animals? Cattle population dynamics in the communal
areas of Zimbabwe. In: Behnke RH, Scoones I, Kerven C (eds) Range ecology at disequilib-
rium. Overseas Development Institute, London, pp. 62–76
Scoones I (1995) Exploiting heterogeneity: habitat use by cattle in dryland Zimbabwe. J Arid
Envir 29:221–237
Seip DR (1992) Factors limiting woodland caribou populations and their interrelationships with
wolves and moose in southeastern British Columbia. Can J Zool 70:1494–1503
Sinclair ARE (1977) The African buffalo. A study of resource limitation in populations. University
of Chicago Press, Chicago
Sinclair ARE, Arcese P (1995) Population consequences of predation-sensitive foraging: the
Serengeti wildebeest. Ecology 76:882–891
Skinner JD (1993) Springbok treks. Trans Roy Soc S Afr 48:291–305
Skinner JD, Louw GN (1996) The springbok. Transvaal Museum Monograph No. 10
Smuts GL (1978) Interrelations between predators, prey, and their environment. BioScience
28:316–320
Solberg EJ, Loison A, Gaillard J-M, Heim M (2004) Lasting effects of conditions at birth on
moose body mass. Ecography 27:677–687
Stenseth NC, Mysterud A, Ottersen G, Hurrell JW, Chan KS, Lima M (2002) Ecology and clima-
tology: ecological effects of climatic fluctuations. Science 297:1292–1298
Stenseth NC, Ottersen G, Hurrell JW, Mysterud A, Lima M, Chan KS, Yoccoz NG, Adlandsvik B
(2003) Studying climate effects on ecology through the use of climate indices: the North
Atlantic Oscillation, El Nino Southern Oscillation and beyond. Proc R Soc Lond B
270:2087–2096
Talbot LM, Talbot MH (1963) The wildebeest in western Masailand, East Africa. Wildl Monogr
No. 12
Tixier H, Duncan P (1996) Are European roe deer browsers? A review of variations in the com-
position of their diets. Rev Ecol (Terre Vie) 51:3–17
Tyson PD, Gatebe CK (2001) The atmosphere, aerosols, trace gases and biogeochemical change
in southern Africa: a regional integration. S Afr J Sci 97:106–118
van de Vijver CADM, Poot P, Prins HHT (1999) Causes of increased nutrient concentrations in
post-fire regrowth in an East African savanna. Plant Soil 214:173–185
Walker BH, Ludwig D, Holling CS, Peterman RS (1981) Stability of semi-arid savanna grazing
systems. J Ecol 69:473–498
Walker BH, Emslie RH, Owen-Smith N, Scholes RJ (1987) To cull or not to cull: lessons from a
southern African drought. J Appl Ecol 24:381–402
Whyte IJ, Joubert SCJ (1988) Blue wildebeest population trends in the Kruger National Park and
the effects of fencing. S Afr J Wildl Res 18:78–87
Williamson D, Mbano B (1988) Wildebeest mortality during 1983 at Lake Xau, Botswana. Afr J
Ecol 26:341–344
Williamson D, Williamson J, Ngwamotsoko KT (1988) Wildebeest migration in the Kalahari. Afr
J Ecol 26:269–280
Wilson VJ (1970) Data from the culling of kudu in the Kyle National Park, Rhodesia. Arnoldia
(Rhodesia) 4:1–26
Ydenberg RC, Prins HHT (1981) Spring grazing and the manipulation of food quality by barnacle
geese. J Appl Ecol 18:443–453
Chapter 7
Species Diversity of Browsing and Grazing
Ungulates: Consequences for the Structure
and Abundance of Secondary Production
Herbert H.T. Prins and Hervé Fritz
7.1 Introduction
There are two fundamentally different ways to look on the way consumers use their
resources. The first is that different species partition the resource in a particular
manner, so that some species get a particular part and other species another part. In
this view, the sum of the shares is equal to the total. The second way of looking at
resource partitioning is shaped by thoughts about niche differentiation: species
differ in their ability to extract particular resources from a continuum, and because
they are different there is an additive effect if different species utilise an area
together. Specialization leads to increased partitioning of resources which leads to
a higher total offtake. We are interested in the question, then, whether increased
species richness of vertebrates that make use of the vegetation—that is, ungulate
grazers and browsers—leads to an increased offtake of the plant biomass in an area,
and subsequently whether, if true, this translates into higher herbivore biomass
and/or productivity.
This question is of interest for several reasons. The first is from a theoretical point
of view. Much work has been done since the niche concept was developed in the
1920s by Grinell, and later examined closely again in the 1950s and 1960s by
Hutchinson and McArthur. The quest for finding order among all the competing
species may have come from a corporatist world view and a harking back to the guild
structure of earlier societal organisation. Yet, the niche concept has been an enormous
stimulus to ecology, and has long been shaping our thinking. In the 1980s and 1990s,
however, the niche concept met increasing resistance among ecologists, culminating
in Hubbell’s neutral theory (Hubbell 2001). This may have been a reflection, again,
of society’s drift towards neo-liberalism and its associated notions about a totally free
market. The neutral theory implies functional redundancy among species, and hence
absence of impacts of changes in diversity on functional processes at the community
or ecosystem scale. At the other extreme, niche theory postulates that all species dif-
fer to some extent in the resources they use. This implies functional complementarity
among species, and hence, for instance, increased productivity and other ecosystem
processes with diversity (Tilman et al. 1997; Loreau 1998). So, theory is moving
hither and thither; a confrontation with facts can be useful to ascertain whether the
© Springer 2008
180 H.H.T. Prins and H. Fritz
direction in which theoreticians are driving us is the right one or not, because only
encounter with reality will help us decide which theory we’d best embrace. However,
as often, apparently opposed theory may in fact apply to different ecological
situations; a mutually exclusive situation does not necessarily exist (Holyoak and
Loreau 2006; Leibold and McPeek 2006).
From a purely thermodynamic and energy-capture-rate viewpoint, one also may
reason that a part of the Earth’s surface has a particular capacity for harnessing the
energy of photons into chemical energy through the action of chlorophyll. That
process depends on the conversion efficiency of plants (actually only about 3%) and
the amount of leafy material. Neither grazing nor browsing will affect the
conversion efficiency per se, but grazing or browsing could increase the amount of
living phytomass (McNaughton et al. 1988; du Toit et al. 1990), and it could be
envisaged that a particular combination of different herbivore species could result
in a higher primary production, and because of that a higher secondary production.
Even if primary production remained constant, different combination of herbivores
may use it more thoroughly, leaving less of it to decomposers, also leading to a
higher biomass and secondary production.
The second reason to be interested in the question of the efficiency of harvesting
has to do with management. Many people in many societies are interested in
whether harvests can be optimised. Much of the Earth’s primary production is
inedible for humans, and we use grazers and browsers to transfer this primary
production into resources that are of direct use to us: meat, milk, hides, bones,
hooves. and organs all have a direct or indirect use in human society. Again, we thus
ask ourselves whether a judicious combination of different herbivore species could
result in a higher secondary production. Much agriculture is done through the
monoculture of a crop, even though De Wit (1960) and others have shown that from
a production point mixed crops are more productive. The reason that farmers
choose for monoculture has to do with management, not with productivity. Our
management question has direct effect on the relevance of biodiversity: is a high
diversity of grazers and browsers ‘good’? We know it is pleasing to see many
different forms of herbivores in an area—indeed, this is the basis for much eco-
tourism—but does high biodiversity have other benefits? In this chapter we will
first ask the question, ‘What causes species richness, especially of ungulates?’, and
then we will try to answer the question ‘Does increasing the species richness of the
herbivore community lead to a higher secondary production and more efficient use
of the vegetation?’ We will pay special attention to whether combining browsers
with grazers leads to a higher offtake than either grazers alone or browsers alone.
In our chapter we will not concentrate on ‘ad hominum’ (ad animalum?) types of
explanation in which every species is so uniquely adapted to its niche that general
patterns cannot be found. Indeed, we keep in mind that ‘fiber digestion is not
significantly different between browsers and grazers, although fiber digestion is
positively related to herbivore size’ (Robbins et al. 1995), and that ‘after controlling
for the effects of body mass, there is little difference in digestive strategy [among]
(African) ruminants with different morphological adaptations of the gut’ (Gordon
and Illius 1994).
7 Species Diversity of Browsing and Grazing Ungulates 181
7.2 Suggested Causes of Species Richness
Finding the cause of species richness is like the quest for the Holy Grail. The question
is difficult to formulate precisely, because it entails evolution and phylogenetics
(see Janis, Chapter 2), special adaptations (see Clauss, Chapter 3), past and present
competition (see Duncan and Poppi Chapter 4; Searle and Shipley, Chapter 5) and
different population dynamics (see Owen-Smith, Chapter 6), yet the literature
discloses different possibilities, and here we single out the following (Box 7.1):
Oindo et al. (2001) and Oindo (2002) looked in great detail through the use of
satellite remote sensing and weather data at the spatial distribution of species
richness in Kenya, Tanzania, and Uganda. They looked at different groups of
organisms, including ungulates, and came to the conclusion that by looking at the
normalised vegetation index (NDVI, a good proxy for phytomass production),
variability in primary production correlates well with species richness. Also Janis
et al. (2000) search for the cause of ungulate species richness in primary produc-
tion, but they think it is linked to the average levels of plant productivity: ‘Both
maximum species richness of all ungulates and the proportion of browsers
declined steadily in the ungulate communities through the middle Miocene, to
levels comparable to those of the present by the late Miocene. We suggest that the
early Miocene [17 Ma] browser-rich communities may reflect higher levels of pri-
mary productivity in Miocene vegetation, compared with equivalent present-day
vegetation types. The observed decline in species richness may represent a gradual
decline in primary productivity, which would be consistent with one current
hypothesis of a mid-Miocene decrease in atmospheric CO2 concentrations from
higher mid-Cenozoic values’. Note that this is a correlative conclusion, and that a
mechanism for the link between species richness and primary productivity is not
suggested. Olff, Ritchie and Prins (2002) modelled a possible cause for species
richness on the basis of existing theory and available data from across Africa.
They then made predictions concerning the found relation to other continents, and
tested it for North American data. They came to the conclusion that ‘More plant-
available moisture reduces the nutrient content of plants but increases productivity,
Box 7.1. Biodiversity, that is, species richness, of ungulate assemblages is

enhanced by:
Low intra-annual variation in NDVI = primary production (Oindo 2002; Oindo
et al. 2001) ),
High inter-annual average NDVI (Oindo 2000; Oindo et al. 2001)
High primary production (Janis et al. 2000)
High soil nutrients in combination with intermediate precipitation (Olff, Ritchie
& Prins 2001)
High spatial heterogeneity of ecosystems (du Toit & Cumming 1999)
whereas more plant-available nutrients increase both of these factors. Because

larger herbivore species tolerate lower plant nutrient content but require greater
plant abundance, the highest potential herbivore diversity should occur in loca-
tions with intermediate moisture and high [soil] nutrients. … Thus gradients of
precipitation, temperature and soil fertility might explain the global distribution of
large herbivore diversity’ (Olff et al. 2002). A fourth explanation for ungulate spe-
cies richness looks at a different putative mechanism, namely spatial heterogene-
ity: ‘This exceptional fauna diversity and herbivore biomass density is directly
linked [italics added] to the high spatial heterogeneity of African savanna
ecosystems. The dependence of herbivore dietary tolerance on body size translates
into important size-related differences between savanna ungulate species in terms
of habitat specificity, geographical range, and the share of community resources
exploited’ (du Toit and Cumming 1999). How this direct link works is not
explained, but that is here beside the point. The point is that species richness is
thought to be caused by something—it is not a random phenomenon. The exact
cause is not clear yet and at this moment the approach of Prins and Olff (1998) and
Olff et al. (2002), which was strongly inspired by Hutchinson (1957), is the most
causal-analytical.
In all studies but that of Janis et al. (2000), the patterns and the likely processes
do not distinguish between browsers and grazers, and mostly concentrate on body
size. There may be commonalities between browsers and grazers that only body size
explains, e.g., volatile fatty acid (VFA) in digestive physiology (Gordon and Illius
1994) but there are differences in body-size distribution between the two dietary
types. Grazers tend to be larger on average than browsers, but body size spans
equally for both dietary groups, with the largest ruminant in fact being a browser
(namely, the giraffe). Body size is unlikely to encompass all traits associated with
browsers and grazers, as many other traits vary between trophic guilds, although
body size certainly also co-varies partly with most of them (reviews Gordon 2003;
Fritz and Loison 2006). Obviously, the different reaction to the distribution of their
primary resources will also affect the relative abundance and diversity of browsers
and grazers, as suggested by Janis et al. (2000), and possibly ultimately secondary
production. As a matter of fact, browser and grazer biomasses respond similarly to
annual rainfall (same slopes) in African savanna ecosystems, but browsers remain
lower in biomass for a given annual rainfall (Fig. 7.1). This is even more true when
mixed-feeders such as elephants are removed from the comparison (mixed-feeders
are split into the grazer and browser component according to their graze/browse
share in the diet). The fact that browse resources are more dispersed and less
abundant in rangelands may explain this pattern.
Interestingly, the numbers of species in the grazer and browser guilds in savannas
show the same quadratic shape with annual rainfall but with a lower species richness
in browsers (Fig. 7.2). This suggest that the environmental determinants of species
richness in grazers and browsers may be similar, but the pool of species may be
smaller in savanna browsers due to the lower browse production in savannas (com-
pared to forest, for instance) and possibly also paleo-historical changes in landscape
that induced more losses in browsers (e.g., Janis et al. 2000 for North America).
4,0
Grazers-HighSNA
3,5
Grazers-MediumSNA
3,0
Biomass of herbivores
Grazers-LowSNA
(Log 10 kg/km 2)
Browsers
2,5
2,0
1,5
1,0
0,5
2,00 2,25 2,50 2,75 3,00 3,25 3,50 3,75
Annual Rainfall (Lo 10 mm )
Fig. 7.1 Relationship between annual rainfall, soil nutrient status and the abundance of the two
major feeding guilds, grazers (squares) and browsers(circles). Soil nutrient availability (SNA) was
only significant for grazers
7.3 The Effect of Species Richness on Ecosystem

Functioning: An Overview
So, species richness of assemblages of browsers, grazers, mixed-feeders, or

combinations thereof is different at different places and at different times (see also
Janis, Chapter 2). What does this mean to ecosystem functioning? In Box 7.2 we
present some hypotheses that have been floating around in the literature already for
a long time but which are still of great interest:
In this chapter we are only interested in the first effect; the increased species
richness of herbivores on nutrient retention, resilience, resistance have to our
knowledge never been tested, and here we do not deal with the buffering effect but
with the increased productivity. Of course, we are interested to know if increased
species richness leads to increased secondary productivity, not on if it increases
primary productivity.
With plants it has been shown about 50 years ago that increased species richness
leads to increased production (De Wit 1960), but with the new interest in questions
concerning ecosystem functioning the same insights are now re-appraised and
reformulated. ‘There is evidence that biodiversity loss can lead to reductions in
biomass production, … Under the unperturbed conditions, the species-poor systems
achieved lower biomass production than the species-rich systems’ (Pfisterer and
Schmidt 2002). Other studies report the same, e.g., ‘Recent studies on grasslands
demonstrate that species losses and subsequent changes in diversity can … alter
14
Low Soil Nut. Avail.
Med Soil Nut. Avail.
12 High Soil Nut. Avail.
Total quadratic fit
Grazers Species Richness
10
2
0 200 400 600 800 1000 1200
Annual Rainfall
10
Low Soil Nut. Avail.
Medium Soil Nut. Avail.
High Soil Nut. Avail.
8
Browsers Species Richness
0 200 400 600 800 1000 1200

Annual Rainfall
Fig. 7.2 The relationship between species richness (number of species) as a function of annual
rainfall and soil nutrient status, for grazers (upper graph) and browsers (lower graph). Only the
quadratic relationship was significant, not the soil nutrient status. Rainfall only explains 10% of
the observed variance in grazers whereas it explains 41% in browsers
Box 7.2. Increased biodiversity, measured as increased species richness, causes:

Increased productivity (Tilman et al. 2001, Engelhard & Ritchie 2002)
Increased nutrient retention (Hiremath & Ewel 2001, Cardinale et al. 2002)
Increased resilience (not supported: Engelhardt & Kadlec 2001: no effect)
Increased resistance (not supported: Pfisterer & Schmidt 2002: even opposite)
Reduced temporal variability in ecosystem processes; buffering effect, includ-
ing an ‘insurance effect’ (probable: Yachi & Loreau 1999, Loreau 2000,
Loreau et al. 2001).
ecosystem functioning (e.g., productivity’ Engelhardt and Kadlec 2001), and ‘Plant
diversity and niche complementarity had progressively stronger effects on ecosystem
functioning … with 16-species plots attaining 2.7 greater biomass than monoculture.
Diversity effects were neither transient nor explained solely by a few productive or
unviable species. … Even the best chosen monocultures cannot achieve greater
productivity or carbon stores than higher-diversity sites’ (Tilman et al. 2001). These find-
ings have been well summarised as ‘Positive short-term effects of species diversity on
ecosystem processes, such as primary productivity and nutrient retention, have been
explained by two major types of mechanisms: (1) functional niche complementarity
(the complementarity effect; Engelhard and Ritchie 2002 call this ‘the niche
differentiation effect’), and (2) selection of extreme trait values (the selection effect;
Engelhard and Ritchie 2002 call this the ‘sampling effect’). In both cases, biodiversity
provides a range of phenotypic trait variation. In the complementarity effect, trait
variation then forms the basis for a permanent association of species that enhance
collective performance. In the selection effect, trait variation comes into play only as
an initial condition, and a selection process then promotes dominance by species with
extreme trait values’ (Loreau 2000; see also Bond and Chase 2002). All these studies
dealt with plants or plankton, not with animals or even vertebrates. Partly that is
because vertebrate studies are more difficult to conduct, but partly it is because plant
ecologists have discovered that good experimental studies that are well designed may
lead to answers much faster than observational studies.
A higher ecosystem nutrient use efficiency (the ratio of net primary productivity to
soil nutrient supply), fostered by higher plant species diversity, is an integrative meas-
ure of ecosystem functioning (Hiremath and Ewel 2001). We maintain that the ecosys-
tem nutrient use efficiency is defined too narrowly in modern ecosystem studies that
are dominated by plant ecologists. We maintain that secondary productivity, that is, the
ratio of secondary primary productivity to nutrient supply from plant resources is also
an integrative measure of ecosystem functioning (and by the same token tertiary pro-
ductivity by predators). So, does higher consumer diversity lead to higher secondary
productivity? As higher diversity at a given trophic level may affect the stocks condi-
tioning the fluxes between ecosystem compartments, as well as productivity, we also
investigated the role of species diversity in the biomass of herbivore assemblages.
7.4 Herbivore Diversity and the Use of Primary Production
One of the fundamental assumptions behind the prediction relating diversity to

biomass or to productivity is that an increase in species richness induces a more
efficient, or more complete, use of the primary production. We thus first investi-
gated whether sympatric species of ungulates used resources differently, before
exploring the possible effect of species diversity on herbivore biomass and
productivity.
7.4.1 Diet Overlap and Feeding Niches
Many studies have been conducted, on domestic herbivores, wild herbivores, and
on combinations of wild and domestic ones, on the issue of diet overlap. These
studies generally point out that there is some degree of niche segregation among
different types of herbivore. Here we give a short overview of different type of
results that have been found.
In semi-arid temperate grassland diet overlap between red deer Cervus elaphus
and cattle varied greatly depending on availability of palatable fractions of herbs,
shrubs and grasses; red deer were better shrub users (28–50% in diet) than cattle
(6–12%) (Pordomingo and Rucci 2000). In a Louisiana pine range white-tailed deer
Odocoileus virginianus and cattle diet overlap was 11–31%. Deer mostly used
browse and herbs, cattle graminoids (Thill and Martin 1986). In California dietary
overlap between black-tailed deer Odocoileus hemionus and elk Cervus elaphus
was lowest in wet winter months (dietary N highest, standing crop lowest), and
overlap highest in dry summer months (dietary N lowest, standing crop highest)
(Gogan and Barrett 1995). In Colorado the diet overlap between mule deer
Odocoileus hemionus and elk was 3% in winter and 48% in summer; between elk
and cattle it was 30–50% in summer, while at the same time of the year it was 12–38%
between mule deer and cattle (Hansen and Reid 1975). Also in Colorado diet overlap
between mule deer and cattle was 2–11%; between mule deer and horse it was
2–11%, which was indicative more of a complementary than of a competitive rela-
tionship (Hubbard and Hansen 1976). Again in Colorado diet overlap between
mule deer and cattle was 1–22%; the authors observed that ‘When cattle are forced
from a grass-dominated diet to browse forage on overgrazed ranges, diet overlap
and forage competition between deer and cattle increase’. (Lucich and Hansen
1981). Diet overlap was also studied among domestic cattle, sheep, bison (Bison
bison), and pronghorn (Antilocapra americana) in Colorado, leading to the conclu-
sion that diet overlap appeared to depend on recent evolutionary history and on
body size, though values were strongly influenced by forage quantity and quality
(Schwartz and Ellis 1981). In West Virginia the diet overlap between cattle and
sheep was 76%, while between cattle and goats 75% (Cox-Ganser 1990).
European studies found the same sort of dietary overlaps between grazers or
browsers. In Northern Fennoscandia diet overlap between moose (Alces alces) and
roe deer (Capreolus capreolus) was 21–34%, between moose and red deer 32%,
between red deer and sheep 59–64%, between sheep and reindeer (Rangifer
tarandus) 55% and finally between sheep and goat it was 77%. Neither difference
in feeding type nor body mass successfully predicted diet overlap (Mysterud 2000).
In Scotland: ‘Deer showed no change in the proportion of grass in their diet in the
presence or absence of sheep, but … the diet of sheep contained a significantly
higher proportion of grasses when they were grazing with red deer (52% versus
38%)’ (Cuartas et al. 2000). Goats grazed Myrica, Juncus and Molinea more than
sheep, while the sheep preferred Caluna which was not significantly grazed by
goats, but the overlap was considerable (Fisher et al. 1994). Also in the Netherlands
overlap between red deer and other ungulates was large (in summer 70% and in
winter 62% for cattle and red deer, and 58% in summer and 77% in winter for
ponies and red deer (Van Wieren 1996). In the French Vosges an analysis of stom-
ach contents showed an overlap in diet between red deer and roe deer ranging
28–55% in winter months and 26–51% in summer months ( Storms et al. 2006). In
the Camargues in France horses and cattle largely overlapped in their niches
(58–77%), both for habitat and food (Ménard et al. 2002).
Latin American studies also find differing degrees of dietary overlap. In
Mexico diet overlap between white-tailed deer and cattle was 51% (Gallina
1993). In the dry areas of Brazil, during the wet season the diets of goats and
sheep was quite different but by the end of the dry-wet transition period intake of
grasses and woody plants was similar. There was a high similarity of diets
(Araujo Filho et al. 1996). Interestingly, there was not a strong dietary overlap
between sheep and the indigenous llama (Lama glama) in Bolivia (Genin et al.
1994); neither was there too much overlap between diets of indigenous
Venezuelan deer (Odocoileus virginianus) with cattle or capybara (Hydrochoerus
hydrochaeris) because 93% of their diet originated from the wooded fringe area.
The capybara did not compete with cattle in the extensive intermediate area, and
the taller and drier herbage was preferred by horses and cattle but not by capybara.
However, in the natural habitat of the capybara, the lowest region, there was
substantial diet overlap (Escobar and Gonzalez 1976).
Studies like these have also been conducted in Africa. Dekker (1997) gives diet
overlaps for Messina in South Africa, and du Toit et al. (1995) found that the dietary
overlap between sheep and goats in the Karoo differed; it was 95–96% during the
growing season, and 79–86% during the dormant season. In Senegal cattle selected
a very different diet from goats, and sheep were intermediate; the differences in diet
among animal species declined in the dry season (Nolan et al. 1996). Other studies
highlighted the diet overlap or dietary difference between domestic species and indig-
enous ones. The diet and feeding height of kudu (Tragelaphus strepsiceros) and
goats and of black rhinoceros (Diceros bicornis) and goats overlapped to a large
extent. Overlap in diet between giraffe (Giraffa camelopardalis) and goats was
extensive but overlap in feeding height was small; goats and eland (Taurotragus
oryx), despite feeding at similar heights, generally consumed different species
(Breebaart et al. 2002). Makhabu (2005) on the Chobe riverfront, Botswana, found
around 20% overlap among elephant (Loxodonta africana) and three other browsers
(giraffe, kudu, impala; Aepyceros melampus) both in dry and wet season, but that the
overlap in plant use ranged from 56% to 76% among the three other species in
the wet season and from 57% to 82% in the dry season. However, although plant
parts use also overlapped in a similar way (49–72%), giraffe, kudu, and impala over-
lapped less in feeding heights, especially in the dry season (4%–32%). In Tanzania
cattle overlapped with zebra (Equus burchellii) in the early wet season and with
wildebeest (Connochaetes taurinus) in the early dry season; in the wet season, cattle
showed overlap in resource use with both zebra and wildebeest (Voeten and Prins
1999). Other studies focussed on natural assemblages without domestic species. In
Kenya diet overlap between ungulates and very small herbivores is low (French
1985), but between large herbivores the overlaps are large. For instance, in Lake
Nakuru National Park, Mwasi (2002) found the following overlaps: between impala
and African buffalo (Syncerus caffer), late wet season 58%, short dry season 81%
and early wet season 75%; between impala and common zebra, late wet season 83%,
short dry season 55%, and early wet season 82%. In Uganda there were significant
seasonal differences in the diet of most of the herbivores, including buffalo, Uganda
kob (Kobus kob), topi (Damaliscus lunatus), warthog (Phacochoerus aethiopicus),
waterbuck (Kobus ellipsiprimnus), and hippopotamus (Hippopotamus amphibus); and
there was greater separation in the longer dry season (Field 1972). In the Democratic
Republic of Congo Hart (1986) found that the diets of all species (six varieties of
duiker and chevrotain) in the upland forest converged when high quality fruits and
seeds were abundant, diets diverged when high quality food was scarce. During
scarcity some species showed habitat segregation, other segregated along lines of
fruit specialisation. Diet overlap occurred in the mixed forest when both food abun-
dance and diversity were low. In Mozambique, Prins et al. (2006) found considerable
overlap between duiker antelopes and suni (Neotragus moschatus) in the wet season
(63–83%) but less in the dry season (21–38%). Overall, only 10% of dietary items
were species-exclusive in any given season. Only a few studies into diet overlap
among different herbivores were conducted in Asia, but they show the same picture.
In Ladakh, for example, Mishra (2001) reported the following overlaps from the
high-altitude grasslands there: between blue sheep (Pseudois nayaur) and domestic
yak, summer 61%, winter 52%; between blue sheep and donkey, summer 43%,
winter 96%; between goat and yak, summer 84%, winter 72%; and finally between
goat and donkey, summer 66%, winter 92%. Also, ibex (Capra sibirica) has a very
similar diet and habitat to goats and sheep in these systems, suggesting competition
to explain its absence from pastoral zones (Bagchi et al. 2004). Conversely, in Nepal,
the overlap in plant use was very low among blue sheep, argali sheep (Ovis ammon
hodgsoni) and domestic goat (1–8%), although broad diet composition in terms of
grass, forbs, and browse were less different (overlap in categories from 20% to
76%); forb and browse species use discriminated the ungulate species (Shrestha
et al. 2005). In the more semi-arid areas in India, the diet similarities among nilgai
(Boselaphus tragocamelus), chital (Axis axis), and chinkara (Gazella bennetti)
were very high in the dry season, but chital used different habitat, more similar to
sambar (Cervus unicolor), being fairly well segregated from the three other species
(Bagchi et al. 2003).
A general picture does not appear from these studies. Most of them have been
descriptive. The general conclusion is that if different herbivore species, whether
they are browsers or grazers, utilise a given area then there is generally considerable
overlap in diet but there is some segregation, too. Many conclusions concerning
competition or the lack thereof have been drawn from these studies in diet overlap
or dietary segregation. The picture emerging from these studies is, however, discon-
certingly unclear. As a general rule it appears to emerge that if there are more her-
bivore species, then a wider array of plant species are being consumed by the
assemblage in total.
7.4.2 Postulated Advantages of Mixed-Species Feeding
It is interesting to note that scientists who have studied niche overlap have often
stressed niche segregation, and from their studies have made inferences about
ecosystem functioning and secondary productivity without providing the necessary
productivity data to underscore their contentions. Milton (2000) formulated it as a
very clear hypothesis as ‘diversification of livestock (through grazer-browser com-
binations) tends to stabilise or enhance utilisable secondary production’ (italics added).
Sometimes this hypothesis is implicit in the studies we review, for example, ‘Diets
differed only by 4-5% during the growing season. This margin was considered too
small to recommend combining small stock breeds in an effort to ensure greater
utilisation efficiency through multiple use of the vegetation’ (du Toit et al. 1995;
italics added). Other studies are much more explicit. For instance, Breebaart et al.
2002) propose ‘a mixed farming system which includes goats, eland and giraffe as
a useful management tool for using savanna vegetation more efficiently’, and
Owen-Smith (1985) concluded that ‘the kudu is a prime candidate for the inclusion
alongside cattle in mixed species ranching enterprises in most regions of savanna
vegetation’, and Genin et al. (1993) wrote the absence of ‘a strong dietary overlap
between [sheep and llama], suggested that mixed grazing could allow a better
utilization of the vegetation’. A study conducted on cattle ranches in Zimbabwe
came to the conclusion that ‘there is a definite need for a browser to utilise woody
plants and to balance the present monospecies ranching system. … it is concluded
[on basis of the evaluation of the attributes of eland and other browsers] that eland
are very well adapted to complement cattle in the ranching industry’ (Lightfoot and
Posselt 1977), and in a South African review it was concluded that ‘The full pro-
duction potential of the thornveld areas can be achieved with cattle farming as the
primary enterprise and with goats playing a secondary role’ (Aucamp 1976).
Jewell (1980) was unambiguous when he stated that ‘natural communities of
game animals exhibit a high standing crop biomass because their ecological separa-
tion, particular in their utilisation of food resources renders many species comple-
mentary. A high density of one species may facilitate energy flow and the
success of another herbivorous species’. Also Nolan et al. (1999) in a review of
87 references explicitly state ‘Complementary grazing behaviour patterns among
different animal types improve individual animal performance and output per unit
of area …’. The conclusion thus seems to be rather exact and unquestionable that
higher consumer diversity leads to higher secondary productivity. However, even
though many animal ecologists have concluded, sometimes speculated, on the issue
of secondary productivity, very few if any controlled studies have been done in
which the herbivore assemblage was manipulated while productivity was measured.
From our review, it is clear that we have to turn to the agricultural literature to
explore this relationship further; and it is also evident that conclusions about a
positive diversity–biomass relationship in herbivores require more rigorous analyses
across sites of varying diversity, controlling for environmental parameters.
7.5 Mammalian Herbivore Species Richness Links

to Secondary Productivity and Biomass
Why would an increased diversity of consumers lead to an increased productivity

of those consumers? There are at least two main mechanisms that could be
involved: one is that the efficiency of biomass transfer from the one trophic level to
the next is influenced by the number of species in the consumer as well as in the
producer guild; second is that the apparent increase in total use of the primary
production, as documented above, increases the consumer biomass irrespective of
transfer efficiency, or increases the availability of plant resources through
facilitation (Cardinale et al. 2002). Agriculture experiments provide simple cases to
explore the functional link between herbivore diversity and secondary productivity,
mainly to test the possible changes in biomass transfer efficiency, as the diversity
of plants and consumers is often too low to explore adequately the effect of niche
complementarities on secondary biomass, and possibly productivity. This latter
issue is best investigated in wild ungulate assemblages, or eventually in pastoral
herds, more diverse and faced with a heterogeneous primary production.
7.5.1 Domestic Herbivore Diversity and Secondary Productivity
We evaluated many agricultural experiments. Table 7.1 shows that very often the
combination of two grazers leads to an increase of total secondary productivity of
the system, but a combination of a browser and a grazer does not (Table 7.2): actu-
ally that combination quite often led to a decrease. Mixed grazing can thus increase
secondary production, but this does not occur always. The exact interaction between
different herbivores and vegetation dynamics appears to be of great importance: an
increase is found when sheep and cattle are grazed together, but it rarely occurs
when sheep and goats or goats and cattle are grazed together. If secondary pro-
ductivity does not merely depend on the combination of different classes of
species (‘browsers’ versus ‘grazers’) but if species-specific idiosyncratic differences
Table 7.1 Effect of mixed grazing by two grazing species (sheep and cattle) on secondary
productivity (kg/ha per year or per grazing season): very frequently, the combination of two grazing
species leads to an increase total productivity
Sheep Sheep + cattle Cattle Reference
++ > + Dickson et al. 1981
+ = + Nolan and Connolly 1989
++ > + Nolan and Connolly 1989
++ > + de Boer and Hanekamp 1992
++ > + Logan et al. 1991
++ = ++ > + Olson et al. 1999
++ > + Martinez et al. 2002
+ = + = + Hamilton 1976
+ < ++ Abaye et al. 1994
Table 7.2 Effect of mixed grazing by grazers (sheep or cattle) and browsers (goats) on secondary
productivity (kg/ha per year or per grazing season): only rarely the combination of a grazer and a
browser lead to an increased total productivity; normally it does not
Goats Goats + cattle Cattle Reference
++ > + Martinez et al. 2002
+ < ++ Donaldson 1979
+ < ++ Leite et al. 1995
Sheep Sheep + goat Goats

+ = + = + Wilson and Mulham 1980 (normal)
++ > + < ++ Wilson and Mulham 1980 (drought)
+ < ++ Donaldson 1979
between different species are important, it is important to look at the exact effects
on the resource when species graze together.
Ecological theory suggests that small grazers outcompete larger ones and that
larger grazers facilitate smaller ones (Illius and Gordon 1987, Prins and Olff 1998,
Huisman and Olff 1998). We evaluated this by assessing which species was gaining
(in terms of productivity) when grazed in combination in comparison to when it
was husbanded in a single-species setting, and which species lost (Table 7.3). There
is no clear picture emerging: sometimes a small species benefits from a large one
but sometimes a large one to the detriment of the smaller; sometimes a grazer ben-
efits and sometimes a browser. The explanation does not lie in (1) functional niche
complementarity or in (2) selection of extreme trait values (Tilman 1999; Loreau
2000; Loreau et al. 2001; Engelhardt and Ritchie 2002). Secondary production
increase is mainly found in systems where grass and clover grow together but rarely
in other areas!
It appears as if the biodiversity effect in these cases works indirectly through
mediating plant competition. If grazer A modifies the competitive interaction
between two resources, in this case grasses and clover, and if the other grazing spe-
cies B makes better use of that second resource than grazer A, and if the productivity
Table 7.3 On basis of individual performance of animals (measured as weight gain per day or weight
reached at the end of the season), it was assessed which species gained from mixed grazing, and which
species lost. The column remarks briefly describe the system in which the experiment was conducted
Who
System benefits? Remarks Reference
Sheep & Goats Sheep Goats increase clover del Pozo et al. 1998
Sheep & Goats Sheep Goats increase clover Hardy and Tainton 1995
Cattle & Sheep Sheep Cattle increase clover McCall et al. 1986
Cattle & Sheep Sheep Lolium-Trifolium Abaye et al. 1994
Cattle & Goats Cattle Goats increase clover Osoro et al. 2000
Cattle & Goats Cattle Mopani veld Donaldson 1979
(If after goat)
Cattle & Goats Goats Mopani veld Donaldson 1979
(If after cattle)
System No Benefit? Remarks Reference

Cattle & Sheep Cattle Normal years, Australia Hamilton 1976
Cattle & Sheep Cattle Sweden Brelin 1979
Cattle & Sheep Cattle Mixed bush grass, RSA Mitchell 1985
Cattle & Sheep Cattle Sourveld, RSA Hardy and Tainton 1995
Cattle & Sheep Cattle Irrigated land, Martinique Mahieu et al. 1997
Cattle & Sheep Cattle Rangeland, Utah Olson et al. 1999
Cattle & Sheep Sheep Rangeland, Utah Olson et al. 1999
Goats & Sheep Sheep Dry lands, Brazil Araujo Filho et al. 1982
Goats & Sheep Sheep Dry lands, Brazil Leite et al. 1995
Goats & Sheep Goats Dry lands, Brazil Leite et al. 1995
System Who Suffers? Remarks Reference

Cattle & Sheep Cattle Cattle forced to poor food Clark 1980
Cattle & Sheep Cattle Martinez et al. 2002
Goat & Cattle Cattle Dry lands, Brazil Araujo Filho et al. 1982
Goat & Sheep Sheep Australia, drought years Wilson and Mulham 1980
of the benefiting species B is higher than the loss of species A, then the total
productivity of A + B can increase. The key lies in understanding the competitive
interaction between the resources (the plant species) that comprise the primary
production. Consumers can shift the competitive balance between the species
comprising the first trophic level, which may affect secondary production, but it is
not a rule that increased diversity of consumers leads to increased productivity of
the consumer assemblage. Agricultural experiments show that different trophic
levels are not governed by the same general rules. There are two caveats though,
one is that agricultural experiments are set in contexts of low diversity/heterogeneity
of the primary production, hence the complementarities are likely to be reduced.
The second is that these experiments use herbivores that have been selected over
hundreds of generations to be extremely efficient in converting plant productivity
into secondary production—and perhaps they are equally efficient! In that case
diversity does not increase productivity. The often observed mixed herding
strategies are then merely a case of risk spreading. An apparent opposition between
agriculture context and theoretical prediction about diversity and ecosystem
functioning is not new, as it also occurred in plant studies, for which conditions or
motivations were not necessarily those relevant for testing the theoretical models
(Vandermeer et al. 2002). The next step is therefore to investigate the patterns
exhibited by wild herbivore assemblages of different species richness.
7.5.2 Diversity–Biomass Relationship in Wild Assemblages
The literature review on diet overlaps from wild herbivore studies makes it likely
that a more complete use of the primary production takes place if there are more
herbivore species, which could thus translate into an increased secondary production.
This is actually supported by the experiments conducted on the Dos Arroyos Ranch
in the Sonoran Desert, which shows that increased diversity seems indeed to lead to
increased secondary productivity (Mellink 1995), and by the results from Fritz and
Duncan (1994), which suggest an effect of species number on the biomass of wild
African ungulate communities. We developed the comparative approach to specifi-
cally test for diversity-production relationship, but as productivity is difficult to
access in wild herbivore assemblages, we mostly investigated patterns relating spe-
cies diversity to biomass. To be able to have enough variations in species richness,
only African savanna ungulate assemblages provided an adequate case study.
In the current theoretical framework, the predictions are that when the consumers
of a given trophic level are generalist, the increase in diversity may not induce an
increase consumer biomass or productivity, as species would not be complementary
(Long and Morin 2005; Jiang and Morin 2005; Gamfeldt et al. 2005). Conversely, if
the consumers are mainly made up of specialists, then theory predicts that diversity
should be positively linked with production as species would then be more comple-
mentary in their resource use. In the context of mammalian herbivore assemblages,
this certainly calls for distinguishing the relative roles of body size (generalists tend
to be bigger) and diet types: grazers (more often generalist), browsers (more special-
ist), and mixed-feeders (the ultimate generalists?). There then seems to be a place for
the concept of ‘feeding types’ (sensu Hofmann 1973 and later work).
We should thus expect that the diversity–production relationship should be observed
in browsers, in which the number of selective specialist species is higher. Accordingly.
the speciation rate (particularly for Tragelaphinae; Vrba 1987) and also the comple-
mentarity of niche is potentially greater as the vertical dimension of the niche can be
discriminating (e.g., du Toit 1990; Makhabu 2005). In large mammalian herbivores,
however, there are species using resources that will rarely if ever be used by other mam-
malian species, but also that will transform the environment because of their body size
(Owen-Smith 1988). The most striking example is the elephant, which is able to con-
sume primary production in the form of branches and bark, uneaten by others, and
which also modifies the environment. The elephant in fact may have enough impact to
cause increased niche diversity. In this example, body size is a trait associated with two
specific properties within the community, namely, a wide dietary niche because body
size allows for the use of poorer quality food, and an impact on the environment that
may promote diversity and abundance to some extent, yet possibly reduce them at very
high densities (e.g., Fritz et al. 2002). Diversity could be positively linked with produc-
tion in a community with elephants, although here the pattern may in fact only be due
to one species, and not to species richness per se. Conversely, as the elephant is the
ultimate generalist, ungulate assemblages may only exhibit a diversity–biomass rela-
tionship once elephants are accounted for.
In a detailed analysis performed on 30 protected areas we show here that the
overall metabolic biomass of herbivores is affected by the number of species in
the system (5% of the observed variance), an effect secondary to rainfall and soil
nutrient availability (overall model R2=0.90) confirming the initial results from
Fritz and Duncan (1994) on a data set including pastoral areas. Interestingly, the
analysis for pastoral sites exclusively (1–6 species) did not show any significant
relationship, which is consistent with most results from agriculture experiments
(see above). When investigating at the feeding guild level, we found that the met-
abolic biomass of grazers was not related to the number of species, as expected
from theory, whether considering the whole community, the community without
4
Biomass of mesoherbivore in the assemblage
3
(Log 10 kg 0.75 km-2)
Mesobrowsers
Mesograzers
0
0 2 4 6 8 10 12 14
Number of species in the guild
Fig. 7.3 Relationship between species diversity (number of species) and the biomass of
mesoherbivores (<1000kg), grazers and browsers, in the assemblages. The relationship is only
significant for mesobrowers. The equation is y = 0.25× + 1.11, r2 = 0.25
elephants, or that without megaherbivore. For the browsers, species richness was
only significant on the metabolic biomass of the mesobrowser guild (Fig. 7.3) and
on the browser guild without elephants; the overall browser biomass was only
influenced by rainfall, as expected from the fact that the biomass of browsers is
largely dominated by elephants (again the ultimate generalist) and that elephant
biomass is exclusively explained by rainfall (Fritz et al. 2002).
Our regional analysis thus show that there may be an effect of species diversity
on herbivore biomass, at least in wild herbivores, and that this relationship supports
theoretical predictions associated with the roles of specialists and generalists in
food web and ecosystem functioning. The results from pastoral areas and agriculture
experiments are in fact also in line with the specialist/generalist predictions, since
domestic species have been selected (at least in extensive farming and pastoral
systems) to use primary production efficiently and are often fairly generalist.
Therefore, it is not surprising that species richness does not promote higher
herbivore biomass in these systems.
7.6 Conclusions
In this chapter we concentrated on the possible role of ungulate species diversity in

promoting higher secondary production or biomass. In savannas where we could per-
form a comparison, browsers’ and grazers’ diversity seem to respond in similar ways
to environmental variables, although grazer species are more numerous, paralleling
the fact that grazers generally contribute a greater share of the biomass in the assem-
blages. In general, reviewing the literature across most continents, we found that
increasing herbivore species increased the use of primary production, at least through
the use of a wider range of plants and plant parts. However, the suggestion that a
greater number of herbivore species increases primary production is weak, and so is
the possibility that herbivore diversity increases efficiency in biomass transfer. The
fact that some combination of herbivores may modify the competitive abilities of
plants and then promote higher secondary productivity should be investigated more
as a process interrelating herbivore diversity and production.
We show that a positive diversity–biomass relationship exists, although in
mammalian herbivores it does not seem as significant as in primary producers. The
diversity–productivity relationship seems circumstantial in domestic assemblages,
but could not be tested or examined on wild assemblages, for which we only
examined standing biomass. The most important point in this comparison regarding
the theoretical framework about diversity, food web, and ecosystem functioning is
that secondary consumers’ levels dominated by specialists will show a relationship,
whereas not if generalists dominate. In showing that specialist browsers seem to be
complementary in their use of their primary resources, we not only confirm field
observations about possible niche complementarity, but also suggest that systems
richer in browsers may use primary production more thoroughly, at least in
savanna-type ecosystems. This echoes the debate on mixed-species production

systems, especially in wild or mixed wild/domestic assemblages (e.g., Cumming
1993), but underlines the point that composition of species has to be carefully
thought about.
The lack of relationship between species diversity and metabolic biomass in
grazers brings into question the function of such a great diversity in African savannas,
and certainly calls for investigating the ‘insurance hypothesis’ that species diversity
buffer the ecosystem processes against environmental fluctuations because different
species respond differently, and hence lead to functional compensations (Yachi and
Loreau 1999). A comparative analyses of three African systems recently showed
that species diversity, indeed, seemed to promote the stability of secondary production
and biomass over long time periods (Prins et al., unpublished results). The same
insurance strategy may well be the reason for multispecies pastoral herds, as
generalist domestic species do not show striking niche complementarity.
References
Abaye AO, Allen VG, Fontenot JP (1994). Influencing of grazing cattle and sheep together and
separately on animal performance and forage quality. J Anim Sci 72:1013–1022
Araujo Filho JA de, Gadelha JA, Viana OJ (1982) Complementary grazing by cattle, sheep and
goat on the “Caatinga” of northeast Brazil. Proc Third International Conference on Goat
Production and Disease, Dairy Goat Journal Pub Co, Scottsdale, AZ pp 532
Araujo Filho, AJ de, Gadelha JA, Leite ER, Souza PZ, Crispim SMA, Rego MC (1996) Botanical
and chemical composition of the diet of sheep and goats grazing together in the Inhamuns
region, Ceara. Rev Soc Brasil Zootech 25:383–395
Aucamp AJ (1976) The role of the browser in the bushveld of the Eastern Cape. Proc of the
Grassland Soc Southern Afr 11:135–138
Bagchi S, Goyal SP, Sankar K (2003) Niche relationships of an ungulate assemblage in a dry
tropical forest. J Mammal 84:981–988
Bagchi S, Mishra C, Bhatnagar YC (2004) Conflicts between traditional pastoralism and conser-
vation of Himalayan ibex (Capra sibirica) in the Trans-Himalayan mountains. Anim Conserv
7:121–128
Bond EM, Chase JM (2002) Biodiversity and ecosystem functioning at local and regional spatial
scales. Ecol Lett 5:467–470
Breebaart L, Brikraj R, O’Connor TG (2002) Dietary overlap between Boer goats and indigenous
browsers in a South African savanna. Afr J Range Forage Sci 19:13–20
Brelin B (1979) Mixed grazing with sheep and cattle compared with single grazing. Swed J Agr
Res 9:113–120
Cardinale BJ, Palmer MA, Collins SL (2002) Species diversity enhances ecosystem functioning
through interspecific facilitation. Nature 415:426–429
Clark L (1980) Sheep and cattle on saltbush. Rural Res 107:13–15
Cuartas P, Gordon IJ, Hester AJ, Perez Barberia FJ, Hulbert IAR (2000) The effect of heather
fragmentation and mixed grazing on the diet of sheep Ovis aries and red deer Cervus elaphus.
Acta Theriol 45:309–320
Cumming DHM (1993) Multispecies systems: progress, prospects and challenges in sustaining
range animal production and biodiversity in East and southern Africa. Proceedings of the
World Conference on Animal Production, Edmonton, Canada, vol 1, pp 145–159
Cox-Ganser JM (1990) Comparative grazing behavior of cattle, goats and sheep. Thesis, West
Virginia University
de Boer J, Hanekamp WJA (1992) Combined grazing system of yearling heifers and sheep.
Rapport Proefstation voor de Rundveehouderij, Schapenhouderij en Paardenhouderij 135
Dekker B (1997) Calculating stocking rates for game ranches: substitution ratios for use in the
Mopani Veld. Afr Jour Range Forage Sci 14:62–67
del Pozo M, Osoro K, Celaya R (1998) The effects of complementary grazing by goats on sward
composition and on sheep performance managed during lactation in perennial ryegrass and
white clover pastures. Small Ruminant Res 29:173–184
De Wit CT (1960) On competition. Landbouwpublicaties, Wageningen, NL
Dickson IA, Frame J, Arnold DP (1981) Mixed grazing of sheep versus cattle only in an intensive
grassland system. Anim Prod 33:265–272
Donaldson CH (1979) Goats and/or cattle on mopani veld. Proc Grassland Soc Southern Afr
14:119–123
du Toit JT (1990) Feeding-height stratification among African browsing ruminants. Afr J Ecol
28:55–61
du Toit JT, Bryant JP, Frisby K (1990) Regrowth and palatability of acacia shoots following
pruning by African savanna browsers. Ecology 71:149–154
du Toit JT, Cumming DHM (1999) Functional significance of ungulate diversity in African
savannas and the ecological implications of the spread of pastoralism. Biodivers Conserv
8:1643–1661
du Toit PCV, Blom CD, Immelman WF (1995) Diet selection by sheep and goats in the arid
Karoo. Afr J Range Forage Sci 12:16–26
Escobar A, Gonzalez JE (1976) Study on the competitive consumption of large herbivores of the
flooded area of the Llanos with special reference to the capybara (Hydrochoerus hydrochaeris).
Agron Trop 26:215–277
Engelhardt KAM, Kadlec JA (2001) Species traits, species richness and the resilience of wetlands
after disturbance. J Aquat Plant Manage 39:36–39
Engelhardt KAM, Ritchie ME (2002) The effects of aquatic plant species richness on wetland
ecosystem processes. Ecology 83:2911–2924
Field CR (1972) The food habits of wild ungulates in Uganda by analyses of stomach contents.
E Afr Wildl J 10:17–42
Fisher GEJ, Scanlan S, Waterhouse A (1994) The ecology of sheep and goat grazing in semi-
natural hill pastures of Scotland. In: ‘t Mannetje L, Frame J (eds) Grassland and society,
Wageningen Pers, Wageningen, NL, pp 286–289
French NR (1985) Herbivore overlap and competition in Kenya rangeland. Afr J Ecol
23:259–286
Fritz H, Duncan P (1994) On the carrying capacity for large ungulates of African savanna ecosystems.
P Roy Soc Lond B Bio 256:77–82
Fritz H, Duncan P, Gordon J, Illius AW (2002) Megaherbivores influence trophic guilds structure
in African ungulate communities. Oecologia 131:620–625
Fritz H, Loison A (2006) Large herbivores across biomes. In: Danell K, Bergström R, Duncan P,
Pastor J (eds) Large herbivore ecology, ecosystem dynamics and conservation. Cambridge
University Press, Cambridge, pp 19–49
Gallina, S (1993) White-tailed deer and cattle diets at La Michellia, Durango, Mexico. J Range
Manage 46:487–492
Gamfeldt L, Hillebrand H, Jonsson PR (2005) Species richness changes across two trophic levels
simultaneously affect prey and consumer biomass. Ecol Lett 8:696–703
Genin D, Villca Z, Abasto P (1994) Diet selection and utilization by llama and sheep in high-
altitude arid rangeland of Bolivia. J Range Manage 47:245–248
Gogan PJP, Barrett RH (1995) Elk and deer diets in a coastal prairie-scrub mosaic, California.
J Range Manage 48:327–335
Gordon IJ (2003) Browsing and grazing ruminants: are they the same beasts? Forest Ecol Manag
181:13–21
Hamilton D (1976) Performance of sheep and cattle grazing together in different ratios. Aust J Exp
Agr Anim Husb 16:5–12
Hansen RM, Reid LD (1975) Diet overlap of deer, elk, and cattle in southern Colorado. J Range
Manage 28:43–47
Hardy MB, Tainton NM (1995) The effects of mixed species grazing on the performance of cattle
and sheep in Highland Sourveld. Afr J Range Forage Sci 12:97–103
Hart, JA (1986) Comparative dietary ecology of a community of frugivorous forest ungulates in
Zaire. Thesis, Michigan State University
Hiremath AJ, Ewel JJ (2001) Ecosystem nutrient use efficiency, productivity, and nutrient accrual
in model tropical communities. Ecosystems 4:669–682
Holyoak M, Loreau M (2006) Reconciling empirical ecology with neutral community models.
Ecology 87:1370–1377
Hubbell SP (2001) The unified neutral theory of biodiversity and biogeography. Princeton
University Press, Princeton, NJ
Hofmann, RR (1973) The ruminant stomach: stomach structure and feeding habits of East African
game ruminants. East African Literature Bureau, Nairobi
Hubbard RE, Hansen RM (1976) Diets of wild horses, cattle, and mule deer in the Piceance basin,
Colorado. J Range Manage 29:389–392
Huisman J, Olff H (1998) Competition and facilitation in multi-species plant–herbivore systems
of productive environments. Ecol Lett 1:25–29
Hutchinson GE (1957) Homage to Santa Rosalia or why are there so many kind of animals? Am
Nat 93:145–159
Illius AW, Gordon IJ (1987) The allometry of food intake in grazing ruminants. J Anim Ecol
56:989–999
Janis CM, Damuth J, Theodor JM (2000) Miocene ungulates and terrestrial productivity: where
have all the browsers gone? Proc Nat Acad USA 97:7899–7904
Jewell PA (1980) Ecology and management of game animals and domestic livestock in African
savannas. In: Human ecology in savanna environments. Academic Press, London, pp
353–381
Jiang L, Morin PJ (2005) Predator diet breadth influences the relative importance of bottom-up
and top-down control of prey biomass and diversity. Am Nat 165:350–363
Leibold MA, McPeek MA (2006) Coexistence of the niche and neutral perspectives in community
ecology. Ecology 87:1399–1410
Leite ER, de Araujo Filho JA, Pinto FC (1995) Combined grazing with goats and sheep in lowered
caatinga: performance of pasture and of animals. Pesqui Agropecu Brasil 30:1129–1134
Lightfoot CJ, Posselt J (1977) Eland (Taurotragus oryx) as a ranching animal complementary to
cattle in Rhodesia. 2. Habitat and diet selection. Rhod Agr J 74:53–61
Logan JL, Jennings PG, McLaren LE (1991) Mixed grazing of cattle and sheep. Proc VI World
Red Poll Congress, Jamaica Agricultural Development Foundation, Kingston, pp 99–101
Long ZT, Morin PJ (2005) Effects of organism size and community composition on ecosystem
functioning. Ecol Lett 8:1271–1282
Loreau M (1998) Biodiversity and ecosystem functioning: a mechanistic model. Proc Nat Acad
Sci USA 95:5632–5636
Loreau, M (2000) Biodiversity and ecosystem functioning: recent theoretical advances. Oikos
91:3–17
Loreau M, Naeem S, Inchausi P et al (2001) Ecology, biodiversity and ecosystem functioning:
current knowledge and future challenges. Science 294:804–808
Lucich GC, Hansen RM (1981) Autumn mule deer foods on heavily grazed cattle ranges in
northwestern Colorado. J Range Manage 34:72–73
Mahieu M, Aumont G, Michaux Y et al (1997) Mixed grazing sheep/cattle on irrigated pastures
in Martinique (FWI). Prod Anim 10:55–65
Makhabu SW (2005) Resource partitioning within a browsing guild in a key habitat, the Chobe
riverfront, Botswana. J Trop Ecol 21:641–649
Martinez A, Osoro K, Lemaire G (2002) Yearling calves live weight gains and productivity under
single or mixed spring grazing with goat or sheep. In: Durand JL, Emile JC, Huyghe C (eds)
Multi-function grasslands: quality forages, animal products and landscapes. Proc 19th
European Grassland Federation Meeting, Versailles, pp 1050–1051
McNaughton SJ, Ruess RW, Seagle SW (1988) Large mammals and process dynamics in African
ecosystems. Biosci 38:794–800
Mellink E (1995) Use of Sonoran rangelands: lessons from the Pleistocene. In: Steadman DW,
Mead JI (eds) Late quaternary environments and deep history: a tribute to Paul S. Martin.
Mammoth Site of Hot Springs, Hot Springs, SD, pp 50–60
Ménard C, Duncan P, Fleurance G, Georges JY, Lila M (2002) Comparative foraging and nutrition
of horses and cattle in European wetlands. J Appl Ecol 39:120–133
Milton SJ (2000) Theme: interactions between diversity and animal production in natural range-
lands. Afr J Range Forage Sci 17:1–3, 7–9
Mishra C (2001) High altitude survival: conflicts between pastoralism and wildlife in the Trans-
Himalaya. Thesis, Wageningen University
Mitchell TD (1985) Goats in land and pasture. In: Copland JW (ed) Goat production and research
in the tropics. Australian Centre for International Agricultural Research, Canberra, pp
115–166
Mwasi SM (2002) Compressed nature: co-existing grazers in a small reserve in Kenya. Thesis,
Wageningen University
Mysterud A (2000) Diet overlap among ruminants in Fennoscandia. Oecologia 124:130–137
Nolan T, Connolly J (1989) Mixed versus mono-grazing by steers and sheep. Anim Prod
48:519–533
Nolan T Pulina G, Sikosana JLN, Connolly J (1999) Mixed animal type grazing research under
temperate and semi-arid conditions. Outlook Agr 28:117–128
Oindo B (2002) Patterns of herbivore species richness in Kenya and current ecoclimatic stability.
Biodivers Conserv 11:1205–1221
Oindo B, Skidmore AK, Prins HHT (2001) Body size and abundance relationship: an index for
diversity for herbivores. Biodivers Conserv 10:1923–1931
Olff H, Ritchie ME, Prins HHT (2002) Global environmental controls of diversity in large
herbivores. Nature 415:901–904
Olson KC, Wiedmeier RD, Browne JE, Hurst RL (1999) Livestock response to multispecies and
deferred-rotation grazing on forested rangeland. J Range Manage 52:462–470
Osoro K, Martinez A, Celaya R, Vassalo JM (2000) The effects of mixed grazing with goats on
performance of yearling calves in perennial ryegrass with clover pastures. In: Rook AJ,
Penning PD (eds) Grazing management: the principles and practice of grazing, for profit and
environmental gain, within temperate grassland systems. BBSRC Institute of Grassland and
Environmental Research, Aberystwyth, UK, pp 115–116
Owen-Smith N (1985) The ecological potential of the kudu for commercial production in savanna
regions. J Grassland Soc Southern Afr 2:7–10
Pfisterer AB, Schmidt B (2002) Diversity-dependent production can decrease the stability of
ecosystem functioning. Nature 416:84–86
Pordomingo AJ, Rucci T (2000) Red deer and cattle diet composition in La Pampa, Argentina.
Prins HHT, de Boer WF, van Oeveren H, Correira A, Mafuca J, Olff H (2006) Co-existence and
niche segregation of three small bovids in southern Mozambique. Afr J Ecol 44:186–198
Prins HHT, Olff H (1998) Species richness of African grazer assemblages: towards a functional
explanation. In: Newbery DM, Prins HHT, Brown ND (eds) Dynamics of tropical communities.
BES Symp vol 37, Blackwell, Oxford, p 449–490
Robbins CT, Spalinger DE, Van Hoven W (1995) Adaptations of ruminants to browse and grass
diets: are anatomical-based browser–grazer interpretations valid? Oecologia 103:208–213
Schwartz CC, Ellis JE (1981) Feeding ecology and niche separation in some native and domestic
ungulates on the shortgrass prairie. J Appl Ecol 18:343–353
Shrestha R, Wegge P, Koirala RA (2005) Summer diets of wild and domestic ungulates in Nepal
Himalaya. J Zool 266:111–119
Storms D, Said S, Fritz H, Hamann J-L, Saint-Andrieux C, Klein F (2006) Influence of hurricane
Lothar on red and roe deer winter diets in the northern Vosges, France. Forest Ecol Manag
237:164–169
Thill RE, Martin A (1986) Deer and cattle diet overlap on Louisiana pine–bluestem range. J Wildl
Manage 50:707–713
Tilman D (1999) The ecological consequences of changes in biodiversity: a search for general
principles. Ecology 80:1455–174
Tilman D, Lehman CL, Thomson KT (1997) Plant diversity and ecosystem productivity: theoreti-
cal considerations. Proc Nat Acad Sci USA 94:1857–1861
Tilman D, Reich PB, Knops J, Wedin D, Mielke T, Lehman C (2001) Diversity and productivity
in a long-term grassland experiment. Science 294:843–845
Vandermeer J, Lawrence D, Symstad A, Hobbie S (2002) Effect of biodiversity on ecosystem
functioning in managed ecosystems. In: Loreau M, Naeem S, Inchausti P (eds) Biodiversity
and ecosystem functioning. Oxford University Press, Oxford, pp 221–233
Van Wieren SE (1996) Do large herbivores select a diet that maximizes short term digestible
energy intake? Forest Ecol Manag 88:149–156
Voeten MM, Prins HHT (1999) Resource partitioning between sympatric wild and domestic her-
bivores in the Tarangire region of Tanzania. Oecologia 120:287–294
Vrba, ES (1987) Ecology in relation to speciation rates: some case histories of Miocene–recent
mammal clades. Evol Ecol 1:283–300
Wilson AD, Mulham WE (1980) Vegetation changes and animal productivity under sheep and
goat grazing on an arid belah (Casuarina cristata)–rosewood (Heterodendrum oleifolium)
woodland in western New South Wales. Aust Rangeland J 2:183–188
Yachi S, Loreau M (1999) Biodiversity and ecosystem productivity in a fluctuating environment:
the insurance hypothesis. Proc Nat Acad Sci USA 96:1463–1468
Chapter 8
Impacts of Grazing and Browsing
by Large Herbivores on Soils
and Soil Biological Properties
Kathryn A. Harrison and Richard D. Bardgett
8.1 Introduction
Herbivores can have a wide range of effects on terrestrial ecosystems. Some of

these effects are direct, such as the removal and consumption of herbage—which
can vary some 100-fold across terrestrial ecosystems from less than 1% to greater
than 60% (McNaughton et al. 1989)—treading on soil and vegetation, and the
return of excreta (Floate 1981). Herbivores also have important indirect effects on
ecosystems, altering rates of nutrient cycling and changing nutrient availability to
plants (Bardgett and Wardle 2003; Bardgett 2005). These indirect effects of herbiv-
ores on ecosystems are mediated by feedbacks that occur between plants and
below-ground decomposer communities, and especially soil microbes, which play
a central role regulating nutrient availability to plants.
This chapter examines some of the many ways that above-ground herbivores might
indirectly influence ecosystem properties through their influence on below-ground
organisms and processes. We also examine how these changes in soil biological proper-
ties may in turn feed back to affect above-ground primary production in both positive
and negative ways. The chapter will highlight the different mechanisms by which
herbivores affect decomposer organisms and their activities over various temporal and
spatial scales, ranging from short-term responses at the individual plant level, to long-
term responses at the level of the plant community. The effects of different types of
herbivory, namely grazing and browsing, on soil biological properties will also be
discussed. The mechanisms highlighted will be illustrated with specific examples from
a range of different ecosystems, thereby providing a global perspective on herbivore
effects on soils.
8.2 Herbivore Effects on Nutrient Dynamics
The effects of herbivory on plant community structure and function have been
widely studied (e.g., McNaughton 1984; Haukioja et al. 1990; Dyer et al. 1993;
Collins et al. 1998; Lehtilä et al. 2000). However, the indirect effects of herbivory
© Springer 2008
202 K.A. Harrison and R.D. Bardgett
on below-ground properties have often been overlooked. In recent years, it has

become apparent that there exists an important link between above-ground and
below-ground processes, and that these links are affected by herbivores. A review
by Bardgett et al. (1998) was one of the first to address this issue, giving an account
of how foliar herbivory influences soil organisms and decomposer food webs.
Historically, much work in this area has been done in grassland ecosystems (Floate
1970a, b; Bardgett et al. 1997; McNaughton et al. 1997; Tracy and Frank 1998),
although work has also been done on salt marshes (Van Wijnen and Van der Wal
1999), oak savanna (Ritchie et al. 1998), boreal forests (Pastor et al. 1993; Stark
et al. 2000), arctic tundra (Van der Wal et al. 2004) and New Zealand forests
(Wardle et al. 2001).
Many of the indirect effects of herbivory have been identified, however, the
processes involved are complex, and effects vary on a case-by-case basis (Bardgett
and Wardle 2003; Bardgett 2005). The majority of effects, such as increased
rhizodeposition and the production of secondary metabolites do, however, culmi-
nate in a soil process response, mediated by soil microbes and microfauna; this
might ultimately alter rates of nutrient cycling. In many terrestrial ecosystems,
nitrogen (N) is the main nutrient limiting plant growth (Vitousek and Howarth
1991). Therefore, the availability and recycling rates of nitrogen are important
factors controlling primary productivity (Van Wijnen et al. 1999).
Previous studies have identified two hypotheses for herbivore effects on soil
microbial properties and nutrient cycling: the accelerating effect and the decelerat-
ing effect (Ritchie et al. 1998; Bardgett et al. 1998; Fig. 8.1). The accelerating
effect predicts that herbivory will stimulate soil microbial activity, thereby promoting
rates of nutrient cycling. This effect is most commonly seen when herbivory takes
the form of grazing on grassland species. In contrast, the decelerating effect
predicts that herbivores will have a negative effect on soil biological properties and
hence ecosystem productivity. Conversely, this effect is most commonly exhibited
when herbivory takes the form of browsing on tree species. In this chapter we
review some of the many ways in which above-ground herbivory can positively or
negatively feed back on plant nutrition and productivity via changes in below-ground
processes, such as nutrient cycling rates and decomposition.
8.3 Positive Feedback Effects of Above-Ground Herbivory
Positive effects of herbivory on soil biota and nutrient cycling occur when dominant
plant species respond to grazing by exhibiting compensatory growth (Augustine
and McNaughton 1998). This mechanism is most common in grasslands of high
soil fertility (Fig. 8.2), where herbivory in the form of grazing positively affects the
decomposer subsystem through preventing colonisation of later successional plants
which produce poorer litter quality, as well as through returning carbon and
nutrients to the soil in labile forms such as dung and urine, and as enhanced
8 Impacts of Grazing and Browsing by Large Herbivores on Soils 203
Fig. 8.1 Feedback loops illustrating (A) the acceleration effect and (B) the deceleration
effect of herbivores on feedbacks between plant species and nutrient cycling. Arrows indicate
net indirect effect of herbivores on the abundance of plants or the rate of the process (from
Ritchie et al. 1998)
A. FERTILE, PRODUCTIVE B. INFERTILE, UNPRODUCTIVE

ECOSYSTEMS ECOSYSTEMS
HERBIVORES
High % of NPP consumed Low % of NPP consumed
High return of labile fecal material to soil Low return of fecal material to soil, high litter return
Cause retardation of succession, leading to Cause acceleration of succession, leading to
domination by plants with high litter quality dominaiton by plants with low litter quality
PLANTS
Fast growing, short lived Slow growing, long lived

High allocation of C to gwoth High C allocation to secondary metabolites
High specific leaf area Low specific leaf area
Short leaf life span Long leaf life span
High forage quality Low forage quality
LITTER
High % N Low % N
Low phenolics High phenolics
Low lignin and structural carbohydrates High lignin and structural carbohydrates
SOIL FOOD WEB
Bacterial-based energy channel Fungal-based energy channel

High density earthworms High density enchytraeid worms
Relatively low density microarthropods High density macro- and microarthropods
SOIL PROCESSES
High bioturbation of soil Low soil mixing

Rapid decomposition and nutrient mineralization Slow decomposition and nutrient mineralization
Low soil C sequestration High soil C sequestration
High nutrient supply rates Low nutrient supply rates
Fig. 8.2 Herbivore effects on ecosystems can vary depending on whether they impact on (A)
fertile systems supporting high herbivory, typically in the form of grazing, or (B) infertile
habitats supporting low herbivory, typically in the form of browsing. Herbivore-driven changes
in plant species composition can indirectly feed back in a positive or negative way to influence
the quality and quantity of resources (i.e., plant litter) entering the decomposed subsystem. The
linkages between below-ground and above-ground systems feedback (dotted line) to the plant
community positively (A) in fertile conditions and negatively (B) in infertile ecosystems
(adapted from Wardle et al. 2004)
rhizodeposition (Holland and Detling 1990; Holland et al. 1996; Bardgett et al.
1997; McNaughton et al. 1997; Frank and Groffman 1998; Hamilton and Frank
2001). The mechanisms by which grazing by herbivores has been shown to
positively feedback to increase nutrient cycling rates will now be described, with
examples given of how these processes function in different ecosystems.
8.3.1 Urine and Dung Deposition
By depositing urine and dung, nitrogen is recycled in forms that are more available to
plants and soil microbes (Frank and Evans 1997). Plant material that is removed by
animals is digested and nutrients are released far quicker into soil from the resulting
faeces than if released from litter, leading to increased nutrient availability and plant
nutrient uptake. In many studies, the deposition of urine and dung has been shown to
directly affect nutrient cycling and availability. For example, Stark et al. (2000) in a
study of the effect of reindeer grazing in Scots pine forests in Finnish Lapland, stated
that urine and faeces enhanced nutrient cycling, strengthening the positive effect that
other identified factors, such as removal of lichen cover, had on nutrient cycling.
As increased soil microbial activity leads to accelerated nitrogen cycling in urine
and faecal patches, a positive feedback mechanism may occur, whereby increased soil
N availability leads to a greater tissue N concentration of plants. This in turn increases
the probability that these areas will be re-grazed and thus receive extra excretory inputs,
further enhancing recycling rates (Frank and Groffman 1998). McNaughton et al.
(1997) found that urination from grazers, such as Thompson’s (Gazella thompsoni) and
Grant’s (Gazella granti) gazelles in Serengeti National Park, Tanzania, enriched soils
with N from urea, leading to a burst of organic matter mineralization that produced
greater available mineral N in the soil than would occur solely from urea addition. This,
together with other factors, such as increased leaf N concentration, led to a two-fold
increase in rates of N mineralization in soils supporting dense resident animal popula-
tions compared with those where animals were uncommon.
The deposition of animal waste by herbivores has been shown to increase
microbial biomass and stimulate microbial activity, which in turn increases nutrient
cycling rates in grassland. (Bardgett et al. 1997, 2001; Tracy and Frank 1998).
In British upland grasslands, urine and dung from grazing cattle and sheep have
been shown to increase soil microbial activity and N cycling, and hence plant
production (Floate 1970a, b). Similarly, Bardgett et al. (2001) found that micro-
bial biomass was maximal at moderate levels of sheep grazing on semi-natural
grassland, which was attributed to increased inputs of labile C to soil from root
exudation and animal wastes, and also changes in litter quality due to vegetation
change. Grazing of Agrostis-Festuca and Nardus dominated hill grasslands in
Britain was also found to stimulate microbial biomass and activity, and the abun-
dance of soil microfauna, which together regulate nutrient cycling (Bardgett et al.
1997). In dry grassland and shrub-grassland in Yellowstone National Park, Frank
and McNaughton (1992) detected a positive association between dung deposition
and above-ground primary production, and suggested that grazing and productivity
are coupled to herbivore-facilitated nutrient cycling in this system. Tracy and
Frank (1998) also found that grazing by elk (Cervus elaphus), bison (Bison
bison), and pronghorn antelope (Antilocarpa americana) in Yellowstone caused
increased microbial biomass and rates of N mineralization in soil. These authors
hypothesized that microbial populations in grazed grassland were sustained
mainly by inputs of labile C from dung deposition and increased root turnover or
root exudation beneath grazed plants. In a tundra ecosystem in Spitsbergen, Van

der Wal et al. (2004) found that soil microbial biomass C and N were stimulated
by increasing the density of reindeer faeces (Rangifer tarandus platyrhynchus)
applied to the ground and that this led to an increase in the standing crop of
grasses; these findings were taken to indicate that large herbivores, such as reindeer,
have the ability to manipulate their own food supply in a positive way.
Although urine and dung deposition undoubtedly have positive effects on below-
ground properties and rates of nutrient cycling, it has been argued that it cannot
explain the widely documented positive effects of herbivores at large spatial scales
(Hamilton and Frank 2001). This is because, typically, waste patches remain local-
ised within an ecosystem, influencing a relatively small surface area (Augustine
and Frank 2001). Also, in some cases, the positive effects of animal wastes are
insufficient to negate other, adverse effects of herbivory on nutrient mineralization
and are therefore not apparent at the ecosystem scale (Pastor et al. 1993; Bardgett
and Wardle 2003; Harrison and Bardgett 2003, 2004).
8.3.2 Alterations in Plant C and N Allocation
The allocation of C in plants is often closely linked to the allocation of N; therefore,

these nutrients must be considered in tandem. The effects of herbivory on plant
nutrient flow are evident as either short-term changes in plant C and N allocation
and root exudation, or as long-term changes in root biomass. The following sum-
marizes specific examples of how these responses may positively affect soil micro-
bial communities and hence rates of nutrient cycling in soil.
8.3.2.1 Short-Term Responses
In the short-term, the quantity of resources supplied to the soil can be altered through
effects of herbivory on plant C and N allocation and on root turnover and exudation
patterns (Bardgett 2005). Above-ground herbivory of grasses has been shown to
increase assimilate allocation to below-ground components of plants and decrease
allocation to the shoots. For example, laboratory experiments on grasses have found
that clipping can lead to an accumulation of assimilates in below-ground organs and
increased root respiration and exudation (Bokhari and Singh 1974; Dyer and Bokhari
1976). Furthermore, increased root exudation, as a result of foliar herbivory, has been
shown to greatly influence soil microbial community structure and stimulate soil
microflora and microfauna and C use efficiency by microbes in the rhizosphere
(Bardgett et al. 1998; Mawdsley and Bardgett 1997; Guitian and Bardgett 2001). This
increased microbial productivity in the rhizosphere following herbivory may also
favour higher level consumers such as enchytraeids and microbe-feeding nematodes
in the soil microfood-web (Wardle 2002). These below-ground effects of herbivory
can feed back positively on plant growth and plant N content. For example, Hamilton
and Frank (2001) found that defoliation of Poa pratensis promoted root exudation of
carbon, which was quickly assimilated by soil microbes increasing their biomass,
in turn increasing soil N mineralization and plant N uptake, thereby increasing
the growth and N status of the re-growing plant. Similarly, simulated herbivory
in the form of browsing of tree seedlings has been shown to lead to enhanced N
mineralization and inorganic N availability in rhizosphere soil, presumably
owing to a stimulation of biological activity resulting from increased rhizodepo-
sition (Ayres et al. 2004).
It has been established, therefore, that herbivores can alter plant C and N alloca-
tion leading to a positive feedback cycle that increases soil microbial activity and
rates of nutrient cycling. Guitian and Bardgett (2000), in microcosm study, exam-
ined the response of three dominant grass species to different intensities of defolia-
tion and found that defoliation of grasses that were tolerant to grazing (e.g., Festuca
rubra and Cynosurus cristatus) led to an increased allocation of resources to
shoots, whereas defoliation of grasses with a low tolerance to grazing (e.g.,
Anthoxanthum odoratum) led to an increase in the relative allocation of resources
below ground. Mikola et al. (2001a, b) also found that defoliation of white clover
(Trifolium repens), perennial ryegrass (Lolium perenne), and ribwort plantain
(Plantago lanceolata) led to an increased allocation of resources to above-ground
growth, which they suggested was typical of species adapted to intense, but infre-
quent defoliation. This implies that grazing causes alterations in the allocation of C
and N within the plant, and that the direction the nutrients are allocated can depend
on species.
8.3.2.2 Long-term Responses
The long-term effects of herbivory on plant C and N allocation are primarily

apparent through alterations in root biomass and morphology. There is mixed
evidence about how herbivory affects root productivity, with positive (Milchunas
and Laurenroth 1993), negative (Guitian and Bardgett 2000; Mikola et al. 2001a;
Ruess et al. 1998) and neutral (McNaughton et al. 1998) effects of herbivory on
root biomass being detected. In the long-term, above-ground herbivory has been
shown to reduce root biomass and change root morphology and architecture
(Bardgett et al. 1998). For example, in a study on prairie dog colonies in Wind
Cave National Park, South Dakota, grazing was found to decrease N allocation to
the roots and decrease root biomass. This had the effect of decreasing N immobiliza-
tion, due to reduced microbial growth, and increasing net N mineralization and plant
available N in colonised prairie dog colonies (Holland and Detling 1990).
Enhancement of root biomass by grazing is also possible and has been shown to
occur with increased rates of soil nutrient cycling. For example, a study carried out
by Chaneton et al. (1996) on temperate sub-humid grassland in Argentina, found that
grazing leads to the allocation of N to below-ground organs, increasing root biomass
and the amount of underground nutrient circulation. Although herbivore-induced
increases in soil nutrient cycling had previously been identified in a Serengeti
grassland (McNaughton et al. 1997), other studies in this ecosystem showed that
root biomass was not affected by grazing as there was no significant difference in
mean root biomass on an annual basis between fenced and un-fenced plots
(McNaughton et al. 1998). Herbivores, therefore, have been shown to have positive,
negative and no effect on root biomass whilst still accelerating nutrient cycling rates.
This indicates that a long-term ecosystem, response to herbivory may depend on the
plant species, be they woody tree species, forbs or grasses, and on the ecosystem in
question, and the relative impacts of other abiotic factors, such as topography and
climate, on below-ground processes such as decomposition and rates of soil nutrient
cycling.
8.3.3 Selective Foraging on Less Nutritious Species
Commonly, herbivores have been shown to selectively forage on nutrient rich

plant species (Pastor et al. 1993; Ritchie et al. 1998; Van Wijnen et al.1999),
which can ultimately lead to decreased rates of nutrient cycling (see Sect.
8.4.1). However, selective foraging on less nutritious species by herbivores has
been shown to increase soil nutrient cycling rates; by selectively feeding on
foliage from plants of lower nutritional content, herbivores by-pass the slower
litter decomposition pathway and deposit more readily decomposable faecal
material on soil, which is more rapidly made available for plant uptake. Stark
et al. (2000) presented data concerning such an incidence. These authors inves-
tigated the effect of reindeer grazing in the lichen-dominated Scots Pine forests
of Finnish Lapland. Lichen produces litter that decomposes slowly and hence,
reindeer grazing reduced the amount of litter that was produced. By ingesting
and metabolising lichen, the reindeer allowed nutrients to be released and
made available for plant growth at a faster rate than if left on the ground to be
decomposed by soil microbes.
8.3.4 Litter Deposition
As herbivores remove vegetation by grazing and browsing, litter production falls

(Ruess and Seagle 1994). Biomass is, therefore, re-directed from the slow decom-
position pathways in soil food webs to the herbivores, which recycle nutrients more
efficiently and return plant-available nitrogen to the soil (in the form of urine and
dung) at a faster rate (Coughenour 1991). By bypassing the decomposition process
in this way, nutrient cycling rates are increased (Hamilton et al. 1998). Reducing
litter build-up, therefore, can increase the productivity of an ecosystem (Milchunas
and Laurenroth 1993). McNaughton et al. (1988) reviewed the effects of grazing
by herbivores within the Serengeti and found that grasslands where the major ungulate
herds concentrated during the wet season had nutrient turnover times of less
than a year. However, fenced plots accumulated substantial quantities of litter,
and turnover times exceeded five years. Therefore, the removal of grazing herbivores
can slow nutrient cycling rates and reduce the productivity of an ecosystem.
8.3.5 Increased Soil Temperature
By removing vegetation, herbivores may increase soil microbial activity and hence
nutrient cycling rates by allowing more light to reach the soil surface leading to an
increase in soil temperature (Pastor et al. 1993). Areas that remain ungrazed have
been found to have lower soil temperatures, due to the thick layer of litter present.
This inhibits microbial activity and reduces net N mineralization (Frank and
Groffman 1998). For example, exclusion of grazing by barnacle geese and reindeer
over a period of seven years at Ny-Ålesund, Spitsbergen, caused an increase in the
thickness of the moss layer, and a reduction in soil temperature of 0.9° C (Van der
Wal et al. 2001). Conversely, areas affected by herbivory have less litter build up (see
Sect. 8.3.5) allowing soil temperatures to rise, which stimulates microbial activity and
increases net N mineralization. Similarly, studies in the high Arctic, Spitsbergen,
have shown that reindeer grazing promotes soil N availability and plant productivity
(Van der Wal et al. 2004). In this study, experimental faecal addition to moss domi-
nated tundra was shown to increase grass growth and microbial biomass in soil, but
also to lead to a reduction in the depth of the moss layer influencing soil temperature.
These results took some three years to develop and the cause for this was thought to
be a direct fertilizing effect of faeces in these strongly nutrient-limited situations, and
to a suppressive effect of faeces on the depth of the moss layer, which strongly regu-
lates soil temperature owing to its ability to hold moisture (Brooker and Van der Wal
2003). Studies on the effects of herbivory carried out on upland steppe in Yellowstone
National Park (Coughenour 1991), in seasonally dry high country in New Zealand
(McIntosh et al. 1997), and in oak savanna in Minnesota (Ritchie et al. 1998), have
also found that grazing elevates soil temperature thereby accelerating decomposition
and mineralization of organic matter by soil biota.
8.4 Negative Feedback Effects of Above-Ground Herbivory
Negative effects of herbivory on soil biota and rates of nutrient cycling most
commonly occur in unproductive ecosystems, where browsing on leaves and
woody material is the dominant form of herbivory (Fig. 8.2). Here, low consump-
tion rates and selective foraging on nutrient-rich plants can lead to the dominance
of defended plants that produce recalcitrant litter (Ritchie et al. 1998). Since most
nutrients will be returned to the soil as recalcitrant plant litter, the net effect of
herbivory in these low productivity ecosystems is to reduce soil biotic activity,
nutrient mineralization, and supply rates of nutrients from soil, despite inputs of
dung and urine (Pastor et al. 1993). Negative feedbacks may also occur where grazing
and inappropriate stocking densities affect soil physical properties leading to

reduced microbial activity and increased nutrient loss through erosion. These
mechanisms will now be discussed in detail and illustrated with specific examples
of ecosystems where these processes have been identified.
8.4.1 Selective Foraging on Nutrient-Rich Tissue
Plant species vary in the quality of foliage that they produce. Species (e.g., some
woody plants and trees) that produce high amounts of defence compounds, which
are less nutritious, also produce litter of poorer quality which is less readily decom-
posed by soil microorganisms. Since herbivore digestion and decomposition are
both regulated by microorganisms with similar enzyme complements, digestion of
plant material with high concentrations of secondary compounds will be slower
than digestion of plant material of a better quality with a greater nutrient content
(Pastor et al. 1993; Chesson 1997). It is reasonable to suggest, therefore, that her-
bivores might selectively browse on plant species with nutrient-rich tissue, and
seek to avoid those species with foliage of high secondary compound concentra-
tion, and hence of a lower nutritional quality (Bardgett et al. 1998). In doing so,
browsers may indirectly alter plant community structure encouraging the domi-
nance of less nutritious plants with lower quality litter, which is slow to decompose.
This shift would have negative impacts on below-ground processes governed by
soil microbes, such as decomposition and soil nutrient cycling rates, ultimately
reducing ecosystem productivity (Pastor et al. 1993; Kielland and Bryant 1998;
Ritchie et al. 1998).
The importance of selective feeding by herbivores for soil processes was
illustrated by a study of moose browsing in the boreal forests of the Isle Royale
National Park in Lake Superior (Pastor et al. 1993). These authors showed that
selective foraging by moose on hardwoods, with nutrient rich tissue, led to the
dominance of less nutritious species such as spruce, which produce litter of lower
quality and hence decomposability. This low quality litter was slow to decom-
pose and built up on the soil surface, leading to a reduction in soil nitrogen min-
eralization (presumably due to low N availability from litter and also reduced soil
temperature, which can lead to a less favourable microclimate for soil microbes
and mesofauna) and hence a decline in the productivity of the ecosystem (Pastor
et al. 1993). Many other studies on browsing and grazing have also found that
selective foraging increases the dominance of less nutritious species and ulti-
mately slows down rates of nutrient cycling. For example, Van Wijnen et al.
(1999) found that grazing by geese, hares and rabbits on areas of salt marsh
encouraged plants such as Limonium vulgare to dominate. L. vulgare is a less
nutritious species with leaves of high tannin content which decompose slowly,
hence decreasing rates of mineralization (Van Wijnen et al. 1999). Herbivory, in
the form of browsing, on oak savanna in Minnesota by deer, rabbits, and a variety
of insects, was also found to decrease nutrient cycling rates due to selective
foraging by herbivores on nutrient-rich tissue (Ritchie et al. 1998). By selectively

foraging on woody plants and legumes, prairie grasses such as Andropogon
geradi and Sorghastrum nutans became dominant. The litter from these grasses
was found to decompose slowly, thereby decelerating nutrient cycling rates in
this ecosystem. Conversely, the exclusion of herbivores in these areas led to an
increase in cover and biomass of legumes and woody plants. The nutrient-rich
tissue of these plants decomposed relatively fast, therefore increasing rates of
nutrient cycling (Ritchie et al. 1998).
Kielland and Bryant (1998) found that selective browsing by moose in Alaskan
taiga led to a replacement of more nutritious deciduous species with less nutritious
evergreens, causing slower nutrient turnover. In situations such as this, a positive
feedback system may subsequently arise, whereby reduced rates of nutrient turno-
ver and nutrient availability (due to the presence of less nutritious species) favour
growth of nutrient-poor species, which reduce their loss of nutrients to the environ-
ment, allowing increased dominance in nutrient-poor habitats. This feedback
strengthens the effects of the herbivores on the ecosystem and could lead to addi-
tional reductions in above-ground production and rates of nutrient cycling (Ritchie
et al. 1998). In New Zealand’s forests, however, although browsing by introduced
mammals (feral goats and deer) significantly altered plant community composition,
reducing nutritious broad-leaved species and promoting other less nutritious types,
subsequent effects on measures of soil nutrient status were highly idiosyncratic, with
an equal number of positive and negative effects being detected across thirty
locations (Wardle et al. 2001).
8.4.2 Production of Secondary Metabolites
As outlined above, one factor determining the palatability and decomposability of plant
material is the concentration of secondary metabolites within the leaves. Browsers can
induce the production of secondary metabolites in foliage, which negatively impacts on
soil biota due to reduced litter quality (Bardgett 2005). For example, severe defoliation
of trees, such as that caused by periodic invertebrate attack, often results in reduced
concentrations of N and increased concentrations of certain secondary metabolites
(e.g., phenolics) in subsequently produced foliage (Rhoades 1985). Findlay et al. (1996)
reported that damage to leaves from actions including herbivory results in nitrogen-
complexation with bound phenolic material. These authors showed that cellular damage
caused by spider mites to seedlings of Populus deltoides increased the concentration of
polyphenols in foliage, resulting in a 50% reduction in the decomposition rate of subse-
quently produced litter. Compounds, such as polyphenols, are relatively resistant to
decomposition as they inhibit the actions of soil microbes (Hättenschwiler and Vitousek
2000), and it has been suggested that they may also act to tie up nitrogen (Findlay et al.
1996). By inhibiting the activity of soil microbes, the presence of high concentrations
of plant secondary metabolites can reduce nitrogen return via N mineralization, therefore
decreasing rates of soil nutrient cycling.
8.4.3 Impact on Soil Physical Properties
Grazing, and inappropriate stocking densities, can have adverse effects on ecosystems,
which include poaching of soils, compaction and erosion. These processes can, in
turn, have negative consequences for both land and water quality as well as the
exchange of greenhouse gases. For example, grazing was found to reduce soil
organic matter C and N content and microbial biomass in a study carried out in
Utah, USA, which looked at the destabilising effects of grazers on soil surfaces
(Neff et al. 2005). These authors found strong evidence that grazing triggers wind
erosion, largely due to the disruption of biological soil crusts and long-term
changes in vegetation cover/composition, and results in significant nutrient loss in
this semi-arid setting.
Herbivore effects on soil physical properties have also been shown to have
negative feedback on soil biota, potentially decreasing nutrient cycling rates.
For example, increased sheep stocking density on an Australian pasture (10,
20, and 30 sheep ha−1) severely reduced numbers of Collembola in the surface
soil (King and Hutchinson 1976; King et al. 1976). Similarly, reductions in
collembollan numbers was associated with increased stocking density of a
lowland perennial ryegrass (Lolium perenne) grassland (Walsingham 1976).
These responses were attributed to changes in soil pore space, which was
greatly reduced with increased sheep grazing. Similarly, browsing by intro-
duced goats and deer in New Zealand’s natural forests was found to have
consistently adverse effects on mesofaunal and macrofaunal groups; this was
attributed to the physical effects of the browsers, namely trampling and scuff-
ing (Wardle et al. 2001). Herbivore induced alterations in soil microclimate
have also been shown to have negative effects on the decomposer subsystem
and hence nutrient cycling rates. Stark et al. (2000) found that removal of the
protective cover of lichens by reindeer grazing led to the exposure of soil
biota to a less favourable microclimate, which was predicted to have a signifi-
cant negative affect on soil microbial processes in dry oligotrophic Scots pine
forest in Fennoscandinavia. This was contrary to the observation that micro-
bial activity was greater in grazed than ungrazed areas (see Sect. 8.3.5 also
for effects of grazing on soil temperature). These authors suggested, however,
that the relative dominance of these positive and negative effects of grazing
may vary depending on season. Reindeer trampling, which compacts the soil
structure, has also been shown to crush plant roots thereby reducing carbon
input to the soil (Stark et al. 2003).
8.5 Conclusions
Herbivores impact on ecosystems via a variety of mechanisms, the dominance of

which determines whether a positive or negative response is seen. These mechanisms
operate at different spatial and temporal scales, for example, at the individual plant
level (via the production of secondary metabolites), through to the ecosystem scale
(via shifts in plant community structure), and at the short-term, for example hours to
weeks (via root exudation) up to the long-term, for example months to years (via
changes in root biomass).
Predicting whether herbivores have a positive or negative affect on processes such
as decomposition and soil nutrient cycling in a particular ecosystem is therefore
difficult, as the mechanisms involved invariably interact with each other, and also
with abiotic factors such as climate and topography (Verchot et al. 2002). However,
it is becoming clearer that it may be possible to predict the impacts herbivores have
on soil nutrient cycling by looking at the type of ecosystem they are part of and the
nature of their herbivory. For example, positive effects of herbivory on soil biota and
nutrient cycling are more likely to occur when dominant plant species respond to
grazing by exhibiting compensatory growth (Augustine and McNaughton 1998). This
appears to be most common in productive grasslands, were herbivory in the form of
grazing positively affects the decomposer subsystem through preventing colonisation
of later successional plants which produce poorer litter quality, as well as through
returning carbon and nutrients to the soil in labile forms as dung and urine, and as
enhanced rhizodeposition (McNaughton et al. 1989; Holland and Detling 1990;
Holland et al. 1996; Bardgett et al. 1997; Frank and Groffman 1998; Hamilton and
Frank 2001; Bardgett and Wardle 2003). In contrast, negative effects of herbivory on
soil properties and ecosystem productivity may be more likely to occur in unproduc-
tive ecosystems, where browsing, rather than grazing, is most common. In these sys-
tems, low consumption rates by herbivores and selective foraging on nutrient rich
plants can lead to the dominance of defended plants that produce recalcitrant litter
(Ritchie et al. 1998). The net effect of herbivory in these low productivity ecosystems
is often to reduce soil biotic activity, nutrient mineralization, and supply rates of
nutrients from soil, despite inputs of dung and urine (Pastor et al. 1993).
In this chapter we have illustrated how above- and below-ground linkages, such
as those mediated by herbivores, can have important consequences on a number of
ecosystem properties. It is clear, therefore, that future research in this area needs to
address such interactions and feedbacks. Herbivores have repeatedly been shown
to have positive, negative, and neutral effects on ecosystem processes, such as
nutrient mineralization and decomposition, and it is this context dependency which
now holds the greatest challenge for research in the future. It is only when we con-
sider the interactions of herbivores and their environment at different temporal and
spatial scales and the interactions of biotic and abiotic factors in the environment
that we will be able to more accurately predict herbivore effects on ecosystems.
References
Augustine DJ, Frank DA (2001) Effects of migratory ungulates on spatial heterogeneity of soil
nitrogen properties in a grassland ecosystem. Ecology 82:3149–3162
Augustine DJ, McNaughton SJ (1998) Ungulate effects on the functional species composition of
plant communities: Herbivore selectivity and plant tolerance. J Wildl Manage 52:1165–1183
Ayres E, Heath J, Possell M, Black HIJ, Kerstiens G, Bardgett RD (2004) Tree physiological
responses to above-ground herbivory directly modify below-ground processes of soil carbon
and nitrogen cycling. Ecol Lett 7:469–479
Bardgett RD (2005) The biology of soil: a community and ecosystem approach. Oxford University
Press. Oxford, UK, pp 242
Bardgett RD, Leemans DK, Cook R, Hobbs P (1997) Seasonality of the soil biota of grazed and
ungrazed hill grasslands. Soil Biol Biochem 29:1285–1294
Bardgett RD, Wardle DA (2003) Herbivore mediated linkages between above-ground and
below-ground communities. Ecology 84:2258–2268
Bardgett RD, Wardle DA, Yeates GW (1998) Linking above-ground and below-ground interactions:
How plant responses to foliar herbivory influence soil organisms. Soil Biol Biochem
30:1867–1878
Bardgett RD, Jones AC, Jones DL, Kemmitt SJ, Cook R, Hobbs P (2001) Soil microbial community
patterns related to the history and intensity of grazing in sub-montane ecosystems. Soil Biol
Biochem 33:1653–1664
Bokhari UG, Singh JS (1974) Effects of temperature and clipping on growth, carbohydrate reserves
and root exudation of western wheatgrass in hydroponic culture. Crop Sci 14:790–794
Brooker R, Van der Wal R (2003) Can soil temperature direct the composition of high Arctic plant
communities? J Veg Sci 14:535–542
Chaneton EJ, Lemcoff JH, Lavado RS (1996) Nitrogen and phosphorus cycling in grazed and
ungrazed plots in a temperate subhumid grassland in Argentina. JAppl Ecol 33:291–302
Chesson A (1997) Plant degradation by ruminants: parallels with litter decomposition in soils.
In: Cadisch G, Giller KE (eds) Driven by nature: plant litter quality and decomposition.
CAB International, Wallingford, UK, pp 47–66
Collins SL, Knapp AK, Briggs JM, Blair JM, Steinauer EM (1998) Modulation of diversity by
grazing and mowing in native tallgrass prairie. Science 280:745–747
Coughenour MB (1991) Biomass and N responses to grazing of upland steppe on Yellowstone’s
northern winter range. J Appl Ecol 28:71–82
Dyer MI, Bokhari UG (1976) Plant–animal interactions: Studies of the effects of grasshopper
grazing on blue grama grass. Ecology 57:762–772
Dyer MI, Turner CL, Seastedt TR (1993) Herbivory and its consequences. Ecol Appl 3:10–16
Findlay S, Carreiro M, Krischik V, Jones CG (1996) Effects of damage to living plants on leaf
litter quality. Ecol Appl 6:269–275
Floate MJS (1970a) Mineralization of nitrogen and phosphorus from organic materials of plant
and animal origin and its significance in the nutrient cycle in grazed upland hills and soils.
J Brit Grassland Soc 25:295–302
Floate MJS (1970b) Decomposition of organic materials from hill soils and pastures II.
Comparative studies on the mineralization of carbon, nitrogen and phosphorus from plant
materials and sheep faeces. Soil Biol Biochem 2:173–185
Floate MJS (1981) Effects of grazing by large herbivores on N cycling in agricultural ecosystems.
In: Clark FE, Rosswall T (eds) Terrestrial nitrogen cycles – processes, ecosystem strategies and man-
agement impacts. Ecol Bull Swedish Nature Science Research Council, Stockholm, pp 585–597
Frank DA, Evans RD (1997) Effects of native grazers on grassland N cycling in Yellowstone
National Park. Ecology 78:2238–2248
Frank DA, Groffman PM (1998) Ungulate vs. landscape control of soil carbon and nitrogen
processes in grasslands of Yellowstone National Park. Ecology 79:2229–2241
Frank DA, McNaughton SJ (1992) The ecology of plants, large mammalian herbivores, and
drought in Yellowstone National Park. Ecology 73:2043–2058
Guitian R, Bardgett RD (2000) Plant and soil microbial response to defoliation in temperate
semi-natural grassland. Plant Soil 220:271–277
Hamilton EW, Frank DA (2001) Can plants stimulate soil microbes and their own nutrient supply?
Evidence from a grazing tolerant grass. Ecology 82:2397–2402
Hamilton EW, Giovannini EW, Moses, MS, Coleman JS, McNaughton SJ (1998) Biomass and
mineral element responses of a Serengeti short-grass species to nitrogen supply and defoliation:
Compensation requires a critical [N]. Oecologia 116:407–418
Harrison KA, Bardgett RD (2003) How browsing by red deer impacts on litter decomposition in
a native regenerating woodland in the Highlands of Scotland. Biol Fert Soils 38:393–399
Harrison KA, Bardgett RD (2004) Browsing by red deer negatively impacts on soil nitrogen avail-
ability in regenerating native forest. Soil Biol Biochem 36:115–126
Hättenschwiler S, Vitousek PM (2000) The role of polyphenols in terrestrial ecosystem nutrient
cycling. TREE 15:238–243
Haukioja E, Ruohomaki K, Senn J, Soumela J, Walls M (1990) Consequences of herbivory in the
mountain birch (Betula pubescens spp tortuosa): Importance of the functional organisation of
the tree. Oecologia 82:238–247
Holland JN, Cheng W, Crossley Jr DA (1996) Herbivore-induced changes in plant carbon allocation:
Assessment of below-ground carbon fluxes using C-14. Oecologia 107:87–94
Holland EA, Detling JK (1990) Plant response to herbivory and below-ground nitrogen cycling.
Ecology 71:1040–1049
Kielland K, Bryant JP (1998) Moose herbivory in Taiga: Effects on biochemistry and vegetation
dynamics in primary succession. Oikos 82:377–383
King LK, Hutchinson KJ (1976) The effects of sheep stocking intensity on the abundance and
distribution of mesofauna in pastures. J Appl Ecol 13:41–55
King LK, Hutchinson KJ Greenslade P (1976) The effects of sheep numbers on associations of
Collembola in sown pastures. J Appl Ecol 13:731–739
Lehtilä K, Haukioja E, Kaitaniemei P, Laine RA (2000) Allocation of resources within mountain
birch canopy after simulated winter browsing. Oikos 90:160–170
Mawdsley JL, Bardgett RD (1997) Continuous defoliation of perennial ryegrass (Lolium perenne)
and white clover (Trifolium repens) and associated changes in the composition and activity of
the microbial population of an upland grassland soil. Biol Fert Soils 24:52–58
McIntosh PD, Allen RB, Scott N (1997) Effects of exclosure and management on biomass and
soil nutrient pools in seasonally dry high county, New Zealand. J Environ Manage
51:169–186
McNaughton SJ (1984) Grazing lawns: animals in herds, plant form and co-evolution. Am Nat
124:863–886
McNaughton SJ, Banyikwa FF, McNaughton MM (1997) Promotion of the cycling of diet-enhancing
nutrients by African grazers. Science 278:1798–1800
McNaughton SJ, Banyikwa FF, McNaughton MM (1998) Root biomass and productivity in a
grazing ecosystem: the Serengeti. Ecology 79:587–592
McNaughton SJ, Oesterheld M, Frank DA, Williams KJ (1989) Ecosystem-level patterns of
primary productivity and herbivory in terrestrial habitats. Nature 341:142–144
McNaughton SJ, Ruess RW, Seagle SW (1988) Large mammals and process dynamics in African
ecosystems. Herbivorous mammals affect primary productivity and regulate recycling balances.
BioSci 38:794–800
Mikola J, Yeates GW, Barker GM, Wardle DA, Bonner KI (2001a) Effects of defoliation intensity
on soil food-web properties in an experimental grassland community. Oikos 92:333–343
Mikola J Yeates GW, Wardle DA, Barker GM, Bonner KI (2001b) Response of soil food-web
structure to defoliation of different plant species combinations in an experimental grassland
community. Soil Biol Biochem 33:205–214
Milchunas DG, Laurenroth WK (1993) Quantitative effects of grazing on vegetation and soil over
a global range of environments. Ecol Monogr 63:327–366
Neff JC, Reynolds, RL, Belnap, J, Lamothe, P (2005) Multi-decadal impacts of grazing on soil
physical and biogeochemical properties in southeast Utah. Ecol Appl 15:87–95
Pastor J Dewey B, Naiman RJ, McInnes PF, Cohen Y (1993) Moose browsing and soil fertility in
the boreal forests of Isle Royale National Park. Ecology 74:467–480
Rhoades DF (1985) Offensive–defensive interactions between herbivores and plants: their
relevance in herbivore population dynamics and ecological theory. Am Nat 125:205–238
Ritchie ME, Tilman D, Knops JMH (1998) Herbivore effects on plant and nitrogen dynamics in
oak savanna. Ecology 79:165–177
Ruess RW, Hendrick RL, Bryant JP (1998) Regulation of fine root dynamics by mammalian
browsers in early successional Alaskan taiga forests. Ecology 79:2706–2720.
Ruess RW, Seagle SW (1994) Landscape patterns in soil microbial processes in the Serengeti
National Park, Tanzania. Ecology 75:892–904
Stark S, Tuomi J, Strömmer R, Helle T (2003) Non-parallel changes in soil microbial carbon and
nitrogen dynamics due to reindeer grazing in northern boreal forests. Ecography 26:51–59
Stark S, Wardle DA, Ohtonen R, Helle T, Yeates GW (2000) The effect of reindeer grazing on
decomposition, mineralisation and soil biota in a dry oligotrophic Scots Pine forest. Oikos
90:301–310
Tracy BF, Frank DA (1998) Herbivore influence on soil microbial biomass and N mineralisation
in a northern grassland ecosystem: Yellowstone National Park. Oecologia 114:556–562
Van der Wal R, Bardgett RD, Harrison KA, Stien A (2004) Vertebrate herbivores and ecosystem
control: cascading effects of faeces on tundra ecosystems. Ecography 27:242–252
Van der Wal R, Brooker R, Cooper E, Langvatn R (2001) Differential effects of reindeer on high
Arctic lichens. J Veg Sci 12:705–710
Van der Wal R, Pearce ISK, Brooker R, Scott D, Welch D, Woodin SJ (2003) Interplay between
nitrogen deposition and grazing causes habitat degradation. Ecol Lett 6:141–146
Van Wijnen HJ, Van der Wal R (1999) The impact of herbivores on nitrogen mineralisation rate:
Consequences for salt-marsh succession. Oecologia 118:225–231
Verchot LV, Groffman PM, Frank DA (2002) Landscape versus ungulate control of gross
mineralisation and gross nitrification in semi-arid grassland of Yellowstone National
Park. Soil Biol Biochem 34:1691–1699
Vitousek PM, Howarth RW (1991) Nitrogen limitation on land and in the sea – how can it occur.
Biogeochem 13:87–115
Walsingham JM (1976) Effect of sheep grazing on the invertebrate population of agricultural
grassland. Proc R Soc Dublin 11:297–304
Wardle DA, Barker GM, Yeates GW, Bonner KI, Ghani A (2001) Introduced browsing mammals
in New Zealand natural forests: above-ground and below-ground consequences. Ecol Monogr
71:587–614
Wardle DA (2002) Communities and ecosystems: linking the aboveground and belowground
components. Monogr Pop Biol 34. Princeton University Press, NJ
Wardle DA, Bardget RD, Klironomos JN, Setälä H, van der Putten WH, Wall DH (2004)
Ecological linkages between aboveground and belowground biota. Science 304:1629–1633
Chapter 9
Plant Traits, Browsing and Grazing Herbivores,
and Vegetation Dynamics
Christina Skarpe and Alison Hester
9.1 Introduction
Large browsing and grazing herbivores can have a profound influence on the
physiognomy, composition, and function of vegetation, from the landscape scale to
a single plant (Hobbs 1996; Augustine and McNaughton 1998). By selective forag-
ing among populations of plant modules and genets and along resource gradients,
herbivores exert differential pressures on plant populations at different spatial and
temporal scales (Kielland and Bryant 1998; Allison 1990; Price 1991; Jia et al.
1995; Danell et al. 2003). In addition, herbivore behavioural costs (predation,
shelter seeking, etc.) can strongly influence locational, and therefore foraging,
choice (Schmitz 2003; Letourneau and Dyer 2004). Such differential herbivory
pressures have consequences for vegetation composition, ecosystem properties and
for plant evolution (Pastor and Naiman 1992; Jefferies et al. 1994; Danell et al.
2003). In this chapter we review some aspects of how large herbivores can modify
vegetation, limiting the discussion to vascular plants and mammalian herbivores
>ca. 5 kg in weight. In order to understand the interactions between large herbivores
and vegetation we first show how various plant traits influence the foraging pattern
by large herbivores, the amount of resources the plant loses in a foraging event, and
the way the plant responds to the loss. This is important for the effects of herbivory
on the competitive hierarchy among plants, which in turn govern vegetation
dynamics and composition. We start with discussing how plant architecture,
including that of trees/shrubs, forbs, and graminoids, influences patterns of forag-
ing by large herbivores, the relative losses of resources by the plant, and the plant
responses to those losses. We then discuss the resource economy of plants, how
resources are acquired, and how they are allocated to growth and storage, and the
implications of plant resource economy for plant performance following herbivory.
This is followed by a discussion of plant strategies to resist herbivory, by either
avoiding being eaten or minimising the negative effects of losses to herbivores. The
significance of animal foraging behaviour is also touched upon, but this forms the
main topic in several other chapters of the book. Finally, we discuss how the dif-
ferences in plant properties and in herbivore utilisation among plant populations at
different scales lead to shifts in competitive relations among plants and subsequently
© Springer 2008
218 C. Skarpe and A. Hester
to shifts in vegetation properties, such as composition of populations, species, and

functional types. Different conceptual models proposed to explain such shifts in
vegetation are discussed.
9.2 Plant Architecture and Herbivory
9.2.1 Introduction
Vascular plant heights span five orders of magnitude, from minute, <1 cm, prostrate
herbs to rainforest trees of more than 50 m. The associated range in biomass, related
more to volume than to height, is even larger. The difference in plant height corresponds
to a similar difference in the spatial distribution of photosynthetic biomass, flowers, and
fruits of potential interest for foraging large herbivores (Harper 1977; Gill 1992, 2006;
Bodmer and Ward 2006). Similarly, meristems are differently distributed in space,
which is of importance for plant survival and resprouting following biomass removal
(Raunkiær 1937; Haukioja and Koricheva 2000). The difference in size and architecture
of plants is related to many other plant attributes, such as life history, longevity, and
reproductive strategies. Size and architecture of plants also influence the proportion of
the biomass that is edible and accessible for terrestrial large mammalian herbivores
(Harper 1977; Crawley 1983; Haukioja and Koricheva 2000).
In order to understand plant strategies and responses in relation to herbivory, it
is important to realise that plants are modular organisms, which are very different
from mammals, for example. A genet, i.e., a plant originating from sexual repro-
duction, normally a seed, can in clonal species give rise to several ramets, looking
more or less similar to the mother plant. Both genets and ramets are composed of
smaller units, that we for simplicity also call modules, like shoots or tillers and
leaves or needles (Harper 1977; Waller 1997). Under different conditions plant
modules may behave either as members of the individual, exchanging resources
with other modules, or as partly independent units, competing between themselves
for resources. Such different behaviour has profound implications for the effect of
herbivory on plants. With strong connectivity, resources can be drawn from many
modules to ameliorate the loss of biomass in any part of the plant (Wilsey 2002).
In contrast, with low connectivity resources are less mobile between modules, and
the distribution of herbivory in the plant becomes much more important.
9.2.2 Trees and Shrubs
Common for all woody species, from minute arctic willows with only shoot tips
and leaves exposed above the moss layer, to giant tropical rainforest trees, is the
relatively large allocation of biomass to woody structures (Hytteborn 1975; Körner
9 Plant Traits, Browsing and Grazing Herbivores, and Vegetation Dynamics 219
1994). In most species, the primary function of woody stems and branches is to
support the canopy and to transport water, mineral nutrients, and photosynthates
between roots and leaves (Devlin 1966). The common explanation for the tall stems
of trees is to lift the canopy higher than its neighbours in order to compete for light.
Alternative explanations, for example for the occurrence of relatively tall trees in
savanna or wooded steppe, where competition for light is negligible once canopies
are above the grass layer, may be to reduce the risk of scorching by frequent grass
fires, or to lift photosynthetic biomass out of reach of ground-living herbivores.
Many species also store nutrients and carbohydrates in stems and branches
(Honkanen et al. 1994; Vanderklein and Reich 1999; Millard et al. 2001). Woody
biomass is relatively inedible for most mammalian herbivores and constitutes a
protected resource of high importance for plant survival and resprouting after
herbivory (van der Meijden et al. 1988).
Woody plants are composed of different kinds of shoots that can be defined in
different ways (Kozlowski 1966). Here we simply separate between long shoots
and short shoots. All new shoots grow from a meristem on an older shoot, often
from the apical meristem in the tip of the old shoot. In addition long shoots bear
several nodes with axillary meristems and leaves (Waller 1997). It is the long
shoots that contribute to the growth and architecture of trees. Long shoots and their
leaves are the modules most frequently browsed by large herbivores (Danell et al.
1994; Shipley et al. 1999), often with effects for the growth form of the tree (see
below). Some tree genera also develop short shoots, which do not elongate, but
produce one short internode and one leaf or a bouquet of leaves, or needles in coni-
fers, each year (Atkinson et al. 1992). In species with plastic or indeterminate
growth, such as most deciduous trees (Atkinson 1992; Raspe et al. 2000), short
shoots may develop into long shoots following damage to other long shoots, e.g.,
by browsing (Danell et al. 1985). Thus, in such cases, short shoots might be
regarded as resting buds with a means to be self-supporting with carbon. A third
type of shoots are basal shoots or suckers, developing at the stem base primarily
after severe damage to the plant, e.g., by herbivores or fire, and having the ability to
develop into new ramets, ensuring persistence of the plant (Bond and Midgley
2001; Gill 2006). Some woody plants have clonal growth, implying that suckers
may sprout from roots at some distance from the mother plant, so that one genet
may give raise to many ramets. This is common in dwarf shrubs, e.g., Vaccinium spp.
(Flower-Ellis 1971) but also occurs in many shrubs (Hester et al. 2006a) and a few
trees, such as aspen (Populus spp.) (Peterson and Jones 1997). Ramets, as well as
shoots and leaves, are modules of a genet. Modularity in woody plants is generally
more complex than in herbs and often displays a lower degree of physiological inte-
gration (Beck et al. 1982; Sachs and Hassidim 1996; Peterson and Jones 1997;
Haukioja and Koricheva 2000). Thus, responses to browsing may be localised to the
modules closest to the damage (Danell et al. 2003), and the response by indi-
vidual modules may differ profoundly from that of the whole tree or clone
(Haukioja and Koricheva 2000).
Canopy architecture, i.e., the position of branches, shoots and leaves within a
canopy, differs between plant species and affects how much light the leaves may
capture (Waller 1997). Canopy architecture may also influence browsing by large
herbivores, as widely dispersed shoots may give low foraging efficiency, whereas
densely packed small shoots can be difficult for a browser to attack (Renaud et al.
2003). As shown later in this chapter, browsing herbivores may strongly change
tree canopy architecture, for example by modifying numbers, distribution, and
morphology of shoots, which, in turn, may affect future herbivory as well as plant
photosynthesis rate, productivity, and competitive ability (Danell et al. 2003;
Hester et al. 2006b). The strongest effect of large herbivores on tree vegetation is
usually through selective browsing on young small individuals, determining the
composition of the mature stand (Vourc’h et al. 2002; Danell et al. 2003).
9.2.3 Herbaceous Plants Other Than Graminoids
Herbaceous plants typically differ from most woody plants in being smaller, con-
sisting of fewer modules and meristems and, it is believed, tend to be more short
lived (Haukioja and Koricheva 2000), although little is known about the lifespan
of most non-woody plants (Inghe and Tamm 1985). Independent of total life span,
most herbs in seasonal climates have periodic shoot reduction to lower heights or
to subterranean parts in the winter or the dry season (Raunkiær 1937). Hence, herbs
tend to be less apparent (sensu Feeny 1976) for large herbivores both in space
and time than are trees or shrubs. On the other hand many trees and shrubs have
the advantage of growing above browse height of most terrestrial herbivores, unlike
most herbaceous plants.
While woody plants allocate most biomass to support structures and roots and
only a small proportion to photosynthetic tissue, herbs typically allocate about half
their biomass to leaves and less than a quarter to stems, which are generally not
strongly lignified (Körner 1994; Porter and Nagel 2000). If the amount of residual,
inedible, biomass is of importance for plant resprouting after herbivory, herbs
would generally have lower capacity for recovery than woody plants. In addition,
herbs would have fewer meristems from where to resprout. On the other hand, the
high allocation to photosynthetic tissue gives herbaceous plants a high potential
growth rate (mass per mass unit) compared to woody plants (Haukioja and
Koricheva 2000).
9.2.4 Graminoids
Grasses are by far the most important family of plants eaten by large herbivores,
and the speciation of grazing large herbivores is closely linked with the development
of grass-dominated vegetation during the Miocene (MacFadden 1997; Pérez-Barbería
et al. 2004; Stromberg 2004). Apart from bamboo (Bambusa spp.), grasses, i.e.,
species of the family Poaceae, are herbaceous non-woody plants, generally from
a few centimetres to a few metres high. In graminoids we include grasses and other
grass-like monocotyledonous plants such as sedges and rushes. A fundamental
difference between grasses and dicotyledonous plants, herbs as well as woody
plants, is that dicotyledons grow from apical meristems that are vulnerable to
herbivory, whereas growth in monocotyledons is from meristems (apical and inter-
calary) generally situated close to the ground. Thus, the latter are mainly below
‘optimal’ grazing height for most large herbivores (Illius and Gordon 1987) and
protected between old leaf sheaths, which make them less accessible (Wolfson
and Tainton 1999; Haukioja and Koricheva 2000).
The basic unit in a grass plant is the tiller, corresponding to the shoot in dicotyle-
donous plants (Wolfson and Tainton 1999). Vegetative grass tillers have a very short
stem with an apical meristem close to the ground. As long as the meristem is not
grazed, the tiller continues its growth after defoliation (Lemaire and Chapman 1996).
What appears to be the stem is actually densely rolled young leaves, one within the
other, from which the leaf blades develop, the oldest leaves at the base and the
youngest at the top. The leaf blade grows from an intercalary meristem at the base,
so that the tip is the oldest part of the leaf. Therefore, as long as the intercalary
meristem is active, the leaf will continue its growth even if much of the blade is
removed (Lemaire and Chapman 1996; Wolfson and Tainton 1999). Only when the
apical meristem of the tiller turns into a reproductive meristem will the stem elongate
into a culm carrying the inflorescence in its top. At that stage the meristem and inflo-
rescences are vulnerable to herbivory, and therefore a high density of grazers eating
inflorescences may almost entirely prevent sexual reproduction in grasses, giving an
advantage to vegetatively reproducing species and genotypes, which subsequently
often dominate in heavily grazed grasslands (Diaz et al. 1992; Briske 1996).
In most species of both perennial and annual grasses the initial tiller develops
lateral basal meristems or axillary buds that can give rise to new tillers sprouting
either immediately or following the death of the old tiller, e.g., after heavy grazing
(Briske 1996). The character of these adventitious tillers determines the growth
form of the plant. In tufted grasses the new tillers grow upwards and eventually form
more or less dense tussocks. In clonal species some of the tillers form rhizomes
or stolons, which develop roots and new tillers, i.e., new ramets, from axillary
buds at the nodes, which in turn may develop new tillers, stolons, or rhizomes
(Wolfson and Tainton 1999). In this way, one genet in many grasses and sedges
consists of numerous ramets with many tillers and may cover a large area. There
is some evidence to suggest that the stoloniferous growth form in grasses is an
escape strategy (see 9.4.2) to grazing by large herbivores (Wolfson and Tainton
1999; Wilsey 2002). The reduction in erect tall grasses and the increase in
decumbent and creeping morphotypes under grazing seem to depend both on the
superiority of low growing genotypes and on phenotypic plasticity in many species
(Georgiadis and McNaughton 1988; Jaramillo and Detling 1988; Painter et al.
1989; Briske 1996). A prostrate growth form allows the plant to maintain a large
proportion of biomass and meristems below the grazing height for most large her-
bivores, facilitating resprouting after grazing (Diaz et al. 1992, Briske 1996,
Wolfson and Tainton 1999). It also enables the plant to spread vegetatively in an
environment where inflorescences may often be grazed, and to quickly colonise

gaps and utilise the spatial heterogeneity in the sward imposed by grazing animals
(Wilsey 2002). At the same time the grazing of (potentially) taller neighbours is a
prerequisite for the plant to avoid being shaded out.
In many ways clonal grasses and other strongly clonal herbs resemble woody
plants in possessing a large number and array of modules and meristems. The
degree of connectivity and exchange of resources between grass modules have been
subject to controversy, but the general conclusion is that they are more developed
in graminoids than in many trees (Marshall and Sagar 1968; Langer 1972;
Danckwerts and Gordon 1987; Peterson and Jones 1997). This provides flexibility
in resource allocation and growth, and clonal grasses may be connected over large
areas exchanging resources between ramets and tillers, enhancing the overall
growth and resistance to grazing (Wilsey 2002). Stoloniferous graminoids are
among the most grazing-tolerant plants, in some ecosystems sustaining consump-
tion of up to 90% of the above-ground production (McNaughton 1985).
9.3 The Chemistry of Plants
9.3.1 Photosynthesis
The most fundamental chemical process in green plants is photosynthesis, by which

chlorophyll is used as a catalyst to bind solar energy and carbon dioxide into
energy-rich carbohydrates (Mooney 1997). A number of different pathways of
photosynthesis has evolved, increasing the effectiveness of solar energy capture
under different conditions. Two basic pathways of photosynthesis with relevance
for herbivory are the C3 and C4 types, in which the compounds produced initially
are 3-phosphoglyceric acid and four-carbon acids, respectively (Ellis et al. 1980).
The C3 pathway dominates in dicotyledonous plants and in non-tropical grasses,
while C4 photosynthesis is common in tropical grasses (Mooney 1997; Wolfson
and Tainton 1999). C4 photosynthesis requires high temperatures and in such
climates its main advantage compared to C3 photosynthesis is the low CO2
compensation point, implying that photosynthesis is efficient even with partly
closed stomata, and low (or non-existing) photorespiration (Mooney 1997; Wolfson
and Tainton 1999). This gives C4 plants higher water-use efficiency (units carbon
fixed per unit water lost) than C3 plants, making them more productive in hot dry
climates than C3 plants (Mooney 1997). However, C4 plants tend to be more ligni-
fied and less digestible than C3 plants, although Wilson and Hacker (1987) showed
that the difference in digestibility between sympatric species differing in photosyn-
thetic pathway was small. There is some evidence that C4 grasses under some
conditions may increase relative to C3 grasses under grazing (Wang 2002). C4
grasses support a large proportion of domestic and natural grazing ecosystems of
the world, e.g., the African and American savannas.
Photosynthesis rate is strongly related to nitrogen (N) concentration in the leaf,

and is generally highest in young leaves, declining with age as stomatal and meso-
phyll resistance increases (Ludlow and Wilson 1971). The photosynthetic rate of
remaining mature leaves may, however, increase following herbivory (Hodgkinson
1974; Painter and Detling 1981; Wolfson and Tainton 1999). This may be caused by
defoliation retarding the physiological aging of remaining leaves (Gifford and
Marshall 1973), by increasing quantity and quality of light for the leaves as shading
tissue is removed (Wallace 1990; Senock et al. 1991) or because competition between
leaves for resources other than light is reduced (Wareing et al. 1968). Also the crea-
tion of strong sinks for photosynthates in resprouting meristems may contribute to
this ‘rejuvenation’ (Wareing and Patrick 1975). However, the opposite response, with
reduced photosynthetic rate of remaining leaves, has also been recorded (Wallace
et al. 1985). As photosynthesis is the basic means of production in the plant, strategies
related to herbivory often aim to protect or quickly restore photosynthetic capacity.
This can be achieved, for example, by allocating resources to vegetative growth at the
expense of reproduction and by developing leaves with high specific leaf area
(Crawley 1983; Bergström et al. 2000; Skarpe and van de Wal 2002).
9.3.2 Energy and Nutrient Reserves
Carbohydrates produced in photosynthesis are stored in the leaf as vacuolar sugar

or starch during the day and are translocated or used locally for growth and
respiration during the night (Chapin et al. 1990). Young developing leaves and
shoots depend initially on carbohydrates transferred from other parts of the plant.
As leaf growth ceases, the leaf’s requirement for carbohydrates declines, and it
becomes self supporting and a net exporter of energy to young growing tissue in the
same shoot or tiller, and eventually to other sinks in the plant (Chapin et al. 1990).
Clearly, these movements of carbohydrates and nutrients within the plant have
implications for the severity, timing and location of damage on the plant from
herbivore impacts (Crawley 1983; Rosenthal and Kotanen 1994; Fitter 1997).
Nutrients are taken up from the soil solution by plant roots, often in association
with mycorrhiza (Begon et al. 1996; Fitter 1997). Nitrogen is often taken up in the
form of nitrate or ammonium. However, Ruess and McNaughton (1984) found that
some African grasses from a heavily grazed area used urea N from the urine of
grazing animals, which apparently stimulated above ground growth in these grasses
more than did ammonium N. Since then it has been found that many plants, some
of them from environments with intense herbivory, are taking up organic N
(Falkengren-Grerup et al. 2000; Bardgett et al. 2003). Nutrients are transferred
from roots to leaves, where they are used for growth and other life functions, such
as formation of proteins (Nambiar 1987; Chapin et al. 1990). Many studies have
shown that herbivores are attracted to plant material with high concentration of
protein N (Danell et al. 1991a, b; Crawley 1997). Such selective removal of plant
material can significantly reduce new growth, through reduced photosynthetic
capacity as well as loss of nutrients and carbohydrate reserves (Crawley 1983;

Rosenthal and Kotanen 1994; Crawley 1997; Millard et al. 2001).
Most perennial plants build up energy and nutrient stores over different time
scales to buffer temporal asynchrony between resource uptake and demand
(Chapin et al. 1990). There are differences in plant strategies, and species described
as adapted to frequent disturbances (e.g., by herbivory) have been found to
invest more in stored nutrient resources than species apparently not adapted
to disturbance (Bloom et al. 1985; Kruger and Reich 1997), and may also re-absorb
a larger proportion of nutrients from leaves before abscission (Pugnaire and
Chapin 1993). Similarly, plants in resource-poor environments invest propor-
tionately more in storage than do plants in resource-rich environments (Millard
and Proe 1991; Millard 1996). In seasonal environments stores of energy in
the form of starch or non-structural carbohydrates are built up late in the
growing season, when growth ceases (Langstrom et al. 1990; Holopainen et al.
1995) and resources are re-absorbed from senescing leaves (Thomas and
Stoddart 1980).
Following Chapin et al. (1990), we define storage as ‘any resources that build up
in the plant and can be mobilized later to support biosynthesis for growth or other
functions’. This includes resources specifically allocated to reserves (sensu Chapin
et al. 1990) in competition with growth and other needs of the plant. It also includes
accumulation of ‘surplus’ resources, e.g., in situations when photosynthesis or nutri-
ent absorption continues while growth is restricted (Chapin et al. 1990; Herms and
Mattson 1992; Steinlein et al. 1993). Resources employed in growth may also be
recycled and used for new growth elsewhere in the plant (Heilmeier et al. 1986).
Opportunity costs for storage, of course, decline if the stored resources have other
functions in the plant, for example, when used in defence compounds that can turn
over and support structural biosynthesis when needed (Coley et al. 1985). Plants
use recently acquired carbohydrates and nutrients before those that are stored, if
both are available, suggesting that there is an opportunity cost to storage, involving
both the growth of storage tissue (rhizomes, tubers, bulbs, etc.) and the filling of
these tissues with storage products (Thomas and Stoddart 1980; Chapin et al. 1990;
Steinlein et al. 1993) and/or the retrieval of the stored products. Woody plants use
stems and branches for storage; many plants additionally use roots or other under-
ground structures such as rhizomes, tubers and bulbs (Coleman et al. 1991; Millard
et al. 2001; Fig. 9.1). Many evergreen species do not seem to store nutrients, but
resources are left in situ in last generation’s leaves/needles, and are translocated
from there to new developing buds (Nambiar and Fife 1991; Eckstein et al. 1998;
Millard et al. 2001), leaving them vulnerable to loss as a result of herbivory
(Millard et al. 2001; Hester et al. 2004). Storage organs are often protected against
large herbivores by chemical defence and/or by being situated underground. Large
herbivores, however, often track the nutrient rich biomass. For example, elephants
have been observed (Skarpe unpublished) to feed on the green parts of grasses like
Panicum repens and Cynodon dactylon in the growing season, discarding under-
ground parts, but in the dry season to kick up and eat the nutrient rich rhizomes
while discarding the wilted shoots.
Storage locations: Vulnerability to mammalian herbivory
A: leaves and buds: High (unless out of reach)

B: woody stem: Low
C: roots and underground organs: Generally low (except for digging/uprooting)
Fig. 9.1 Plant storage locations and consequences for herbivory and its impacts
9.4 Plant Resistance
9.4.1 Introduction
Plants have ‘evolved’ various strategies which minimise the negative effect of
herbivory on plant fitness (Rosenthal and Kotanen 1994; Strauss and Agrawal
1999). Rosenthal and Kotanen (1994) called all these plant strategies ‘resistance’
and within that concept distinguished ‘avoidance strategies’, including escape and
defence strategies, and ‘tolerance strategies’.
Central to most theories on plant resistance is the assumption that all resistance
traits imply a cost for the plant, directly in terms of resource allocation and/or
indirectly in the form of ecological costs (Koricheva 2002b; Strauss et al. 2002;
Hester et al. 2006b). Only if the fitness cost of resistance is less than the cost for
the herbivory it prevents or ameliorates, will plants with resistance traits have
higher fitness and a competitive advantage compared to plants without such traits
in the same situation (Jokela et al. 2000). Thus, the interactions between resource
availability or uptake, resource loss to herbivores, and plant resistance strategies are
dynamic in space and time, complex and poorly understood (Stamp 2003; Hester
et al. 2006b). However, there is no doubt that differences in resistance strategies in
relation to prevalent herbivory regimes are a powerful means by which large
browsing and grazing herbivores influence competitive hierarchies between plants
both within and across species (Strauss et al. 2002; Cipollini et al. 2003; Hester
et al. 2006b), and, hence, vegetation composition.
Avoidance and tolerance can represent alternative evolutionary responses to
herbivory (van der Meijden et al. 1988; Juenger and Lennartsson 2000; Fig. 9.2),
and are often negatively related in comparisons among species. It was long assumed
that the two strategies are mutually exclusive, but recent studies have shown that
this is not necessarily the case (Tiffin and Rausher 1999; Mauricio 2000).
9.4.2 Escape Strategies
Plants can avoid herbivory by spatial or temporal escape. Milchunas and Noy-Meir
(2002) distinguish between internal and external avoidance, including escape,
mechanisms. Internal escape mechanisms encompass morphological traits, as dis-
cussed above, such as short stature, maintaining a large proportion of the biomass
RESISTANCE
AVOIDANCE TOLERANCE
ESCAPE DEFENCE INTRINSIC EXTRINSIC
Internal External Morphological Physiological
Physical Chemical
Fig. 9.2 Plant resistance to herbivores: conceptual strategies (derived from Rosenthal and Kotanen
1994; Strauss and Agrawal 1999; Kotanen and Rosenthal 2000; Milchunas and Noy-Meir 2002),
modified from Hester et al. (2006)
below herbivory level (depending on the size of the herbivore) or keeping most
edible biomass above reach for terrestrial herbivores. The latter strategy, however,
requires that young plants survive herbivory to attain tall growth (Vila et al. 2002).
Phenological traits making the plant less apparent (sensu Feeny 1976) or attractive
to large herbivores include deciduousness, the reduction of the whole plant to
underground organs, and survival only as seed. Storage of resources in woody
stems or underground is also a kind of escape strategy. Ward and Saltz (1994) report
that gazelles digging for lily bulbs in the Negev drive a selection of the lilies to
grow to even greater depths to escape herbivory.
In environments with high herbivory pressure, sensitive plant species mainly
occur in inaccessible places such as rock outcrops or steep slopes, thus using exter-
nal/locational escape from herbivory. Grazing-sensitive species may have survived in
such refugees for thousands of years in heavily livestock grazed areas, for example in
the Middle East (Noy-Meir and Seligman 1979; Milchunas and Noy-Meir 2002)
and probably in areas with intense natural herbivory (Skarpe et al. 2004). Such
populations may serve as sources of propagules for surrounding sink populations
that may not be able to reproduce themselves except in narrow ‘windows of oppor-
tunity’ when herbivore densities are reduced following, for example, disease or
drought (Milchunas and Noy-Meir 2002; Skarpe et al. 2004). Plants may also
escape herbivory by association with either less palatable or more palatable species,
depending on the foraging pattern of the herbivore (see 9.6.4; Hjältén et al. 1993a;
Olff et al. 1999; Hester et al. 2006b).
9.4.3 Structural and Chemical Defences
Structural defences, mainly spines, prickles and thorns of different types and origin,
are hypothesised to have evolved as a defence against foraging by large herbivores
in many parts of the world (Milewski et al. 1991; Myers and Bazely 1991; Grubb
1992; Karban et al. 1999). Spinescence occurs in a wide variety of taxa and habi-
tats, and is particularly abundant in arid and semi-arid regions and in nutrient-rich
areas of small extent (Grubb 1992; Ward and Young 2002). Spinescence has been
shown to reduce short-term intake rate of many browsers (Cooper and Owen-Smith
1986, 2001) by reducing bite rate and/or bite size (Belovsky et al. 1991, Skarpe et al.
2006). Particularly, spines tend to change foraging mode from twig biting and leaf
stripping to the less detrimental picking of leaves from between the spines (Cooper
and Owen-Smith 1986; Gowda 1996). Thus, while not actually preventing browsing,
spinescence may reduce plant losses to large herbivores.
A large number of chemical compounds in plants have been shown to have
deterrent effects on large herbivore foraging. Indeed, variation in plant defences is
a major factor governing ‘palatability’ and animal forage choice (Stamp 2003). A
plethora of hypotheses have been presented to explain the evolution of defences,
under what conditions plants should invest in defences and in what types of
defences (Feeny 1976; Bryant et al. 1983; Coley et al. 1985; Milewski et al. 1991;
Myers and Bazely 1991; Herms and Mattson 1992; Grubb 1992; Karban et al.
1999; Jokela et al. 2000; Koricheva 2002a). According to their effects on animals,
Feeny (1976) divided chemical defences into; quantitative defences’, where the
deterrent effect increases with the concentration of the defence compound, and
‘qualitative defences’, where the deterrent effect is already high at very low con-
centrations. However, Foley et al. (1999) questioned the validity of this division.
Chemical defences may be constitutive or induced (Rosenthal and Janzen 1979;
Palo and Robbins 1991; Foley et al. 1999). It is hypothesised that constitutive
defences are an evolutionary response to intense herbivory and act as deterrents,
reducing the likelihood and severity of attack. Induced chemical defences are
hypothesised to be adaptive, phenotypically plastic, responses which reduce her-
bivory only when needed, in contrast to the ‘continuous’ nature of constitutive
defences. Particularly in environments with stochastic or periodic herbivory,
induced defences may be less costly for the plant than constitutive defences
(Åström and Lundberg 1994; Cippolini et al. 2003; Hester et al. 2006b).
Feeney’s (1976) definition of quantitative defences includes many of the car-
bon-based compounds, e.g., different types of phenolics and related substances
(Palo and Robbins 1991). Compounds of the phenolic family are widespread in
plants of widely different taxonomic and geographic origin, including trees,
forbs, and grasses (Cooper et al. 1988; Sunnerheim et al. 1988; du Toit et al.
1991). Some phenolics undoubtedly reduce large herbivore foraging (Bryant et
al. 1989; Cooper and Owen-Smith 1985; Woodward and Coppock 1995), and
there is evidence that they are selected for by plants under heavy herbivory. These
types of chemical defences are mainly found in plants evolved in nutrient-poor
environments (Coley et al. 1985; Bryant et al. 1983). Such plants are inherently
slow growing and have little ability to compensate for lost biomass, and are there-
fore predicted to invest in defence (Coley et al. 1985). It has been suggested that
the concentration of carbon-based defences fluctuates with the difference between
carbon gained by the plant in photosynthesis and carbon used for growth, and
hence may be sensitive to differences in growth rate in space and time (Herms
and Mattson 1992; Stamp 2003). It is hypothesised that plants in resource-rich
environments often do not avoid herbivory but develop tolerance traits (see Sect.
9.4.4). Defence compounds, if they occur in such environments, are often effec-
tive in low concentration (Feeney’s ‘qualitative defences’), such as many alka-
loids (Coley et al. 1985). These types of defence are generally not solely carbon
based, but contain substances such as nitrogen. For fast-growing plants needing
all the carbon they acquire for growth, this is ‘cheaper’ than defences which
require high concentrations to be efficient. A problem with chemical plant
defences is the ability of herbivores to develop behavioural and physiological
counter adaptations (Iason 2005; Iason and Villalba 2006). For example, many
browsers have tannin binding proteins in their saliva, lessening the digestibility-
reducing effect of ingested tannins (Robbins et al. 1987; Austin et al. 1989;
Juntheikki 1996). More specific adaptations also exist; for example, Pass et al.
(2002) found liver enzyme adaptations to eucalypt constituents by various spe-
cialist-feeding marsupials.
9.4.4 Plant Tolerance
Plants in resource-rich environments often do not avoid herbivory, but develop

tolerance traits to minimise the harmful effects of herbivory. Also in situations
where the plant defence is ineffective in preventing loss of biomass, for example,
if the herbivore(s) is(are) insensitive to the defence, or if trampling is an important
source of damage, tolerance may be a more advantageous strategy than avoidance
(Strauss and Agrawal 1999; Jokela et al. 2000). Tolerance is defined here as ‘the
strategy by plants to minimise the reduction in fitness following herbivory’
(Rosenthal and Kotanen 1994; Briske 1996; Strauss and Agrawal 1999; Kotanen
and Rosenthal 2000). Tolerance traits include high and flexible rates of nutrient
absorption, photosynthesis and growth, and numerous protected meristems
(Bradshaw 1965; Rosenthal and Kotanen 1994; Hester et al. 2006b). Like avoid-
ance, tolerance to herbivory involves costs for the plant, e.g., the building and
maintenance of stores of energy and nutrients as well as of dormant buds that can
be activated following herbivory (McNaughton 1983; van der Meijden et al. 1988;
Bilbrough and Richards 1993). Herbivory-tolerant plants are generally palatable
to large herbivores, and have in some cases been found to increase in resource-rich
environments following herbivory or intense trampling (Olofsson et al. 2001) and
may form preferentially grazed patches where nutrient cycling may also be
enhanced by the greater herbivore presence creating a positive feedback loop (sec-
tion 9.5.5; Olofsson et al. 2001; Harrison and Bardgett this volume). Tolerance in
plants is assumed to have little direct effect on herbivore fitness, thus is considered
unlikely to trigger counter-adaptations in herbivores (Rosenthal and Kotanen
1994; Roy and Kirchener 2000).
9.5 Effects of Herbivory on Plants
9.5.1 Introduction
Foraging by large herbivores can have a wide variety of effects on plants, from
instantaneous death to increased growth and competitive ability (Crawley 1983).
As discussed above, herbivory interacts with plant architecture, photosynthesis,
dynamics of energy and nutrients, and expressions of defence and tolerance
(Crawley 1983; Hester et al. 2006b). Effects of herbivory on plants are also influ-
enced by external factors, including previous and subsequent disturbance such as
fire or drought, inter- and intraspecific competition and resource availability. The
ultimate effect on plant performance depends on the impact by the herbivore on
plant fitness, which is difficult to quantify, and is often substituted by measures of
biomass, production or reproductive output (Crawley 1997). Herbivore differen-
tial foraging among plants and populations of plants and the variation in plant
responses to being eaten can have a strong effect on the competitive relations
among plants. Thus, by increasing the inter- and intraspecific variation in fitness
of plants (Danell et al. 2003), large herbivores can have a profound influence both
on vegetation composition and plant evolution (Vourc’h et al. 2001, 2002; Danell
et al. 2003).
9.5.2 Reserve Dynamics
Plants growing in seasonal environments do not generally respond to herbivory

occurring during the dormant season until the beginning of the following growing
season, unless the impact is severe enough to cause desiccation and death, but
may respond immediately to grazing/browsing taking place during the growing
season (Senn and Haukioja 1994; Hester et al. 2004; Gill 2006). Sprouting of new
shoots and leaves following dormancy and damage inflicted during the dormant
period is generally initiated using stored resources (Chapin et al. 1990; Millard
et al. 2001). Many studies have shown that the response of trees to damage by
herbivores depends greatly on the ability of the trees to use stored nutrient
resources for growth (Haukioja et al. 1990; Honkanen et al. 1999; Millard et al.
2001; Sect. 9.3.2 this chapter). Millard et al. (2001) stressed the importance of
the location of resources in the plant for the preservation of resources for future
use, linking this to the considerable difference between evergreen and deciduous
species in their ability to resprout following herbivory (Bryant et al. 1983; Danell
et al. 1991b, 1997). As many evergreens do not store nutrients, but resources
remain in situ in the latest generation of leaves or needles (Sect. 9.3.2), such
resources are often lost to herbivores, preventing their recycling to new growth.
Millard et al. (2001) suggest that the often high level of defences in evergreen
leaves and needles may have evolved more to protect the nutrient resources than
the leaves themselves.
9.5.3 Sprouting and Resprouting
In addition to its dependence on availability of stored resources, resprouting

following herbivory requires buds that escape the herbivore and can be activated for
new growth. Regrowth following herbivory includes refoliation, sprouting of new
shoots, or the production of whole new ramets (Miquelle 1983; Danell et al. 1997;
Bond and Midgley 2001). Plants vary in their ability to develop new leaves on the
defoliated positions (Miquelle 1983). For refoliation to take place, a certain mini-
mum degree of defoliation is often required, and refoliation is more likely to occur
early in the growing season, when the expected productive lifetime of the leaf is
relatively long, than late in the season (Miquelle 1983; Danell et al. 1994; Skarpe
and van der Wal 2002; Hester et al. 2004, 2005). Likewise, resprouting of shoots is
generally most intense following herbivory early in the growing season or during
dormancy, as plants have more time to compensate before the end of the growing
season (Gill 1992; Bergström and Danell 1995; Hester et al. 2005; Guillet and
Bergström 2006). When browsing removes leading shoots the apical dominance
is reduced (Aarssen 1995), resulting in the development of many lateral shoots,
which reduces the height of the plant but often increases its lateral spread compared to
unbrowsed individuals (Belovsky 1984; Hester et al. 2004; Makhabu et al. 2006a).
High or medium intensity twig browsing in woody species generally removes sig-
nificant proportions of meristems present, in many species resulting in fewer shoots
in the following growing season (Danell et al. 1994; Bergström et al. 2000; Hester
et al. 2004). As a result there is less competition for resources among individual
shoots, and shoots can become larger and have higher nutrient concentrations than
those on undamaged trees (Danell et al.1994; Bergström et al. 2000; Rooke et al.
2004). This may also be influenced by increased root:shoot ratio following the
browsing of shoots (McNaughton 1984; Danell and Bergström 1989). The reduc-
tion in defence compounds sometimes observed in such shoots may be a result of
resources being allocated for fast growth at the expense of defence, and/or the
breakdown of existing defence compounds and their components subsequently
used for growth (Coley et al. 1985). Conversely, leaf stripping of trees during the
growing season has been shown to result in an increase in number of shoots the
following season, but a decrease in shoot size (Danell et al. 1994). Following
severe damage, e.g., by fire or very intense browsing, some trees, particularly
young ones, and many herbs may respond by sprouting basal shoots from the lower
part of the stem (Bond and Midgley 2001). Such shoots have the capability to
develop into new ramets and enhance the persistence of the plant (Bond and
Midgley 2001). Thus, large herbivores may significantly change abundance, mor-
phology, and chemistry of shoots, and hence interfere with their own future food
resource as well as with plant fitness and thereby with vegetation composition.
In grasses the apical meristems of vegetative tillers often survive grazing, as
described in Sect. 9.2.4, and can produce new leaves following a grazing event
(Lemaire and Chapman 1996; Crawley 1997; Fig. 9.3). Many grasses develop new
tillers from lateral buds on the basal part of the stem following grazing, whether or
not the apical meristem is consumed (Briske 1996). The activation of such buds is
influenced by light quality, which varies with the density of the grass canopy above
the meristem (Deregibus et al. 1985). The ability to sprout new tillers from lateral
buds also varies with species, growth form, and time of grazing (Briske 1996). It is
generally strong during much of the growing period in rhizomatous carpet-forming
species, often occupying grazed environments, and poorer in erect tussock species
(Briske 1996). In clonal species the resprouting modules may develop into leaf
bearing tillers, stolons, or rhizomes. The formation of new tillers following grazing,
in combination with often reduced internode length, may make a grazed sward
lower but denser, with higher biomass per volume and higher leaf-to-stem ratio
compared to an ungrazed sward (Hodgson 1981). Such grazed swards provide a
larger bite size and higher energy and nutrient intake rate for grazers than ungrazed
swards, and therefore may be maintained by herbivores (Hodgson 1981; Bircham
and Hodgson. 1983; Jaramillo and Detling 1988).
Fig. 9.3 Schematic picture of clonal grass plant with stolons, rhizomes, fertile and vegetative
tillers, and positions of meristems
9.5.4 Repeated Herbivory
The reduced height of browsed trees (Sect. 9.5.4 this chapter) leads to a larger
proportion of shoots and leaves remaining available within browsing height for
terrestrial herbivores (Hester et al. 2000; Makhabu et al. 2006a). Improved accessi-
bility, together with the larger size and generally higher nutritional value of
regrowth shoots, has often been observed to increase the probability of browsing
for a previously browsed plant (Löyttyniemi 1985; Danell and Bergström 1987;
Welch et al. 1991; Bergström et al. 2000; Moore et al. 2000; Makhabu and Skarpe
2006). However, reduced preferences have been recorded in other studies (Danell
et al. 1997; Duncan et al. 1998). It is not always clear why these differences are
found, but the occurrence of induced defences may offer one explanation (Bryant
et al. 1983). Increased browsing of previously browsed plants may result in a ‘feeding
loop’, where browsing-induced changes in the plant lead to further browsing and
further changes in the plant (du Toit et al. 1990). The impact of browsing megaher-
bivores, such as the African elephant (Loxodonta africana) on trees has been shown
to sometimes lead to the formation of low, intensely coppiced trees or stands of
trees with high production of preferred browse, forming what might be called a
‘browsing lawn’, analogous to ‘grazing lawns’ (see below; Jachmann and Bell
1985; Makhabu et al. 2006a). Despite the differences in observed preference or
avoidance of previously browsed plants, browsing generally increases variation
among plants in food quality for herbivores (Danell et al. 2003). Rebrowsing
implies that the targeted trees suffer repeated damage and may eventually die, but
that a smaller proportion of the population is browsed than would be expected if
browsing was randomly distributed. Conversely, a larger proportion of the

population than expected should be browsed if previously browsed trees were
avoided. Obviously, both scenarios would impact vegetation structure and compo-
sition. If the differences between trees leading to browsing in the first place are
heritable, rebrowsing could lead to a genetic change in the tree populations, with
potential effects for future vegetation composition and herbivory (Vourc’h et al.
2001, 2002). However, surprisingly little is known about what internal and external
factors govern the decision by the browser to feed from a certain plant in the first
place (Bergström and Danell 1987; Welch et al. 1991; Duncan et al. 1998).
In grasslands, the repeated utilisation by grazers of previously grazed patches with
vigorously resprouting, grazing-tolerant grasses, i.e., ‘grazing lawns’, has long been
documented (Lock 1972; McNaughton 1984; Hodgson 1981; Coppock et al. 1983;
McIntyre and Tongway 2005). The short-cropped sward of such repeatedly grazed
patches may have less biomass per area but higher biomass density (biomass per
volume), lower proportion of standing dead, higher leaf to stem ratio, and higher
concentration of nutrients and minerals than grasses in less grazed patches (Hodgson
1981; Bircham and Hodgson 1983; McNaughton 1984). Repeated grazing in nutrient
rich environments has also been shown to enhance soil nutrient availability, presum-
ably due to increased cycling of nutrients caused by higher quality litter and the
deposition of urine and faeces (Hodgson 1981; Ruess and McNaughton 1984;
Harrison and Bardgett this volume). While the ‘grazing lawn’ concept is patch based,
and a grazing lawn may include many individual plants, genets and ramets, and also
different species, the ‘browsing lawn’ concept for woody species generally includes
a single tree or a stand of trees of the same species (Jachmann and Bell 1985; du
Toit et al. 1990; Makhabu et al. 2006a). In contrast to grazing lawns, most studies of
repeated browsing have found reduced litter quantity and quality and lower soil N
concentrations in the presence of browsing compared to the situation without browsing
(Pastor et al. 1993; Ritchie et al. 1998; Persson et al. 2005). Thus, the nutrient dynamics
and persistence of a ‘browsing lawn’ may be fundamentally different from that of a
‘grazing lawn’ (Naiman et al. 2003; Fig. 9.4).
9.5.5 Compensatory Growth
Plants are defined as compensating for herbivory when the fitness of browsed or
grazed plants is the same as of undamaged plants, undercompensating or overcom-
pensating is defined as when damaged plants have lower and higher fitness, respec-
tively, than undamaged plants (McNaughton 1983; Belsky 1986; Maschinski and
Whitham 1989; Whitham et al. 1991; Noy-Meir 1993). In most studies biomass is
discussed instead of fitness, as this is easier to measure (Strauss and Agrawal
1999). There are many records of ‘overcompensation’ in terms of above-ground
biomass by grasses and sedges in nutrient rich environments (McNaughton 1983;
Paige and Whitham 1987; Hik and Jefferies 1990), and rather fewer for woody
species (Dangerfield and Modukanele 1996). However, it has been questioned
Fig. 9.4 “Browsing lawn” intensely resprouting mopane trees (Coelophospermum mopane)
repeatedly browsed by elephants (this species is not much browsed by other browsers).
whether the observed regrowth should strictly be defined as compensatory growth

(Järemo et al. 1996); in fact whether overcompensation really occurs or not may be
a matter of definition, and this is still the subject of much controversy (Belsky
1986; Crawley 1997; Hester et al. 2006b).
Tolerance traits, including fast (re)growth, stored resources, and numerous
meristems, are to a large extent species specific, and are prerequisites for resprouting
and compensatory growth (Strauss and Agrawal 1999; Juenger and Lennartsson
2000). Generally, fast growing species, particularly short-lived herbs and grasses,
have a greater ability to regrow after biomass loss than do relatively slow growing
woody species (Whitham et al. 1991; Bergström and Danell 1987), although trees
may compensate over a longer time scale (Haukioja and Koricheva 2000). Total
biomass of current season’s shoots on trees browsed (or clipped) the previous year
Fig. 9.4 b (continued) ‘Grazing lawn’ short-cropped repeatedly grazed sward in the foreground with
wildebeest (Connochaetes taurinus) and Grant’s gazelle (Gazella granti); Serengeti, Tanzania
is usually smaller or similar to that on unaffected trees (Hjältén et al. 1993b; Danell
et al. 1997; Bergström et al. 2000; Hester et al. 2004), rarely larger (Dangerfield
and Modukanele 1996, Hester et al. 2004). Generally, the more severe the damage
the poorer is the ability of the plant to compensate. However, severity of browsing
interacts with plant age and physiological (phenological) stage and with resource
availability to affect compensatory ability (Bullock et al. 2001; Gill 1992; Guillet
and Bergström 2006). Damage during the dormant period in seasonal ecosystems
generally has least effect on plant growth, as many plant resources are stored in
stems and roots at that time, although this depends on species (Gill 1992; Millard
et al. 2001). Browsing during the growing season may have greater effect on the
plant, but early herbivory may cause the smallest reduction in growth, as plants
have a longer time to compensate before the end of the growing season (Bergström
and Danell 1995; Hester et al. 2005; Guillet and Bergström 2006). In perennial
plants, older individuals, with more developed root systems and nutrient stores, are
considered more likely to compensate well for herbivory than younger plants
(Guillet and Bergström 2006). Plants with high resource availability and ‘tolerance’
traits are likely to have greater flexibility in growth and nutrient uptake than plants
in resource-poor environments, thus regrowing faster after herbivore damage
(Coley et al. 1985; Rosenthal and Kotanen 1994; Danell et al. 1997). Furthermore,
for trees, plants with high resource availability are likely to grow above browsing
height for most herbivores faster than trees in nutrient-poor environments, which
will suffer browsing for a longer period (Danell et al. 1991b, 1997). However,
plants with greater resource availability and faster growth are likely to suffer greater
losses to herbivores whilst still within browsing height (Price 1991; but see
Makhabu et al. 2006b). Danell et al. (1991b), for example, found that Scots pine
(Pinus sylvestris) along a resource gradient lost most biomass to browsers in the
resource-rich part of the gradient, but suffered most reduction in growth as a result
of herbivory in the resource-poor area.
9.5.6 Regeneration and Persistence
Herbivory may interfere with sexual reproduction in plants, either indirectly by

changing physiology and allocation of resources, or directly by consumption of flower
buds during the dormant season and flowers and fruits during the growing season.
Large herbivores may also facilitate reproduction in plants by pollinating flowers (du
Toit 2003; Skarpe personal observations) and disperse seeds attached to their bodies
or in their digestive tracts (Bodmer and Ward 2006). Most often, herbivory causes
plants to allocate resources to vegetative growth at the expense of sexual reproduc-
tion, which favours species that are able to reproduce vegetatively (Briske 1996;
Crawley 1997). Many herbivory-tolerant grasses, herbs, and dwarf shrubs are clonal,
and spread vegetatively with rhizomes and/or stolons (Briske 1996; Crawley 1997)
which gives them an advantage in vegetation under herbivory. Few trees are clonal,
in the sense that they spread vegetatively, but many trees are able to sprout new ramets
from meristems on the lower part of the stem or upper parts of the roots following
severe damage, e.g., from fire or browsing (Bond and Midgley 2001; Bergström et
al. 2000). While sprouting ability in juveniles can be regarded as a recruitment
strategy, maintained sprouting behaviour in mature trees is termed a persistence trait
(Bond and Midgley 2001), promoting survival in environments where sexual regen-
eration may be a rare event. Thus, many herbivory-tolerant plants can persist for a
considerable time, even up to thousands of years (Kay 1997; Piggott 1993), without
sexual reproduction, either by reproducing vegetatively, persisting from suckers or
simply by having a long lifespan. Populations of large herbivores generally fluctuate,
and may decline temporarily, e.g., from disease, temporary drought, or icing events,
providing a ‘window of opportunity’ for plants to reproduce, although the success
depends on the length of the vulnerable stage of the plant in relation to the duration
of the ‘window’. The result may be distinct cohorts of plants having established dur-
ing such periods of low herbivore density (Prins and van der Jeugd 1993).
9.6 Herbivore Foraging Behaviour
9.6.1 Introduction
Modes of foraging as well as degree of selectivity vary between preferential

browsers (eating woody plants and forbs) and preferential grazers (feeding on
graminoids), between ruminants requiring comparatively nutrient rich food and hind
gut fermenters, which can sustain themselves on poorer food, and among herbivores
with different body sizes (Bell 1971; Jarman and Sinclair 1979; Demment and van
Soest 1985; Gordon 2003; Makhabu 2005; Pastor et al. 2006a). This influences the
distribution and abundance of herbivore species at different spatial scales (Olff et al.
2002) and, hence, the type and degree of impact on the vegetation. For example, one
may expect hindgut fermenting megaherbivores to dominate in a nutrient-poor
savanna, such as the Chobe in northern Botswana, whereas smaller ruminants would
play a more important role in nutrient-rich systems like the Serengeti. Sensitivity to
different plant defences may also vary with animal body size and digestive system
(Palo 1987; Stokke 1999). Grazers and browsers tend to encounter very different
food resources, and, subsequently, have developed different foraging modes. It has
been hypothesised that understanding bite size and its implications is fundamental
for understanding the foraging behaviour of large herbivores (Gordon 2003).
Grasses constitute a food resource of comparatively homogeneous quality, and bite
size is closely associated with sward density. Wilmshurst et al. (1999) and Fryxell
et al. (2004, 2005) found that antelopes in the Serengeti closely track the grass
height, thus maximising their energy intake rate. In contrast, food for browsers varies
strongly in ‘available’ bite size, nutrient concentration, and defences (Spalinger and
Hobbs 1992; Gross et al. 1993; Bergman et al. 2000; Gordon 2003). Thus, food
selection in browsers tends to be more complex than for grazers and is less well
understood (Spalinger and Hobbs 1992; Gross et al. 1993; Gordon 2003).
9.6.2 Hierarchical Foraging
The direct effects of herbivores on vegetation composition are caused by the

interplay between selectivity of animal foraging and plant competitive ability.
Selective foraging has often been described as a nested hierarchy of decisions taken
by the animal at different spatial scales. This implies that palatable plants in
selected feeding sites are more heavily used by the herbivore than plants of the
same species outside such sites (Senft et al. 1987; Skarpe et al. 2000, 2007; Boyce
et al. 2003; Fryxell et al. 2004). Selected feeding sites may vary over time, as ani-
mals track food quality or availability, e.g., along an altitudinal gradient (Albon and
Langvatn 1992; Mysterud et al. 2001), or in large-scale migrations (Murray 1995).
Selection criteria may be similar at different scales (Schaefer and Messier 1995;
Skarpe et al. 2006), for example when areas with highest availability of the
preferred food are selected, whereas other studies have found that physical properties
of the landscape are more important for habitat selection at larger spatial scales
(Senft et al. 1987; Boyce et al. 2003). Selection may be related to occurrence of,
for instance, key resources, exposure, snow conditions, or insects. In areas adjacent
to permanent water in a dry climate or salt- or clay licks, herbivory may be intense
both on preferred and less preferred plants (Senft et al. 1987). Burnt areas may also
suffer intense herbivory due to young and nutrient-rich resprouting vegetation and,
in grazing systems, low proportions of dead material (Hobbs and Gimingham 1987;
Moe et al. 1990). There may also be intense interactions between fire in grasslands
and large grazing herbivores (Archibald et al. 2005).
9.6.3 Large Herbivores and Predators
In much of Europe and North America wild ungulates have increased dramatically in
recent decades with changing environments, limited hunting, and few large predators
(Fuller and Gill 2001; Côté et al. 2004). The re-introduction or natural increase of
predators such as wolves in some areas has led to changes in patterns of herbivory,
dependent more on non-lethal behavioural effects than to predation, but still causing
a decrease in ungulate populations (Lima 1998; Brown et al. 1999, 2001; Ripple and
Beschta 2004). Such behavioural changes relate to trade-off decisions between forage
optimisation and the high cost of vigilance in a predator rich habitat (Illius and
Fitzgibbon 1994). Large herbivores have been found to avoid areas with high preda-
tion risk, e.g., core areas of wolf territories or main wolf track routes, and to select
either open habitats to facilitate detection of predators, areas with good cover to hide
from predators, and/or areas with good escape possibilities (Ripple and Beschta 2003,
2004; Fortin and Beyer 2005). There is evidence that the presence of carnivores pro-
foundly changes spatial habitat use by large herbivores, with concomitant reduced
pressure on small areas with preferred vegetation. In North America such behavioural
changes in elk (Cervus elaphus) and moose (Alces alces) have been reported to
reduce the browsing pressure on some heavily preferred species, such as Populus
tremuloides and Salix spp. growing in dense vegetation with low visibility and/or
constituting an attraction for predators seeking herbivore aggregations (Fortin and
Beyer 2005; Ripple and Beschta 2005). Such behavioural differences in large herbiv-
ores between habitats with and without predators should be present also in, for exam-
ple, the predator-rich African savanna ecosystem where small, fenced reserves often
are devoid of large predators, but have to our knowledge not been studied there.
9.6.4 The Importance of Neighbours
At smaller scales the position of a plant relative to other plants of higher or lower
palatability can influence the probability of being eaten positively or negatively
(see above) depending upon what scale the herbivore makes foraging decisions.
When the herbivore makes decisions at the stand level, a palatable or less palatable
plant may escape herbivory when growing among unpalatable neighbours, but may
suffer increased risk of herbivory when growing among tasty neighbours (Hjältén
et al.1993a; Palmer et al. 2003). When the herbivore makes decisions at a finer
scale, highly palatable plants are sought out and eaten, which may reduce herbivory
on less palatable neighbours (Hjältén et al. 1993a). McNaughton (1978) found that
a palatable species was protected from grazing by African buffalo (Syncerus caffer)
and wildebeest (Connochaetus taurinus) when growing together with unpalatable
neighbours, whereas more selective Thompson gazelles (Gazella thomsonii) and
zebra (Equus burchelli) actively picked out the palatable species.
9.7 Large Herbivore Effects on Vegetation
9.7.1 Introduction
The importance of large herbivores in modifying vegetation and ecosystems

depends less on the average proportion of plant biomass removed and more on ani-
mal selectivity and exploitation of heterogeneity in plant populations, at many
scales of space and time (Hobbs 1996; Augustine and McNaughton 1998).
Generally, we suggest that the key factor governing direct impacts by large herbiv-
ores on vegetation composition and processes is the interplay between animal dif-
ferential foraging and competitive relations between plants. Generally, herbivory
reduces the competitive strength of a plant, but positive effects and no effects have
also been recorded (Grime 1979; Tilman 1988; Stuart-Hill and Tainton 1989;
Crawley 1997; Olofsson et al. 2002; Hester et al. 2006b). Although herbivory may
kill small plants, e.g., seedlings, instantly, the more common effect is a reduction
in growth and resource uptake. This affects competitive hierarchies between plants,
which can result in the replacement of strongly impacted plants by species which
are less affected (Hobbs 1996; Augustine and McNaughton 1998; Hester et al.
2006b). Indirectly, large herbivores also strongly influence vegetation and ecosys-
tems by interfering with nutrient cycling via changes in above- and below-ground
litter quantity and quality and by the deposition of faeces and urine (Pastor and
Naiman 1992; Pastor et al. 1993, 2006b; Ritchie et al. 1998; van der Wal et al.
2004;.Persson et al. 2005) Such effects are dealt with elsewhere (Harrison and
Bardgett this volume).
9.7.2 Herbivory and Composition of Plant Populations
Large herbivores feed selectively among plants within a population, exploiting

phenotypic differences, for example, in plant vigour and concentration of nutrients
and defence compounds (Price 1991; Duncan et al. 1998; Vourc’h et al. 2001;
Danell et al. 2003). The selective competitive advantage of resistance traits within
a population has attracted considerable recent interest (Weis and Hochberg 2000).
Generally, genets with resistance traits gain a competitive advantage when costs of
herbivory (in plants without resistance traits) are high, costs of resistance are low,
and the growth rate of the species is relatively low (Hjältén et al. 1993b; Verwijst
1993; Shabel and Peart 1994; Kelly 1996; Augner et al. 1997). Using a modelling
approach, Weis and Hochberg (2000) found that the advantage of resistance traits
was more obvious with asymmetric than with symmetric competition. Thus, as
herbivory often increases variation in plant size and other plant qualities (Danell et
al. 2003), it may strongly influence the composition of plant populations. Over
longer time scales, this can also influence the genetic properties of the population.
Vourc’h et al. (2001), for example, found that black tailed deer (Odocoileus hemi-
onus sitkensis) selectively browsed plants of Thuja plicata with low concentrations
of defence compounds, which favoured the persistence of plants with high concen-
trations of defences. However, Vila et al. (2002) found no differences in chemistry
or preference by deer (Odocoileus hemionus sitkensis) between individual trees in
a population of Picea sitchensis, and concluded that escape from browsing is only
a matter of age.
In dioecious plants, differences in reproductive investment may lead to differ-
ences between sexes in growth rate and chemistry, causing herbivores to distin-
guish between sexes. There is much evidence for preferential selection of male
plants over female plants (Alliende 1989; Danell et al. 1991a; Hjältén 1992; Ågren
et al. 1999). This may lead to skewed sex ratios in favour of females, as often
observed (Alliende 1986; Crawford and Balfour 1990; Danell et al. 1991a;
Dormann and Skarpe 2002). However, the correlation between sex-related foraging
preferences and plant productivity or concentration of defences is not always strong
(Danell et al. 1991a; Hjältén 1992; Dormann and Skarpe 2002), and factors such as
phenological differences between sexes, resource availability, latitude or altitude
may influence the results strongly (Danell et al. 1991a, Ågren et al. 1999).
9.7.3 Herbivory and the Composition of Plant Communities
Although the general ‘rules’ are the same, the interplay between herbivory and
interactions among plant species is often more complex than that within species.
This is because of the generally larger variation in animal selectivity and plant
response to herbivory among species than within. There may also be more complex
interactions among plants of different species than among plants of the same spe-
cies, including associations among plants leading to increased or reduced suscepti-
bility to herbivory, as discussed above (9.6.4).
The sensitivity of plant community composition, in terms of taxa or combinations
of plant traits (‘functional types’) to herbivory, has been suggested to vary with
potential primary production or resource availability (Milchunas and Lauenroth 1993;
Grime et al. 1997). Milchunas and Lauenroth (1993) compared data on grazed and
ungrazed vegetation in a worldwide data set from 236 sites of grass-dominated vege-
tation. They found that the degree of change in species composition with grazing
regime depended first of all on resource (water) availability or potential net primary
productivity, thereafter on evolutionary history of grazing and only in third place on
current herbivory intensity. Many plant traits related to shortage of resources or
‘stress’ (sensu Grime 1979) may also function as ‘neutral resistance’ to herbivory
(Edwards 1989). For example, in dry grasslands the adaptation to drought is
largely convergent with adaptation to grazing (e.g., short stature, clonality, narrow
hard leaves, and high root/shoot ratio for better competition for soil resources).
In contrast, in a sub-humid grassland the high biomass production and competi-
tion for light favours tall, broad-leaved species with a low root/shoot ratio,
which makes the plants more vulnerable to herbivory (Milchunas et al. 1988;
Milchunas and Lauenroth 1993; Grime et al. 1997). Under such conditions tall
tussock or bunch grasses frequently decrease under herbivory when in competition
with shorter, often clonal, species (Mitchley 1988; Belsky 1992; Briske 1996;
Skarpe 1991a).
Not only the degree but also the type of vegetation change in relation to herbivory
varies with resource availability, as that influences the relative competitive
advantage of different plant resistance traits (Coley et al. 1985; Herms and Mattson
1992; Grime et al. 1997; Stamp 2003, Sect. 4 in this chapter). In resource-poor
environments in particular, herbivory may lead to a decrease in much-eaten plant
species and an increase in defended unpalatable plants (Coley et al. 1985). In natural
systems this would imply a reduction in future grazing or browsing, as animals
would select other foraging areas or suffer reduced densities (Senft et al. 1987;
Hobbs 1996). However, this depends on tolerance of the herbivore to a reduction
in food quality, and may operate with considerable time lag. In domestic systems
with high animal densities and low mobility such vegetation change may be obvious,
leading to a reduction in forage quality and/or quantity (Stoddart and Smith 1955;
Skarpe 1991a). Similar development can take place in natural systems where animals
are attracted to resources other than food, e.g., water or licks (Senft et al.1987). In
nutrient-rich systems grazing may lead to a competitive advantage to plants with
tolerance traits, and may create a feedback loop whereby grazing promotes nutrient
cycling and enhanced forage quality leading to repeated grazing (McNaughton
1984, Hobbs 1996; Olofsson 2001, Sect. 9.5.4 of this chapter; Fig. 9.5). To what
extent and under what conditions browsing may lead to a corresponding increase
in trees with tolerance traits is not well understood (Jachmann and Bell 1985;
Naimann et al. 2003).
The variation in vegetation composition in response to herbivory is suggested to
be smaller in plant communities and floristic regions with a long evolutionary and
historical exposure to herbivory than in communities and floristic regions without
such a history (Milchunas and Lauenroth 1993; Milchunas et al. 1988; Ward 2004,
2006). Grubb (1992), for example, showed that functionally similar vegetation
types with a different history of herbivory differed in the relative frequency of
thorny (a herbivory-related trait) species, and that even the proportion of thorny
Fig. 9.5 Left: Feedback loop enhancing repeated herbivory in resource-rich environment. Right:
Herbivory leading to reduced forage quality and a decrease in herbivory in resource-poor
environment, modified from Skarpe et al. (2004).
species within a plant genus or family differed with different history of herbivory.
In African savannas and North American prairies with a long history of grazing,
species turnover and changes in primary production following recent changes in
grazing regime have been shown to be moderate (O’Connor 1993; Milchunas and
Lauenroth 1993). In western North America and the South American pampas,
which have had few large herbivores for a considerable time before the introduction
of livestock, grazing has more often caused a reduction in primary production and
ground cover of the vegetation and an increase in ruderal and alien species (Mack
and Thompson 1982; Milton et al. 1994; Rusch and Oesterheld 1997).
As discussed in the first section of this chapter, plant architecture is of key
importance for herbivore impacts. Contrary to grasses and forbs, trees and shrubs
often grow out of reach for browsers. Thus, it is mainly juveniles, seedlings, and
saplings that are vulnerable to browsing impacts. Many studies have shown how
ungulates can completely eliminate seedlings or saplings of certain species making
regeneration possible only in limited periods, ‘windows of opportunity’, when her-
bivore populations are low for one reason or another (Gill 1992; Prins and van der
Jeugd 1993; Kay 1997; Engelmark et al. 1998; Skarpe et al. 2004; Côté et al. 2004).
Repeatedly browsed young trees in a forest may either get killed by the herbivore
or suffer reduced competitive ability relative to other woody or herbaceous plants
(Gill 1992; Hester et al. 1996; Hulme 1996). Browser impacts on competitive rela-
tions at the seedling and sapling stage are fundamentally important in determining
the species composition of the mature tree layer, which, in turn, generally domi-
nates the ecosystem processes in the forest for a considerable period of time
(Hulme 1996; Weisberg and Bugmann 2003; Senn and Suter 2003; Gill 2006).
The recent high densities of ungulate browsers in much of Europe and North
America are considered to be the main cause of large-scale shifts in tree species
composition, population structure, and dynamics in many areas (Allison 1990;
Thompson et al. 1992; Thompson and Curran 1993; Kay 1997; Kielland and Bryant
1998; Weisberg and Bugmann 2003; Pastor et al. 2006b; Tremblay et al. 2007).
For some species, re-expansion can be rapid after reduction or removal of herbivores,
but for other species it is likely that beyond certain thresholds, a return to a more
dominant state will not occur without some other perturbation or intervention
(Tremblay et al. 2006). As evergreen species have much of their nutrient resources
accessible for large herbivores throughout the year (see above), they may be par-
ticularly vulnerable to browsing. This may be a reason for the widespread replace-
ment of evergreen Tsuga canadensis with Acer saccarum in the USA (Frelich and
Lorimer 1985; Rooney et al. 2000). However, changes in the opposite direction
have also been recorded, for example, where the evergreen Picea abies was found
to gain dominance over deciduous Betula spp. as a result of much heavier browsing
pressure by moose on Betula than on the heavily defended Picea abies (Engelmark
et al. 1998). Tree species regenerating from a ‘seedling bank’, i.e., a population of
suppressed seedlings and saplings that are released after a specific event such as
wind-throw, can be particularly exposed to repeated browsing (Tremblay et al.
2006). However, changes in forest composition are also affected by many factors
other than browsing, working at different spatial and temporal scales, including sil-
vicultural practices, suppression of fires, and cascading effects of the decline or
recovery of large predators (Kay 1997; Senn and Suter 2003; Hobbs 2003; Ripple
and Beschta 2004). There is still a lack of predictive understanding of the relative
importance of browsers in forest dynamics, for example, the relationship between
population dynamics of ungulates and of trees, and processes determining browsing
patterns, plant responses at the population and landscape level, and herbivore–
plant–soil interactions driving vegetation change (Tremblay et al. 2004, 2006; Gill
2006; Pastor et al. 2006b).
Herbivore mediated interactions among species may also lead to changes in
vegetation physiognomy and spatial diversity. To what extent large herbivores once
modified primeval forest into more open parkland is debatable (Bradshaw and
Mitchell 1999; Birks 2005; Mitchell 2005; Vera et al. 2006). However, in transition
zones between forest and open vegetation such as tundra, steppe, or savanna, large
herbivores alone or in interaction with fire may increase the openness of forest or
woodland (Laws 1970; Caughley 1976; Dublin et al. 1990; van Langevelde et al.
2003). Megaherbivores, by their sheer size and strength, may be of particular
importance in reducing the cover or biomass of trees (Caughley 1976; van de
Koppel and Prins 1998). However, recent research has emphasised the importance
of interactions among herbivores, rather than single-species effects, in driving veg-
etation dynamics (van de Koppel and Prins 1998). There is also evidence that
smaller herbivore species are often responsible for the lack of tree regeneration in
many areas (Belsky 1984; van de Koppel and Prins 1998; Prins and van de Jeugd
1993; Rutina et al. 2004). How profound such dynamics may be is illustrated by an
example from northern Botswana (Skarpe et al. 2004), where there is evidence that
a raised alluvial area close to permanent water shifted from open flats to tall wood-
land following the dramatic decline in large herbivores around 1900 caused by
rinderpest and excessive hunting for ivory. It then changed to thicket vegetation
from about the 1950s following the recovery of the herbivore populations, and is
now covered by increasingly open shrub vegetation. A number of studies have
suggested that interactions among elephants, vegetation, and, to varying extent,

fires can lead to cyclic transitions between grassland and woodland (Caughley
1976; Dublin et al. 1990). However, van de Koppel and Prins (1998), modelling the
outcome of interactions between vegetation and a large and a small herbivore,
found that oscillations between woodland and grassland were only likely under
certain conditions involving an efficient large herbivore and a smaller herbivore.
In contrast to large herbivores causing increasingly open vegetation, heavy live-
stock grazing in arid and semi-arid rangelands with summer rain frequently leads
to an increase in woody vegetation—‘bush encroachment’ or ‘shrub encroachment’
(van Vegten 1983; Skarpe 1990b). Many factors may contribute to the increase in
woody vegetation (Archer et al.1995), but the significance of heavy grazing seems
undeniable (van Vegten 1983; Skarpe 1990b). It is important in this context to sep-
arate between the often dense regeneration stands of some Acacias, for example,
which will self-thin over time, and the bush encroachment stands which do so only
to a limited extent (Skarpe 1991b). The classical explanation of the expansion of
woody plants once the grass layer is severely damaged is the ‘the two soil layer
model’ (Walter 1939; Walker et al. 1981; Walker and Noy-Meir 1982). This is
based on the assumption that woody species and grasses largely use different soil
layers, and that the decrease in competitive power of shallow rooted grasses pro-
motes deeper rooted woody species. However, this model has been challenged, and
it is obvious that the spatial separation of tree and grass roots is far from complete
(Stuart-Hill and Tainton 1989). Woody species often have deep roots enabling
them to exploit resources unavailable for grasses. However, many encroacher tree/
scrub species also have a remarkably large proportion of shallow roots (van Vegten
1983; Knoop and Walker 1985; Skarpe 1990b), possibly competing with grasses
for quickly recycled nutrients from dung and urine (Tolsma et al. 1987a, b). A new
model for bush encroachment, with important management implications, has been
presented by Ward (2006) and its main departure from the above models relates to
the conditions promoting initiation of bush encroachment.
9.8 The Theory of Vegetation Change
9.8.1 Introduction
Classical succession theory (Clements 1916) predicts deterministic, unidirectional

change between different states along a continuum from early successional to ‘climax’
vegetation. Increasing herbivory has been assumed to push the equilibrium towards
the early successional end of the gradient (retrogressive succession), whereas a
release of herbivory would lead to a progressive succession towards the ‘climax’
end. There have been a wide variety of developments of, and challenges to, this
basic theory. Here we outline just a few of the major ones relating to grazing
management.
9.8.2 Succession Theories
The range succession model is one derivative of classical succession theory and this
has been widely used by range managers across the globe since about the 1940s
(Stoddart and Smith 1943, 1955; Ellison 1960; Briske et al. 2003). The basic
assumption of the model is that grazing slows down or prevents natural succes-
sional change from one state to another. Heavy grazing and/or drought is assumed
to return communities to early successional states, and late successional states
being achieved by removal of grazing, with a whole continuum of intermediate
successional states in-between. Therefore the underlying assumption is that grazing
levels can be manipulated to maintain or create ‘desirable equilibrium’ range condi-
tions, as required (Westoby et al 1989; Briske et al 2003).
Classical succession theory, although widely and usefully applied in a range of
systems, was found to have many predictive weaknesses, with many examples
where the assumptions made did not hold and the predicted patterns of change did
not occur (e.g., Turner 1971; Noble and Slatyer 1980, West et al 1984; Walker et al
1986; Grime et al 1988; Ellis and Swift 1988; Westoby et al. 1989). One major
weakness of classical successional theories is that they have, as an underlying con-
cept, the assumption that ‘equilibrium’ is a normal or desirable state, and systems
are moved between different states by various predictable processes. Much ecosys-
tem description is also based on this assumption (Sullivan 1996). In systems where
control variables are relatively consistent and predictable, then this conceptual
approach of equilibrium dynamics works pretty well, but in more unpredictable sys-
tems, such as semi-arid areas, where rainfall in particular is highly variable and
unpredictable, non-equilibrium is in fact more the ‘norm’ and attempts to manage
for ‘desirable equilibrium states’ can be disastrous (Westoby et al. 1989; Behnke and
Scoones 1993; Sullivan 1996; Richardson et al 2005). Various definitions of equi-
librium and non-equilibrium exist in ecological literature (e.g. Briske et al. 2003;
Illius and O’Connor 2004), the former using the term ‘persistent non-equilibrium’ to
describe systems with high climatic variability. Two widely used non-equilibrium
models are state-and-transition and state-and-threshold, as described below.
9.8.3 The State-and-Transition Model
The state-and-transition (ST) model was proposed as an alternative to classical suc-

cession theory, since it accommodates non-linear and non-equilibrium theories of
plant community dynamics (Westoby et al. 1989). In brief, this model defines stable
states and transition probabilities between those states, thus enabling evaluation of
different hypotheses about vegetation processes and multidirectional responses to
different management actions, for example. ST models have been widely used for
grazing management research and application in rangeland systems, often com-
bined with ordination techniques (Gauch 1982; Foran et al. 1986; Friedel 1991;
Filet 1994; Allen-Diaz and Bartolome 1998; Stringham et al. 2003; Hill et al.
2005). Allen-Diaz and Bartolome (1998) provide a good summary of comparison

with classical succession theory, both for their own sagebrush–grass rangeland
dataset and for a range of other studies. This paper highlights how the main limita-
tion of classical succession theory is simply the emphasis on broad, linear transi-
tions. The ST models, on the other hand, give better predictive power but, to
achieve this, they require much more detailed data (both time- and site-specific) on
transitions in all directions. Such detailed data is rarely fully available for any sys-
tem and so this limits the usefulness of this modelling approach (Briske et al. 2003;
Hill et al. 2005; Bestelmeyer et al. 2006), although it is still widely used and
actively applied by both managers and modellers (Kurz et al. 2000; McIntosh et al.
2003; Letnic et al. 2004).
9.8.4 State-and-Threshold and Catastrophe Theories
Transition models fail in systems where discontinuous, irreversible changes take

place (Briske et al. 2003; Warren 2005). For this reason Laycock (1991) developed
the state-and-threshold model, which allowed for the possibility of systems failing
to return to a previous state, even when the cause of change from that state had been
removed. Discontinuous, irreversible changes are, by their nature, difficult to pre-
dict, unlike the transition model approach. However, the main criticism of this
approach was that the state-and-threshold model failed to incorporate the underly-
ing continuous vegetation processes inherent in most systems (i.e., the strength of
the previous models). It is seen that none of the models thus far gave a single
approach for continuous and discontinuous dynamics (Lockwood and Lockwood
1993; Richardson et al. 2005) The search for a unified approach resulted in the
application of catastrophe theory to ecological systems (Loehle 1989; Lockwood
and Lockwood 1991; Rietkerk et al. 1996), which was developed by Thom (1975)
as a mathematical framework to model sudden, discontinuous phenomena, but one
which mathematically could also encompass more continuous dynamics, as per the
more classical transition models (Saunders 1980; Hooley and Cohn 2003). In brief,
catastrophe theory describes the ‘qualitatively different discontinuities of a system’
(Lockwood and Lockwood 1993), e.g., rangeland condition classes, which are
driven by specific combinations of ‘control variables’, such as rainfall or tempera-
ture, or woodland understorey communities, driven by combinations of light and
soil factors (Hooley and Cohn 2003). Lockwood and Lockwood (1993) give a good
summary of two expressions of catastrophe theory: ‘fold-catastrophe’ and ‘cusp-
catastrophe’; the latter has been more widely and usefully applied in biological sci-
ences. The appeal of catastrophe theory lies in its simplification of control variables
to five or less, rather than the multiple combinations of ‘state variables’ required to
model such systems mechanistically—catastrophe theory does not require a mecha-
nistic understanding of these state interactions. However, this has also led to criti-
cism of this approach as providing little insight into the complexity of mechanisms
driving a system, which is arguably important in the development of a sound,
predictive understanding of how a system works. However, the necessity of such a

detailed understanding is debatable (particularly for management purposes), since
many systems can be well modelled with relatively few main drivers (Berryman
and Stanseth 1984; Lockwood and Lockwood 1993; Hooley and Cohn 2003;
Tremblay et al. 2004).
9.9 Conclusions
In this chapter we have outlined a range of factors which are important in defining
vegetation dynamics in response to foraging by large mammalian herbivores. We
have discussed properties of plants that make them differ in attractiveness and
availability to large herbivores, and how animals exploit the variation in such prop-
erties among plant modules and genets. We have also seen how plants vary in their
ability to deal with herbivory. We conclude that the impacts by large herbivores on
vegetation composition are determined by the interplay between the differential
foraging by large herbivores and differences in the relative competitive strength of
plants. This interplay between plants and large herbivore communities is highly
dynamic and interactive, affecting and being affected by cascading effects involv-
ing other trophic levels and abiotic factors. While the research on interactions
between plants and large herbivores has made considerable progress during the last
decades, the function of these interactions in an ecosystem context is still little
understood, particularly for browser–woody vegetation systems.
References
Aarssen LW (1995) Hypothesis for the evolution of apical dominance in plants: implications for
the interpretation of overcompensation. Oikos 74:149–156
Ågren J, Danell K, Elmqvist T, Ericson L, Hjältén J (1999) Sexual dimorphism and biotic interac-
tions. In: Geber MA, Dawson TE, Delph LE (eds) Gender and sexual dimorphism in flowering
plants. Springer, Berlin Heidelberg New York
Albon SD, Langvatn R (1992) Plant phenology and the benefits of migration in a temperate ungu-
late. Oikos 65:502–513
Allen-Diaz B, Bartolome J (1998) Sagebrush–grass vegetation dynamics: comparing classical and
state–transition models. Ecol Applic 8:795–804
Alliende MC (1986) Growth and reproduction in a dioecious tree, Salix cinerea. Thesis, Univ of
Wales
Alliende MC (1989) Demographic studies of a dioecious tree. II. The distribution of leaf predation
within and between trees. J Ecol 77:1048–1058
Allison TD (1990) The influence of deer browsing on the reproductive biology of Canada yew
(Taxus canadensis Marsh.). 1. Direct effects on pollen, ovule, and seed production. Oecologia
83:523–529
Archer S, Schimel DS, Holland EA (1995) Mechanisms of shrubland expansion – land-use, climate
or CO2. Clim Change 29:91–99
Archibald S, Bond WJ, Stock WD, Fairbanks DHK (2005) Shaping the landscape: fire-grazer
interactions in an African savanna. Ecol Appl 15:96–109
Åström M, Lundberg P (1994) Plant defence and stochastic risk of herbivory. Evol Ecol 8:288–298
Atkinson MD (1992) Biological flora of the British Isles. No. 175. Betula pendula Roth (B. verrucosa
Ehrh.) and B. pubescens Ehrh. J Ecol 80: 837–870
Augner M, Tuomi J, Rout M (1997) Effects of defoliation on competitive interactions in European
white birch. Ecology 78:2369–2377
Augustine DJ, McNaughton SJ (1998) Ungulate effects of the functional species composition of
plant communities: herbivore selectivity and plant tolerance. J Wildl Manage 62:1165–1183
Austin PJ, Suchar LA, Robbins CT, Hagerman AE (1989) Tannin-binding proteins in saliva of
deer and their absence in saliva of sheep and cattle. J Chem Ecol 15:1335–1347
Bardgett RD, Streeter TC, Bol R. (2003) Soil microbes compete effectively with plants for
organic-nitrogen inputs to temperate grasslands Ecology 84:1277–1287
Beck CB, Schmid R, Rothwell GW (1982) Stelar morphology and the primary vascular system of
seed plants. Bot Rev 48:691–815
Begon M, Harper JL, Townsend CR (1996) Ecology: individuals, populations and communities,
3rd edn. Blackwell, Oxford
Behnke RH, Scoones I (1993) Rethinking range ecology: implications for range management in
Africa. In: Behnke RH, Scoones I, Kerven C (eds). Range ecology at disequilibrium: new
models of natural variability and pastoral adaptation in African savannas. Overseas
Development Institute, London, pp 1–30
Bell RHV (1971) A grazing ecosystem in the Serengeti. Sci Am 225:86–93
Belovsky GE (1984) Moose and snowshoe hare competition and a mechanistic explanation from
foraging theory. Oecologia 61:150–159
Belovsky GE, Schmitz OJ, Slade JB, Dawson TJ (1991) Effects of spines and thorns on Australian
arid zone herbivores of different body masses. Oecologia 88:521–528
Belsky AJ (1984) Role of small browsing mammals in preventing woodland regeneration in the
Serengeti National Park, Tanzania. Afr J Ecol 22:271–279
Belsky AJ (1986) Does herbivory benefit plants? A review of the evidence. Am Nat
127:870–892
Belsky AJ (1992) Effects of grazing, disturbance and fire on species composition and diversity in
grassland communities. J Veg Sci 3:187–200
Bergman CM, Fryxell JM, Gates CG (2000) The effect of tissue complexity and sward height on
the functional response of wood bison. Func Ecol 14:61–69
Bergström R, Danell K (1987) Moose winter feeding in relation to morphology and chemistry of
six tree species. Alces 22:91–112
Bergström R, Danell K (1995) Effects of simulated summer browsing by moose on leaf and shoot
biomass of birch, Betula pendula. Oikos 72:132–138
Bergström R, Skarpe C, Danell K (2000) Plant responses and herbivory following simulated
browsing and stem cutting of Combretum apiculatum. J Veg Sci 11:409–414
Berryman AA, Stanseth NC (1984) Behavioural catastrophes in biological systems. Behav Sci
29:127–137
Bestelmeyer BT, Trujillo DA, Tugel AJ, Havstad KM (2006) A multi-scale classification of veg-
etation dynamics in arid lands: what is the right scale for models, monitoring and restoration?
J Arid Env 65:296–318
Bilbrough CJ, Richards JH (1993) growth of sagebrush and bitterbrush following simulated
winter browsing: mechanisms of tolerance. Ecology 74:481–492
Bircham JS, Hodgson J (1983) The influence of sward condition on rates of herbage growth and
senescence under continuous stocking management. Grass Forage Sci 38:323–331
Birks HJB (2005) Mind the gap: How open were European primeval forests. Trends Ecol Evol
20:154–156
Bloom AJ, Chapin FS III, Mooney HA (1985) Resource limitation in plants – an economic analogy.
Ann Rev Ecol Syst 16:363–392
Bodmer R, Ward D (2006) Frugivory in large mammalian herbivores. In: Danell K, Bergström R,
Duncan P, Pastor J (eds) Large herbivore ecology and ecosystem dynamics. Cambridge Univ
Bond WJ, Midgley JJ (2001) Ecology of sprouting in woody plants: the persistence niche. Trends
Ecol Evol 16:45–51
Boyce MS, Mao JS, Merrill EH, Fortin D, Turner MG, Fryxell J, Turchin P (2003) Scale and het-
erogeneity in habitat selection by elk in Yellowstone National Park. Ecoscience 10:421–431
Bradshaw AD (1965) Evolutionary significance of phenotypic plasticity in plants. Adv Genet
13:115–155
Bradshaw RHW, Mitchell FJG (1999) The palaeoecological approach to reconstructing former
grazing–vegetation interactions. Forest Ecol Manag 120:3–12
Briske DD (1996) Strategies of plant survival in grazed systems: a functional interpretation.
In: Hodgson J, Illius AW (eds) The ecology and management of grazing ecosystems. CAB
International, Wallingford, UK, pp 37–68
Briske DD, Fuhlendorf SD, Smeins FE (2003) Vegetation dynamics on rangelands: a critique of
the current paradigms. J Appl Ecol 40:601–614
Brown JS, Laundré JW, Gurung M (1999) The ecology of fear: optimal foraging, game theory,
and trophic interactions J Mammal 80:385–399
Brown JS, Kotler BP, Bouskila A (2001) Ecology of fear: Foraging games between predators and
prey with pulsed resources. Ann Zool Fenn 38:71–87
Bryant JP, Chapin FS, Klein DR (1983) Carbon/nutrient balance of boreal plants in relation to
vertebrate herbivory. Oikos 40:357–368
Bryant JP, Kuropat PJ, Cooper SM, Frisby K, Owen-Smith N (1989) Resource availability
hypothesis of plant antiherbivore defence tested in a South African savanna ecosystem. Nature
340:227–229
Bullock JM, Franklin J, Stevenson MJ, Silvertown J, Coulson SJ, Gregory SJ, Tofts R (2001)
A plant trait analysis of responses to grazing in a long-term experiment. J Appl Ecol
38:253–267
Chapin FS III, Schulze ED, Mooney HA (1990) The ecology and economics of storage in plants.
Annu Rev Ecol Syst 21:423–447
Caughley G (1976). The elephant problem - an alternative hypothesis. E Afr Wildl J
14:265–283
Cipollini D, Purringnton CB, Bergelson J (2003) Costs of induced responses in plants. Basic Appl
Ecol 4:79–89
Clements FE (1916) Plant succession. Carnegie Institute, Washington, DC
Coleman GD, Chen THH, Ernst SG, Fuchigami H (1991) Photoperiod control of poplar bark stor-
age protein accumulation. Plant Physiol 96:686–692
Coley PD, Bryant JP, Chapin FS (1985) Resource availability and plant antiherbivore defence.
Science 230:895–899
Cooper SM, Owen-Smith N (1985) Condensed tannins deter feeding by browsing ungulates in a
South African savanna. Oecologia 67:142–146
Cooper SM, Owen-Smith N, Bryant JP (1988) Foliage acceptability to browsing ruminants in
relation to seasonal changes in leaf chemistry of woody plants in a South African savanna.
Coppock DL, Detling JK, Ellis JE, Dyer MI (1983) Plant–herbivore interactions in a North
American mixed-grass prairie. I. Effects of black-tailed prairie dogs on intraseasonal above-
ground plant biomass and nutrient dynamics and plant species diversity. Oecologia 56:1–9
Côté SD, Rooney TP, Tremblay JP, Dussault C, Waller DM (2004) Ecological impacts of deer
overabundance Annu Rev Ecol Evol 35:113–147
Crawford RMM, Balfour J (1990) Female-biased sex ratios and differential growth in arctic
willows. Flora 184:291–302
Crawley MJ (1983) Herbivory - the dynamics of animal–plant interactions. Studies in ecology,
vol 10. Blackwell, Oxford
Crawley MJ (ed) (1997) Plant ecology. Blackwell, Oxford

Danckwerts JE, Gordon AJ (1987) Long-term partitioning, storage and remobilisation of 14C
assimilation by Lolium perenne (cv. Melle). Ann Bot 59:55–66
Danell K, Bergström R (1987) Effects of simulated winter browsing by moose on morphology and
biomass of two birch species. J Ecol 75:533–544
Danell K, Bergström R (1989) Winter browsing by moose on two birch species: impact on food
resources. Oikos 54:11–18
Danell K, Huss-Danell K, Bergström R (1985) Interactions between browsing moose and two
species of birch in Sweden. Ecology 66:1867–1878
Danell K, Hjältén J, Ericson L, Elmqvist T (1991a) Vole feeding on male and female willow
shoots along a gradient of plant productivity. Oikos 62:145–152
Danell K, Niemela P, Varvikko T, Vuorisalo T (1991b) Moose browsing on Scots pine along a
gradient of plant productivity. Ecology 72:1624–1633
Danell K, Bergström R, Edenius L (1994) Effects of large mammalian browsers on architecture,
biomass, and nutrients of woody plants. J Mammal 75:833–844
Danell K, Haukioja E, Huss-Danell K (1997) Morphological and chemical responses of mountain
birch leaves and shoots to winter browsing along a gradient of plant productivity. Ecoscience
4:296–303
Danell K, Bergström R, Edenius L, Ericsson G (2003) Ungulates as drivers of tree population
dynamics at module and genet levels. Forest Ecol Manag 181:67–76
Dangerfield JM, Modukanele M (1996) Overcompensation by Acacia erubescens in response to
simulated browsing. J Trop Ecol 12: 1–4
Demment MW, van Soest PJ (1985) A nutritional explanation for body-size patterns of ruminant
Deregibus VA, Sanchez RA, Casal JJ, Trlica MJ (1985) Tillering responses to enrichment of red
light beneath the canopy in a humid natural grassland. J Appl Ecol 22:199–206
Devlin RM (1966) Plant physiology. Reinhold, New York
Diaz S, Acosta A, Cabido M (1992) Morphological analysis of herbaceous communities under
different grazing regimes. J Veg Sci 3:689–696
Dormann CF, Skarpe C (2002) Flowering, growth and defence in the two sexes: consequences of
herbivore exclusion for Salix polaris. Funct Ecol 16:649–656
Dublin HT, Sinclair ARE, McGlade J (1990). Elephants and fire as causes of multiple stable states
in the Serengeti-Mara woodlands. J Anim Ecol 59:1147–1164
Du Toit JT (2003) Large herbivores and savanna heterogeneity. In: du Toit JT, Rogers KH, Biggs
HC (eds) The Kruger experience: ecology and management of savanna heterogeneity. Island
Press, Washington, DC, pp 292–309
Du Toit JT, Bryant JP, Frisby K (1990) Regrowth and palatability of Acacia shoots following
pruning by African savanna browsers. Ecology 71:149–154
Du Toit E.W, Wolfson MM, Ellis RP (1991) The presence of condensed tannins in the leaves of
Eulalia villosa. J Grassland Soc S Afr 8:74–76
Duncan A J, Hartley SE, Iason GR (1998) The effect of previous browsing damage on the mor-
phology and chemical composition of Sitka spruce (Picea sitchensis ) saplings and on their
subsequent susceptibility to browsing red deer (Cervus elaphus). Forest Ecol Manag
103:57–67
Eckstein RL, Karlsson PS, Weih M (1998) The significance of resorption of leaf resources for
shoot growth in evergreen and deciduous woody plants from a subarctic environment. Oikos
81:567–575
Edwards PJ (1989) Insect herbivory and plant defence theory. In: Grubb PJ, Whittaker JB (eds)
Toward a more exact ecology. Blackwell, Oxford, pp 275–297
Ellis JE, Swift DM (1988) Stability of African pastoral ecosystems: alternate paradigms and
implications for development. J Range Manage 41:450–459
Ellis RP, Vogel JC, Fuls A (1980) Photosynthetic pathways and the geographical distribution of
grasses in South West Africa/Namibia. S Afr J Sci 76:307–314
Ellison L (1960) Influence of grazing on plant succession of rangelands. Bot Rev 26:1–78
Engelmark O, Hofgaard A, Arnborg T (1998) Successional trends 219 years after fire in an old
Pinus sylvestris stand in northern Sweden. J Veg Sci 9:583–592
Falkengren-Grerup U, Mansson KF, Olsson MO (2000) Uptake capacity of amino acids by ten
grasses and forbs in relation to soil acidity and nitrogen availability. Environ Exp Bot
44:207–219
Feeny P (1976) Plant apparency and chemical defence. Rec Adv Phytochem 10:1–40
Filet PG (1994) State and transition models for rangelands. 3. The impact of the state and transi-
tion model on grazing lands research, management and extension: a review. Trop Grasslands
28:214–222
Fitter AH (1997) Acquisition and utilization of resources. In: Crawley MJ (ed) Plant ecology, 2nd
edn. Blackwell, Oxford
Flower-Ellis JGK (1971) Age structure and dynamics in stands of bilberry (Vaccinium myrtillus L.).
Royal Coll Forestry, Stockholm
Foley WJ, Iason GR, MacArthur C (1999) Role of plant secondary metabolites in the nutritional
ecology of mammalian herbivores. How far have we come in 25 years? In: Young HJG, Fahey
GC (eds) Nutritional ecology of herbivores, Am Soc Anim Sci, Savoy, IL, pp 130–209
Foran BD, Bastin G, Shaw KA (1986) Range assessment and monitoring in arid lands: the deriva-
tion of functional groups to simplify vegetation data. J Environ Manage 27:85–97
Fortin D, Beyer HL (2005) Wolves influence elk movements: behavior shapes a trophic cascade
in Yellowstone National Park. Ecology 86:1320–1330
Frelich LE, Lorimer CG (1985) Current and predicted long-term effects of deer browsing in
hemlock forests in Michigan, USA. Biol Conserv 34:99–120
Friedel MH (1991) Range condition assessment and the concept of thresholds: a viewpoint.
Fryxell JM, Wilmshurst JF, Sinclair ARE (2004) Predictive models of movement by Serengeti
grazers. Ecology 85:2429–2435
Fryxell JM, Wilmshurst JF, Sinclair ARE, Haydon DT, Holt RD, Abrams PA (2005). Landscape
scale, heterogeneity, and the viability of Serengeti grazers. Ecol Lett 8:328–335
Fuller RJ, Gill RMA (2001) Ecological impacts of increasing numbers of deer in British wood-
land. Forestry 74:193–199
Gauch HG (1982) Multivariate analysis in community ecology. Cambridge Univ Press, Cambridge
Georgiadis NJ, McNaughton SJ (1988) Interactions between grazers and a cyanogenic grass,
Cynodon plectostachyus. Oikos 51:343–350
Gifford RM, Marshall C (1973) Photosynthesis and assimilate distribution in Lolium multiflorum
following differential tiller defoliation. Aust J Biol Sci 26:517–526
Gill RMA (1992) A review of damage by mammals on north temperate forests: 1 Deer. Forestry
65:145–169
Gill RMA (2006) The influence of large herbivores on tree recruitment and forest dynamics.
In: Danell K, Bergstrom R, Duncan P, Pastor J (eds) Large herbivore ecology, ecosystem
dynamics and conservation. Cambridge Univ Press, Cambridge, pp 170–202
Gordon IJ (2003) Browsing and grazing ruminants: are they different beasts? Forest Ecol Manag
181:13–21
Gowda JH (1996) Spines of Acacia tortilis: what do they defend and how? Oikos 77:279–284
Grime JP (1979) Plant strategies and vegetation processes. Wiley, Chichester
Grime JP, Hodgson JG, Hunt R (1988) Comparative plant ecology: a functional approach to
common British species. Unwin Hyman, London.
Grime JP, Thompson K, Hunt R et al (1997) Integrated screening validates primary axes of
specialisation in plants Oikos 79:259–281
Gross JE, Shipley LA, Hobbs NT, Spalinger DE, Wunder BA (1993) Functional response of
herbivores in food-concentrated patches; tests of a mechanistic model. Ecology 74:778–791
Grubb P (1992) A positive distrust in simplicity. J Ecol 80:585–610
Guillet C, Bergström R (2006) Compensatory growth of fast-growing willow (Salix) coppice in
response to simulated large herbivore browsing. Oikos
Harper JL (1977) Population biology of plants. Academic Press, London
Harrison AK, Bardgett RD (2007) Impacts of grazing and browsing by large mammals on soils
and soil biological properties. (this volume)
Haukioja E, Koricheva J (2000) Tolerance to herbivory in woody vs. herbaceous plants. Evol Ecol
14:551–562
Haukioja E, Ruohomäki K, Senn J, Suomela J, Walls M (1990) Consequences of herbivory on the
mountain birch (Betulas pubescens ssp tortuosa): importance of the functional organisation of
the tree. Oecologia 82:238–247
Heilmeier H, Schulze E-D, Whale DM (1986) Carbon and nitrogen partitioning in the biennial
monocarp Arctium tomentosum Mill. Oecologia 70:466–467
Herms DA, Mattson WJ (1992) The dilemma of plants – to grow or defend. Quart Rev Biol
67:283–335
Hester AJ, Mitchell FJG, Kirby KJ (1996). Effects of season and intensity of sheep grazing on tree
regeneration in a British upland woodland. Forest Ecol Manag 88:99–106
Hester AJ, Edenius L, Buttenshøn RM, Kuiters AT (2000). Interactions between forests and
herbivores: the role of controlled grazing experiments. Forestry 73:381–391
Hester AJ, Millard P, Baillie GJ, Wendler R (2004). How does timing of browsing affect above-
and below-ground growth of Betula pendula, Pinus sylvestris and Sorbus aucuparia? Oikos
105:536–550
Hester AJ, Lempa K, Neuvonen S, Høegh K, Feilberg J, Arnpórsdóttir S, Iason GR (2005) Birch
sapling responses to severity and timing of domestic herbivore browsing – implications for
management. In: Wielgolaski FE (ed) Plant ecology, herbivory and human impact in Nordic
mountain birch forests. Ecological studies, vol 180. Springer, Berlin Heidelberg New York, pp
139–155
Hester AJ, Scogings PF, Trollope WSW (2006a) Long-term impacts of goat browsing on bush-
clump dynamics in a semi-arid subtropical savanna. Plant Ecol 183:277–290
Hester AJ, Bergman M, Iason GR, Moen R (2006b). Impacts of large herbivores on plant com-
munity structure and dynamics. In: Danell K, Bergström R, Duncan P, Pastor J (eds) Large
herbivore ecology and ecosystem dynamics. Cambridge Univ Press, Cambridge, pp 97–141
Hik DS, Jefferies RL (1990) Increases in the net above-ground primary production of salt-marsh
forage grass: a test of the predictions of the herbivore-optimization model. J Ecol
78:180–195
Hill MJ, Roxburgh SH, Carter JO, McKeon GM (2005) Vegetation state change and consequent
carbon dynamics in savanna woodlands of Australia in response to grazing, drought and fire:
a scenario approach using 113 years of synthetic annual fire and grassland growth. Aust J Bot
53:715–739
Hjältén J (1992) Plant sex and hare feeding preferences. Oecologia 89:253–256
Hjältén J, Danell K, Lundberg P (1993a) Herbivore avoidance by association – vole and hare
utilization of woody plants Oikos 68:125–131
Hjältén J, Danell K, Ericson L (1993b) Effects of simulated herbivory and intraspecific competi-
tion on the compensatory ability of birches. Ecology 74:1136–1142
Hobbs NT (1996) Modification of ecosystems by ungulates. J Wildl Manage 60:695–713
Hobbs NT (2003) Challenges and opportunities in integrating ecological knowledge across scales.
Forest Ecol Manage 181:223–238
Hobbs RJ, Gimingham CH (1987) Vegetation, fire and herbivore interactions in heathland. Adv
Ecol Res 16:88–173
Hodgkinson KC (1974) Influence of partial defoliation on photosynthesis, photorespiration and
transpiration by Lucerne leaves at different ages. Aust J Plant Physiol 1:561–578
Hodgson J (ed) (1981) Sward measurement handbook. British Grassland Society, Cirencester, UK
Holopainen J, Rikala R, Kainulainen P Oksanen J (1995) Resource partitioning to growth, storage
and defence in nitrogen-fertilized Scots pine and susceptibility of the seedlings to the tarnished
plant bug Lygus rugulipennis. New Phytol 131:521–532
Honkanen T, Haukioja E, Suomela J (1994) Effects of simulated defoliation and debudding on
needle and shoot growth in Scots pine (Pinus sylvestris): implications of plant source/sink
relationships for plant–herbivore studies. Funct Ecol 8:631–639
Honkanen T, Haukioja E, Kitunen V (1999) Responses of Pinus sylvestris branches to simulated

herbivory are modified by tree sink/source dynamics and by external resources. Funct Ecol
13: 126–140
Hooley JL, Cohn EVJ (2003) Models of filed layer interactions in an experimental secondary
woodland. Ecol Model 169:89–102
Hulme PE (1996) Natural regeneration of yew (Taxus baccata L.): microsite, seed or herbivore
limitation? J Ecol 84:853–861
Hytteborn H (1975) Deciduous woodland at Andersby, eastern Sweden. Above-ground tree and
shrub production. Acta Phytogeograph Suecica 61:96
Iason GR (2005) The role of plant secondary metabolites in mammalian herbivory: ecological
perspectives. Proc Nutr Soc 64:123–131
Iason GR, Villalba JJ (2006) Behavioral strategies of mammal herbivores against plant secondary
metabolites: the avoidance–tolerance continuum. J Chem Ecol 32:1115–1132
Illius AW, Fitzgibbon C (1994) Costs of vigilance in foraging ungulates Anim Behav 47:481–484
Illius AW, Gordon IA (1987) The allometry of food intake in grazing ruminants. J Anim Ecol
56:989–999
Illius A, O’Connor TG (2004) The definition of non-equilibrium and the role of key resource – an
ecological perspective. In: Vetter S (ed) Rangelands at equilibrium and non-equilibrium. Univ
of Western Cape, Cape Town, p 16
Inghe O, Tamm CO (1985) Survival and flowering of perennial herbs IV. Oikos 45:400–420
Jachmann H, Bell RHV (1985) Utilization by elephants of the Brachystegia woodlands of the
Kasungu National Park, Malawi. J Afr Ecol 23:245–258
Jaramillo VJ, Detling JK (1988) Grazing history, defoliation and competition: effects on short
grass production and nitrogen accumulation. Ecology 69:1599–1608
Järemo J, Nilsson P, Tuomi J (1996) Plant compensatory growth: herbivory or competition? Oikos
77:238–247
Jarman PJ, Sinclair ARE (1979) Feeding strategy and the pattern of resource partitioning in ungu-
lates. In: Sinclair ARE, Norton-Griffiths M (eds) Serengeti – dynamics of an ecosystem. Univ
of Chicago Press, Chicago, pp 130–163
Jefferies RL, Klein DR, Shaver GR (1994) Vertebrate herbivores and northern plant communities
– reciprocal influences and response. Oikos 71:193–206
Jia J, Niemelä P, Danell K (1995) Moose Alces alces bite diameter selection in relation to twig
quality on four phenotypes of Scots pine Pinus sylvestris. Wildl Biol 1:47–55
Jokela J, Schmid-Hempel P, Rigby MC (2000) Dr. Pangloss restrained by the Red Queen – steps
towards a unified defence theory. Oikos 89:267–274
Juenger T, Lennartsson T (2000) Tolerance in plant ecology and evolution: toward a more unified
theory of plant–herbivore interaction Evol Ecol 14:283–287
Juntheikki MR (1996) Comparison of tannin-binding proteins in saliva of Scandinavian and North
American moose (Alces alces). Biochem Syst Ecol 24:595–601
Karban R, Agrawal AA, Thaler JS, Adler LS (1999) Induced plant responses and information
content about risk for herbivory. Trends Ecol Evol 14:443–447
Kay CE (1997) Is aspen doomed? J Forest 95:434
Kelly JK (1996) Kin selection in the annual plant Impatiens capensis. Am Nat 147:899–918
Kielland K, Bryant JP (1998) Moose herbivory in taiga: effects on biogeochemistry and vegeta-
tion dynamics in primary succession. Oikos 82:377–383
Knoop WT, Walker BH (1985) Interactions of woody and herbaceous vegetation in a southern
African savanna. J Ecol 73:235–253
Koricheva J (2002a) The carbon–nutrient balance hypothesis is dead; long live the carbon–nutrient
balance hypothesis? Oikos 98:537–539
Koricheva J (2002b) Meta-analysis of sources of variation in fitness costs of plant antiherbivore
defences. Ecology 83:176–190
Körner C (1994) Biomass fractionation in plants: a reconsideration of definitions based on plant
functions. In: Roy J, Garnier E (eds) A whole plant perspective on carbon–nitrogen interactions.
SPB Academic Publishing, The Hague, pp 175–185
Kotanen PM, Rosenthal JP (2000) Tolerating herbivory: does the plant care if the herbivore has a
backbone? Evol Ecol 14:537–549
Kozlowski TT (1966) Shoot growth in woody plants. Botan Rev 30:335–392
Kruger EL, Reich PB (1997) Responses of hardwood regeneration to fire in mesic forest openings.
III. Whole plant growth, biomass distribution, and nitrogen and carbohydrate relations. Can J
Forest Res 27:1841–1850
Kurz WA, Beukema SJ, Klenner W, Greenough JA, Robinson DCE, Sharpe AD, Webb TM
(2000) TELSA: the tool for exploratory landscape scenario analyses. Comput Electron Agr
27:227–242
Langer RHM (1972) How grasses grow. Studies in biology 34. Edward Arnold, London
Langstrom B, Tenow O, Ericsson A, Hellgvist C, Larsson S (1990) Effects of shoot pruning on
stem growth, needle biomass, and dynamics of carbohydrates and nitrogen in Scots pine as
related to season and tree age. Can J Forest Res 20:514–523
Laws RWJ (1970) Elephants as agents of habitat and landscape change in East Africa. Oikos
21:1–15
Laycock WA (1991) Stable states and thresholds of range condition in North America rangeland:
a viewpoint. J Range Manage 44:427–435
Lemaire G, Chapman D (1996) Tissue flows in grazed plant communities. In: Hodgson J, Illius
AW (eds) The ecology and management of grazing systems. CAB International, Wallingford,
UK, pp 3–36
Letnic M, Dickman CR, Tischler MK, Tamayo B, Beh C-L (2004) The responses of small mammals
and lizards to post-fire succession and rainfall in arid Australia. J Arid Environ 59:85–114
Letourneau DK, Dyer LA et al. (2004) Indirect effects of a top predator on a rain forest understory
plant community. Ecology 85:2144–2152
Lima SL (1998) Non-lethal effects of the ecology of predator–prey interactions. Bioscience
48:25–34
Lock JM (1972) The effects of hippopotamus grazing on grasslands. J Ecol 60:445–467
Lockwood JA, Lockwood DR (1993) Catastrophe theory: a unified paradigm for rangeland eco-
system dynamics. J Range Manage 46:282–288
Loehle C (1989) Catastrophe theory in ecology: a critical review and an example of the butterfly
catastrophe. Ecol Model 49:125–152
Ludlow MM, Wilson GL (1971) Photosynthesis of tropical pasture plants III. Leaf age. Aust J
Biol Sci 24:1077–1087
Löyttyniemi K (1985) On repeated browsing of Scots pine saplings by moose (Alces alces). Silva
Fenn 19:387–391
MacFadden BJ (1997) Origin and evolution of the grazing guild in New World terrestrial mam-
mals. Trends Ecol Evol 12:182–187
Mack RN, Thompson JN (1982) Evolution in steppe with few large hooved mammals. Am Nat
119:757–773
Makhabu SW (2005) Resource partitioning within a browsing guild in a key habitat, the Chobe
Riverfront, Botswana. J Trop Ecol 21:641–649
Makhabu SW, Skarpe C (2006) Rebrowsing by elephants three years after simulated browsing on
five woody plant species in northern Botswana. S Afr J Wildl Res 36:99–102
Makhabu SW, Skarpe C, Hytteborn H (2006a) Elephant impact on shoot distribution on trees and
on rebrowsing by smaller browsers. Acta Oecol 30:136–146
Makhabu SW, Skarpe C, Hytteborn H, Mpufu ZD (2006b) The plant vigour hypothesis revisited
– how is browsing by ungulates and elephant related to woody species growth rate? Plant Ecol
184:163–172
Marshall C, Sagar GR (1968) The distribution of assimilates in Lolium multiflorum Lam following
differential defoliation. Ann Bot 32:715–719
Maschinski J, Whitham TG (1989) The continuum of plant responses to herbivory: the influence
of plant association, nutrient availability and timing. Am Nat 134:1–19
Mauricio R (2000) Natural selection and the joint evolution of tolerance and resistance as plant
defences. Evol Ecol 14:491–507
McIntosh BS, Muetzelfeldt RI, Legg CJ, Mazzoleni S, Csontos P (2003) Reasoning with direction
and rate of change in vegetation state transition modelling. Environ Modell Softw
18:915–927
McIntyre S, Tongway D (2005) Grassland structure in native pastures: links to soil surface condi-
tion. Ecol Manage Restor 6:43–50
McNaughton SJ (1978) Serengeti ungulates: feeding selectivity influences the effectiveness of
plant defense guilds. Science 199:806–807
McNaughton SJ (1983) Compensatory growth as a response to herbivory. Oikos 40:329–336
McNaughton SJ (1984) Grazing lawns, animals in herds, plant form, and coevolution. Am Nat
124:863–886
McNaughton SJ (1985) Ecology of a grazing ecosystem: the Serengeti. Ecol Monogr
55:259–294
Milchunas DG, Lauenroth WK (1993) Quantitative effects of grazing on vegetation and soils over
a global ranger of environments. Ecol Monogr 63:327–366
Milchunas DG, Noy-Meir I. (2002). Grazing refuges, external avoidance of herbivory and plant
diversity. Oikos 99:113–130
Milchunas DG, Sala OE, Lauenroth WK (1988) A generalised model of the effects of grazing by
Milewski AV, Young TP, Madden D (1991) Thorns as induced defences: experimental evidence.
Millard P (1996) Ecophysiology of the internal cycling of nitrogen for tree growth. J Plant Nutr
Soil Sci 159:1–10
Millard P, Proe MF (1991) Leaf demography and the seasonal internal cycling of nitrogen in syca-
more (Acer pseudoplatanus L.) seedlings in relation to nitrogen supply. New Phytol
117:587–596
Millard P, Hester AJ, Wendler R, Baillie G (2001) Remobilization of nitrogen and the recovery
of Betula pendula, Pinus sylvestris, and Sorbus aucuparia saplings after simulated browsing
damage. Funct Ecol 15:535–543
Milton SJ, Dean WRJ, du Plessis MA, Siegfried WRA (1994) A conceptual model of arid range-
land degradation. The escalating cost of declining productivity. Bioscience 44:70–76
Miquelle DG (1983) Browse regrowth and consumption following summer defoliation by moose
J Wildl Manage 47:17–24
Mitchell FJG (2005) How open were European primaeval forests? Hypothesis testing using pal-
aeoecological data. J Ecol 93:168–177
Mitchley J (1988) Control of relative abundance of perennials in chalk grassland in southern
England. II. Vertical canopy structure. J Ecol 76:341–350
Moe SR, Wegge P, Kapela EB (1990) The influence of man-made fires on large wild herbivores
in Lake Burungi area in northern Tanzania. Afr J Ecol 28:35–43
Mooney HA (1997) Photosynthesis. In: Crawley MJ (ed) Plant ecology, Blackwell, Oxford,
pp 345–374
Moore NP, Hart JD, Kelly PF, Langton SD (2000) Browsing by fallow deer (Dama dama) in
young broadleaved plantations: seasonality, and the effects of previous browsing and bud
eruption. Forestry 73:437–445
Murray MG (1995) Specific nutrient requirements and migration of wildebeest In: Sinclair ARE,
Arcese P (eds) Serengeti II - Dynamics, management and conservation of an ecosystem. Univ
of Chicago Press, Chicago, pp 231–256
Myers JH, Bazeley D (1991) Thorns, spines, prickles and hairs: are they stimulated by herbivory
and do they deter herbivores? In: Tallamyr DJ, Raup MJ (eds) Phytochemical induction by
herbivores. Academic Press, New York, pp 326–343
Mysterud A, Langvatn R, Yoccoz NG, Stenseth NC (2001). Plant phenology, migration and
geographical variation in body weight of a large herbivore: the effect of a variable topography.
Naiman RJ, Braack L, Grant R, Kemp AC, du Toit JT, Venter FJ (2003) Interactions between
species and ecosystem characteristics. In: du Toit JT, Rogers KH, Biggs HC (eds) The Kruger
experience – ecology and management of savanna heterogeneity. Island Press, Washington,

DC, pp 221–241
Nambiar EKS (1987) Do nutrients retranslocate from fine roots? Can J For Res 17:913–918
Nambiar EKS, Fife DN (1991) Nutrient retranslocation in temperate conifers. Tree Physiol
9:185–207
Noble IR, Slatyer RO (1980) The use of vital attributes to predict successional changes in plant
communities subject to recurrent disturbances. Vegetatio 43:5–21
Noy-Meir I (1993) Compensating growth of grazed plants and its relevance to the use of rangelands.
Ecol Appl 3:32–34
Noy-Meir I, Seligman NG (1979) Management of semi-arid ecosystems in Israel. In: Walker BH
(ed) Management of semi-arid ecosystems. Elsevier, Amsterdam, pp 398–424
O’Connor TG (1993) The influence of rainfall and grazing on the demography of some African
savanna grasses: a matrix modelling approach. J Appl Ecol 30:119–132
Olff H, Vera FWM, Bokdam J (1999) Shifting mosaics in grazed woodland driven by the alternation
of plant facilitation and competition. Plant Biol 1:127–137
Olff H, Ritchie ME, Prins HH. (2002). Global environmental controls of diversity in large herbivores.
Nature 415:901–904
Olofsson J, Kitti H, Rautiainen P, Stark S, Oksanen L (2001) Effects of summer grazing by reindeer
on composition of vegetation, productivity and nitrogen cycling. Ecography 24:13–24
Olofsson J, Moen J, Oksanen L (2002) Effects of herbivory on competition intensity in two
arctic–alpine tundra communities with different productivity. Oikos 96:265–272
Paige KN, Whitham RG (1987) Overcompensation in response to mammalian herbivory: the
advantage of being eaten. Am Nat 129:407–416
Painter EL, Detling JK (1981) Effects of defoliation on net photosynthesis and regrowth of western
wheatgrass. J Range Manage 34:68–71
Painter EL, Detling JK. Steingraeber DA (1989) Grazing history, defoliation, and frequency-
dependent competition: effects on two North American grasses. Am J Bot 76:1368–1379
Palmer SCF, Hester AJ, Elston DA, Gordon IJ, Hartley SE (2003) The perils of having tasty
neighbors: grazing impacts of large herbivores at vegetation boundaries. Ecology
84:2877–2890
Palo RT (1987) Chemical defence in woody plants and the role of digestive system of herbivores.
In: Provenza FD, Flinders JF, McArthur ED (eds) Proceedings from a symposium on plant–
herbivore interactions. US Dept Agr Forest Serv, Gen Tech Report INT-222
Palo RT, Robbins CT (eds 1991) Plant defences against mammalian herbivory. CRC Press, Boca
Raton, FL
Pass GJ, McLean S, Stupans I, Davies NW (2002) Microsomal metabolism and enzyme kinetics
of the terpene p-cymene in the common brushtail possum (Trichosurus vulpecula), koala
(Phascolarctos cinereus) and rat. Xenobiotica 32:383–397
Pastor J, Naiman RJ (1992) Selective foraging and ecosystem processes in boreal forests. Am Nat
139:690–705
Pastor J, Dewey B, Naiman RJ, McInnes PF, Cohen Y (1993) Moose browsing and soil fertility
in the boreal forests of Isle Royale National Park. Ecology 74:467–480
Pastor J, Cohen J, Hobbs NT (2006a) The role of large herbivores in ecosystem nutrient cycles.
In: Danell K, Bergström R, Duncan P, Pastor J (eds) Large herbivore ecology and ecosystem
dynamics. Cambridge University Press, Cambridge, pp 289–325
Pastor J, Danell K, Bergström R, Duncan P (2006b). Themes and future directions in herbivore–
ecosystem interactions and conservation. In: Danell K, Bergström R, Duncan P, Pastor J (eds)
Large herbivore ecology and ecosystem dynamics. Cambridge University Press, Cambridge,
pp 468–478
Pérez-Barbería FJ, Elston DA, Gordon IJ, Illius AW (2004) The evolution of phylogenetic differ-
ences in the efficiency of digestion in ruminants. Pro Royal Soc Lond B Biol Sci
271:1081–1090
Persson I-L, Pastor J, Danell K, Bergström R (2005) Impact of moose population density on the
production and composition of litter in boreal forests. Oikos 108:297–306
Peterson CJ Jones RH (1997) Clonality in woody plants: a review and comparison with clonal
herbs. In: De Kroon H, Groenendael J (eds) The ecology and evolution of clonal growth in
plants. Backhuys, Leiden, pp 263–289
Piggott CD (1993) Are the distribution of species determined by failure to set seed? In: Marshall C,
Grace J (eds) Fruit and seed production. Cambridge University Press, Cambridge, pp 203–216
Porter H, Nagel O (2000) The role of biomass allocation in the growth response of plants to
different levels of light, CO2, nutrients and water: a quantitative review. Aust J Plant Physiol
27:595–607
Price PW (1991) The plant vigour hypothesis and herbivore attack. Oikos 62:244–251
Prins HHT, van der Jeugd HP (1993) Herbivore population crashes and woodland structure in
East Africa. J Ecol 81:305–314
Pugnaire FI, Chapin FS III (1993) Controls over nutrient resorption from leaves of evergreen
Mediterranean species. Ecology 74:124–129
Raunkiær C (1937) Plant life forms. Clarendon, Oxford
Raspe O, Finlay C, Jacquemart A-L, (2000) Biological flora of the British Isles, no. 214. Sorbus
aucuparia L. J Ecol 88: 910–930
Renaud PC, Verheyden-Tixier H, Dumont B (2003) Damage to saplings by red deer (Cerevus
elaphus): effect of foliage height and structure. Forest Ecol Manage 181:31–37
Richardson FD, Hahn BD, Hoffman MT (2005) On the dynamics of grazing systems in the
semi-arid succulent Karoo: the relevance of equilibrium and non-equilibrium concepts to
the sustainability of semi-arid pastoral systems. Ecol Model 187:491–512
Rietkerk M, Ketner P, Stroosnijder L, Herbert HT (1996) Sahelian rangeland development:
a catastrophe? J Range Manage 49:512–519
Ripple WJ, Beschta RL (2003) Wolf reintroduction, predation risk, and cottonwood recovery in
Yellowstone National Park. Forest Ecol Manage 184:299–313
Ripple WJ, Beschta RL (2004) Wolves and the ecology of fear: can predation risk structure
ecosystems? BioScience 54:755–766
Ripple WJ, Beschta RL (2005) Willow thickets protect young aspen from elk browsing after wolf
reintroduction. West N Am Nat 65:118–122
Ritchie ME, Tilman D, Knops JMH (1998) Herbivore effects on plant and nutrient dynamics in
oak savannah. Ecology 79:165–177
Robbins CT, Mole S, Hagerman AE, Hanley TA (1987) Role of tannins in defending plants
against ruminants; reduction in dry matter digestion? Ecology 68:1606–1615
Rooke T, Bergström R, Skarpe C, Danell K (2004) Morphological responses of woody species to
simulated twig-browsing in Botswana. J Trop Ecol 20:281–289
Rooney TP, McCormick RJ, Solheim SL, Waller DM (2000) Regional variation in recruitment
of hemlock seedlings and saplings in the upper Great Lakes, USA. Ecol Applic
10:1119–1132
Rosenthal GA, Janzen DH (1979) Herbivores: their interaction with plant secondary metabolites.
Academic press, New York
Rosenthal JP, Kotanen PM (1994) Terrestrial plant tolerance to herbivory. Trends Ecol Evol
9:145–148
Roy BA, Kirchener JW (2000) Evolutionary dynamics of pathogen resistance and tolerance.
Evolution 54:51–63
Ruess RW, McNaughton SJ (1984) Urea as a promotive coupler of plant–herbivore interactions.
Rusch GM, Oesterheld M (1997) Relationship between productivity, and species and functional
group diversity in grazed and non-grazed Pampas grassland. Oikos 78:519–526
Rutina LP, du Toit JT, Moe SR, Hytteborn H (2004) Browsing ungulates limit tree seedling
survival in the Chobe riparian zone, northern Botswana. In: Rutina LP (ed) Impalas in an
elephant-impacted woodland: browser-driven dynamics of the Chobe riparian zone, northern
Botswana. Thesis, Agricultural Univ of Norway
Sachs T, Hassidim M (1996) Mutual support and selection between branches of damaged plants.
Vegetatio 127: 25–30
Saunders PT (1980) An introduction to catastrophe theory. Cambridge University Press, Cambridge

Schaefer JA, Messier F (1995) Habitat selection as a hierarchy: the spatial scales of winter foraging
by muskoxen. Ecography 18:333–344
Schmitz OJ (2003) Top predator control of plant biodiversity and productivity in an old-field
ecosystem. Ecol Lett 6:156–163
Senft RL, Coughenour MB, Baily DW, Rittenhouse LR, Sala OE, Swift DM (1987) Large herbivore
foraging and ecological hierarchies. Landscape ecology can enhance traditional foraging
theory. BioScience 37:789–799
Senn J, Haukioja E (1994) Reactions of the mountain birch to bud removal: effects of severity and
timing, and implications for herbivores. Funct Ecol 8:494–501
Senn J, Suter W (2003) Ungulate browsing on silver fir (Abies alba) in the Swiss Alps: beliefs in
search of supporting data. Forest Ecol Manage 181:151–164
Senock RS, Sisson WB, Donart GB (1991) Compensatory photosynthesis of Sporobolus flexuosus
(Thurb.) rydb following simulated herbivory in the northern Chihuahua desert. Bot Gaz
152:275–281
Shabel AB, Peart DR (1994) Effects of competition, herbivory and substrate disturbance on
growth and size structure in pin cherry (Prunus pensylvanica) seedlings. Oecologia
98:150–158
Shipley LA, Illius AW, Danell K, Hobbs NT, Spalinger DE (1999) Predicting bite size selection
of mammalian herbivores: a test of a general model of diet optimization. Oikos 84:55–68
Skarpe C (1990) Shrub layer dynamics under different herbivore densities in an arid savanna,
Botswana. J Appl Ecol 27:873–885
Skarpe C (1991a) Impact of grazing in savanna ecosystems. Ambio 20:351–356
Skarpe C (1991b) Spatial patterns and dynamics of woody vegetation in an arid savanna. J Veg
Sci 2:565–572
Skarpe C, van der Wal R (2002) Effects of simulated browsing and length of growing season on
leaf characteristics and flowering in a deciduous arctic shrub, Salix polaris. Arct Antarct
Alpine Res 34:282–286
Skarpe C, Bergström R, Bråten A-L, Danell K (2000) Browsing in a heterogeneous savanna.
Ecography 23:632–640
Skarpe C Aarrestad PA, Andreassen, HP et al (2004) The return of the giants: ecological effects
of an increasing elephant population. Ambio 33:276–282
Skarpe C, Jansson I, Seljeli L, Bergström R, Eivin Røskaft E (2007) Browsing by goats on three
spatial scales in a semi-arid savanna. J Arid Environ 68:480–491
Spalinger DE, Hobbs NT (1992) Mechanisms of foraging in mammalian herbivores: new models
of functional response. Am Nat 140:325–348
Stamp N (2003) Out of the quagmire of plant defence hypotheses. Quart Rev Biol 78:23–55
Steinlein T, Heilmeier H, Schulze E-D (1993) Nitrogen and carbohydrate storage in biennials
originating from habitats of different resource availability. Oecologia 93:374–382
Stoddart LA, Smith AD (1943) Range management. McGraw-Hill, New York
Stoddart LA, Smith AD (1955) Range management, 2nd edn. McGraw-Hill, New York
Stokke S (1999) Sex differences in feeding-patch choice in a megaherbivore: elephants in Chobe
National Park, Botswana. Can J Zool 77:1723–1732
Strauss SY, Agrawal AA (1999) The ecology and evolution of plant tolerance to herbivory. Trends
Strauss SY, Rudgers JA, Lau JA, Irwin RE (2002) Direct and ecological costs of resistance to her-
bivory. Trends Ecol Evol 17:278–281
Stringham TK, Krueger WC, Shaver PL (2003) State and transition modelling: an ecological process
approach. J Range Manage 56:106–113
Stromberg CA (2004) Using phytolith assemblages to reconstruct the origin and spread of grass-
dominated habitats in the great plains of North America during the late Eocene to early
Miocene. Palaeogeogr Palaeocl 207:239–275
Stuart-Hill GC, Tainton NM (1989) The competitive interaction between Acacia karroo and the
herbaceous layer and how this is influenced by defoliation. J Appl Ecol 26:285–298
Sullivan S (1996) Towards a non-equilibrium ecology perspectives from an arid land. J Biogeog
23:1–5
Sunnerheim K, Palo RT, Theander O, Knutsson PG (1988) Chemical defence in birch.
Platyphylloside: a phenol from Betula pendula inhibiting digestibility. J Chem Ecol
14:549–560
Thom R (1975) Structural stability and morphogenesis. An outline of a general theory of models.
Benjamin, Reading, MA
Thomas H, Stoddart JL (1980) Leaf senescence. Ann Rev Plant Physiol 31:83–111
Thompson ID, Curran WJ (1993) A reexamination of moose damage to balsam fir – white spruce
forests in Newfoundland: 27 years later. Can J Zool 23:1388–1395
Thompson ID, Curran WJ, Hancock JA, Butler CE (1992) Influence of moose browsing on succes-
sional forest growth on black spruce sites in Newfoundland. For Ecol Manage 47:29–37
Tiffin P, Rausher MD (1999) Genetic constraints and selection acting on tolerance to herbivory in
the common morning glory Ipomoea purpurea. Am Nat 154:700–716
Tilman D (1988) Plant strategies and the dynamics and structure of plant communities. Princeton
University Press, Princeton, NJ
Tolsma DJ, Ernst WHO, Verwei RA, Vooijs R (1987a) Seasonal variation of nutrient concentra-
tions in a semi-arid savanna ecosystem in Botswana J Ecol 75:755–770
Tolsma DJ, Ernst WHO, Verwei RA (1987b) Nutrients in soil and vegetation around two artificial
waterpoints in eastern Botswana J Appl Ecol 24:991–1000
Tremblay JP, Hester AJ, McLeod J, Huot J (2004) Choice and development of decision support
tools for the sustainable management of deer–forest systems. Forest Ecol Mgmt 191:1–6
Tremblay JP, Huot J, Potvin F (2007) Density-related effects of deer browsing on the regeneration
dynamics of boreal forests. J Appl Ecol 44:552–562
Tremblay JP, Huot J, Potvin F (2006) Divergent nonlinear responses of the boreal forest field layer
along an experimental gradient of deer densities. Oecologia 150:78–88
Turner GT (1971) Soil and grazing influences on a salt-desert shrub range in western Colorado.
Vanderklein DW, Reich PB (1999) The effect of defoliation intensity and history of photosynthesis,
growth and carbon reserves of two conifers with contrasting life spans and growth habits.
New Phytol 144:121–132
van de Koppel J, Prins HHT (1998) The importance of herbivore interactions for the dynamics of
African savanna woodlands: an hypothesis J Trop Ecol 14:565–576
van der Meijden E, Wijn M, Verkaar HJ (1988) Defence and regrowth, alternative plant strategies
in the struggle against herbivores. Oikos 51:355–363
van der Wal R, Bardgett RD, Harrison KA, Stien A. (2004) Vertebrate herbivores and ecosystem
control: cascading effects of faeces on tundra ecosystems. Ecography 27:245–252
van Langevelde F, van de Vijver CADM, Kumar L, van de Koppel J et al (2003) Effects of fire
and herbivory on the stability of savanna ecosystems. Ecology 84:337–350
van Vegten JA, (1983) Thornbush invasion in eastern Botswana. Vegetatio 56:3–7
Vera FWM, Bakker ES, Olff H (2006) Large herbivores: missing partners of western-European
light-demanding tree and shrub species? In: Danell K, Bergström R, Duncan P, Pastor J (eds)
Large herbivore ecology and ecosystem dynamics. Cambridge University Press, Cambridge,
pp 203–231
Verwijst T (1993) Influence of the pathogen Melampsora epitea on intraspecific competition in a
mixture of Salix viminalis clones. J Veg Sci 4:717–722
Vila B, Vourc’h G, Gillon D, Martin JL, Guibal F (2002) Is escaping deer browse just a matter of
time in Picea sitchensis? Trees 16:488–496
Vourc’h G, Martin JL, Duncan P, Escarré J, Clausen T (2001) Defensive adaptations of Thuja pli-
cata to ungulate browsing: a comparative study between mainland and island populations.
Vourc’h G, Vila B, Gillon D, Escarré J, Guibal F, Fritz H, Clausen T, Martin JL (2002)
Disentangling the causes of damage variation by deer browsing on young Thuja plicata. Oikos
98:271–283
Walker BH, Noy-Meir I (1982) Aspects of the stability and resilience of savanna ecosystems.
New York
Walker BH, Ludwig D, Holling CS, Peterman RM (1981) Stability of semi-arid savanna grazing
systems. J Ecol 69:473–498
Walker BH, Matthews DA, Dye PJ (1986) Management of grazing systems – existing versus an
event-orientated approach. S Afr J Sci 82:172
Wallace IL (1990) Comparative photosynthetic responses of big bluestem to clipping versus graz-
ing J Range Manage 43: 58–61
Waller DM (1997) The dynamics of growth and form. In: Crawley MJ (ed) Plant ecology, 2nd edn.
Blackwell, Oxford
Walter H (1939) Grassland, Savanne und Busch der arideren Teile Afrikas in ihrer ökologischen
Bedingtheit. Jahrb Wissenschaftl Bot 87:750–860
Wang RZ (2002) Photosynthetic pathway type of forage species along grazing gradient from the
Sognen grassland, northeastern China. Photosynthetica 40:57–61
Ward D, Young TP (2002) Effects of large mammalian herbivores and ant symbionts on con-
densed tannins of Acacia drepanolobium in Kenya. J Chem Ecol 28:913–929
Ward D (2004) The effects of grazing on plant biodiversity in arid ecosystems. In: Shachak M,
Picket STA, Gosz JR, Perevolotsky A (eds) Biodiversity in drylands: towards a unified frame-
work. Oxford Univ Press, Oxford, pp 233–249
Ward D (2006) Long term effects of herbivory on plant diversity and functional types in arid
ecosystems. In: Danell K, Bergström R, Duncan P, Pastor J (eds) Large herbivore ecology
and ecosystem dynamics. Cambridge Univ Press, Cambridge, pp 142–169
Ward D, Saltz D (1994). Foraging at different spatial scales: Dorcas gazelles foraging for lilies in
the Negev Desert. Ecology 75:48–58
Wareing P, Patrick J (1975) Source–sink relationship and partition of assimilates in the plant. In:
Cooper JP (ed) Photosynthesis and productivity in different environments. Cambridge Univ
Wareing P, Khalifa MM, Trehane KJ (1968) Rate-limiting processes in photosynthesis at saturat-
ing light intensities. Nature 220:453–457
Warren A (2005) The policy implications of Sahelian change. J Arid Environ 63:660–670
Weis AE, Hochberg ME (2000) The diverse effects of intraspecific competition on the selective
advantage to resistance: a model and its predictions. Am Nat 156:276–292
Weisberg PJ, Bugmann H. (2003) Forest dynamics and ungulate herbivory: from leaf to landscape.
Forest Ecol Manage 181:1–12
Welch D, Staines BW, Scott D, French DD, Catt DC (1991) Leader browsing by red and roe deer
on young Sitka spruce trees in western Scotland I. Damage rates and the influence of habitat
factors. Forestry 64:61–82
West NE, Provenza FD, Johnson PS, Owens MK (1984) Vegetation change after 13 years of
livestock grazing exclusion on sagebrush semidesert in west central Utah. J Range Manage
37:262–264
Westoby M, Walker B, Noy-Meir I (1989) Opportunistic management for rangelands not at
equilibrium. J Range Manage 42:266–274
Whitham TG, Maschinski J, Larson KC, Paige KN (1991) Plant responses to herbivory: the
continuum from negative to positive and underlying physiological mechanisms. In: Price P,
Lewinsohn T, Fernandes W, Benson W (eds) Plant–animal interactions: evolutionary ecology
in tropical and temperate regions. Wiley New York, pp 227–256
Wilmshurst JF, Fryxell JM, Farm BP, Sinclair ARE, Henschel CP (1999) Spatial distribution of
Serengeti wildebeest in relation to resources. Can J Zool 77:1223–1232
Wilsey B (2002) Clonal plants in a spatially heterogeneous environment: effects of integration on
Serengeti grassland response to defoliation and urine-hits from grazing animals. Plant Ecol
159:15–22
Wilson JR, Hacker JB (1987) Comparative digestibility and anatomy of some sympatric C3 and
C4 arid zone grasses. Aust J Agric Res 38:287–295
Wolfson MM, Tainton NM (1999) The morphology and physiology of the major forage plants.
In: Tainton N (ed) Veld management in South Africa, Univ of Natal Press, Pietermauritzburg,
pp 54–90
Woodward A, Coppock DL (1995) Role of plant defence in the utilisation of native browse in
southern Ethiopia. Agroforest Syst 32:147–161
Chapter 10
The Impact of Browsing and Grazing
Herbivores on Biodiversity
Spike E. van Wieren and Jan P. Bakker
10.1 Biodiversity and Large Mammalian Herbivores
Biodiversity definable as the variety of life in all its forms. Biodiversity encompasses
the entire biological hierarchy. Two different hierarchical schemes are frequently
used to classify biological entities (Sarkar and Margules 2002): a spatial hierarchy
and a taxonomic hierarchy. The spatial hierarchy runs from biological molecules
and macromolecules, through cell organelles, cells, individuals, populations and
meta-populations, communities, ecosystems ultimately to the biosphere; and the
taxonomic hierarchy from alleles through loci, linkage groups, genotypes, sub-
species, species, genera, families, orders, classes, phyla, and kingdoms. Each
level is highly heterogeneous and, even more important, most entities are poorly
defined. For example, there is still much confusion about the species concept, to
mention one of the key entities in biology (Sarkar and Margules 2002).
Furthermore biodiversity also comprises the relationships between entities within
a hierarchy. Through all the confusion of the concept of biodiversity it is very
difficult to operationalise and, therefore, various proximates have been developed
which allow the estimation of (subsets of) biodiversity in the field. According to
Ricotta (2005): ‘biodiversity may be defined simply as a set of multivariate sum-
mary statistics for quantifying different characteristics of community structure’.
The most widely used biodiversity indicators are character or trait diversity,
species diversity and species assemblage diversity. The rationale of trait diversity is
that evolutionary mechanisms usually impinge directly on traits of individuals
(Vane-Wright et al. 1991; Sarkar and Margules 2002). Trait, however, is a difficult
term and largely determined by pragmatic considerations and, therefore, not very
precise. The use of the species assemblage indicates that what is important is the
variety of biotic communities with their associated pattern of interactions, some-
times referred to as functional biodiversity. Focusing on communities will
automatically take care of species. Despite all the discussion around the concept,
species diversity is the most commonly used indicator of biodiversity, even though we
know that there is much more to biodiversity than the variety of species. A few
reasons why species richness is used, is that it is thought to correlate with many
© Springer 2008
264 S.E. van Wieren and J.P. Bakker
measures of ecological diversity and that there is a positive relationship with trait
richness and topographic diversity (Sarkar and Margules 2002).
A simple, much frequently used, index is the number of species in a location
(Desrochers and Anand 2005). This is a crude way to estimate diversity and
alternative diversity measures have been developed which combine species richness
with relative distribution of species abundances. The Shannon entropy (H) and the
Simpson diversity (1/D) indices are currently the most widely used (Pielou 1975;
Shannon and Weaver 1949; Simpson 1949). These established measures have
recently been extended to incorporate additional information related to ecological
complexity (see Desrochers and Anand 2005 for review). Within the recognition of
the effect of habitat heterogeneity on biodiversity, ‘total’ diversity can be partitioned
into within- and between-habitat diversity (Whittaker 1972). Thus total diversity
(γ-diversity) can be estimated by computing H of all sampling units pooled together;
within-habitat diversity (α-diversity) is estimated from the H of all sampling units,
and between-habitat diversity is the difference between γ-diversity and α-diversity
(Desrochers and Anand 2004). From this the quantity β-diversity can be computed
as the ratio of γ-diversity and α-diversity. This measure is useful as it quantifies
habitat heterogeneity across habitats, and can be used in environmental assessment
because it indicates the degree to which habitats have been partitioned by species
(Desrochers and Anand 2004; Ricotta 2005).
In the studies with regard to the effects of grazing and browsing on biodiversity,
the most widely used indices are almost identical to the most widely used ones. These
are trait diversity (phenotypic and genotypic), species diversity (species richness and
species composition), and community diversity (structure and composition).
Evolutionary relationships between mammalian herbivores and biodiversity.
Large mammalian herbivores have been around for some time (Janis this volume)
and we can ask to what extent this group has contributed to biodiversity as described
above. As real proof is absent we can only speculate and indicate probabilities. First
of all the group itself is a contribution to terrestrial animal diversity; more than 200
species of ungulates exist now, but summed over the Tertiary a multiple of this
number has populated the earth, evolving and going extinct in an irregular fashion.
(Hernández Fernández and Vrba 2005). On a higher trophic level we can safely
assume that the large herbivores have provided exclusive niches for a number of the
larger predator species, among them lions and tigers. Large herbivores and large
carnivores most likely co-evolved during the Tertiary.
Given the long evolutionary time scale and the wide distribution of large herbivores
across the major biomes, they must have exerted strong selective forces on, in particular,
the plant world. It is generally acknowledged that herbivory has led to a number of
plant responses which can be seen as adaptations to predation. Examples are the
development of morphological structures such as thorns and hairs, and chemical
defenses, although the latter probably predate the mammal era by a long time
(Lindroth 1988). Thus the first, and maybe the largest, effects would have been on
traits of plants inducing larger phenotypic variation and also, perhaps, a larger
genotypic variation. It is not known if this selection has been strong enough to lead
to speciation and thus to new species although angiosperms and mammalian herbivores
10 The Impact of Browsing and Grazing Herbivores on Biodiversity 265
co-evolved for a long time during the Tertiary (Tiffney 2004). The development of
large fleshy fruits which need to be predigested and dispersed by large mammals is
an indication that some plant species are dependent, and can maybe only exist, in the
presence of large mammals.
Several species of temperate North American and European thorny shrubs may
have evolved under browsing by now extinct large Pleistocene herbivores, since
the presence of taxa such as Pinus, Crataegus, Rhamnus, Rhus, and Juniperus
goes back through the Pleistocene even as far as the early Tertiary (Tallis 1991).
These ecological anachronisms are often hard to prove. Large herbivores such as
cattle and horses, domesticated from their extinct ancestors aurochs and tarpan,
respectively, make it possible to test hypotheses in the recruitment strategy of
more preferred woody species that are spatially associated with thorny shrubs
under grazed conditions. A cross-site comparison of four floodplain woodlands
in north-western Europe showed that sessile oak (Quercus robur) can regenerate
in the presence of large herbivores through spatial association with Blackthorn
(Prunus spinosa), a clonal thorny shrub (Bakker et al. 2004). An experiment with
transplanted oak seedlings revealed that oak seedlings grew best in grassland
exclosures and on the edge of thorny shrubs that received most light. Oak survival
was strongly reduced in oak woodland with low light availability. However, in
sites with high rabbit (Oryctolagus cuniculus) density no young trees were found.
Rabbits graze both on young ramets of Blackthorn and on young oaks. It was
concluded that the process of associational resistance did not work against rab-
bits, as they consume the oak seedlings and the thorny shrubs that have no thorns
when they are young. Under low rabbit densities the young ramets of Blackthorn
can escape from grazing, and once established, they give shelter for the oak
(Bakker et al. 2004).
It is likely that large herbivores have (had) an effect on community diversity.
Even if no new species have evolved in the presence of large herbivores, it can be
assumed that at least locally they have changed the structure and species composi-
tion of communities. As such they have contributed to community diversity. Some
species, such as elephant, moose and beaver, have been credited with even more
effective power in the sense that they are thought to be able to control vegetation
development on a larger landscape scale (Owen-Smith 1988; McInnes et al. 1992;
Naiman et al. 1986). If such effects should exist in the natural world, then some
large herbivores would be able to affect biodiversity on the ecosystem level.
In our view the effects of large mammalian herbivores on biodiversity are
probably relatively modest, mainly acting through increasing trait diversity and
through the reshuffling of species from the regional species pool by creating new
communities. This does not exclude, however, the enormous potential for change
in species composition that can be achieved with grazing, as will be demonstrated
below. Most of the remainder of this chapter is about the application of grazing and
browsing in man-modified and/or man-controlled systems and, as such, the exam-
ples differ from the natural world. Most grazing is carried out with a preconceived
human management goal and it is important to realize that the results of most of the
grazing research should be viewed in this context.
Table 10.1 Plant traits which can be affected by herbivory, with indication whether they are
aimed at defence (d), escape (e), or tolerance (t)
Trait
Thorns/spines (d)
Trichomes (e.g., hairs) (d)
Chemicals (e.g., tannins, alkaloids, silicate) (d)
Low stature (height/biomass) (e)
Lower reproductive potential (< flowers, seeds) (e)
Prostrate growth form (e)
More rosettes (e)
Increased tillering (t)
Increased photosysnthetic rates (t)
10.2 Effects on Traits
The main short-term and direct mechanism by which herbivores affect plants is
through defoliation. Plants react to this kind of attack by various kinds of resistance
mechanisms. Plants may try to avoid damage via defence or escape in space and
time, or they may tolerate herbivore damage, once attacked. Various plant traits
may be affected by herbivory and a list of important ones is given in Table 10.1.
An example of induced defense by herbivory is the variation in spinescence in
plants (Young et al. 2003; Obeso 1997). Young et al. (2003) studied the effects of
large herbivores on spine length of Acacia drepanolobium in East Africa by simulated
and natural herbivory. New spines in plots from which herbivores were excluded were
35–40% shorter than spines in plots exposed to all browsers (Fig. 10.1). On low
Large mammalian herbivores excluded

2 Large mammalian herbivores present
Megaherbivores also present
Thorn length(cm)
1.5
0.5
0
Low High
Branch height
Fig. 10.1 Mean spine (Thorn) lengths of low (∼1 m) and high (∼2 m) branches on Acacia.
drepanolobium trees accessible to different herbivores in an East African savanna (from Young
et al. 2003)
branches, spine lengths were low in total exclusion plots, and longer in plots that
allowed wildlife and plots with both megaherbivores (elephants, giraffes). On higher
branches, recently produced spines were similar in length in total exclusion plots and
plots that allowed wildlife, but were longer in plots in which megaherbivores were
allowed. Apart from these induced effects, Young et al. (2003) also showed that simu-
lated large mammal browsing induced greater spine length on trees that had reduced
spine length after several years of herbivore exclusion (Fig. 10.2). Interestingly, the
induced responses were highly localized and the main variation in natural browsing
appeared to be associated with escape in space from browsers by the production of
branches outside the reach of browsers, as the trees grow. This escape from herbivory
by individual branches high above the ground is seen as an important force in the
development of the inducibility of spine length in Acacia spp. (Young et al. 2003).
An important aspect of tolerance with respect to herbivory is a change in photo-
synthetic rates. Widely different effects have been reported: from extreme under-
compensation to even overcompensation (Rosenthal and Kotanen 1994; Strauss and
Agrawal 1999). This induction of the productivity trait may lead to the development
3.0 Before simulated herbivory
After simulated herbivory

Mean spine length (cm)
2.0
1.0
0.0
Shoot removal Controls
Treatment
Fig. 10.2 Reinduction of spine length in simulated herbivory experiments on Acacia trees that had
been protected from large mammalian herbivory for five years in an East African savanna. Spine
lengths before (open bars) and one year after (solid bars) simulated herbivory. Control branches
were either on a paired fork on the same branch or on a branch on the opposite side of the tree crown.
Experimental shoot removal resulted in new spine three times longer than before the experiment, but
no significant response on other branches on the same tree (after Young et al. 2003)
(or expression) of different ecotypes within species, which behave differently under
the same grazing treatment. A nice example is the work at Wind Cave National Park
on the physiological responses of prairie grass populations to responses to prairie-dog
grazing (Holland and Detling 1990; Fig. 10.3). This work shows that long-term
grazing in two ecotypes of Agropyron smithii results in a biphasic response curve
for above- and below-ground production and for gross and net mineralized N as a
function of grazing intensity. Furthermore, the model results suggest that differences
in the long-term quasi-equilibrium conditions develop between the two ecotypes,
even though both populations show the biphasic response. The populations adapted
to grazing contrasted sharply with the less adapted population in terms of grazing-
induced potential and N-restricted plant production. The grazing-adapted population
shifted its optimal growth conditions at high grazing intensities, whereas the non-
grazing-adapted population showed growth limitations at much lower grazing
intensities (Fig. 10.3).
Given the above examples, it is clear that herbivory can induce the expression of
phenotypic variation of many traits. This may lead to various phenotypes (and gen-
otypes) being simultaneously present in a given area in contrast to a lower pheno-
typic variation present in ungrazed situations (Loreti et al. 2001). It has also been
proposed that resistance to grazing is more varied in species with a long grazing
history (Wilsey et al. 1997; McNaughton 1984; Painter et al. 1993), while in species
with a short evolutionary history of grazing no, or very little, variation was found
(Jaindl et al. 1994; Loreti et al. 2001).
Recently it has been found that herbivory can have effects on plant mating systems.
Resource limitation caused by herbivory can negatively affect flower production,
flowering phenology, and seed mass and number (Steets and Ashman 2004).
Herbivory can also, through leaf damage, reduce the number of flowers on a plant that
open synchronously, and reduce flower morphology and nectar reward (Steets and
Ashman 2004; Sharaf and Price 2004; Vazquez and Simberloff 2004). Through these
effects, phenotypic variation in mating trait expression might change, but effects
might also cascade through the system, especially by changing the pollinator pool.
This is especially apparent in systems not accustomed to a long history of grazing and
where introduced herbivores have strong negative effects on plant species which are
important for the community of flower visitors (Vazquez and Simberloff 2004).
Although there is a growing consensus that tolerance and defence are evolving
traits under selection from herbivores in natural plant populations (Strauss and
Agrawal 1999), and there are many plant systems for which genetic variation in
tolerance to herbivory have been reported (Strauss and Agrawal 1999), it is still not
clear to what extent trait variation indeed evolved as the result of selective pressures
of vertebrate herbivory, let alone that this selection has led to novel genotypes. In
fact, many authors stress that the traits affected by herbivory are basic characteris-
tics of plants with important roles in growth and reproduction. Some traits may be
by-products of selection for other types of damage: the protected apical meristems,
for instance, may have evolved in response to drought and fire (Stowe et al. 2000;
Coughenour 1985; Rosenthal and Kotanen 1994). The potential selective pressures
exerted by ungulates resulting in speciation seem rather limited.
PLANT POPULATIONS AND NITROGEN CYCLING
OFF-COLONY ECOTYPE
500 30
N-Limited Grazing-Limited
A Potential Production
400
N-Restricted Production
Available N
20
300
200
Above ground production (g/m2/y)
10
Plant available (g/m2/mo.)

100
0 0
ON-COLONY ECOTYPE
500 30
N-Limited Grazing-Limited
B
400
20
300
200
10
100
0 0
0.0 0.2 0.4 0.6 0.8
Grazing intensity
Fig. 10.3 Productivity estimates for two South Dakota western wheatgrass populations growing
in the vicinity of prairie dog colonies, as depicted by Century Model studies. A shows results from
populations not located on prairie dog colonies, B from those on such colonies. A pronounced
biphasic relationship exists for the B population potential production, not so for the A population
in this regard. Both populations show biphasic responses for N dynamics, but in the interaction
between N-limited and grazing-limited responses, the A population shows a shift to the left as a
function of grazing intensity, while the B population shows a shift to the right. This suggests that
western wheatgrass has developed ecotypic variation in response to stresses arising during its
long-term grazing history (from Holland et al. 1992)
10.3 Effects on Plant Communities
Grasslands. The main mechanisms by which foraging herbivores affect plants are
selectivity and trampling. Differences in quality and quantity between plant species
lead to preferential grazing selectively affecting plant traits such as tolerance and
defence. More intense grazing (and thus trampling) increases the number of
vegetation gaps, changing the colonization matrix of the system. All these differential
influences may lead to a change in vegetation pattern and species composition
(Sternberg et al. 2000).
Because of the above mechanisms, site selection occurs which can be enhanced
by feedback mechanisms. Through grazing, plant quality can become enhanced
leading to re-utilization of the patch (Bakker et al. 1983; Weber et al. 1998; Adler
et al. 2001). Patch size depends on grazing intensity and type of grazer. Sheep
create patches of about 60 cm across (Bakker et al. 1984), while cattle patches can
be up to several metres (Fuls 1992). The resulting micro patterns can be relatively
fixed in time (Bakker 1989). An increase in size of patches can lead to the formation
of grazing lawns. Frequently created by (very) large grazers, these can provide
good grazing conditions for smaller herbivores (Verweij et al. 2006).
Selective grazing can also lead to changes in plant species composition. Some
species are expected to be more affected than others when being grazed,
depending on species specific characteristics. Tall annual grasses are frequently
negatively affected because their regeneration capacity decreases, while
hemicryptic species are less affected. They can withstand heavy grazing as their per-
ennating buds are buried near the soil surface and they frequently have strong
physical and chemical defenses (Sternberg et al. 2000). In relation to grazing,
plants have been characterized in functional groups such as decreasers and
increasers. Figure 10.4 gives an example of how a shift in functional groups can
occur under different grazing intensities.
Most of the important grazing studies have been carried out on climatically deter-
mined, natural, grasslands and these will be considered first. Climatically determined
grasslands (including savanna and shrub steppe) cover about 25% of the earth’s land
surface (Lauenroth 1979). They receive 250–1000 mm of annual precipitation and
have mean annual temperatures between 0° and 26° C.
The effects of grazing in grasslands can be best understood when viewed as
superimposed on general relationships between plant species richness and plant
biomass. This general relationship is hump-back shaped and based on resource
limitation at the lower biomass level and competition for light at the high biomass
level with maximum species richness at intermediate biomass levels correspond-
ing with a moderate competition or disturbance level gradient (Grime 1973;
Frank 2005). Grazing can be superimposed on this model because herbivores may
increase or decrease species richness depending on grazing intensity and the
amount of biomass and selective grazing on a dominant species. A number of
models have been developed to explain the response of grasslands to large gener-
alist herbivores. Among them are the range succession model (Dyksterhuis
Fig. 10.4 Contribution of grazing response groups to species density for three categories of
grazing. Grazing has a significant effect on species density in this design. The study was
conducted within the southeast Queensland Bioregion, Australia, with grassy eucalypt woodland
as the dominant vegetation type (McIntyre and Martin 2001)
1949), the predation hypothesis (Paine 1966), the intermediate disturbance

hypothesis (Grime 1973), the Huston hypothesis, the Milchunas hypothesis
(Milchunas et al. 1988), and the state-and-transition model (Westoby et al. 1989;
Laycock 1991). Of these hypotheses, only the Milchunas hypothesis will be
discussed here because it satisfactorily explains a wider range of real situations
than do the other models (Cingolani et al. 2005). In this model, recent insights
from the state-and-transition models can be incorporated; this will be discussed
in the final paragraph of this chapter.
Milchunas et al. (1988) developed a generalized model of the effects on
grassland community structure which proposed that interactions occurring along
gradients of evolutionary history of grazing and environmental moisture were
primary factors that explained changes in plant species diversity, changes in species
composition, and invasion by exotics across gradients in grazing intensity. The
direct act of grazing by large herbivores represents a loss of organs to individual
plants and an alteration of canopy structure to the assemblage of plants. Plant reac-
tion to the grazing event will depend upon the ability of individuals to compensate
for lost organs and the relative impact of the removal on competitive relationships
in the canopy. Canopy development and above-to-belowground biomass ratios
increase with increasing moisture. Therefore, a given percentage removal from the
canopy will have a greater effect in subhumid than in arid environments both in
terms of the intensity of plant interactions in the canopy and percentage removal of
total plant biomass. Grazing should have greater affect on species composition in
more humid areas because adaptations of tall growth forms capable of competing
for light in a dense canopy are opposite to those that provide resistance to or
avoidance of grazing. In contrast, plant adaptations to frequent loss of organs from
drought or herbivory are similar, and under these arid and semiarid conditions com-
petition is primarily for belowground resources. The evolutionary history was
brought into the model because it was hypothesized that increasing grazing history,
over evolutionary time, results in greater capacities for regrowth following herbivory,
favouring prostrate growth forms. In communities of short evolutionary history, and
high moisture content, grazing causes rapid shifts in the dynamic balance between
suites of species adapted to either grazing avoidance/tolerance or competition in the
canopy. Milchunas et al. (1988) presented the model and the experimental evidence
for four boundary cases (Fig. 10.5). The model of Milchunas has been confirmed
Fig. 10.5 Plant diversity of grassland communities in relation to grazing intensity along gradients
of moisture and of evolutionary history of grazing. Increments on the diversity axis are equal in
all cases, but equal specific values are not implied; that is, relative response, not absolute diversity
is implied (from Milchunas et al. 1988)
in a number of studies (Milchunas and Lauenroth 1993, Sternberg et al. 2000; Hunt
2001; see Cingolani et al. 2005 for recent review). The model has been expanded
in the sense that for the axis ‘moisture’ one can substitute ‘annual net primary
production’ or ‘productivity’ (Milchunas et al. 1988; Cingolani et al. 2005; Frank
2005). It is relevant to note that in all models plant diversity decreases, sometimes
strongly, when grazing intensity increases beyond some optimum. It seems that
currently most grazing systems of the world (if not all) are situated somewhere at
the right hand of any of these curves, which means that most are overgrazed.
Herbivores can affect the rate of succession in grasslands and alter the composition
of some successional stages. Salt marshes typically go from an early successional
stage with species such as Puccinellia maritima and Plantago maritima to a late
successional stage with a dominance of Elymus athericus and Atriplex portulacoides
(Kuijper and Bakker 2005). Small herbivores such as geese and hares preferentially
forage on early successional species which can tolerate grazing well. At the same
time these grazers prevent colonization of the later successional species thereby
retarding the succession by decades (Van der Wal et al. 2000; Kuijper and Bakker
2005). They also create a mid-successional stage, not present in the absence of
herbivores, with a high abundance of grazing-tolerant species (Festuca rubra,
Puccinellia maritima, Plantago maritima, Triglochin maritima). They thus add, albeit
temporarily, an extra species diversity element to the system.
Most of the grasslands in Europe, Japan, eastern North America, and areas of
Australia and Asia are of anthropogenic origin. They have been derived from
forests and have undergone huge changes (Prins 1998). It is likely that in most cases
the species composition no longer bears resemblance to the original situation.
Important changes have been a large increase in light reaching the ground layer,
with an accompanying increase in the biomass of the field layer, and a decrease in
vertical vegetation structure. Although it is almost impossible to estimate the
change in biodiversity, we can safely assume that much biodiversity has been lost
in the process as so many species are confined to the tree canopy, the various layers
in a forest, and to dead wood. The focus of the effects of grazing in anthropogenic
grasslands thus has always been on limited subsets of biodiversity (flowering
plants, butterflies, dung beetles, ground breeding birds). Within this context, the
many grazing studies (summarized by Bakker 1998) have produced an amazing
variety of results. A few studies have found an increase in species richness at inter-
mediate grazing intensity (Sternberg et al. 2000; Bakker et al. 1997) but in most
studies the results are not clear (Bakker 1998; Bullock et al. 2001; Luoto et al.
2003). Contributing to the inconsistency is the great variety in grazing treatments
(winter grazing, summer grazing, year round grazing, rotational grazing, different
species of herbivores) imposed on plant communities, the short duration of many
experiments, and the probably inherent variability that can be expected in these
altered systems. Much work has been done on the effects when grazing is totally
removed. Because of changes in agricultural practices, many agricultural grasslands
have been abandoned and generally adverse effects (in selected functional
groups) have been found. The herb layer frequently becomes dominated by tall
grasses and herbs while plant species diversity declines; the aboveground standing
crop increases accompanied by litter accumulation (Bakker 1998; Luoto et al. 2003).
These effects are also in line with the Milchunas et al. (1988) hypothesis. The
frequently occurring development of vegetation communities towards the original
state through bush encroachment and tree regeneration are also generally considered
as undesirable with respect to biodiversity targets (Bakker 1998; Verdu et al. 2000).
It should be noted that restoration attempts can be seriously hampered by seemingly
permanent changes in the state of the system (see final section, 10.7).
Woody vegetation. Very little is known of the effects of large herbivores in
forests under natural conditions. In the few remnants of natural temperate forests,
the density of grazers and browsers is quite low but nevertheless we can expect
some effects of browsing—and trampling in particular. The most important ones
are modification of the structure and composition of the plant community and
effects on the succession rate (Pastor et al. 1997). Forest herbivores will concentrate
foraging on available regeneration gaps where they may be able to extend the time
to gap closure, enhance ground cover and diversity (Côté et al. 2004). As browsing
animals are highly selective, some species in the gaps will be more affected than
others, and this is likely to influence the ultimate composition and the vertical
structural component of the forest.
It has been hypothesized that, under natural conditions, large herbivores were
able to maintain large open areas in temperate forests leading to much more open
landscapes than in the absence of grazers (Vera 2000). If so, they should truly be
regarded as keystone species. This hypothesis, however, is not very well supported
by evidence from pollen analysis (Mitchell 2005) and is difficult to test.
By far the greater part of the forests of the world are heavily modified by man.
Tree planting after logging has led to early successional forests on a grand scale in
the temperate and boreal regions. This has increased the food base for deer tremen-
dously, which in turn has led to an explosion in deer numbers, both in Europe and
North America (Fuller and Gill 2001). It is not surprising that under these changed
conditions herbivores will have profound effects on forest structure and composi-
tion with cascading effects on other taxa. The results of many effect studies have
been reviewed recently (see Côté et al. 2004) and here only a short summary of the
main type of changes is given. Heavy browsing reduces twig density, resulting in a
more open canopy. Because more light reaches the forest floor, the production of
shrubs and herbs increases, decomposition rates change, changing field layer
vegetation affecting vertebrate communities (Persson et al. 2000; Suominen et al.
1999). Deer may cause a shift in the canopy composition by selective browsing,
generally in the direction of the less preferred, more browsing-tolerant species
(Table 10.2). When browsing pressure increases further, the rate and direction of
forest succession can be seriously affected and succession can be stalled.
All the above effects are viewed as negative, both from (timber) production and
conservation perspectives. Given the poor state of these systems are in, there is no
easy solution. The most frequently mentioned counteractive measure is control of
deer numbers (Côté et al. 2004). This is not always easy to do and there may be asso-
ciated negative side effects (changes in habitat use, effects on predators, increasing
vigilance; see final section, 10.7). Protection of forest for some time may also help.
Table 10.2 Compositional shifts in dominant tree species induced by deer browsing in boreal and
temperate forests (from Côté et al. 2005)
Former dominant New dominant
Balsam fir (Abies balsamea) White spruce (Picea glauca)
Birch (Betula spp.) Norway spruce (Picea abies)
Eastern hemlock (Tsuga canadensis) Sugar maple (Acer saccharum)
Mixed hardwoods Black cherry (Prunus serotina)
Oak (Quercus spp.) Savanna type system
Scots pine (Pinus sylvestris) Hardwoods and Norway spruce
If the system is allowed to consist again of the mature-age component for a large
part, this will slowly turn the tide.
A combination of these measures is an option as well. Simply letting nature run
its course can be an option, too, but in some cases this can take a long time,
resulting in extinction of taxa of conservation interest in the meantime. It is increas-
ingly recognized, however, that a simple population approach to deer management
is not sufficient and that an approach should be taken that considers whole-
ecosystem effects (Côté et al. 2004; McShea et al. 1997; Gaillard et al. this volume).
Much more work should be done to study relationships between community com-
position across taxa and various population sizes of deer to understand the full
range of deer impacts on biodiversity (Kramer et al. 2003).
10.4 Effects on Invertebrates
Invertebrates comprise a huge group occupying a large number of niches and it

would be expected that there are many ways in which large herbivores affect this
group. The studies that have been done in this field always have to focus, by
necessity and through preference, on a limited subset of taxa, which makes it
difficult to judge the overall interactions of invertebrates with large herbivores.
Herbivores may affect invertebrates directly and indirectly through changes in, for
example, quality and quantity of litter, dead wood, flowers, seeds, cover, egg laying
sites, nesting sites, and vegetation structure. In natural systems we would expect large
herbivores to affect insect diversity predominantly through the presence of their liv-
ing and dead bodies, their excreta, and through local effects on vegetation structure
and the creation of gaps. There are many intimate associations between large herbiv-
ores and commensals and parasites. These can be species-specific such as the deer
bot-fly (Cephenomya auribarbis) and the deer ked (Lipoptena cervi; Hutson in
Stewart 2001). Dead ungulates attract carrion-feeding invertebrates, such as rove
beetles as well as assorted flies, some of them being specific to large carcasses. There
exists a whole community of coprophilous insects feeding and laying their eggs on the
faeces of large herbivores, including Atomaria peltataeformes and Aphodius nemoralis
(Kronblad 1971; Peterson 1998; both cited by Persson et al. 2000). The most obvious
taxa in dung are dung beetles (Geotrupidae) and scarab beetles (Scarabidaeidae).
Although many dung-associated species are rather unspecific, some species are
reported to be associated with only primarily one species of ungulate (Duncan et al.
2001). Large changes in ungulate density will affect the relative abundance of those
species associated with them (Putman et al. 1989). Dung-associated invertebrates are
an important source of food for a number of predatory animals and, as such, large
changes in abundance of the coprophilous community can have cascading effects on,
e.g., tawny owls (Strix aluco) and different species of woodland bats (in the UK,
Duncan et al. 2001).
Through trampling and local intensive grazing, large herbivores can create gaps
which are usually favourable for ants, as they provide potential nest sites and
adequate microclimatic conditions (see Gonzales-Megia et al. 2004). In forests,
herbivores may also create gaps or keep gaps open for a prolonged time. This will
benefit insect diversity because the rarest species in woodlands are associated with the
opposite ends of the succession spectrum: open clearings and mature or senescent
habitats, especially dead and decaying wood (Stewart 2001).
Given the above effects, we can assume that large herbivores, by their presence
and activities, do contribute positively to invertebrate diversity. The main
mechanism through which this has happened is increase in (mainly habitat) hetero-
geneity. If grazing pressure increases, however, this positive force can be seriously
compromized because vegetation structure (both horizontally and vertically) will
be simplified, direct competition for food will increase, and other important niches
may disappear (for instance dead wood). There are many studies that report a
decline in at least some (but frequently in many) invertebrate groups as a result of
heavy grazing pressure (Kearns 1997; Kruess and Tscharntke 2002; Sudgen 1985;
Vazquez and Simberloff 2004; Souminen et al. 1999; Gonzalez-Megias et al. 2004;
Duncan et al. 2001; Hutchinson and King 1980). Kruess and Tscharntke (2002)
studied the insect diversity on intensively and less-intensively grazed pastures and
on 5- to 10-year-old ungrazed grasslands in northern Germany and related diverse
taxa to vegetation structure. Here we give their results for butterfly adults and lepi-
dopteran caterpillars (Fig. 10.6). The total number of butterfly species and
lepidopteran caterpillars increased with decreasing grazing intensity. For both these
groups, but also for other groups studied, mean canopy height was the best
predictor of both species richness and abundance. It has to be noted that the less-
intensive grazing treatment must still be considered quite high from a conservation
point of view (1.4 cattle/ha from 1 May to 15 November). From the above example
it may seem that cessation of grazing will lead to the highest insect diversity but, in
fact, most workers conclude that insect diversity in grasslands (and probably also
in forests) is highest at moderate herbivore densities (see Di Giulio et al. 2001;
Swengel and Swengel 2001). Indeed, cessation of livestock grazing after intensive
grazing, may be beneficial for invertebrates in the short term, when plants can
flower. However, long-term cessation of grazing resulted in the disappearance of
characteristic halobiontic invertebrates in salt marshes in Germany (Andresen
et al. 1990) and France (Pétillon et al. 2005). They were replaced by ruderal species
typical for tall inland forb communities. As such, we can link overall insect
Number of species
20 20
a) b)
10 b 10 b
b
5 5 a
a a
1 1
I E U I E U
Grazing intensity Grazing intensity
20 20
Number of species
c) d)
10 10
5 5
1 1
0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80
Mean canopy height (cm) Mean canopy height (cm)
Fig. 10.6 Effects of grazing intensity (I. Intensively grazed pastures; E extensively grazed pastures;
U ungrazed grasslands) on butterfly adults and lepidopteran caterpillars: a) mean (±1 SE) number of
adult species (V2,15 = 13.9, p < 0.001, n = 18, and b) mean (± 1 SE) number of caterpillar species
(F2.15 = 11.35, p < 0.001, n = 18. Different letters above bars indicate significant differences (Tukey’s
honest significant difference). Correlation between mean canopy height and c) number of adult
butterfly species (Y = 2.9–55.5/X, F1.16 = 28.6, r2 = 0.64, p < 0.001, n = 18 and d) number of
caterpillar species (Y = 2.8 – 54.6/X, F2.16 = 10.47, r2 = 0.55, p < 0.001, n = 18) (from Krauss and
Tscharntke 2002)
diversity to the general model of Milchunas et al. (1988; Fig. 10.5) and agree with
Hutchinson (1959) and Hunter and Price (1992) that a high α-plant diversity is
associated with a high α-diversity in higher trophic levels. We can also conclude
that, although some invertebrate groups benefit from heavy grazing, in many
systems insect diversity has declined considerably because of present densities of
large herbivores.
10.5 Effects on Birds
As with plants and insects, we have little information on the overall effects of large
herbivores in natural systems. Large herbivores provide carrion and dung which,
through the invertebrates that live in and on them, provide feeding opportunities for
a number of bird species, notably corvids and raptors (Duncan et al. 2001). Large
herbivores carry parasites and disturb vegetation while moving, providing niches
for commensals such as the cattle egret (Bubulcus ibis) and cowbirds which feed on
insects or parasites carried by the herbivores (Mayfield 1965). On a larger scale,
herbivores may have an impact on the composition and structure of the vegetation,
Table 10.3 Distribution of western European species of birds and mammals (ungulates excluded)
across preferred habitat types; species can be related to more than one habitat type. 57 species of
mammals, 257 species of birds (after Van Wieren 1998)
Birds
Mammals Nesting site Feeding sitea
Bare soil 18 40
Short herbaceous 5 62 57
Tall herbaceous 22 81 47
Shrub 27 20 12
Woodland 41 76 30
a
Aquatic species excluded
with all the cascading effects on other taxa, but this depends very much on the den-
sity of herbivores in a particular system.
As almost all natural grasslands have been modified by man and are now
grazed by livestock, most studies have focused on the changes brought about by
different grazing intensities. Birds are particularly sensitive to changes in the
vegetation structure but also to changes in plant species composition affecting
food supplies and nesting opportunities (see also Table 10.3). Many studies have
shown that cattle grazing reduces the richness and densities of avian communities
in North American prairie habitats (Kantrud 1981; Chamberlain et al. 2000;
Gates and Donald 2000; Söderström et al. 2001). It has also been shown that
waterfowl nest density and nesting success are significantly reduced in grazed
prairies compared to ungrazed ones (Klett et al. 1988; Kirby et al. 1992).
Sometimes overall species richness remains the same but a major shift in species
composition can be noted. As with invertebrates, heavy grazing leads to a reduc-
tion in vegetation heterogeneity with negative consequences for many ground
dwelling bird species, e.g., nest losses through trampling (Beintema and Müskens
1987; Barker et al. 1999; Grant et al. 1999), exposure to predation pressure (Brua
1999; Wilson and Hartnett 2001), or a decrease of large insects for the larger
insectivores (Söderström et al. 2001).
Most grasslands in Europe and many in North America are man-modified
although they frequently have evolved under grazing pressure for centuries. A
native flora and fauna has developed in these grasslands, which has adapted to
grazing disturbance. Due to changes in agricultural practices the greater part of
these grasslands have disappeared, or have become much more intensively used
or have been abandoned, with a later decline in typical farmland birds (Newton
1998). When abandoned, the bird species richness in these habitats can increase
quickly due to an increase in bushes and trees (Van Wieren 1998) but frequently
this direction of change is not much appreciated as the focus is on typical farm-
land birds (Söderström et al. 2001). When looking at total bird species richness
it seems that a low to moderate grazing pressure yields the highest number of
species in grasslands. In the Argentine pampa, moderate grazing showed a higher
density, species richness, and diversity of species when compared to ungrazed or

heavily grazed areas (Zalba and Cozzani 2004). This general finding can be
attributed to the higher vegetation heterogeneity in moderately grazed grasslands.
Another example of this can be found on salt marshes. Salt marshes are natural
grasslands which are generally grazed by livestock in the summer months. In
many areas changes in grazing management have led to either abandonment or
intensification of the grazing pressure (Norris et al. 1998). Cessation of grazing
leads to a tall rank sward while heavy grazing results in a short, uniform dense
sward. Both developments are detrimental for some ground breeding birds such
as, for instance, the redshank Tringa totanus, an important meadow bird in north-
western Europe. When relating breeding density of redshank to grazing intensity,
Norris et al. (1998) found the highest breeding density on low to moderately
grazed swards. Again, a moderate grazing intensity provided the greatest hetero-
geneity in the ground vegetation (Norris et al. 1998).
Salt marshes are also important for avian herbivores such as geese. They prefer
the high quality low growing plants on the lower salt marshes, while the taller,
later successional stages are much less attractive because of the development of
tall, less-preferred plant species. When these older salt marshes are grazed they
become attractive to geese due to the development of short high quality swards
(Van der Graaf et al. 2002; Stock and Hofeditz 2000), a clear example of grazing
facilitation. As most wintering geese in western Europe rely on agricultural
grasslands adjacent to the coast, livestock grazing also plays an important
facilitating role (Bos et al. 2005).
As birds are particularly sensitive to changes in vegetation structure, especially in
the understorey (Miyashita et al. 2004; Fuller 2001), we would expect large changes
in the bird communities associated with forests when the density of large herbivores
increases. In a controlled browsing experiment De Celesta (1994) studied the rela-
tionships between deer density and bird communities in northern hardwood forest in
the US. A clear non-linear, negative relationship was found between bird diversity
and deer abundance. Declines of 2% and 37% in species richness and abundance of
intermediate canopy nesters between lowest and highest deer densities were found.
The density threshold for the major changes was between 7.9 and 14.9 deer/km2. Also
in other hardwood forest increased deer densities led to lower species richness and
abundance of canopy feeders (McShea and Rappole 2000). Sometimes no differences
between density treatments were found because of species replacement effects
(McShea and Rappole 2000; Moser and Witmer 2000; Fuller 2001). This indicates
that, like insects, birds occupy a wide variety of niches and that each landscape has
its own particular type of bird community. The pasture woodlands found in western
Britain for instance, carry a whole suite of species, very typical for that habitat,
including pied flycatcher (Ficedula hypoleuca), redstart (Phoenicurus phoenicurus),
and tree pipit (Anthus trivialis; Fuller 2001). For both grasslands and forests we can
conclude that heavy grazing and browsing leads to a decrease in bird species richness
while low to moderate herbivore pressure has the highest species richness and diversity.
Avian herbivores such as geese are facilitated by moderate to heavy grazing pressures
in grasslands.
10.6 Effects on Mammals
Large herbivores in grassland systems. Generally more than one species of large
herbivore can be found in any system. In Africa especially, species richness can be very
high, up to 31 species in the larger national parks. To what extent does the presence
of one species affect the presence of others? Species within an ungulate community
frequently differ in many respects (size, morphology, digestive physiology) which
results in the various species occupying different parts of the ecosystem: resource
partitioning. Resource partitioning can simply be the result of adaptation to a
particular niche but may also have been (partly) driven by competition which is
frequently a selective force shaping the composition and structure of ungulate com-
munities; although evidence for this is still scarce (Arsenault and Owen-Smith
2002). Murray and Illius (2000) studied the foraging behaviour of sympatric topi
(Damaliscus lunatus) and wildebeest (Connochaetes taurinus) in East Africa and
found that the very selective grazing style of the smaller species (topi) was able to
deplete the green leaf component in the tall sward to the detriment of forage quality
for the larger species. Apart from competition, facilitative interactions have also
been proposed to have shaped ungulate communities. Facilitation may arise when
one species makes more grass accessible to another species, e.g., by reducing
canopy height and removing stems (habitat facilitation), or when grazing by one
species stimulates grass regrowth, thereby enhancing the nutritional quality for
another species (feeding facilitation; Arsenault and Owen-Smith 2002). Although
these mechanisms are not mutually exclusive, habitat facilitation has been the most
frequently observed in the field. For example, grazing hippopotamus (Hippopotamus
amphibius) can transform tall grasslands into extensive grazing lawns (Owen-
Smith 1988) which may be essential to grazing mesoherbivores such as the western
kob (Kobus kob kob) in Cameroon (Verweij et al. 2006). The same hippo, however,
may exert competitive effects on grazers which prefer tall grass. Elimination of
hippos from a part of Queen Elisabeth National Park, Uganda, was followed by a
substantial increase in elephant, buffalo, and waterbuck. Following the recovery of
the hippos, the numbers of these other species declined again in the region
(Arsenault and Owen-Smith 2002). Population responses are thus possible but to
what extent both competition and facilitation within natural ungulate communities
have shaped the composition of the regional species pool, or are confined only to
local effects, is not known (Prins and Olff 1998).
As described above, it is not surprising that the effects of changing herbivore
numbers or the introduction of domestic livestock on habitat modification and habitat
fragmentation can have far-reaching consequences for the functioning and structure
of ecosystems, including effects on native herbivore species and small mammal
communities. This is especially apparent when the density of very large species
changes. For example, the effects of crowding by elephants in many parts of Africa,
can lead to either open parkland or shrubland and positively affect grazers such as
oryx (Oryx gazella) and zebra (Equus burchelli), while browsers such as the lesser
kudu (Tragelaphus imberbis) and gerenuk (Litocranius walleri) decline (Arsenault
and Owen-Smith 2002). Where graziers with their livestock have encroached on
national grasslands, native large herbivore communities have suffered tremendous
losses, both in species richness and numbers. Different mechanisms have been
responsible for this decline but two important ones stand out: a direct effect through
overhunting and a more indirect effect through competition by livestock for food.
Numbers of bison on the Great Plains in North America were estimated at between
30 and 60 million before the 1800s. Overhunting then reduced them to a mere few
thousand animals by the early twentieth century (Flores 1991). Competition with
livestock is generally through modification and combined use of the food base and it
is to be expected that grazer species will be more affected than browsers, especially
if the species are of similar size. Mishra et al. (2002) studied ungulate assemblages in
the Indian Trans-Himalaya where potentially seven species of wild ungulates had to
co-exist with five species of domestic livestock. A number of combinations of two
species have shown great overlap in size, feeding habits, while frequently one is the
domestic form of the also-present wild type (Table 10.4). It was found that at least
four species of the original assemblage of wild species were missing from the region
and this could be attributed to competitive exclusion by the similar domestic species.
Mishra et al. (2001) also found evidence that the (Trans)-Himalaya is severely over-
stocked, a condition which has been frequenly described for many other parts in the
world (Wilson and Macleod 1991; du Toit and Cummings 1999).
Despite the many competitive interactions, facilitation of native species by
introduced livestock has also been reported. Interestingly, these were mainly
between a larger domestic grazer (notably cattle) and smaller medium sized grazers
such as red deer (cattle – red deer Cervus elaphus, Gordon 1988; cattle – wapiti
Cervus elaphus, cattle – red deer and wild boar Sus scrofa, Kuiters et al. 2005). We
can hypothesize that the large domestic grazer here plays the role of a similar wild
Table 10.4 Pair-wise weight ratios for most similar pairs (smallest
ratios) of wild and domestic grazers in the Trans-Himalaya (from:
Mishra et al. 2002)
Species pairs Weight ratio
1 Tibetan Argali – Donkey 1.03
2 Chiru – Goat 1.08
3 Kiang – Domestic Yak 1.08
4 Chiru – Sheep 1.09
5 Kiang – Horse 1.11
6 Ibex – Donkey 1.18
7 Yak – Domestic Yak 1.39
8 Kiang – Cow 1.44
9 Bharal – Sheep 1.57
10 Bharal – Goat 1.62
11 Bharal – Donkey 1.64
12 Yak – Horse 1.67
13 Yak – Dzomo 1.82
14 Yak – Cow 2.16
species originally present in the system, notably the bison in North America and the
extinct aurochs in western Europe. Facilitative effects are most apparent at low to
moderate grazing intensity, but are quickly replaced by competitive interactions as
densities of livestock increase (McCullough 1999).
Small mammals. Large herbivores can affect small mammals by changing vege-
tation structure and composition thereby affecting the living conditions for this
group of species. When large grazers enhance the structure and quality of the grass
sward, small herbivores such as rabbits (Oosterveld 1983) and hare (Kuijper 2005)
can be facilitated. At higher densities other lagomorph species such as the European
brown hare (Lepus europaeus; Frylestam 1976) and the Basin pigmy rabbit
(Brachylagus idahoensis; Siegel et al. 2004) become negatively affected as a result
of decreased grass cover, direct disturbance, and destruction of burrows.
Apart from the food base, rodents are particularly sensitive to cover (Leirs et al.
1996; Bowland and Perrin 1989), because lower cover levels increases predation
risk (Goheen et al. 2004). The effects of large herbivores on rodents is not always
clear. Both positive effects (Keesing and Crawford 2001; Jones and Longland 1999)
and negative effects (Keesing 1998, 2000; Jones and Longland 1999) have been
reported; the effects being dependent on the type of habitat and species. Granivorous
species seem to be more negatively affected than herbivorous species such as the
field mouse (Microtus agrestis) at low densities (Beever and Brussard 2004),
although Steen et al. (2005) could not detect an effect of sheep grazing on the bank
vole (Clethrionomys glareolus). High grazing intensities generally have negative
effects on both species richness and abundance of rodent communities (Steen et al.
2005; Goheen et al. 2004). Although very few studies have been conducted where
rodent response has been estimated at various grazing intensities, there is some
evidence that species richness and abundance is higher at low grazing intensity than
when grazing is completely absent (Steen et al. 2005; Kerley 1992; Smit et al.
2001). At low grazing intensity suitable habitat conditions may be created consist-
ing of alternating short, grazed patches, with higher, ungrazed patches.
Large herbivores in woodlands. Not many studies have been conducted on the
effects of large herbivores on mammals in forests (Côté et al. 2004). If, as is
assumed here, densities of large herbivores in natural forest ecosystems are
relatively low, their impacts on other mammals will most likely be modest. Most
forests, however, are heavily modified, to the detriment of many forest mammal
species (Lessard et al. 2005; Côté et al. 2004). In most European forests large her-
bivore species richness has decreased, like that of the large predator community. At
the same time numbers of a few deer species have exploded throughout the northern
hemisphere as the result of an enormous increase in the early successional stages
represented by commercial timber production (Côté et al. 2004). Some late
successional species such as the woodland caribou (Rangifer tarandus caribou)
have decreased as result of changes in the forest. Negative effects of high deer
numbers on abundance, diversity, and species richness have been reported for small
mammals (Moser and Witmer 2000), while McShea and Rappole (2000) found
higher rodent abundance in areas from which deer were excluded. As in grasslands,
heavy grazing generally has had a negative effect on small mammals in the UK’s
New Forest (Putman et al. 1989). Introduced exotic herbivores can have large
negative effects on the native ones. In Argentina, many areas formerly occupied by
huemul (Hippocamelus bisulcus) are now occupied by red deer (Povilitis 1998).
Feral pigs have modified entire communities and ecosystems around the world
through their digging and rooting activities (Mack and D’Antonio 1998). If forests
become increasingly more open or are converted into grassland, it can be expected
that this will be accompanied by a general decrease in mammal species richness
because a majority of mammals are related to wooded habitats (Table 10.3).
Effects on mammalian predators. Large herbivores are important prey for a suite
of large predators and, as such, they contribute to mammalian diversity. A number
of the smaller predators are carrion feeders (fox, badger, jackal) and many of them
feed on invertebrates living on large herbivore dung or carcasses. Large changes in
large herbivore abundance can thus affect the predator guild as well. As shown
above, heavy grazing and browsing have generally negative effects on bird and
small mammal species composition and it can be expected that these effects will
cascade through the food chain. Heavy grazing and browsing reduced the diversity
and abundance of rodents of the New Forest and this had a significant effect on the
foraging behaviour, diet, population density, and breeding success of the fox
(Vulpes vulpes) and the badger (Meles meles), as on a number of avian predators
(Tubbs and Tubbs 1985; Putman 1986). Conversely, a cessation of grazing had the
opposite effect (Petty and Avery 1990).
If large herbivore densities increase substantially, they may affect other species
of large herbivores through apparent competition. This occurs when the presence of
multiple non-competing prey species elevates predator abundance above levels
maintained by single prey species, which increases predation pressure on multiple
prey assemblages (Morin 1999). A good example of this is the increase of moose
in the northern forests of North America which led to an increase in wolves, which
also prey on woodland caribou. The increased wolf population has now a much
greater impact on the more vulnerable caribou, because of a much greater abun-
dance of the alternative prey species, the moose, threatening the caribou population
with extinction (Lessard et al. 2005). Introduced herbivores may have similar
effects when an increase in native predator abundance, because of the increased
prey basis, leads to increased predation on native prey species (Vazquez 2002).
10.7 Large Herbivores and Biodiversity
Synthesis. When evaluating the overall effects of large herbivores on diversity, we

can conveniently come back to the theoretical model of Milchunas et al. (1988;
Fig. 10.5) because this model is still valid (see above and the recent review by
Cingolani et al. 2005). In fact, the model can also be extended to include life
forms other than the plants for which it was developed; the model seems to hold
for birds and mammals as well as for some groups of invertebrates. When looking
carefully at the results, it seems, however, that for most animal groups the hump
lies a bit more to the left than for plants, and that for animals maximum diversity
is found at a lower level of herbivory. This latter has also been suggested for
semi-natural grasslands (Vickery and Herbert 2001; Luoto et al. 2003). The
model can also be generalized with respect to its X-axis which originally is a
moisture gradient but can be indicated as a nutrient or primary productivity gradi-
ent (Olff et al. 2002). Therefore, we can conclude that in most systems there will
be an optimum level of herbivory whereby biodiversity will be maximal. This
level will not be the same in all systems because, apart from abiotic factors, the
length of the evolutionary history the system has experienced with large herbiv-
ores is of great significance. Systems with a long grazing and browsing history
are much more adapted to herbivory than systems with only a short history con-
taining more grazing-tolerant species and species which exhibit compensatory
growth (Adler et al. 2004). Biodiversity in these systems changes less with vary-
ing herbivore densities while those with only a short history are more vulnerable
to change (Fig. 10.5).
Problems. The biodiversity in the world would be highest when each system is
managed to be positioned on top of its hump (see Fig. 10.5). However, from the
foregoing it may be clear that this generally is not the case and in reality systems
can be found on either side of the curve, although too high a grazing pressure is
much more common than too low. In some systems, notably in western Europe and
the Mediterranean, much of the agricultural land is being abandoned with generally
negative consequences for biodiversity (Verdu et al. 2000; Newton 1998). It has to
be noted, though, that these evaluations are always made with respect to only a
selected subset (e.g., dung beetles, flowering plants, meadow birds) and it can not
be ruled out that total biodiversity would actually increase when these systems
revert back to the original state, especially in that ancient woodlands and forests
were very species rich (Niemala 1997).
The real problems are that in the greater part of the world manipulation of both
the species composition and the abundance of the large herbivores (both native and
domesticated) has led to high, sometimes very high, densities. Although it has been
stated that livestock have done more damage than all chainsaws and bulldozers
combined (Noss 1994), it has to be stressed that other man-related activities have
contributed (e.g., clearing, lumbering, mowing, burning, fertilizing), especially in
the semi-arid systems (De Pietri 1992). Although the negative effects of heavy
grazing pressure on biodiversity have been highlighted, there are many other factors
that play a part, e.g., decreasing secondary production, soil erosion, increasing sur-
face water runoff, a decrease in landscape appreciation and hence a lower recrea-
tional value, and an increase in invasive species (Heitschmidt et al. 2004).
Before trying to formulate some solutions, one particular and nagging problem
has to be discussed. In the past few decades, more and more systems are suffering
from an increase in grazing-tolerant shrubs which can be either exotics or native
species. This can happen when grazing pressure is lowered (Manier and Hobbs
2006; Coughenour 1991; Verdú et al. 2000) but more frequently when herbivore
pressure has become very high (Veblen et al. 1989; Archer et al. 1987; Vazquez and
Simberloff 2004; George and Bazzaz 1999; Raffaele and Veblen 2001; Weber et al.
1998; Sternberg et al. 2000). The shrub problem is thus a worldwide phenomenon
and it has been suggested that a strong increase in shrub development only happens
when grazing exceeds a threshold level (Weber et al. 1998). Attempts to counteract
these developments have generally failed, and this has led to the notion that systems
may have irreversibly changed, the return to the previous (‘natural’) systems now
closed because these systems have moved to another stable state. This finding
underscores recent theoretical insights that many systems are not characterized by
one single climax state but can produce, as the result of self-organizing and positive
feedback processes, an array of stable states, each of which can be highly resilient
and resistant to change (Peterson et al. 1999; Fig. 10.7). In this sense large herbiv-
ores can also act as switches able to move systems towards alternative successional
pathways which can be stable. Such non-linear dynamics have been described in
rangeland pastures (Laycock 1991; Lockwood and Lockwood 1993), savanna-
woodland systems (Dublin 1995; Rietkerk et al. 2002), and temperate and boreal
forests (Pastor et al. 1993). If this assertion is correct then it will be hard indeed to
reverse developments causing decreased biodiversity.
Possible solutions for an overstocked world. This much is clear: the major road
to improvement of most systems is a substantial decrease in large herbivore
pressure. This is true not only when biodiversity is the goal but also when the aim
is for development of a more sustainable rangeland agriculture system (Heitschmidt
et al. 2004). In systems where production, in some form, is an important objective,
more management activities are acceptable than in systems where restoration of
biodiversity is the main focus (national parks, reserves). In calculating a more
a
Plant abundance
Fig. 10.7 Three hypothetical relationships between the

abundance of a forage plant and deer browsing pressure.
(a) Deer have only modest and monotonic effects on the
c population. (b) A reversible threshold exists beyond which
plant abundance drops precipitously. (c) Browsing beyond a
certain threshold point causes a nonlinear decline that is not
simply reversible. The plant population requires a large (or
prolonged) reduction in browsing as well as a disturbance factor
that promotes an increase of its abundance to recover. This
requirement indicates an ‘alternate stable state’. Arrows indicate
Browsing pressure dynamic changes at various points (from Côté et al. 2005)
optimal herbivore density in production systems, reference could be made to the

(evolutionary) grazing history of the system, sensu Milchunas et al. (1988; Fig. 10.5).
Systems with a long history of herbivory can sustain higher densities and a greater
variation in density than those with only a short history. As we know that the
systems with a long history of grazing may have (had) high densities of large
herbivores, it is puzzling that even those systems can become overstocked and
degraded. It is suggested here that this may not simply be an effect of too many
animals (with additional supplementary feeding and continuous water supply), but
could also be the result of constant heavy grazing in almost every corner of an area.
There is hardly any part of a grazed area which experiences a temporary relaxation
of grazing and for many species this is simply too much. McNaughton (1984) found
that, in systems with a long history of grazing, there exist two pools of species: one
short and one tall. This suggests that these systems have apparently evolved under
fluctuating densities of large herbivores and this could be a key factor in restoration
management. Thus, apart from a general decrease in numbers, a greater spread of
grazing intensity in time and space would improve biodiversity (Gordon 2006).
This calls for larger areas to be grazed by free ranging herds.
When restoration for conservation is the main goal, various management
strategies are open, mostly aiming at attacking problems to help nature follow its
own course again and thus of a temporal nature. Sometimes a non-intervention
policy can be applied; this may be an option even when a transition phase has to be
accepted where unwanted developments take place. Restoration generally should
not be aimed at zero grazing because there are very few situations where large
herbivores would not naturally be present and thus this would lead to a lower
biodiversity. In western Europe attempts are being made to restore ecosystems by
restoring native communities of large herbivores, referred to as naturalistic grazing,
but the lands set aside for these efforts are said to be too small (see report on
Naturalistic Grazing by Hodder et al. 2005).
Nevertheless, given the earlier-mentioned possibility of irreversible changes, a
reversal to some earlier state may not happen and become impossible. Additional
measures such as culling, burning, bush clearance, or predator control may help
here but there is no guarantee that the problem will thereby permanently be solved,
and furthermore these measures can be costly and produce unwanted side effects.
It is not possible to create a situation where there will be only winners. But when
the main goal is biodiversity, the best option seems to be herbivore densities that
are in close agreement with the historical herbivore density under which the system
evolved, with the lowest possible burden of additional management measures.
References
Adler LS, Karban R, Strauss SY (2001) Direct and indirect effects of alkaloids on plant fitness via
herbivory and pollination. Ecology 82:2032–2044
Adler P, Milchunas D, Lauenroth W, Sala O, Burke I (2004) Functional traits of graminoids in
semi-arid steppes: a test of grazing histories. J Appl Ecol 41:653–663
Andresen H, Bakker JP, Brongers M, Heydemann B, Irmler U (1990) Long-term changes of salt
marsh communities by cattle grazing. Vegetatio 89:137–148
Archer S, Garrett MG, Detling JK (1987) Rates of vegetation change associated with prairie dog
(Cynomys ludovicianus) grazing in North American mixed-grass prairie. Vegetatio 72:159–166
Arsenault R, Owen-Smith N (2002) Facilitation versus competition in grazing herbivore assem-
blages. Oikos 97:313–318
Bakker JP (1989) Nature management by grazing and cutting. On the ecological significance of
grazing and cutting regimes applied to restore former species-rich grassland communities in
the Netherlands. Kluwer, Dordrecht
Bakker JP (1998) The impact of grazing to plant communities. In: Wallis de Vries MF, Bakker JP,
Van Wieren SE (eds) Grazing and conservation management. Kluwer, Dordrecht, pp
137–184
Bakker JP, De Leeuw J, Van Wieren SE (1984) Micro-patterns in grassland vegetation created and
sustained by sheep-grazing. Vegetatio 55:153–161
Bakker JP, Esselink P, Van der Wal R, Dijkema KS (1997) Options for restoration and manage-
ment of coastal salt marshes in Europe. In: Urbanska KM, Webb NR, Edwards PJ (eds)
Restoration ecology and sustainable development. Cambridge Univ Press, Cambridge, pp
286–322
Bakker ES, Olff H, Vandenberghe C, De Maeyer K, Smit R, Gleichman JM, Vera FWM (2004)
Ecological anachronisms in the recruitment of temperate light-demanding tree species in
wooded pastures. J Appl Ecol 41:571–582
Barker AM, Brown NJ, Reynolds CJM (1999) Do host-plant requirements and mortality from soil
cultivation determine the distribution of graminivorous sawflies on farmland? J Appl Ecol
36:271–282
Beever EA, Brussard PF (2004) Community and landscape-level responses of reptiles and small
mammals to feral-horse grazing in the Great Basin. J Arid Environ 59:271–297
Beintema AJ, Müskens GJDM (1987) Nesting success of birds in a Dutch agricultural grassland.
Bos D, Loonen M, Stock M, Hofeditz F, Van Der Graaf S, Bakker JP (2005) Utilisation of Wadden
Sea salt marshes by geese in relation to livestock grazing. J Nat Conserv 15:1–15
Bowland AE, Perrin MR (1989) The effect of overgrazing on the small mammals in Umfolozi
Game Reserve. Z Saeugetierkd 54:251–260
Brua RB (1999) Ruddy duck nesting success: do nest characteristics deter nest predation? Condor
101: 867–870
Bullock JM, Franklin J, Stevenson MJ, Silvertown S, Coulson SJ, Gregory SJ, Tofts R (2001) A
plant trait analysis of responses to grazing in a long-term experiment. J Appl Ecol
38:253–267
Chamberlain DE, Fuller RJ, Bunce RGH, Duckworth JC, Shrubb M (2000) Changes in the abun-
dance of farmland birds in relation to the timing of agricultural intensification in England and
Wales. J Appl Ecol 37:771–788
Cingolani AM, Noy-Meir I, Dýaz S (2005) Grazing effects on rangeland diversity: a synthesis of
contemporary models. Ecol Appl 15:757–773
Côté S, Rooney TP, Tremblay JP, Dussault C, Waller DM (2004) Ecological impacts of deer over-
abundance. Annu Rev Ecol Evol Syst 35:113–147
Coughenour MB (1985) Graminoid responses to grazing by large herbivores: adaptations,
exaptations and interacting processes. Ann Mo Bot Gard 72:852–863
Coughenour MB (1991) Spatial components of plant–herbivore interactions in pastoral, ranching,
and native ungulate ecosystems. J Range Manage 44:530–542
De Pietri J (1992) Alien shrubs in a national park: can they help in the recovery of natural
degraded forest? Biol Conserv 62:127–130
Desrochers RE, Anand M (2004) From traditional diversity indices to taxonomic diversity indices.
Int J Ecol Environ Sci 30:93–99
Di Giulio M, Edwards PJ, Meister E (2001) Enhancing insect diversity in agricultural grasslands:
the roles of management and landscape structure. J Appl Ecol 38:310–319
Dublin HT (1995) Vegetation dynamics in the Serengeti–Mara ecosystem: the role of elephants,
fire and other factors. In: Sinclair ARE, Arcese P (eds) Serengeti II: dynamics, management
and conservation of an ecosystem. Univ of Chicago Press, Chicago, pp 71–90
Duncan AJ, Hartley SE, Thurlow M, Young S, Staines BW (2001) Clonal variation in monoter-
pene concentrations in Sitka spruce (Picea sitchensis) saplings and its effect on their suscepti-
bility to browsing damage by red deer (Cervus elaphus). Forest Ecol Manage 148:259–269
du Toit JT, Cummings DHM (1999) Functional significance of ungulate diversity in African savan-
nas and the ecological implications of the spread of pastoralists. Biodiv Conserv 8:1643–1661
Dyksterhuis EJ (1949) Condition and management of rangeland based on quantitative ecology.
Flores D (1991) Bison ecology and bison diplomacy: the southern plains from 1800 to 1850. J Am
Hist 78:465–485
Frank DA (2005) The interactive effects of grazing ungulates and aboveground production on
grassland diversity. Oecologia 143:629–634
Frylestam B (1976) Effects of cattle grazing and harvesting of hay on density and distribution of
an European hare population. In: Pielowski Z, Pucek Z (eds) Ecology and management of
European hare populations. Polish Hunting Assoc, Warsaw, pp 199–203
Fuller RJ (2001) Responses of woodland birds to increasing numbers of deer: a review of evidence
and mechanisms. Forestry 74:289–298
Fuller RJ, Gill RMA (2001) Ecological impacts of increasing numbers of deer in British
woodland. Forestry 74:193–199
Gates S, Donald PF (2000) Local extinction of British farmland birds and the prediction of further
loss. J Appl Ecol 37:806–820
George LO, Bazzaz FA (1999) The fern understory as an ecological filter: emergence and estab-
lishment of canopy-tree seedlings. Ecology 80:833–845
Goheen JR, Keesing F, Allan BF, Ogada DL, Ostfeld RS (2004) Net effects of large mammals on
Acacia seedling survival in an African savanna. Ecology 85:1555–1561
Gordon IJ (1988) Facilitation of red deer grazing by cattle and its impact on red deer performance.
Gordon IJ (2006) Restoring the function of grazed ecosystems. In: Danell K, Bergström R,
Duncan P, Pastor J, Olff H (eds) Large herbivore ecology and ecosystem dynamics. Cambridge
Univ Press, Cambridge, pp 449–467
Grant MC, Orsman C, Easton J et al (1999) Breeding success and causes of breeding failure of
curlew Numenius arquata in Northern Ireland. J Appl Ecol 36:59–74
Grime JP (1973) Control of species diversity in herbaceous vegetation. J Environ Manage
1:151–167
Heinken T, Raudnitschka D (2000) Do wild ungulates contribute to the dispersal of vascular plants
in central European forests by epizoochory? Forstwiss Centralbl 121:179–194
Heitschmidt RK, Vermeire LT, Grings EE (2004) Is rangeland agriculture sustainable? J Anim Sci
82(E Suppl):138–146
Hernández Fernández M, Vrba ES (2005) A complete estimate of the phylogenetic relationships
in Ruminantia: a dated species-level supertree of the extant ruminants. Biol Rev 80:269–301
Hodder KH, Bullock JM, Buckland, P, Kirby KJ (2005) Large herbivores in the wildwood and
modern naturalistic grazing systems. English Nature Res Reports Nr 648, Peterborough, UK
Holland EA, Detling JK (1990) Plant response to herbivory and belowground nitrogen cycling.
Ecology 71:1040–1049
Hunt LP (2001) Heterogeneous grazing causes local extinction of edible perennial shrubs: a
matrix analysis. J Appl Ecol 38:238–252
Hunter MD, Price PW (1992) Playing chutes and ladders: heterogeneity and the relative roles of
bottom-up and top-down forces in natural communities. Ecology 73:724–732
Hutchinson GE (1959) Homage to Santa Rosalia, or why are there so many kinds of animals? Am
Nat 93:145–159
Hutchinson KJ, King KL (1980) The effect of sheep stocking level on invertebrate abundance,
biomass and energy utilization in a temperate sown grassland. J Appl Ecol 17:369–387
Jaindl RG, Doescher P, Miller RF, Eddleman LE (1994) Persistence of Idaho fescue on degraded
rangelands: adaptation to defoliation or tolerance. J Range Manage 47:54–59
Jones AL, Longland WS (1999) Effects of cattle grazing on salt desert rodent communities. Am
Midl Nat 141:1–11
Kantrud H (1981) Grazing intensity effects on the breeding avifauna of North Dakota native
grasslands. Can Field Nat 95:404–417
Keesing F (1998) Ecology and behaviour of the pouched mouse, Saccostomus mearnsi, in central
Kenya. J Mammal 79:919–931
Keesing F 2000 Cryptic consumers and the ecology of an african savanna. BioScience
50:205–215
Keesing F, Crawford T (2001) Impacts of density and large mammals on space use by the pouched
mouse (Saccostomus mearnsi) in central Kenya. J Trop Ecol 17:465–472
Kerley GIH (1992) Ecological correlates of small mammal community structure in the semi-arid
Karoo, South Africa. J Zool 227: 17–27
Kirby RE, Ringelman JK, Anderson DA, Sodja RS (1992) Grazing on National Wildlife Refuges:
do the needs outweigh the problems? Trans N A Wildl Nat Res Conf 57:611–626
Kiviniemi K (1996) A study of adhesive seed dispersal of three species under natural conditions.
Acta Bot Neerl 45: 73–83
Klett AT, Shaffer TL, Johnson DH (1988) Duck nest success in the meadow pothole region.
J Wildl Manage 52:431–440
Kramer K, Groen TA, Van Wieren SE (2003) The interacting effects of ungulates and fire on forest
dynamics: an analysis using the model FORSPACE. Forest Ecol Manage 181:205–222
Kruess A, Tscharntke T (2002) Contrasting responses of plant diversity to variation in grazing
intensity. Biol Conserv 106:293–302
Kuijper DPJ (2005) Small herbivores losing control – plant–herbivore interactions along a natural
productivity gradient. Thesis, Univ of Groningen
Kuijper DPJ, Bakker JP (2005) Top-down control of small herbivores on salt-marsh vegetation
along a productivity gradient. Ecology 86:914–923
Kuiters AT, Groot B, Geert WTA, Lammertsma DR (2005) Facilitative and competitive interac-
tions between sympatric cattle, red deer and wild boar in Dutch woodland pastures. Acta
Theriol 50:241–252
Laycock WA (1991) Stable states and thresholds of range condition on North American range-
lands: a viewpoint. J Range Manage 44:427–433
Leirs H, Verheyen W, Verhagen R (1996) Spatial patterns in Mastomys natalensis in Tanzania
(Rodentia, Muridae). Mammalia 60:545–555
Lessard RB, Martell S, Walters CJ, Essington TE, Kitchell JFK (2005) Should ecosystem
management involve active control of species abundances? Ecol Soc 10:1–23
Lindroth RL (1988) Adaptations of mammalian herbivores to plant chemical defenses. In: Spencer
KC (ed) Chemical mediation of coevolution. Academic Press, San Diego, pp 425–445
Lockwood JA, Lockwood DR (1993) Catastrophe theory: a unified paradigm for rangeland eco-
system dynamics. J Range Manag 46:282–88
Loreti J, Oesterheld M, Sala O (2001) Lack of intraspecific variation in resistance to defoliation
in a grass that evolved under light grazing pressure. Plant Ecol 157:195–202
Luoto M, Pykälä J, Kuussaari M (2003) Decline of landscape-scale habitat and species diversity
after the end of cattle grazing. J Nat Conserv 11:171–178
Mack MC, D’Antonio CM (1998) Impacts of biological invasions on disturbance regimes. Trends
Manier DJ, Hobbs NT (2006) Large herbivores influence the composition and diversity of shrub-
steppe communities in the Rocky Mountains, USA. Oecologia 146:641–651
Mayfield HA (1965) The brown-headed cowbird with old and new hosts. Liv Bird 4:13–28
McCullough DR (1999) Density dependence and life-history strategies of ungulates. J Mammal
80:1130–1146
McInnes PF, Naiman RJ, Pastor J, Cohen Y (1992) Effects of moose browsing on vegetation and
litter of the boreal forest, Isle Royale, Michigan, USA. Ecology 73:2059–2075
McIntyre S, Martin TG (2001) Biophysical and human influences on plant species richness in
grasslands – comparing variegated landscapes in sub-tropical and temperate regions. Austral
Ecol 26:233–245
McIntyre S, Heard KM, Martin TG (2003) The relative importance of cattle grazing in subtropical
grasslands: does it reduce or enhance plant biodiversity? J Appl Ecol 40:445–457
McNaughton SJ (1984) Grazing lawns: animals in herds, plant form, and coevolution. Am Nat
124:863–886
McShea WJ, Rappole JH 2000 Managing the abundance and diversity of breeding bird populations
through manipulation of deer populations. Conserv Biol 14:1161–70
McShea WJ, Underwood HB, Rappole JH (eds; 1997) The science of overabundance: deer ecol-
ogy and population management. Smithson Inst Press, Washington, DC
Milchunas DG, Lauenroth WK (1993) Quantitative effects of grazing on vegetation and soils over
a global range of environments. Ecol Monogr 63:327–366
Milchunas DG, Lauenroth WK, Sala OE (1988) A generalized model of the effects of grazing by
Mishra C, Prins HHT, Van Wieren SE (2001) Overstocking in the Trans-Himalayan rangelands of
India. Environ Conserv 28:279–283
Mishra C, Van Wieren SE, Heitkönig IMA, Prins HHT (2002) A theoretical analysis of competi-
tive exclusion in a Trans-Himalayan large-herbivore assemblage. Anim Conserv 5:251–258
Mitchell FJG (2005) How open were European primeval forests? Hypothesis testing using palae-
oecological data. J Ecol 93:168–177
Miyashita T, Takada M, Shimazaki A (2004) Indirect effects of herbivory by deer reduce abun-
dance and species richness of web spiders. Ecoscience 11:74–79
Morin PJ (1999) Community ecology. Blackwell, Malden, MA, USA
Moser BW, Witmer GW (2000) The effects of elk and cattle foraging on the vegetation, birds, and
small mammals of the Bridge Creek Wildlife Area, Oregon. Int Biodeter Biodegr 45:151–157
Murray MG, Illius AW (2000) Vegetation modification and resource competition in ungulates.
Oikos 89:501–508
Naiman RJ, Melilo JM, Hobbie JE (1986) Ecosystem alteration of boreal forest streams by beaver
(Castor canadensis). Ecology 67:1254–1269
Newton I (1998) Bird conservation problems resulting from agricultural intensification in Europe.
Ecol Appl 11:307–322
Niemala J (1997) Invertebrates and boreal forest management. Conserv Biol 11:601–610
Norris K, Brindley E, Cook T, Babbs S, Forster Brown C, Yaxley R (1998) Is the density of
redshank Tringa totanus nesting on saltmarshes in Great Britain declining due to changes in
grazing management? J Appl Ecol 35: 621–634
Noss RF (1994) Cows and conservation biology. Conserv Biol 8:613–616
Obeso JR (1997) The induction of spinescence in European holly leaves by browsing ungulates.
Plant Ecol 129:149–156
Olff H, Ritchie ME, Prins HHT (2002) Global environmental controls of diversity in large
herbivores. Nature 415: 901–904
Oosterveld P (1983) Eight years of monitoring of rabbits and vegetation development on aban-
doned arable fields grazed by ponies. Acta Zool Fenn 174:71–74
Cambridge Univ Press, Cambridge
Paine RT (1966) Food web complexity and species diversity. Am Nat 100:850–860
Painter EL, Detling JK, Steingraeber DA (1993) Plant morphology and grazing history:
relationships between native grasses and herbivores. Vegetatio 106:37–62
Pastor J, Dewey B, Naiman RJ, McInnes PF, Cohen Y (1993) Moose browsing and soil fertility
in the boreal forest of Isle Royale National Park. Ecology 74:467–480
Pastor J, Moen R, Cohen Y (1997) Spatial heterogeneities, carrying capacity, and feedbacks in
animal–landscape interactions. J Mammal 78:1040–1052
Persson I-L, Danell K, Bergstrom R (2000) Disturbance by large herbivores in boreal forests with
special reference to moose. Ann Zool Fenn 37:251–263
Peterson GD, Allen CR, Holling CS (1999) Ecological resilience, biodiversity and scale.
Ecosystems 1:6–18
Pétillon J, Ysnel F, Canard A, Lefeuvre JC (2005) Impact of an invasive plant (Elymus athericus)
on the conservation value of tidal salt marshes in western France and implications for manage-
ment: responses of spider populations. Biol Conserv 126:103–117
Petty SJ, Avery MJ (1990) Forest bird communities. Occasional paper 26. Forestry Commission,
Edinburgh
Pielou EC (1975) Ecological diversity. Wiley, New York
Povilitis A (1998) Characteristics and conservation of a fragmented population of huemul
Hippocamelus bisulcus in central Chile. Biol Conserv 86:97–104
Prins HHT (1998) Origins and development of grassland communities in northwestern Europe. In:
Wallis DeVries MF, Bakker JP, Van Wieren SE (eds) Grazing and conservation management.
Kluwer, Boston, pp 55–106
Prins HHT, Olff H (1998) Species-richness of African grazer assemblages: towards a functional
explanation. In: Newberry DM, Prins HHT, Brown N (eds) Dynamics of tropical communities.
Blackwell, Oxford, pp 449–490
Putman RJ (1986) Competition and coexistence in a multispecies grazing system. Acta Theriol
31:271–291
Putman RJ, Edwards PJ, Mann JCE, How RC, Hill SD (1989) Vegetational and faunal changes in
an area of heavily grazed woodland following relief of grazing. Biol Conserv 47:13–32
Raffaele E, Veblen TT (2001) Effects of cattle grazing on early regeneration of matorral in north-
west Patagonia, Argentina. Nat Area J 21:243–249
Ricotta C (2005) Through the jungle of biological diversity. Acta Biotheor 53:29–38
Rietkerk M, Van Boerlijst MC, Van Langevelde F et al (2002) Self-organization of vegetation in
arid ecosystems. Am Nat 160:524–530
Rosenthal J, Kotanen PM (1994) Terrestrial plant tolerance to herbivory. Trends Ecol Evol
9:117–157
Sarkar S, Margules C 2002 Operationalizing biodiversity for conservation planning. J Bioscience
27(S2):299–308
Shannon CE, Weaver W (1949) The mathematical theory of communication. Univ of Illinois
Press, Urbana, IL
Sharaf KE, Price MV (2004) Does pollination limit tolerance to browsing in Ipomopsis aggregata?
Simpson EH (1949) Measurement of diversity. Nature 163:688
Smit R, Bokdam J, den Ouden J, Olff H, Schot-Opschoor H, Schrijvers M (2001) Effects of intro-
duction and exclusion of large herbivores on small rodent communities. Plant Ecol
155:119–127
Söderström B, Pärt T, Linnarsson E (2001) Grazing effects on between-year variation of farmland
bird communities. Ecol Appl 11:1141–1150
Steen H, Mysterud A, Austrheim G (2005) Sheep grazing and rodent populations: evidence of
negative interactions from a landscape scale experiment. Oecologia 143:357–364
Steets JA, Ashman TL (2004) Herbivory alters the expression of a mixed-mating system. Am J
Bot 91:1046–1051
Sternberg M, Gutman M, Perevolotsky A, Ungar ED, Kigrl J (2000) Vegetation response to graz-
ing management in a Mediterranean herbaceous community: a functional group approach. J
Appl Ecol 37:224–237
Stewart AJA (2001) The impact of deer on lowland woodland invertebrates: a review of the evi-
dence and priorities for future research. Forestry 74:259–70
Stock M, Hofeditz F (2000) Der Einfluss des Salzwiesen-Managements auf die Nutzung des
Habitates durch Nonnen- und Ringelgänse. In: Stock M, Kiehl K (eds) Die Salzwiesen der
Hamburger Hallig. Landesamt Nationalpark Schleswig-Holsteinisches Wattenmeer, Tönning,
DE, pp 43–55
Stowe KA, Marquis RJ, Hochwender CG, Simms EL (2000) The evolutionary ecology of toler-
ance to consumer damage. Annu Rev Ecol Syst 31:565–595
Strauss SY, Agrawal AA (1999) The ecology and evolution of plant tolerance to herbivory. Trends
Sudgen EA (1985) Pollinators of Astragalus monoensis Berneby (Fabaceae): new host records;
potential impact of sheep grazing. Great Basin Nat 45:299–312
Suominen O, Danell, K, Bergström R (1999) Moose, trees, and ground-living invertebrates:
indirect interactions in Swedish pine forest. Oikos 84:215–226
Swengel AB, Swengel SR (2001) Effects of prairie and barrens management on butterfly faunal
composition. Biodivers Conserv 10:1757–1785
Tallis JH (1991) Plant community history. Long-term changes in plant distribution and diversity.
Chapman and Hall, London
Tiffney BH (2004) Vertebrate dispersal of seed plants through time. Annu Rev Ecol Evol S 35:1–29
Tubbs CR, Tubbs JM (1985) Buzzards, Buteo buteo, and land use in the New Forest, Hampshire,
England. Biol Conserv 31:41–65
Van der Graaf AJ, Bos D, Loonen MJJE, Engelmoer M, Drent RH (2002) Short-term and long-
term facilitation of goose grazing by livestock in the Dutch Wadden Sea area. J Coastal
Conserv 8:179–188
Van der Wal R, Van Wijnen H, Van Wieren SE, Beucher O, Bos D (2000) On facilitation between
herbivores: how Brent Geese profit from Brown Hares. Ecology 81:969–980
Vane-Wright RI, Humphries CJ, Williams PM (1991) What to protect: systematics and the agony
of choice. Biol Conserv 55:235–254
Van Wieren SE 1998 Effects of large herbivores upon the animal community. In: Wallis DeVries
MF, Bakker JP, Van Wieren SE (eds) Grazing and conservation management. Kluwer, Boston,
pp 185–214
Vazquez DP (2002) Multiple effects of introduced mammalian herbivores in a temperate forest.
Biol Invasions 4:175–191
Vazquez DP, Simberloff D (2004) Indirect effects of an introduced ungulate on pollination and
plant reproduction. Ecol Monogr 74:281–308
Veblen TT, Mermoz M, Martýn C, Ramilo E (1989) Effects of exotic deer on forest regeneration
and composition in northern Patagonia. J Appl Ecol 26:711–724
Vera FWM (2000) Grazing ecology and forest history. CABI International, Wallingford, UK
Verdú JR, Crespo MB, Galante E (2000) Conservation strategy of a nature reserve in Mediterranean
ecosystems: the effects of protection from grazing on biodiversity. Biodivers Conserv
9:1707–1721
Verweij RJT, Verrelst J, Loth PE, Heitkönig IMA, Brunsting AMH (2006) Grazing lawns
contribute to the subsistence of mesoherbivores on dystrophic savannas. Oikos 114:108–116
Vickery PD, Herkert JR (2001) Recent advances in grassland bird research: where do we go from
here? Auk 118:11–15
Weber GE, Jeltsch F, Van Rooyen N, Milton SJ (1998) Simulated long-term vegetation response
to grazing heterogeneity in semi-arid rangelands. J Appl Ecol 35:687–699
Westoby M, Walker B, Noy-Meir I (1989) Opportunistic management for rangelands not at
equilibrium. J Range Manage 42:266–274
Whittaker RH (1972) Evolution and measurement of species diversity. Taxon 21:213–251
Wilsey BJ, Coleman JS, McNaughton SJ (1997) Effects of elevated CO2 and defoliation on
grasses: a comparative ecosystem approach. Ecol Appl 7:844–853
Wilson AD, Macleod ND (1991) Overgrazing—present or absent. J Range Manage
44:475–482
Wilson GWT, Hartnett DC (2001) Effects of ungulate grazers on arbuscular mycorrhizal fungal
communities in tallgrass prairie.Annu Rev Ecol Evol S 35:435–466
Young TP, Stanton ML, Christian CE (2003) Effects of natural and simulated herbivory on spine
lengths of Acacia drepanolobium in Kenya. Oikos 101:171–179
Zalba SM, Cozzani NC 2004 The impact of feral horses on grassland bird communities in
Argentina. Anim Conserv 7:35–44
Chapter 11
Managing Large Herbivores in Theory
and Practice: Is the Game the Same
for Browsing and Grazing Species?
Jean-Michel Gaillard, Patrick Duncan, Sip E. Van Wieren, Anne Loison,

François Klein, and Daniel Maillard
11.1 Introduction
Management of large wild herbivores sometimes includes action on the availability

of resources and/or on the abundance of predators, but the management of most
populations of these animals is limited to the choice of hunting quotas (Williams et al.
2002). The dynamics of the target populations, in interaction with their habitats, is
therefore a central process in management. The dynamics of populations is driven
by their demographic parameters, i.e., the rates of reproduction and survival.
Demographic rates vary according to the sex and age of the individuals, and also
show different degrees of temporal and spatial variation among populations
(Gaillard et al. 2000; Coulson et al. 2001). In this chapter we first address the
question: do browsing and grazing large herbivores have different demographic
rates driving the dynamics of their populations?
Management of most populations of large herbivores being, as stated, based on
the choice of quotas for offtake by hunting, a reliable assessment of trends in popu-
lation size is required. Total counts of individuals in the whole population area or
sample counts of plots/transects have been the most popular way for assessing
population sizes (e.g., Caughley 1977; Seber 1982; for deer see Cederlund et al.
1998 and for mountain ungulates Loison et al. 2006). In the context of game species
such as these, hunting-related methods such as drives (McCullough 1979), observa-
tions from vantage points (Ratcliffe 1987), and number of animals seen during
hunts (Ericsson and Wallin 1999) are often used. However, such counts are often
imprecise, and biased (Williams et al. 2002). Caughley (1977) pointed out that it is
difficult to achieve a coefficient of variation of less than 20% for counts of animals.
The roe deer (Capreolus capreolus), the most abundant deer species in Europe
(Andersen et al. 1998), is particularly difficult to count and underestimates of
density can be as high as 50% (Cederlund et al. 1998). Since grazers and browsers
live in different habitats (Jarman 1974), they pose different problems in estimation
of trends in abundance, due in particular to differences in visibility between habitats
and to differences in variance because of contrasting group sizes.
Another problem in the management of large herbivores is that the response of
populations to a given offtake varies according to the relative importance of
© Springer 2008
294 J.-M. Gaillard et al.
density-dependent and density-independent processes acting on the population

(Caughley 1977). Populations of animals in good condition (because resources are
abundant) respond differently from populations of animals in poor condition to the
same offtake (Caughley 1977). Monitoring populations of large herbivores should,
therefore, be rooted in the concept of density dependence, the functional depend-
ence of a demographic rate on changes in population size (Williams et al. 2002).
This provides a measure of the population–environment relationships, as illustrated
in Fig. 11.1. Density dependence is a commonly observed process in populations
of large herbivores (Saether 1997; Gaillard et al. 2000). It is widely accepted that
populations of large herbivores vary in numbers according to a generalized logistic
model (also called theta logistic model; Lande et al. 2003) that includes non-linear
density-dependent changes in population growth (Gilpin and Ayala 1973; Fowler
1981). At very low density, the population shows a ‘colonising’ status and the
growth rate is close to maximal (r-max sensu Caughley 1977). The population’s
productivity is, however, quite low because of the small size of the population. As
the size of the population increases, small decreases in individual performance lead
the growth rate of the population to decrease slowly, while the population produc-
tivity increases. When the population size reaches 50–90% (depending on the gen-
eration time, Fowler 1988) of the carrying capacity (defined here as the population
size at which the population no longer increases), the population productivity peaks
(MSY—maximum sustainable yield—in Caughley 1977). After this point, the
1.0
0.9
0.8
0.7
Relative value
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Time (years)
Fig. 11.1 Growth, individual growth and size of a theoretical population according to a theta
logistic model. The black full line corresponds to changes of the population size over time, the
grey full line corresponds to the variation of the population productivity according to the population
size, and the dotted line corresponds to the variation of the individual performance according to
the population size
11 Managing Large Herbivores in Theory and Practice 295
performance of the individuals decreases markedly with declining body condition

at high population density, and the productivity of the population declines, too. The
demographic status of the population is then close to saturation. The pattern
described here can happen at very different absolute densities in different areas
because population density relative to habitat quality, rather than to absolute den-
sity, influences life-history traits (Van Horne 1983).
Populations of large herbivores have increased tremendously in the temperate
countries (Gill 1990) and high-density populations are now widespread (McShea
et al. 1997; Warren 1997). Their management, therefore, requires knowledge of the
relationships between the animals and their resources as well as an assessment of
the size of the population.
We show that a set of ecological indicators of the temporal changes in abun-
dance and of the relationships between the population and the habitat can be a reli-
able basis for monitoring. This approach has been developed independently in
France for the roe deer, and for other large herbivores in other countries (du Toit
2002). We address the issue of whether the same methods of monitoring and man-
agement are appropriate for grazers and browsers in Sects. 11.3 Monitoring, and
11.4 Management.
11.2 The Dynamics of Grazer and Browser Populations
Several recent papers have described differences in biological traits among feeding
types in large herbivores. From these, Gordon (2003) concluded there is little sub-
stantive evidence for differences in morphology and physiology between feeding
types once body mass has been accounted for. However, most previous comparative
studies of grazers and browsers have dealt with morphology and physiology, and
very few have focussed on life-history traits (but see Saether and Gordon 1994 for
a notable exception). According to the life history theory (Stearns 1992), life his-
tory tactics of vertebrates range along a fast-slow continuum involving the associa-
tion between relatively large reproductive output, short lifespan, and early age at
first reproduction (see Stearns 1983 on mammals, Gaillard et al. 1989 on mammals
and birds, Shine and Charnov 1992 on reptiles, Rochet et al. 2000 on fishes). We
aim to test whether grazers and browsers have different positions on this fast-slow
continuum using published demographic data. We focussed on the reproductive
traits (age at first reproduction and litter size), adult survival, and on generation
time, which latter provides a direct measure of the speediness of life history
(Gaillard et al. 2005). We did not include juvenile survival because juvenile sur-
vival varies greatly among both years within populations and populations within a
given species (see Gaillard et al. 1998a, 2000 for reviews), so that a large number
of possible confounding effects would need to be accounted for to get a reliable
comparison among the survival of juveniles of the different feeding types. Although
the influence of phylogeny has been discussed in the context of comparisons
between grazers and browsers (see, e.g., Perez-Barberia and Gordon 2000 versus
Mysterud 1998), we did not analyse phylogenetic influences. The usefulness of

correcting for phylogenetic inertia has been challenged (e.g., Price 1997; Ricklefs
and Starck 1996), leading many authors to perform both a standard analysis using
the raw data and a comparative analysis accounting for phylogenetic inertia (inde-
pendent contrasts in most cases; see Fisher and Owens 2000; Nagy and Bradshaw
2000; Perez-Barberia and Gordon 2001 for examples). As the correction for phylo-
genetic inertia increases the p-values of the statistical tests, but does not affect the
parameter estimates, we will consider the size of the effects more than the signifi-
cance of the statistical tests. To test for life-history differences among feeding
types, we split the species of large herbivores into three categories of feeding types:
browsers with a diet including less than 10% grass, intermediate feeders with a diet
including between 10 and 90% grass, and grazers with a diet of more than 90% of
grass; in agreement with most recent analyses focussing on feeding types (e.g., Van
Wieren 1996; Perez-Barberia and Gordon 2000). We used available data bases
(restricted to ruminant species) to assess the influence of feeding types on life his-
tory traits. To account for the marked allometric relationships that occur in most
life history traits (Peters 1983; Calder 1984; Brown and West 2000), we performed
an ANCOVA with the log-transformed life history trait as the dependent variable,
the species-specific body mass (after log-transformation) as the covariate, and the
feeding type as the factor. We ran five models in each case: (1) the model including
an interaction term between feeding types and body mass that corresponds to dif-
ferent allometric relationships among feeding types; (2) the model including an
additive effect between feeding types and body mass that corresponds to a similar
allometric relationship, but with differences in the life history trait at a given mass
among feeding types; (3) the model including the allometric relationship only (i.e.,
no influence of feeding types); (4) the model including differences among feeding
types only (i.e., no influence of body mass); and (5) the model including a constant
value of the life history trait (i.e., no influence of feeding types or body mass).
According to the recommendations of Burnham and Anderson (1998), we used the
model with the lowest Akaiké Information Criterion (AIC) as the best model (best
compromise between precision and accuracy).
Differences in reproductive traits among feeding types. We used the species-
specific data available in Van Wieren (1996) to assess the influence of feeding types
on age at first breeding (first recorded matings, N = 140 species) and on fecundity
(N = 141), principally bovids and cervids from a wide variety of habitats in Africa,
Eurasia, and the Americas. For age at first breeding, the best model included addi-
tive effects of body mass and feeding types (Table 11.1). As expected, the age at
first breeding was allometrically related to body mass, although the slope was much
less than the expected value of 0.25 for biological time (slope 0.147, SE = 0.019).
For a given body mass browsers bred for the first time earlier than intermediate
feeders (difference 0.269, SE = 0.058) and grazers (difference 0.349, SE = 0.066).
The additive effects of body mass and feeding types accounted for 61.8% of the
variability observed in age at first breeding. For the fecundity (measured as the
number of offspring produced per year), the same model was retained (Table 11.1).
As expected, fecundity decreased allometrically with increasing body mass
Table 11.1 Model selection for the analysis of the variation in some life history traits. The table
reports the AIC value of each model fitted (1: Interactive effects between feeding type and body
mass, 2: additive effects of feeding type and body mass, 3: effect of body mass (allometric effect),
4: effect of feeding type, 5: constant value for the life history trait). The selected model (lowest
AIC value) occurs in italics
Model Life history trait 1 2 3 4 5
Age at first breeding 30.61 27.63 53.99 74.79 156.41
Fecundity 90.43 86.57 88.86 111.73 146.45
Adult survival 49.58 45.84 44.27 43.86 42.27
Generation time 11.28 7.98 8.33 23.36 20.18
although the slope differed from −0.25, the expected allometric exponent for fre-
quency (slope −0.130, SE =0.024). For a given body mass browsers tended to have
higher fecundity than intermediate feeders (difference 0.095, SE = 0.072) and had
higher fecundity than grazers (difference 0.202, SE = 0.081). The additive effects of
body mass and feeding types accounted for 37.5% of the variability observed in
annual fecundity.
Differences in adult survival among feeding types. We used the species-specific
data available in Gaillard et al. (2000) to assess the influence of feeding types on
adult survival (N = 24 species). The model with a constant adult survival (2.42,
SE = 0.11 on a logit scale corresponding to a survival rate of 0.92) was selected
(Table 11.1). The additive effects of body mass and feeding types only accounted
for 9.63% of the variability observed in adult survival.
Differences in generation time among feeding types. We used the data in Sinclair
(1996) and Gaillard et al. (2000) to assess the influence of feeding types on generation
time (N = 26 species). The best model included additive effects of body mass and feed-
ing types (Table 11.1). As expected, the generation time was allometrically related to
body mass, with a slope very close to the expected value of 0.25 for biological times
(slope 0.227, SE = 0.050). For a given body mass browsers had shorter generation time
than intermediate feeders (difference 0.251, SE = 0.132) and grazers (difference 0.223,
SE = 0.134). The additive effects of body mass and feeding types accounted for 50.4%
of the variability observed in generation time (Fig. 11.2).
These analyses show that browsers have faster life histories for a given body
mass than intermediate feeders and grazers. A faster life history can involve an
earlier age at first breeding, a higher fecundity rate, and/or a shorter lifespan: we
show that browsers have faster life histories than grazers and intermediate feeders
because they breed earlier and produce more offspring. Thus, a 100 kg browser
breeds for the first time at 1.15 years of age and produces an average of 1.37 off-
spring/year, whereas a grazer of the same mass breeds for the first time at 1.62
years of age and produce an average of 1.12 offspring/year (see Table 11.2). As
expected from scaling theory (Brown and West 2000), the reproductive output of
large herbivores decreased with increasing body mass at similar rates in all feeding
types (Table 11.2). Adult survival of large herbivores, on the other hand, was quite
constant at 0.92 and did not depend either on body mass or on feeding type. Such
a constancy of adult survival fits with previous analyses of large herbivores that
2.6
2.4
2.2
Ln (Generation Time)
2.0
1.8
1.6
1.4
1.2
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
Ln (Body mass )
Fig. 11.2 Allometric relationship between generation time and body mass in ruminants in relation
to the feeding type. The lines (full line for browsers, dashed line for intermediate feeders, and
dotted line for grazers) that correspond to the selected model (i.e., additive effects of body mass
and feeding types) and the data points (filled circles for browsers, open squares for intermediate
feeders, and open circles for grazers) are shown
Table 11.2 Expected age at first breeding (AFB, in years) and annual fecundity (Fec, in number of
offspring born) for large herbivores of different body mass and belonging to different feeding types.
The estimates correspond to the models selected in each analysis (see text for further details)
Browsers Intermediate feeders Grazers
AFB Fec AFB Fec AFB Fec
25 kg 0.934 1.636 1.222 1.488 1.324 1.337
50 kg 1.034 1.495 1.353 1.360 1.466 1.222
100 kg 1.145 1.367 1.498 1.243 1.623 1.117
500 kg 1.451 1.109 1.898 1.008 2.056 0.906
reported high survival rates resilient to environmental variations (Gaillard et al.

2000; Gaillard and Yoccoz 2003). The shorter generation time of browsers
compared to intermediate feeders and grazers is thus due only to a higher reproduc-
tive output. While this shorter generation time we report for browsers increases the
relative influence of reproductive traits on population growth rate (Gaillard et al.
2005), our analyses demonstrate the existence of a single general demographic
pattern common to all large herbivores. Indeed, a 2 kg browser is the fastest large
herbivore, in terms of its demography, that can exist. For such an animal, the
expected generation time estimated from our selected model is 2.57 years.
According to the direct mathematical link that relates elasticities of demographic
parameters (i.e., the relative potential impact of demographic parameters on
population growth) and generation time (see Lebreton and Clobert 1991, Gaillard
et al. 2005), any population with a generation time longer than 2 years will be more
sensitive to a given change in adult survival than to the same change in recruitment.
This means that the population growth of all large herbivores, independently of size
and feeding type, must be more sensitive to a given change in adult survival than
to the same change in a reproductive rate.
The physiological mechanisms behind this difference in generation time remain
to be elucidated: the meta-analyses that have so far been conducted do not prove
that the digestive ability of browsers is greater than grazers and intermediate feeders,
and it is not established that selective feeding by browsers on plants with high levels
of cell contents provides them with richer diets (higher protein contents and/or
lower cell-wall constituents; Duncan, A. and Poppi, this volume). However, the
quality of the data is not good enough to rule out these possible mechanisms.
The life-history differences between browsers and other feeding categories we
report here are likely to have general consequences for the status of populations
of large herbivores in the future, in the context of global change. The marked
increases in CO2 concentrations (IPCC 2001) may favour woody plants within
savannas (Bond and Midgley 2000), so browsers may increase relative to large her-
bivores with other feeding types in this habitat (see Janis et al. 2000 and Gordon
and Prins this volume). However, climate change will not affect the CO2
concentrations alone, and may decrease the area of savannas, so that the loss of
favourable habitat may overcompensate the benefit of increased CO2 concentra-
tions. On the other hand, the increase of woodland areas in western Europe should
benefit browsers. The success of roe deer (Andersen et al. 1998) is consistent with
this view. With a faster turnover, browsers should be more resilient to overhunting
than grazers and intermediate feeders. A comparative analysis of hunting bags over
a large geographical scale in relation to feeding types would be required to test for
such a prediction. Lastly, the faster turnover of browsers as compared to large her-
bivores of other feeding types may confer browsers faster speciation rates. Vrba
(1987) has identified some links between ecological features and speciation rates
that could indicate that such differences between browsers and grazers could occur;
this prediction, however, requires testing explicitly.
11.3 Monitoring
The relationships between the animals and their habitats should be monitored on
the basis of ‘indicators of ecological change’ (sensu Cederlund et al. 1998). As
defined by Waller and Alverson (1997), these are ‘efficient and reliable indicators
capable of serving as “early warnings signs” of impending ecological change’ (see
also Cairns et al. 1993). Such indicators include measures of trends in abundance
of the condition of the animals and of the quality of their habitats. Dale and Beyeler
(2001) have reviewed all criteria that an ecological indicator should meet: ‘be easily
measured, be sensitive to stresses on the system, respond to stress in a predictable
manner, be anticipatory, predict changes that can be averted by management

actions, be integrative, have a known response to disturbances, anthropogenic
stresses, and changes over time, and have low variability in response’.
As no single indicator fulfils all these criteria, a manageably small set of indica-
tors needs to be derived that, together, meet these criteria (Dale and Beyeler, 2001).
Moreover, because we cannot possibly monitor every component of an ecological
system, we are constrained to identify a small number of indicators that characterise
the relevant parts of the system.
Monitoring the abundance of grazers and browsers. The concepts outlined
above are equally valid for grazers and browsers. However, large herbivores of
different feeding types live in different habitats (Hofmann 1989; Fritz and Loison
2006), where visibility is different. Grazers typically live in open areas that have
less variable, and overall greater, visibility than the often quite closed areas such as
forests where browsers live. This means that undercounting might be less of a prob-
lem, and bias error might be lower in counts of grazers as compared to browsers.
However, this does not mean that total counts of grazers are reliable. For instance,
both the number of ibex Capra ibex (Gaillard et al. 2003) and chamois Rupicapra
rupicapra (Loison et al. 2006) counted in open mountainous areas, and the number
of African ungulates counted from the air in savanna (Jachmann 2002) have
been reported to be severely underestimated. However, the rate of underestimation
when counting grazers should be less than that when counting browsers. For
instance, an underestimation by a factor of 3 or more has been commonly reported
for roe deer (Andersen 1953).
As a corollary of differences in habitat openness, group sizes also differ among
feeding types (Jarman 1974), with browsers living in smaller groups than grazers
(Fritz and Loison 2006). As the variance of abundance estimate increases with
increasing group size (Burnham et al. 1980), the precision of counts, all other
things being equal, should be better for browsers than for grazers.
As visibility and group size differ markedly among feeding types and strongly
influence the sampling design of population counts, the methods used to monitor
trends in the abundance of browsers and grazers should be different.
For browsers like roe deer, the Kilometric Index (measured as the number of
animals observed per km of foot transect; Vincent et al. 1991) is closely correlated
with true changes in the abundance of deer (as assessed by using Capture-Mark-
Recapture methods; see Gaillard et al. 1993) in two forests where the exploitation
of trees for timber caused considerable changes in visibility over 20 years. The
passage of hurricane Lothar in late 1999 led to the breaking or uprooting of a third
of the trees in these forests: there may have been an effect of lower visibility on the
index in the three years after the hurricane, but this did not cause any bias in
the long term. Visibility that varies spatially is not a major problem, provided
that the mean value remains reasonably constant over time. This approach is now
widely used for browsers (Maillard et al. 1999). If measures of absolute densities
are required, ‘distance’ methods are insensitive to changes in visibility of the
animals (Burnham et al. 1980), as the distances at which they are sighted are taken
into account.
For grazers, total or sample counts (most often aerial counts; Jachmann 2001)
are usually performed. Although they can provide data for detailed demographic
analyses (Owen-Smith and Mason 2005; Owen-Smith and Mills 2006), the
accuracy of these procedures has not yet been assessed. However, available
empirical evidence indicates that size (larger is better) and colour (high contrast
between coat colour and ground colour is better) influence the reliability of aerial
counts of large herbivores (Redfern et al. 2002; Jachmann 2002).
Monitoring the condition of grazers and browsers. The phenotypic quality of
individuals varies strongly with population density (Fowler 1987), and hence
offers a potential tool to monitor the population status of large herbivores. For
instance, the body mass of fawns in winter (Vincent et al. 1995; Gaillard et al.
1996), the cohort jaw length (Hewison et al. 1996), and the hind foot length of
fawns (Toïgo et al. 2006; Zanneïse et al. 2007) have all been reported to decrease
markedly with increasing population density of roe deer and are now used as
indicators of population changes in a management context (e.g., Blant and
Gaillard 2004 in Switzerland). The winter body mass of fawns is widely used as
a short-term measure, since this is related to events (such as rainfall) in the previ-
ous summer, and the amount of acorn mast in autumn (Kjellander et al. 2006).
The length of the hind foot of fawns integrates events over a longer time-scale,
and is very convenient to measure. For longer time horizons the jaw length is a
useful index of the quality of individuals (Hewison et al. 1996). Resource
shortage in early life leads to permanent effects on the size of many mammals,
including red deer Cervus elaphus (Mysterud et al. 2002) and roe deer (Pettorelli
et al. 2002). This causes variation in performance among year classes and thus to
delayed density-dependent effects on the dynamics of their populations (‘cohort
variation’; Albon et al. 1987).
The reproductive performance of individuals in the population can be moni-
tored by the number of offspring per female (Vincent et al. 1995) or as the number
of fawns per reproducing female (i.e., females with fawns at heel; Boutin et al.
1987). However, for polytocous species such as roe deer, the pattern of variation
of reproductive performance as assessed by sampling females can be blurred by
differential family effects in relation to cohort quality, since in poor cohorts fawns
born in the same family commonly have similar fates (i.e., both die or both survive
in most cases), whereas there is no family effect during good years (Gaillard et al.
1998b). Further work is therefore required to assess the suitability of such
indicators for management.
The abundance of resources, and the level of their use, can be measured easily
on woody plants (Aldous 1944). However, there are strong biases due to differences
among observers in their estimations of plant cover (Morellet 1998). The browsing
index has, therefore, been developed (Morellet et al. 2001); this is calculated for
each plant species using sample plots of 1 m2 as the proportion of plots in which
that species was browsed. Data on the presence/absence of plant species and of
signs of browsing on them in quadrants of 1 m2 have been found to be much more
repeatable, and therefore better indices of the abundance and use of resources
(Morellet et al. 2001; Morellet et al. 2003).
11.4 Management
For the management of populations of large herbivores to succeed, it needs to be

based on (Nichols et al. 1995; Cederlund et al. 1998):
– an assessment of the current state of the animal–habitat relationships based on
monitoring ecological changes
– an analysis of the range of possible alternatives to population counts, based on the
system’s current state, on ecological principles (such as density dependence; see
Fig. 11.1), and on the benefits and costs of the possible states to human society
– a choice of management actions by the different stake-holders based on a vision
of the optimal future state of the animal-habitat relationships, taking into account
uncertainty as to the consequences of management actions in complex systems
Such management is ‘adaptive’; its essence is the use of a wide range of indices
of animal abundance and condition, and ideally of the abundance of resources and the
impact of the animals on resources and on human activities. Managers learn about
systems as they manage them (Lancia et al. 1996). Adaptive management (Walters
1986) deals with scientific uncertainty by incorporating a set of models representing
the competing hypotheses about system responses to management (Runge and
Johnson 2002). With adaptive management the information on the system response
to management is gathered continuously so that the information is used to
improve biological understanding and to inform future decision-making (Nichols et al.
1995; Shea et al. 1998; Williams et al. 2002). Variations on the theme of adaptive
management are used in many wildlife management areas (du Toit 2002).
The two essential decisions in the management of populations of large herbiv-
ores concern (1) the objective and (2) the quota. The objective is often decided
simply on the basis of the objectives of hunters, but should be based on a wide
range of considerations of the benefits and costs of high/low densities of large her-
bivores for society as a whole (taking into account ethical, social and economic
dimensions) as proposed in Cederlund et al. (1998).
Having decided on the objective, which could be to increase the population,
decrease it, or maintain the current size, the quota needs to be decided. Given that
the approach based on indicators of ecological changes proposed above do not pro-
vide information on population sizes, the quota will initially have to be chosen on
the basis of expert opinion, and it will be necessary to take into account the size of
the target species and its feeding type. The potential population productivity of
small and large grazers and browsers are given in Fig. 11.3; intermediate feeders
will have intermediate productivities. As expected, it appears that the population
productivity is higher for a given relative population size of small than of large
species and of browsers than of grazers. Small browsers are thus expected to be
much less sensitive to hunting and/or poaching impact than large grazers.
The quota should then be applied for three years or more, without change, in
order to get estimates of the ecological indicators for a given hunting pressure, so
that the success of the management can be judged from the monitoring. In the initial
0.12
0.10
25 kg Grazer
Population productivity
500 kg Grazer
0.08
25 kg Browser
500 kg Browser
0.06
0.04
0.02
0.00
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Relative Population Size
Fig. 11.3 Population productivity (measured as the population increase over two consecutive
years) as a function of population size for small (25 kg) and large (500 kg) browser and grazer. To
get estimates of population productivities, we used a Leslie demographic model parameterized
with the demographic parameters expected for 25 and 500 kg browsers and grazers (as derived
from the allometric relationships presented in the text). We thus obtained estimates of population
growth (r=0.324 for a 25 kg browser, 0.275 for a 25 kg grazer, 0.275 for a 500 kg browser, and
0.181 for a 500 kg grazer). Then we used the Fowler’s equation relating the relative population
size at which the maximum sustainable yield occurs in a population (R), its population growth and
its generation time (Fowler 1988). We obtained very similar R values among the four case studies
(from 0.536 for a large grazer to 0,560 for a small browser). We thus used a constant R of 0.55.
We then calculated the population productivity as a function of N, r, and R by using the theta
logistic function (Lande et al. 2003)
stages, the management of populations of large herbivores cannot be more accurate

than a trial-and-error process dictates, however, as managers gain knowledge of the
population–environment system over the years, management will become more
finely tuned. Such an approach is thus based on long-term monitoring: the longer
the system is monitored, the better it is understood.
11.5 Conclusions
The main source of variation in the life history of large herbivores is the body size
of the animal. However, the feeding type, i.e., grazing or browsing, also influences
the turnover of the population, being faster for browsers and slower for grazers, but
the relatively more important role of adult survival as compared to recruitment
parameters on the population growth rate of large herbivores does not change.
Feeding type strongly affects the use of habitat and the size of the groups the
animals live in. This means that different methods need to be used for the monitoring
and management of grazers and browsers. Management should be adaptive, based
on the biology of the animals, and appropriate monitoring of ecological and
socio-economic changes in the animal–habitat relationships.
References
Albon SD, Clutton-Brock TH, Guinness FE (1987) Early development and population dynamics
in Red Deer. II. Density-independent effects and cohort variation. J Anim Ecol 56:69–81
Aldous SE (1944) A deer browse survey method. J Mammal 25:130–136
Andersen J (1953) Analysis of a Danish roe-deer population (Capreolus capreolus L) based upon
the extermination of the total stock. Dan Rev Game Biol 2:127–155
Andersen R, Duncan P, Linnell JDC (1998) The European roe deer: the biology of success.
Scandinavian Univ Press, Oslo
Blant M, Gaillard JM (2004) Use of biometric body variables as indicators of roe deer (Capreolus
capreolus) population density changes. Game Wildl Sci 21:21–40
Bond WJ, Midgley GF (2000) A proposed CO2-controlled mechanism of woody plant invasion
in grasslands and savannas. Glob Change Biol 6:865–869
Boutin JM, Gaillard JM, Delorme D, Van Laere G (1987) Suivi de l’évolution de la fécondité chez
le chevreuil (Capreolus capreolus) par l’observation des groupes familiaux. Gibier Faune
Sauvage 4:255–265
Brown JH, West GB (2000) Scaling in biology. Oxford Univ Press, New York
Burnham KP, Anderson DR (1998) Model selection and inference: a practical information-theoretic
approach. Springer, Berlin Heidelberg New York
Burnham KP, Anderson DR, Laake JL (1980) Estimation of density from line transect sampling
of biological populations. Wildlife Monogr No 72
Cairns J, McCormick PV, Niederlehner BR (1993) A proposed framework for developing indica-
tors of ecosystem health. Hydrobiologia 236:1–44
Calder WA (1984) Size, function and life history. Harvard Univ Press, Cambridge, MA
Caughley G (1977) Analysis of vertebrate populations. Wiley, Chichester, NY
Cederlund G, Bergqvist J, Kjellander P, Gill R, Gaillard J-M, Boisaubert B, Ballon P, Duncan P
(1998) Managing roe deer and their impact on the environment: maximising the net benefits
to society. In: Andersen R, Duncan P, Linnell JDC (eds) The European roe deer: the biology
of success. Scandinavian Univ Press, Oslo, pp 337–372
Coulson T, Catchpole EA, Albon SD, Morgan BJT, Pemberton JM, Clutton-Brock TH, Crawley
MJ, Grenfell BT (2001) Age, sex, density, winter weather, and population crashes in Soay
sheep. Science 292:1528–1531
Dale VH, Beyeler SC (2001) Challenges in the development and use of ecological indicators. Ecol
Indic 1:3–10
Duncan AJ, Poppi DP, this volume
du Toit JT (2002) Wildlife harvesting guidelines for community-based wildlife management: a
southern African perspective. Biodivers Conserv 11:1403–1416
Ericsson G, Wallin K (1999) Hunter observations as an index of moose Alces alces population
parameters. Wildlife Biol 5:177–185
Fisher DA, Owens IPF (2000) Female home range size and the evolution of social organization in
macropod marsupials. J Anim Ecol 69:1083–1098
Fowler CW (1981) Density dependence as related to life history strategy. Ecology 62:602–610
Fowler CW (1987) A review of density-dependence in populations of large mammals. In:
Genoways HH (ed) Current mammalogy. Plenum, New York, pp 401–441
Fowler CW (1988) Population dynamics as related to rate of increase per generation. Evol Ecol
2:197–204
Fritz H, Loison A (2006) Large herbivores across biomes. In: Danell K, Duncan P, Bergström R,
Pastor J (eds) Large herbivore ecology, ecosystem dynamics and conservation. Cambridge
Univ Press, Cambridge, pp 19–49
Gaillard J-M, Yoccoz NG (2003) Temporal variation in survival of mammals: a case of environ-
mental canalization? Ecology 84:3294–3306
Gaillard J-M, Pontier D, Allainé D, Lebreton JD, Trouvilliez J, Clobert J (1989) An analysis of
demographic tactics in birds and mammals. Oikos 56:59–76
Gaillard J-M, Delorme D, Boutin JM, Van Laere G, Boisaubert B, Pradel R (1993) Roe deer sur-
vival patterns: a comparative analysis of contrasting populations. J Anim Ecol 62:778–791
Gaillard J-M, Delorme D, Boutin JM, Van Laere G, Boisaubert B (1996) Body mass of roe deer
fawns during winter in two contrasting populations. Journal of Wildlife Management
60:29–36
Gaillard J-M, Festa-Bianchet M, Yoccoz NG (1998a) Population dynamics of large herbivores:
variable recruitment with constant adult survival. Trends Ecol Evol 13:58–63
Gaillard J-M, Andersen R, Delorme D, Linnell JDC (1998b) Family effects on growth and sur-
vival of juvenile roe deer. Ecology 79:2878–2889
Gaillard J-M, Festa-Bianchet M, Yoccoz NG, Loison A, Toïgo C (2000) Temporal variation in
fitness components and population dynamics of large herbivores. Annu Rev Ecol Syst
31:367–393
Gaillard J-M, Loison A, Toïgo C (2003) Variation in life history traits and realistic population
models for wildlife management. In: Festa-Bianchet M, Apollonio M (eds) Animal behavior
and wildlife conservation. Island Press, Washington, DC, pp 115–132
Gaillard J-M, Yoccoz NG, Lebreton JD, Bonenfant C, Devillard S, Loison A, Pontier D, D Allainé
(2005) Generation time: a reliable metric to measure life history variation among mammalian
populations. Am Nat 166:119–123
Gill R (1990) Monitoring the status of European and North American cervids. GEMS information
series global environment monitoring system. UN Environment Programme, Nairobi, Kenya
Gilpin ME, Ayala FJ (1973) Global models of growth and competition. Proc Nat Acad Sci USA,
70:3590–3593
Gordon IJ (2003) Browsing and grazing ruminants: are they different beasts? Forest Ecol Manage
181:13–21
Gordon IJ, Prins HHT, this volume
Hewison AJM, Vincent J-P, Bideau E, Angibault J-M, Putman RJ (1996) Variation in cohort
mandible size as an index of roe deer (Capreolus capreolus) densities and population trends.
J Zool 239:573–581
IPCC (2001) Working Group I third assessment report, climate change 2001: the scientific basis.
http://www.gcrio.org/online.html
Jachmann H (2001) Estimating abundance of African wildlife: an aid to adaptive management.
Springer, Berlin Heidelberg New York
Jachmann H (2002) Comparison of aerial counts with ground counts for large African herbivores.
Janis CM, Damuth J, Theodor JM (2000) Miocene ungulates and terrestrial primary productivity:
where have all the browsers gone? Proc Nat Acad Sci USA 97:7899–7904
Jarman PJ (1974) The social organisation of antelope in relation to their ecology. Behaviour
48:215–265
Kjellander P, Gaillard J-M, Hewison AJM (2006) Density-dependent responses of fawn cohort
body mass in two contrasting roe deer populations. Oecologia 146:521–530
Lancia RA, Braun CE, Collopy MW, Dueser RD, Kie JG, Martinka CJ, Nichols JD, Nudds TD,
Porath WR, Tilghman NG (1996) ARM! For the future: adaptive resource management in the
wildlife profession. Wildlife Soc B 24:436–442
Lande R, Engen S, Saether B-E (2003) Stochastic population dynamics in ecology and conservation.
Oxford Univ Press, Oxford
Lebreton JD, Clobert J (1991) Bird population dynamics, management and conservation: the role
of mathematical modelling. In: Perrins CM, Lebreton JD, Hirons GJM (eds) Bird
population studies: their relevance to conservation and management. Oxford Univ Press, Oxford,
pp 105–125
Loison A, Appolinaire J, Jullien J-M, Dubray D (2006) How reliable are total counts to detect
trends in population size of chamois Rupicapra rupicapra and R. pyrenaica. Wildlife Biol
12:77–88
Maillard D, Gaultier P, Boisaubert B (1999) Revue de l’utilisation des différentes méthodes de
suivi des populations de chevreuils en France. Le bulletin mensuel de l’office national de la
chasse 244:30–37
McCullough DR (1979) The George Reserve deer herd: population ecology of a K-selected spe-
cies. Univ of Michigan Press, Ann Arbor, MI
McShea WJ, Underwood HB, Rappole JH (1997) The science of overabundance: deer ecology
and population management. Smithsonian Institution Press, Washington, DC
Morellet N (1998) Des outils biométriques appliqués aux suivis des populations animales:
l’exemple des cervidés. Thesis, Univ of Lyon
Morellet N, Champely S, Gaillard J-M, Ballon P, Boscardin Y (2001) The browsing index: new
tool uses browsing pressure to monitor deer populations. Wildlife Soc B 29:1243–1252
Morellet N, Ballon P, Boscardin Y, Champely S (2003) A new index to measure roe deer
(Capreolus capreolus) browsing pressure on woody flora. Game Wildlife Sci 20:155–173
Mysterud A (1998) The relative roles of body size and feeding type on activity time of temperate
ruminants. Oecologia 113:442–446
Mysterud A, Langvatn R, Yoccoz NG, Stenseth NC (2002) Large-scale habitat variability,
delayed density effects and red deer populations in Norway. J Anim Ecol 71:569–580
Nagy KA, Bradshaw SD (2000) Scaling of energy and water fluxes in free-living arid-zone
Australian marsupials. J Mammal 81:962–970
Nichols JD, Johnson FA, Williams BK (1995) Managing North American waterfowl in the face
of uncertainty. Annu Rev Ecol Syst 26:177–199
Owen-Smith N, Mason DR (2005) Comparative changes in adult vs. juvenile survival affecting
population trends of African ungulates. J Anim Ecol 74:762–773
Owen-Smith N, Mills MGL (2006) Manifold interactive influences on the population dynamics
of a multispecies ungulate assemblage. Ecol Monogr 76:73–92
Perez-Barberia FJ, Gordon IJ (2000) Differences in body mass and oral morphology between the
sexes in the Artiodactyla: evolutionary relationships with sexual segregation. Ecol Evol Res
2:667–684
Perez-Barberia FJ, Gordon IJ (2001) Relationships between oral morphology and feeding
style in the Ungulata: a phylogenetically controlled evaluation. P Roy Soc Lond B Bio
268:1023–1032
Peters RH (1983) The ecological implication of body size. Cambridge Univ Press, Cambridge
Pettorelli N, Gaillard J-M, Van Laere G, Duncan P, Kjellander P, Liberg O, Delorme D, Maillard
D (2002) Variations in adult body mass in roe deer: the effects of population density at birth
and of habitat quality. P Roy Soc Lond B Bio 269:747–753
Price T (1997) Correlated evolution and independent contrasts. Philos T Roy Soc B
352:519–529
Ratcliffe PR (1987) The management of red deer in the commercial forests of Scotland related to
population dynamics and habitat changes. Thesis, Univ of London
Redfern JV, Viljoen PC, Kruger JM, Getz WM (2002) Biases in estimating population size from
an aerial census: a case study in the Kruger National Park, South Africa. S Afr J Sci
98:455–461
Ricklefs RE, Starck JM (1996) Applications of phylogenetically independent contrasts: a mixed
progress report. Oikos 77:167–172
Rochet MJ, Cornillon PA, Sabatier R, Pontier D (2000) Comparative analysis of phylogenetic and
fishing effects in life history patterns of teleost fishes. Oikos 91:255–270
Runge MC, Johnson FA (2002) The importance of functional form in optimal control solutions of
problems in population dynamics. Ecology 83:1357–1371
Saether B-E (1997) Environmental stochasticity and population dynamics of large herbivores: a
search for mechanisms. Trends Ecol Evol 12:143–149
Saether, B-E and I J Gordon 1994 The adaptive significance of reproductive strategies in ungu-
lates. P Roy Soc Lond B Bio 256:263–268
Seber GAF (1982) The estimation of animal abundance and related parameters, 2nd edn. Griffin,
London
Shea K, NCEAS working group (1998) Management of populations in conservations, harvesting
and control. Trends Ecol Evol 13:371–375
Shine R, Charnov EL (1992) Patterns of survival, growth, and maturation in snakes and lizards.
Am Nat 139:1257–1269
Sinclair ARE (1996) Mammal populations: fluctuation, regulation, life history theory and their
implications for conservation. In: Floyd RB, Sheppard AW, De Barro PJ (eds) Frontiers of
population ecology. CSIRO Publishing, Melbourne, pp 127–154
Stearns SC (1983) The influence of size and phylogeny on patterns of covariation among life his-
tory traits in the mammals. Oikos 41:173–187
Stearns SC (1992) The evolution of life histories. Oxford Univ Press, Oxford
Toïgo C, Gaillard J-M, Van Laere G, Hewison AJM, Morellet N (2006) How does environmental
variation influence body mass, body size and body condition? Roe deer as a case study.
Ecography 29:301–308
Van Horne B (1983) Density as a misleading indicator of habitat quality. J Wildlife Manage
47:893–901
Van Wieren SE (1996) Digestive strategies in ruminants and nonruminants. Thesis, Wageningen
Agricultural Univ, Wageningen, Netherlands
Vincent J-P, Gaillard J-M, Bideau E (1991) Kilometric index as biological indicator for monitor-
ing forest roe deer populations. Acta Theriol 36:315–328
Vincent J-P, Bideau E, Hewison AJM, Angibault JM (1995) The influence of increasing density
on body weight, kid production, home range and winter grouping in roe deer (Capreolus
capreolus). J Zool 236:371–382
Vrba ES (1987) Ecology in relation to speciation rates: some case histories of Miocene–recent
mammal clades. Evol Ecol 1:283–300
Waller DM, Alverson WS (1997) The white-tailed deer: a keystone herbivore. Wildlife Soc B
25:217–226
Walters CJ (1986) Adaptive management of renewable resources. MacMillan, New York
Warren RJ (1997) The challenge of deer overabundance in the 21st century. Wildlife Soc B
25:213–214
Williams BK, Nichols JD, Conroy MJ (2002) Analysis and management of animal populations.
Academic Press, San Diego
Zanneïse A, Baïsse A, Gaillard JM, Hewison AJM, Saint-Hilaire K, Toïgo C, Van Laere G,
Morellet N (2007) Hind foot length: an indicator for monitoring roe deer populations at a
landscape scale. Wildlife Soc B (In press)
Chapter 12
Grazers and Browsers in a Changing
World: Conclusions
Iain J. Gordon and Herbert H.T. Prins
12.1 Introduction
The world on which we live is constantly changing, slowly through geological

and evolutionary change and rapidly as the result of rainfall events or droughts,
storms, hurricanes and cyclones, cold snaps or heat waves, and diurnal changes
in light levels. Some of these changes have only local effects but others have
global effects. In recent times the world has undergone a dramatic global change
in form of the levels of CO2 in the atmosphere, much of it derived from
anthropogenic sources (King 2005). Currently, the level of CO2, 380 ppm, has not
been experienced by the biota on the planet for over 650,000 years (Siegenthaler
et al. 2005). This increased CO2 level is leading to changes in climate with
consequences for rainfall and temperature patterns across the globe (IPCC 2001)
with knock-on effects for vegetation community composition, and plant chemistry.
For example, it is predicted that there will be a shift in the balance from grass to
tree dominance in many savanna systems (Higgins et al. 2000; Bond et al. 2003).
These changes in tree/grass ratios are happening rapidly (e.g., Lewis 2002) and
will provide both opportunities and constraints both for humans in managing
their domestic livestock species and also for wild herbivores. Much of this change
is also brought about because of changes in land use over the last century with
much land being taken out of agriculture, particularly in the temperate regions of
the northern hemisphere (e.g., Ford 1971; Kjekshus 1977; Prins and Gordon,
Chap. 1 this book). Whatever the exact causes or the interaction between them,
an increase in the amount of browse in the landscape clearly will increase the
amount of food available for browsers, whilst a reduction in the amount of grass,
because of competition with woody species, will reduce the amount of food
available for grazers. However, the system responses are unlikely to be that
simple, because of such effects as non-linearity in the system and positive and negative
feedbacks (e.g., Van de Koppel et al. 2002, van Langevelde et al. 2003). As a
consequence systems responses will be difficult to predict at the scale at which
they are managed and managers, therefore, need to be able to manage systems for
unpredictable dynamics; that is, adopting risk averse, adaptive management
principles (see Convention on Biological Diversity 2001). Scientists will have to
© Springer 2008
310 I.J. Gordon and H.H.T. Prins
work with managers to develop modelling frameworks that incorporate both hard
and soft (expert opinion) knowledge. Scientists will also have to use what knowledge
we currently have to make ‘best bet’ predictions as to what the responses of
systems might be to changes in management, and not fall back on the ‘we don’t
yet know enough’ argument to continue to collect data before giving advice.
Putting our money where our mouth is, in this chapter we will outline the likely
biophysical responses of vegetation systems to CO2-induced climate change, in
terms of the quantity and quality of plant material available to grazers and browsers.
We then assess the ways in which browsers and grazer populations are likely to
respond to these changes in food quantity and quality and the consequences for
the large herbivore community structure. Finally we give some advice as to how
browsers and grazers might be managed in changing landscapes with particular
emphasis on the collaboration that has to be built between managers and scientists
to support the development of adaptive systems levels management approaches.
12.2 Responses in Plant Species Composition
Increases in CO2 allow shrubs and trees to grow more quickly, and increase their
water use efficiency (Drake et al. 1997; Ehleringer et al. 1997), relative to grasses,
leading to shrub and tree encroachment into areas that have long been grasslands
(Bond et al. 2003). Elevated CO2 levels may cause higher allocation of carbohydrates
to the roots of woody species. This results in, for instance, larger stores of starch
resulting in faster regrowth after the dormant season (Bond et al. 2003). Because of
this faster growth, woody species have an increased chance to escape a fire trap
(Higgins et al. 2000) where fire can kill saplings before their vulnerable growing
points have reached a height above the flames. Once saplings and young trees or
shrubs are outside this zone, they are not reset by every fire to a prostate post-fire
form again, and can start expanding their crown, thus suppressing grass growth.
This in turn leads to a reduced fire frequency and fire intensity, which feeds back
into expanding woody cover.
Grasslands comprise approximately 30% of the global vegetation communities
(Parton 1995), savannas about 20% (Sankaran et al. 2005), whilst forests cover 30%
(FAO 2005). With changes in the circulating CO2 levels in the atmosphere there will
be changes in the balance of trees and grass in many terrestrial ecosystems leading
to a reduction in the area of the globe covered by grassland and an increase in the
area covered by forest. Two types of systems are most prone to change; the first
are savannas because in these systems woody species are interspersed with grass at
all possible scales, and transition from mixed grass–tree systems to tree-dominated
ones can thus happen over large areas very fast (e.g., Stuart-Hill and Tainton 1999).
Moreover, many savannas appear to have an unstable balance between woody
species and grasses (Sankaran et al. 2005). It is noteworthy that contrary to the
generally believed notion that forests are decreasing in the tropics, they are in fact
spreading again in West and Central Africa. Two wet years in a row are often
12 Grazers and Browsers in a Changing World: Conclusions 311
enough to allow forest margins to expand into an area that was under savanna
before. Only an increased disposal of manpower can keep secondary forests at bay.
The second system type prone to change, is man-made agricultural grassland that
has been converted from original natural forest. Large parts of the eastern United
States, nearly the whole of Europe and also Japan have been cleared of forest in the
service of agriculture. De-intensification of agricultural practices, then, can lead to
the rapid re-emergence of woody species and forest (see Chap. 1, this book).
It has been predicted that the C4 grasses will not respond as effectively as C3
grasses to changes in atmospheric CO2 (Bowes 1993; Ainsworth and Long 2005;
and see Wand et al. 1999). However, the extent to which C3 and C4 grass species
will respond to changes in atmospheric CO2 levels will depend upon the water
availability. Under well-watered conditions, C3 plants show increased photosynthesis
and growth. Well-watered C4 plants exhibited increased photosynthesis in response
to increasing CO2, but total mass and leaf area were unaffected (Ward et al. 1999).
In response to drought, C3 plants drop a large amount of leaf area and maintain
relatively high leaf water potential in the remaining leaves, whereas C4 plants retain
greater leaf area, but at a lower leaf water potential, which suggests that C4 species
may have an advantage over C3 species in response to increasing atmospheric CO2
and more frequent and severe droughts.
12.3 Responses in Plant Chemistry
Plant metabolism is centred around carbon, oxygen, and water. Plants get their
carbon and oxygen primarily from the air; thus the concentrations of carbon and
oxygen in the atmosphere play a major part in the chemistry of plants, most
particularly in the photosynthesis rates. Increased CO2 levels potentially leads to
increased levels of photosynthesis (called the CO2 fertilisation effect) and improve
their water-use efficiency (because plants can restrict stomata, reducing transpiration,
while fixing the same amount of CO2) but the extent to which this potential is
realised depends on the effects of other limiting factors, such as water and nutrients,
including N and P. Therefore, the response of plants to elevated CO2 will depend
on local climatic circumstances, e.g. the degree of aridity and soil nutrient status
(Kimball 1983; Nowak et al. 2004).
Increased CO2 in both grass and browse plants generally leads to increases in
C/N ratios and sugar concentrations, and to decreases in nitrogen and phosphorus
concentrations (Kinney et al. 1997; Coley et al. 2002; Goverde et al. 2004; Hattas
et al. 2005). Thus, whilst nitrogen concentration is reduced, the concentration of
soluble carbohydrates is increased. Other responses are less predictable, with some
authors describing no changes in plant secondary compounds to increases in CO2
(Goverde et al. 2004), whereas others find an increase in concentrations of certain
secondary compounds (tannins and ellagitannins, Kinney et al. 1997; leaf phenols,
Coley et al. 2002) and, depending on the species, even variation in response
(Kinney et al. 1997; Coley et al. 2002).
Since concentrate selectors tend to be energy limited while bulk or roughage

feeders are nitrogen limited (Sinclair 1977; Prins and Beekman 1989; Prins 1996),
the changes in vegetation quality associated with elevated levels of atmospheric
CO2 are likely to lead to improved levels of nutrition for browsers but not for
roughage grazers. All other things being equal, increased overall levels of browser
populations are to be expected. If the ratio of nutrients/secondary compounds
increases, a relative increase in the nutritional value of browse is likely, resulting in
an increase in the nutritional ecology of browsers relative to grazers.
Within the class of roughage grazers there is likely to be a difference in response
between ruminants and hind-gut fermenters; roughage hindgut-fermenting grazers
(equids, white rhinoceros) are perhaps less severely nitrogen limited than ruminant
bulk grazers (many bovids and some cervids). Hindgut-fermenters are often seen as
a sort of evolutionary dead end; but as Janis pointed out in Chapter 2 of this book,
this opinion is not supported by analysis of the fossil record. We believe that within the
class of bulk or roughage grazers, elevated CO2 levels may even tip the balance of
herbivore community composition towards hindgut-fermenters and away from the
large ruminant roughage grazers.
It does not appear there will be any great differences between browse and grass
in their response to the increased CO2, in terms of quality, and so we might expect
similarities in the nutritional responses among browsers and populations of selec-
tive grazers. Yet, in areas where C3 and C4 grasses co-exist, we expect a shift
towards vegetations dominated by C4 grasses. Such a shift counterbalances a
trend towards increased soluble carbohydrates in the plants, making the vegetation
as a whole less attractive to both selective grazers and roughage grazers.
12.4 Responses in Terms of Population Dynamics
With changes in vegetation community structure, that is, the evolving dominance
of ecosystems by shrubs and trees, there may be a shift towards browsers dominating
the majority of herbivore communities. In turn, there is a possibility that browsers
may limit the degree to which shrubs and trees survive in systems, counteracting the
CO2-generated propensity for shrubs and trees to increase. Ultimately, the response
will depend on the degree to which the large herbivore/plant ecosystems are food or
predator limited. To date there is little evidence for large herbivores being pred-
ator limited (Mduma et al. 1999), even where they are confined to islands (Vucetich
and Peterson 2004), and there is ample evidence for a relationship between vegetation
biomass/productivity and herbivore population growth rates/density (Coe et al.
1976; Fritz and Duncan 1994). There is a great deal of evidence that herbivore popu-
lation size is positively correlated with the quantity of vegetation biomass available
(Coe et al. 1976; Fritz and Duncan 1994). There is also much evidence that herbivores
are bottom-up controlled (Drent and Prins 1987).
However, whilst herbivore populations may respond positively to increases in
vegetation productivity and nutritional quality, it is not clear whether this will result
in top-down control of vegetation composition by herbivores (Bond 2005). As such,

it is unlikely that any increase in herbivore population size will have an overall
effect on vegetation composition other than on a local scale (Palmer et al. 2003). It
is most likely, therefore, that vegetation composition will change towards one that
is dominated by trees and shrubs and away from grasses. What effect is this likely
to have on the structure of the large herbivore community?
This will depend on the degree to which herbivores are coupled to the dynamics of
the vegetation. Where the coupling is strong (e.g., in temperate systems) the potential
exists for herbivores to control vegetation structure and composition. Arid and semi-arid
grazing systems, however, are prone to the effects of highly variable rainfall, with
droughts causing frequent episodic mortality in herbivore populations. This has led to
the suggestion that these are actually nonequilibrium or decoupled systems, in which
animal impacts on plants are strongly attenuated or absent (Behnke and Scoones;
1992; Scoones 1994). Illius and O’Connor (1999) argued that even in arid and
semi-arid systems animal numbers are regulated in a density-dependent manner by the
limited forage available in key resource areas, which are utilized in the dry season, but
this re-emphasizes the point that herbivores and vegetation are decoupled (not linked
through density dependent effects) in the biggest part of the population’s range.
12.5 Reponses in Herbivore Community Structure
So, worldwide we expect forests to expand in those areas where people do not
actively suppress its regrowth. We thus predict that the wetter parts of the tropics
(with more than 600 mm annual rainfall: Sankaran et al. 2005), and the wetter parts
of the temperate zone (with more than about 400 mm/yr rainfall, which demarcates
the natural steppe formation) will experience this tendency towards increased forest
cover. Browsers will benefit from this trend; mixed feeders may benefit too, roughage
grazers will not.
It appears easier to make predictions about trends in the types of food than about
the chemical composition of the different food stuffs for the different classes of
herbivores. We predict that for browsing concentrate selectors, food quality will
increase. We also predict that in areas that are dominated by C4 grasses (steppes
and tropical savannas), roughage grazers will not benefit from elevated CO2 levels.
Because they are nitrogen-limited; for this class of grazers the situation will
become even less favourable. This will also apply to grazers in C3-grass-dominated
vegetation (in montane tropical areas and mesic temperate areas). It might be that
selective grazers, such as hares or oribi, will be favoured by increased CO2 levels
because of elevated soluble carbohydrates. However, in the contact zones where
C4-grass-dominated vegetation borders on C3-grass-dominated vegetation, we
predict an expansion of C4 grasses to the detriment of all grazing species that are
adapted to using C3 grasses (see Table 12.1).
It makes sense to suggest that selective grazing species of the Ruminantia
are energy-limited as are their browsing counterparts. Not many herbivores with a
314
Table 12.1 Predicted changes in the numerical abundance of classes of mammalian large herbivores in response to increased carbon dioxide levels. Two
effects are foreseen, namely changes in food quality and changes in the balance between woody species and grasses. Browsers in grassland may appear a
contradiction in terms, but browsers may be living on herbs or make use of small pockets of woody species in a generally grass-dominated biome
Species typical for tropical Species typical for Transition from C4 to C3 Species typical for Species typical for temperate
open forests tropical grasslands grasslands temperate grasslands forests
Roughage grazers Rare, and will become rarer Dominant, but Rare (since the beginning Rare (since the Extinct (European wild
will decrease of the Holocene), and beginning of ‘forest’ horse and aurochs
will further decrease the Holocene), are extinct). Remaining
but ‘southern species’ and will further species are rare, and will
could invade decrease become rarer
Selective grazers Rare, but could increase Common, but Rare, but could increase Will increase Absent
could increase and ‘southern species’
could invade
Mixed grazers Common and can further Common, but Rare, but could increase Will increase Will increase
increase could increase and ‘southern species’
could invade
Browsers Common and will further Rare, but could Rare, but could increase Will increase Will increase
increase increase and ‘southern species’
could invade
I.J. Gordon and H.H.T. Prins
mass greater than that of rodents fall into that group, but the oribi is a candidate.
Large non-ruminant concentrate selectors do also exist and most of these are either
browsers that include fallen fruits in their diets (like the tapiroids) or mixed-feeders
with a strong leaning towards omnivory (suids and tayassuidae). In this respect,
lagomorphs (hares, picas, and rabbits), warthogs, and hippos deserve special inter-
est as to whether they are energy-limited or nitrogen-limited. The evolution of
coprophagy in lagomorphs may indicate that they have ‘solved’ nitrogen limitation
(Hirakawa 2001). If this is true, then with rising CO2 levels we predict an increase
of the densities of lagomorphs and duiker antelopes; we keep our powder dry for
the warthog and the hippo. We predict that, everything else being equal, suids,
tayassuidae, tapiroids, and browsing rhinoceroses will increase.
As pointed out already, roughage-eating hindgut-fermenting grazers (equids,
white rhinoceros) are less severely nitrogen-limited than ruminant bulk grazers
(many bovids and some cervids). On the basis of that, we cautiously predict a
shift in the class of roughage grazers from artiodactyls towards equids and rhi-
nos. However, such a shift in numerical abundance will possibly be hampered
by lack of sufficient biological variety within the group of hindgut-fermenters
remaining for a future renaissance of this group of animals. Indeed, the north-
ern white rhino (Ceratotherium simum) went extinct in 2005, and all wild
equids except for the Burchell’s zebra (Equus burchellii) are vulnerable or
threatened. But other grass-eating rhinos went extinct before, like the steppe
rhinoceros (Dicerorhinus kirchbergensis) and the woolly rhino (Coelodontata
antiquitatis). North American horses went extinct recently, and so did E. mau-
ritianum in North Africa and E. hydruntinus in Europe. There is thus much less
variety left than in the artiodactyl group.
The same lack of potential may apply to the rebound we predict for browsers
or for that most versatile group, the mixed feeders. In the temperate zone, there is
a definite lack of biological variation to capitalise on the more favourable
circumstances. Not only did the North American cameloids go extinct, but perhaps
the animal whose all-round adaptations are missed most sorely is the extinct
straight-tusked elephant (Loxodonta atlanticum / Palaeoloxodon antiquus) of
the last interglacial of the temperate zone. In the tropics there are two species
that typically may benefit from both trends (increase of woody cover and soluble
carbohydrates). These are the Asian elephant (Elephas maximus) and the African
elephant (Loxodonta africana). Ironically, through their desire for ivory and
their need to protect their crops, humans keep elephant population numbers low,
but through their use of fossil fuels, people probably create better conditions for
elephants. We do not believe that in the modern world we will quickly see
new elephant populations from either African or Asian stock taking an abode
in the new forests of Europe, Siberia, or North America. The northern forests
thus will lack bulldozer herbivores to facilitate a secure livelihood for smaller species,
necessitating perhaps continued management. In Table 12.2 we have summarized
our predictions about changes in community structure as a result of the trends
we discern.
Table 12.2 Predicted changes in community structure as a result of an increased tendency

towards woody encroachment and expanse of forests, and to increased availability of soluble
carbohydrates
Typical for the
Predicted community trend Typical for the tropics temperate zone
Browsers will increase Greater kudu, lesser kudu Moose
Concentrate Browsers Duiker antelopes Mountain hare
selectors will Grazers Oribi; perhaps warthog and Rabbits, picas,
increase hippopotamus other hares
Omnivores Pigs, peccaries, tapirs, Wild boar
browsing rhinoceroses
Mixed feeders will More typical grazer Asian elephant, impala White-tailed deer,
increase or stay will stay the same red deer
the same More typical browser African elephant Roe deer
will increase
Bulk or roughage Ruminants will Asian buffalo, African American bison,
grazers will decrease buffalo, wildebeest, European
decrease or wisent, fallow
increase deer, Père
David’s deer
Hind-gut fermenters Zebra species, white rhino Kiang
will increase
12.6 Ways of Managing Browsers and Grazers
12.6.1 Managing for Variability Rather than Stability
Natural grazing systems are typified by dynamic herbivore populations (Prins and
Douglas-Hamilton 1990 and Saether 1997; Clutton-Brock and Pemberton 2004).
The present management ethos which advocates managing herbivore populations
for stability (du Toit et al. 2003), often at levels well below carrying capacity, is,
therefore, ‘unnatural’. For example, for over 50 years the management of the
Kruger National Park in South Africa was predicated on the basis of culling large
herbivore populations to maintain predetermined levels. This management strat-
egy does not reflect the natural dynamics of this semi-arid grazing system where
climatic variability, predator/prey interactions, and disease would have meant
dramatic annual, decadal, and centurial fluctuations in the numbers of large her-
bivores (Owen-Smith and Ogutu 2003). This would have lead to increased diver-
sity of vegetation in the park, reflecting periods of low grazing pressure when tree
recruitment, for example, would have been high, and periods of high grazing
pressure when trees and shrubs would be been rare in the system and grasslands
dominant (e.g., Prins and van der Jeugd 1993). In an area the size of the Kruger
National Park it should be possible to restore this temporal variation in herbivore
grazing pressure by using spatially variable population management policies.
However, in many other parklands, especially in Europe, this management

opportunity does not exist. In the European context, the management of large
herbivore populations will require cooperative management policies which link
groups of (national) parks together allowing the build-up of large herbivore
densities in some parks whilst reducing densities in others. This situation could
be reversed at specific time intervals to allow ecosystem dynamics to mimic natu-
ral boom and bust cycles, with movement of animals between parks maintaining
genetic integrity of the metapopulation (Hanski and Gilpin 1997). Whilst this
strategy would require human intervention, it mimics the natural cycles of inter-
actions between plants and herbivores, allowing plant adapted to both severe and
lax grazing pressures to coexist.
12.6.2 Gardening Versus Laisser Faire
The European and American models of nature conservation differ in that the latter
views a natural area as one that is left alone to reach some hypothesised wilderness
equilibrium, whereas the former takes a much more interventionist approach in which
nature has to be managed primarily for specific interests, e.g., birds, biodiversity,
habitats of conservation importance, Scottishness! Each of the approaches has its
benefits and drawbacks. In East Africa, the laisser faire policy has been for years
the dominant one, while in southern Africa the interventionist ‘gardening’ approach
was dominant for many decades. In West Africa, the laisser faire approach was a
matter of fact, not of choice. It is poignant that with the abolishment of apartheid, the
interventionist gardening approach has been imported into East Africa; specialists
from South Africa and Zimbabwe now take up practice there. On the other hand, in
Germany, France, and the Netherlands there is increased public interest in applying
the American ‘wilderness’ concept to local nature management. Even in Japan, in the
northeastern corner of Hokkaido Island, the management is keenly interested in
cooperating with Yellowstone National Park to apply management principles for
increased ‘wilderness’—as is also happening in Russia’s Far East.
With changes in land use occurring in both the developing and the developed
world, there may be opportunities for both approaches since land abandonment in the
developed world may allow large areas to be left to nature, whereas in the developing
world, where the populations of rural areas are likely to expand dramatically over
the next 50 years, there can be expected to be a requirement to strictly manage
wildlife to reduce interactions with human agricultural interests. However, it may also
be the case that in developed countries urban populations will call for strong control
of wildlife as contact is made with predators in the countryside (on weekend visits),
or as wildlife comes into conflict with people in peri-urban areas (e.g., traffic
accidents, diseases such as Lyme disease). Incongruously, it may be that places such
as Europe and North America will become the bastions for predators whilst in Africa
and Asia predators are extirpated from much of their range as they come into conflict
with ever expanding human populations.
12.7 Where Do We Go From Here?
As we have outlined in previous sections of this chapter, there are likely to be major
changes in the structure and composition of the vegetation communities across
large parts of the globe over the next 50 years. In turn this will lead to changes in
the population densities and community composition of the large herbivores that
rely on the vegetation. What can we do, as scientists to help managers to make
decisions in this changing world? Firstly, we will have to make predictions about
what is likely to happen. This can be at the qualitative level such as we have
described above, however, we are likely to be asked to make more quantitative
predictions in order to inform decisions about such things as what is the vegetation
composition likely to be in a certain area in 20 years, what population levels of
herbivores could these changed landscapes hold, what culling policies might need
to be adopted in order to manage vegetation composition/structure to meet other
biodiversity objectives?
A large number of models are available that predict different components of the
system response, from those dealing with changes in plant growth in relation to CO2
levels, through changes in tree/grass ratios, to herbivore population dynamics
models (Illius and O’Connor 1999). However, given the complexity of the responses
of these multi-dimensional systems it is our view that complex mechanistic models
are likely to be too unwieldy to provide realistic quantitative outputs for management.
We advocate the use of relatively simple Bayesian approaches that take into account
the levels of understanding of system linkages and the strength of those linkages.
This will allow scientists to work with managers to develop models of systems with
the degree of certainty associated with the quantitative predictions.
12.7.1 Landscape Scale Experiments
Some exciting new initiatives have recently been taking place in Russia and the
USA; parks are being established where the extinct large mammal communities
are being replaced by functionally similar species (Zimov 2005; http://www.
faculty.uaf.edu/fffsc/park.html). We have to use the kinds of predictions we pro-
vide in Tables 12.1 and 12.2 if we, as ecologists, are to help the developers of
these parks decide how to introduce and manage species in a changing world. The
predictions we have made about the changes in population density and commu-
nity structure as a result of changes in CO2 are not amenable to classic replicated
experiments, however, the need remains to provide management advice at the
landscape scale. This often forces a paradigm shift where the ecologist has to
adopt a logical stance closer to that of the forensic scientist and address questions
more closely aligned to the particular management issue at hand rather than any
general scientific posture. We would suggest that model-based approaches,
including Bayesian methods, offer powerful ways of addressing such questions,
but the status of the conclusions is different from those that a truly replicated and
scientifically controlled experiment would provide (Hobbs and Hilborn 2006). In
cases where only observational data are available, some model-based approach is
inescapable. We see no clear demarcation between model-based approaches and
classical experimental approaches, particularly in ecology, since both rely on
underlying assumptions about repeatability in the natural world, which in prac-
tice may not be entirely correct. In the end, the new “Pleistocene” Parks offer us
a great opportunity to test hypotheses and provide guidance for adaptive learning
management where scientists and managers work hand in hand to manage our
natural resource for the future.
12.8 Conclusions
The world in which we live is changing rapidly. Rising levels of CO2 and changes
in land use patterns will lead to changes in vegetation productivity and vegetation
community composition and structure. Herbivore populations and communities
will respond to these changes. Unless society is going to adopt a laisser faire
attitude to these changes (in which case there could be significant societal and
economic consequences) then managers will be expected to make decisions which
will affect landscapes in the long term. If science is to play a part in determining
the decisions that are made, then we have to develop partnerships with managers in
which we work in collaboration to use our knowledge to develop predictions of
how systems will change and to gather data to improve our understanding of how
these systems operate. This is a challenge facing us all—and scientists must meet
the challenge if they want to remain relevant and valued by society.
References
Ainsworth EA, Long SP (2005) What have we learned from 15 years of free-air CO2 enrichment
(FACE)? A meta-analytic review of the responses of photosynthesis, canopy. New Phytol
165:351–371
Behnke RH, Scoones I (1992) Rethinking range ecology: implications for rangeland management
in Africa. London: International Institute for Environment and Development
Bond WJ (2005) Large parts of the world are brown or black: a different view on the ‘Green
World’ hypothesis. J Veg Sci 16:261–266
Bond WJ, Midgley GF, Woodward FI (2003) The importance of low atmospheric CO2 and fire in
promoting the spread of grasslands and savannas. Glob Change Biol 9:973–982
Bowes G (1993) Facing the inevitable - plants and increasing atmospheric CO(2). Annu Rev Plant
Phys 44:309–332
Coe MJ, Cumming DH, Phillipson J (1976) Biomass and production of large African herbivores
in relation to rainfall and primary production. Oecologia 22:341–354
Coley PD, Massa M, Lovelock CE, Winter K (2002) Effects of elevated CO2 on foliar chemistry
of saplings of nine species of tropical tree. Oecologia 133:62–69
Convention on Biol Divers, Secretariat (2001) Handbook of the convention on biological diversity,
section ecosystem approach. CBD/UNEP, Montreal
Drent RH, Prins HHT (1987) The herbivore as prisoner of its food supply. In: van Andel J, Bakker J,
Snaydon RW (eds) Disturbance in grasslands; species and population responses. Junk
Publishing, Dordrecht, pp 133–149
FAO (2005) http://www.fao.org/forestry/foris/data/fra2005/kf/common/GlobalForestA4-
ENsmall.pdf
Ford J (1971) The role of trypanosomiases in African ecology: a study of the tsetse fly problem.
Oxford University Press, Oxford
Fritz H, Duncan P (1994) On the carrying capacity for large ungulates of African savanna ecosys-
tems. P Roy Soc Lond B Bio 256:77–82
Goverde M, Erhardt A, Stocklin J (2004) Genotype-specific response of a lycaenid herbivore to
elevated carbon dioxide and phosphorus availability in calcareous grassland. Oecologia
139:383–391
Hattas D, Stock WD, Mabusela WT, Green IR (2005) Phytochemical changes in leaves of sub-
tropical grasses and fynbos shrubs at elevated atmospheric CO2 concentrations. Global Planet
Change 47:181–192
Higgins SI, Bond WJ, Trollope WSW (2000) Fire, resprouting and variability: a recipe for grass–
tree coexistence in savanna. J Ecol 88:213–229
Hirakawa, H (2001) Coprophagy in leporids and other mammalian herbivores. Mammal Rev
31:61–80
Hobbs NT, Hilborn R (2006) Alternatives to statistical hypothesis testing in ecology: a guide to
self teaching. Ecol Appl 16:5–19
Illius AW, O’Connor TG (1999) On the relevance of nonequilibrium concepts to arid and semiarid
grazing systems. Ecol Appl 9:798–813
IPCC (2001) Climate Change 2001: impacts, adaptation and vulnerability-contribution of Working
Group II to the IPCC Third Assessment. Cambridge Univ Press, Cambridge
Kimball BA (1983) Carbon dioxide and agricultural yield – an assemblage and analysis of 430
prior observations. Agron J 75:779–788
King D (2005) Climate change: the science and the policy. J Appl Ecol 42:779–783
Kinney KK, Lindroth RL, Jung SM, Nordheim EV (1997) Effects of CO2 and NO3-availability on
deciduous trees: phytochemistry and insect performance. Ecology 78:215–230
Kjekshus H (1977) Ecology control and economic development in East African history: the case
of Tanganyika, 1850–1950. Heinemann, London
Lewis D (2002) Slower than the eye can see: environmental change in northern Australia’s cattle
lands – a case study from the Victoria River district, Northern Territory. Tropical Savannas
CRC, Darwin
Mduma SAR, Sinclair ARE, Hilborn R (1999) Food regulates the Serengeti wildebeest: a 40-year
record. J Anim Ecol 68:1101–1122
Nowak RS, Ellsworth DS, Smith SD (2004) Functional responses of plants to elevated atmospheric
CO2 – do photosynthetic and productivity data from FACE experiments support early predic-
tions? New Phytol 162:253–280
Parton WJ (1995) Impact of climate change on grassland production and soil carbon worldwide.
Global Change Biol 1:13–22
Prins HHT (1996) Ecology and behaviour of the African buffalo: social inequality and decision-
making. Chapman and Hall, London
Prins HHT, Beekman JH (1989) A balanced diet as a goal of grazing: the food of the Manyara
buffalo. Afr J Ecol 27:241–259
Prins HHT, Douglas-Hamilton I (1990) Stability in a multi-species assemblage of large herbivores
in East Africa. Oecologia 83:392–400
Prins HHT, van der Jeugd HP (1993) Herbivore population crashes and woodland structure in East
Africa. J Ecol 81:305–314
Sankaran M, Hanan NP, Scholes RJ et al (2005) Determinants of woody cover in African
savannas. Nature 436:846–849
Scoones I (1994) Living with uncertainty: new directions for pastoral development in Africa.
International Institute for Environment and Development, London
Siegenthaler U, Stocker TF, Monnin E et al (2005) Stable carbon cycle–climate relationship
during the late Pleistocene. Science 310:1313–1317
Sinclair ARE (1977) The African Buffalo: a study in resource limitations of populations. Univ of
Chicago Press, Chicago
Stuart-Hill GC, Tainton NM (1999) Savanna: geographical distribution and extent. In: Tainton
NM (ed) Veld management in South Africa. Univ of Natal Press, Pietermaritzburg, pp
312–317
Van de Koppel J, Rietkerk M, van Langevelde F et al (2002) Spatial heterogeneity and irreversible
vegetation change in semi-arid grazing systems. Am Nat 159:209–218
van Langevelde F, van de Vijver CADM, Kumar L et al (2003). Effects of fire and herbivory on
the stability of savanna ecosystems. Ecology 84:337–350
Vucetich JA, Peterson RO (2004) The influence of top-down, bottom-up and abiotic factors on the
moose (Alces alces) population of Isle Royale. P Roy Soc Lond B Bio 271:183–189
Wand SJE, Midgley GF, Jones MH, Curtis PS (1999) Responses of wild C4 and C3 grass
(Poaceae) species to elevated atmospheric CO2 concentration: a meta-analytic test of current
theories and perceptions. Global Change Biol 5:723–741
Subject Index
A Lophodont, 27, 28
Agricultural Plagiolophodont, 27, 28
production, 5, 9 Selenodont, 27, 28, 33
subsidies, 7 Chemical defences, 127, 227, 228
Allometry, 134 Chewing
Alternate stable state, 285 investment, 123, 124, 138
Annual rainfall, 151–153, 162, 170, 182–184, rate, 123, 124, 138
313 Climate change, 8, 299, 310
Anthropogenic grassland, 273 Clone, 219
Apical meristem, 89, 107, 219, 221, 231, 268 CO2, 181, 299
Avoidance Community
animals, 139, 226 diversity, 264, 265
plants, 225, 226, 229, 232, 272 structure, 201, 206, 210, 213, 263, 271,
310, 312, 313, 315, 316, 318
Complementary grazing, 189
B Concentrate selectors, 90, 312, 313,
Basal metabolic rate, 57 315, 316
Biodiversity, 7, 10, 180, 183, 185, 191, Convergent evolution, 63, 64, 67, 71, 77
263–286, 317, 318 Conversion efficiency, 180
Biomass Craniodental morphology, 26, 29, 36, 71
ratio, 271 Cropping time, 120–123, 138
Bite size, 119, 122, 123, 133, 136–138, 227, Cytochrome P450, 111
231, 237
Body mass, 48, 52, 61, 62, 64, 66–68, 71, 72,
94, 95, 123, 124 D
Defence
chemical, 127, 224, 227, 228
C constitutive, 228
Census, 5, 166, 167 induced, 228, 232
Cheek teeth-crown height qualitative, 228
Brachydont, 27, 28 quantitative, 228
Hypselodont/evergrowing, 25 structural, 120–122, 227
Hypsodont, 25–29 Demography, 7, 298
Cheek teeth–occlusal morphology Density dependence, 164, 294, 302
Bilophodont, 26, 27, 33 Dental
Bunodont, 26, 27, 33 mesowear, 29, 57
Bunolophodont, 24, 27 microwear, 29, 35, 38
Lophed, 26–28 Detoxification, 34, 53, 56, 107, 109–111
323
324 Subject Index
Diet Grazing history, 268, 269, 272, 286

optimisation, 139 Grazing lawn, 134, 232, 233, 235, 270, 280
overlap, 186–189, 193
Digestibility, 49–51, 54, 57, 58, 71, 77, 91, 92, 95
Disease, 2, 15, 164, 227, 236, 316, 317 H
Dispersion, 118, 120, 121, 125, 127, 133 Habitat heterogeneity, 264, 276
Herbage
nutritive value, 21, 90, 117
E physical characteristics, 102
Ecosystem Hierarchy theory, 126
functioning, 183, 185, 189, 193, 195 Hindgut fermenter, 26, 34, 41, 64, 312, 315
structure, 9 Home range, 126, 160, 162, 163, 170
Enamel, 27–30, 50, 54, 56, 57, 59, 70, 73, 124 Hypsodonty, 28–31, 34–36, 39, 54, 57,
Encounter rate, 124, 125, 134 66–68, 70–72
Equilibrium, 169, 244, 245, 317
Evapo-transpiration, 9
Evolutionary history, 21–41, 134, 186, 241, I
268, 271, 272, 284 Ingesta retention, 54, 58, 62, 64–67, 70, 77
Insurance hypothesis, 196
Intake rate, 21, 117–119, 121, 122, 133, 134,
F 137, 227, 231, 237
Facilitation, 124, 125, 190, 279–281
Feeding
guild, 134, 183, 194 K
niche, 52, 64, 77, 89, 93, 97, 108, 109, 186 Key resources, 238
stations, 126, 135–138
Fibre
cellulose, 41, 50, 51, 56, 90, 93, 118, 121 L
lignin, 50, 53, 90, 95, 118, 121, 130, 131, 204 Land abandonment, 7–11, 13–16, 317
neutral detergent fibre, 91 Land use intensification, 8
Fire, 2, 8, 15, 118, 119, 151, 153, 158, 168, 170 Life history, 218, 295–297, 299, 303
Flower, 218, 236, 266, 268, 275, 276 Litter accumulation, 274
Food intake, 54, 55, 57, 58, 77 Litter size, 163, 295
Forage Liver, 53, 56, 93, 107, 111, 228
production, 153 Livestock, 3, 5, 56, 109, 111, 168, 169, 189, 227,
quality, 128, 170, 204, 241, 242, 280 242, 244, 276, 278–282, 284, 309
Foregut fermenter, 65
Functional
group, 270, 273 M
niche, 185, 191 Mandible, 26, 38, 54, 55, 57, 72
redundancy, 179 Marginal Value Theorem, 135
Functional response, 117–120, 122, 138 Masseter, 29, 54, 57, 66, 67, 71, 73
Mega herbivore, 33, 40, 168, 195, 232, 237,
243, 266, 267
G Memory, 124, 125
Gain function, 123, 137, 138 Meristem, 89, 107, 118, 127, 132, 218–223,
Game farms, 10 229, 231, 232, 234, 236, 268
Genet, 218, 219, 221 Migration, 35, 139
Genotype, 106, 129–132, 221, 263, 268 Mixed farming system, 189
Global change, 299, 309 Mixed-feeders, 35, 40, 182, 183, 193, 315
Grass Mixed-species production systems, 196
C3 vs. C4, 51, 54, 72, 73, 311–314 Morphology, 26, 29, 30, 35, 36, 47, 50,
Grasslands, 3, 4, 21, 30–32, 34, 35, 40–42, 54, 56, 57
62, 101 Muzzle width, 53, 55, 66–68
Subject Index 325
N crashes, 150, 165, 166, 170, 171

Neutral theory, 179 density, 150, 159–162, 164, 170,
Niche 294, 295, 301
complementarity, 185, 191, 195, 196 dynamics, 118, 119, 149–171, 181, 243,
concept, 179 312, 313, 318
differentiation, 179, 185 eruptions, 150
theory, 179 explosion, 1, 239
Nitrogen, 4, 49, 90, 99, 110, 128–132, 134 growth, 3, 8, 150, 163, 164, 166, 294, 299,
Non-linear dynamics, 285 303, 312
Normalised vegetation index (NDVI), 181 management, 302, 303, 316
Nutrient regulation, 163–165
retention, 183, 185 structure, 150, 164, 242
use efficiency, 185 synchrony, 150, 169, 171
Nutritive value, 28, 90, 91, 93, 96, 111, 117 Predation risk, 117, 238, 282
Predation, 139, 164–166, 168, 169, 171, 264,
271, 278, 283
O Preferential grazing, 270
Omasum, 60, 62, 74, 90 Primary production, 180, 181, 186, 190, 192,
193, 195, 201, 205, 240, 242, 273
Productivity gradient, 284
P Protected areas, 167, 170, 194
Passage rate, 41, 93, 94, 100–103 Protein, 4, 49, 52, 53, 56, 67, 90, 91, 93, 103,
Pastoralism, 2 105, 107–109, 135, 138, 153, 159,
Patches 223, 228
optimisation models, 138 Protozoa, 58, 90
perception, 130, 132–137
Pectin, 49, 50
Persistence, 219, 231, 233, 236, 240 R
Phenology, 129, 163, 268 Rainfall, 29, 129, 150, 151, 153, 155, 156,
Phenotypic variation, 264, 268 162, 164, 166, 168–170, 182, 184,
Photosynthesis, 30, 220, 222–224, 228, 229, 311 194, 195, 245, 313
Photosynthetic rate, 223, 267 Ramet, 218, 221, 222, 230, 231, 233, 265
Phylogenetic control, 63, 64, 66, 67, 71, 72, 77 Regional species pool, 265, 280
Phylogeny, 33, 64, 71, 100, 104, 295 Remote sensing, 181
Plant communities, 2, 16, 126, 128, 240–244, Resilience, 183, 185
270–275 Resistence
Plant diversity, 32, 185, 272, 273, 277 plants, 130, 225–229, 240, 241
Plant nutrient allocation Resource partitioning, 179, 280
long-term responses, 207–208 Retention time, 58, 60, 70, 71, 77, 100–101,
short-term responses, 206–207 104, 110
Plant secondary metabolites Reticulum, 77, 78
alkaloids, 107 Root
detoxification, 107 biomass, 206–208, 213
mimosine, 109 exudation, 205–207, 213
post absorptive metabolism, 110–111 Roughage feeders, 312
Plant-herbivore interactions, 118, 119 Ruderal species, 276
Plasticity Rumen
phenotypic, 221, 228 capacity, 64
Population detoxification, 109–111
abundance, 150, 165, 293 fermentation rate, 91, 93, 96, 103, 104
biomass, 129, 130, 132, 135, 136, 150, fibre digestibility, 103, 104
153, 156, 179, 182, 183, 186, 189, fluid, 60, 98, 110
190, 193–195, 207, 211, 212, 220, function, 78, 98, 110
221, 227, 243, 270, 272, 312 microbes, 90, 98, 104
326 Subject Index
Rumen (continued) richness, 10, 11, 15, 37, 179–185, 190,

musculature, 96, 102 193, 195, 264, 270, 273, 279, 280
N recycling, 104 Starch, 8, 50, 90, 105, 223, 224, 310
papillae, 60, 96 State-and-transition, 245, 246, 271
stratification, 92 Steppe, 4, 31, 40, 41, 150, 209, 219,
surface area, 73, 102 313, 315
Ruminant, 26, 28, 33, 34, 41, 42, 54, 58, Structural defences, 120–122, 227, 288
60–62, 64, 67, 71–78, 89, 90, Succession, 15, 204, 244–246, 270, 273, 274,
100, 102, 103, 105, 106, 108, 276, 285
109, 124, 149, 153, 182, 237, Swards, 121, 122, 128, 129, 133, 134, 136,
312, 313, 316 138, 231, 279
S T
Saliva flow, 93, 103, 110 Tannin degrading bacteria, 98, 99, 110
Salivary glands, 53, 56, 60, 71, 103 Tannins, 50, 91, 93, 107–110, 159, 228
Salivary tannin-binding proteins, 67, 97, Tensile strength, 121
107–109 Territory, 161–163, 238
Savanna, 31, 35, 36, 38, 40, 42, 129, 130, 150, Thornveld, 189
152, 156, 158, 168, 170, 182, 189, Tolerance
195, 196, 219, 237, 243, 275, 299, animals, 182, 241
300, 310, 311 plants, 229
Seasonal variation, 63 Trait
Secondary compounds, 34, 49, 56, 107–111, diversity, 263–265
128, 210, 311, 312 variation, 185, 268
Secondary metabolites, 89, 91–93, 107, 109, Trampling, 212, 229, 270, 274, 276, 278
112, 131, 202, 211, 213
Secondary productivity, 183, 185, 189–191,
193, 195 U
Selective foraging, 155, 208, 210, 211, 213, Ungulate, 4, 5, 21–26, 29, 30, 32–41, 57, 64,
217, 237 65, 67
Sex ratio, 163, 240 Urban areas, 5–7, 317
Shrub, 9, 48, 89, 92, 107, 131, 136, 158, 163,
186, 217–220, 236, 244, 265, 278,
285, 310, 312, 316 V
Social facilitation, 124, 125 Vegetation dynamics, 170, 190, 217–247
Soil community structure, 206, 210, 212 Volatile fatty acid
Soil nutrient cycling proportions, 105
accelerating effects, 202
decelerating effects, 202
Spatial heterogeneity, 126, 133, 181, 182, 222 W
Spatial variations Water infiltration, 9
browse, 130–132 Whole-ecosystem effect, 275
grass, 128–130 Wildlife sanctuaries, 6
Species Woody encroachment, 8–15, 316
assemblage, 1, 263, Woody species, 8–11, 13, 218, 231, 233, 234,
diversity, 179–196, 263, 264, 271, 273 244, 265, 309–311
Species Index
A Equids, 33–36, 39–42, 67, 70, 315

American bison, 4, 316
Argali, 188
Aurochs, 4, 265, 282, 314 G
Gaur, 2
Gayal, 2, 4
B Gerenuk, 280
Banteng, 2, 3 Giraffe, 61, 62, 78, 122, 157, 164, 166–168,
Bharal, 281 187–189
Bison, 4, 16, 40, 42, 136, 149, 162, 186, 206,
281, 282
Black rhinoceros, 56, 58, 108, 187 H
Black wildebeest, 4 Hare wallaby, 4
Black-tailed deer, 186 Hippopotamus, 188, 280, 316
Blue sheep, 188 Hyraxes, 22, 24, 26, 70, 124
Bluebuck, 4
Boar, 15, 281, 316 I
Buffalo, 2, 3, 5, 62, 162, 164, 167, 188, 239, Ibex, 188, 300
280, 316 Impala, 149, 161, 164, 166, 188, 316
C K
Camel, 2–4, 24 Kangaroo, 3, 27, 29, 106, 124
Caribou, 165, 282, 283 Kiang, 281, 316
Chevrotain, 188 Kob, 163, 188, 280
Chiru, 281 Kouprey, 4
Chital, 188 Kudu, 56, 62, 122, 136, 137, 157, 164, 166,
168, 188, 189, 280, 316
D
Deer, 3, 4, 7, 15, 16, 24, 40, 41, 78, 101, 108, L
123, 132, 149, 158, 161–163, 165, 169 Lama, 2, 13,
Dromedary, 4 Llama, 40, 41, 187, 189
Duiker, 188, 315, 316
M
E Macropods, 56, 59, 61, 64, 66, 67, 77, 121, 149
Eland, 162, 163, 166, 187, 189 Moose, 15, 25, 78, 122, 132, 136, 137, 158,
Elephant, 21, 23, 28, 41, 70, 77, 168, 182, 187, 162, 163, 165, 168, 186, 210, 211,
193–195, 224, 234, 265, 280, 315, 316 238, 243, 265, 283, 316
327
328 Species Index
Mule deer, 108, 161, 162, 165, 186 S

Sambar, 188
Suids, 39, 41, 61, 71, 315
N Suni, 188
Nilgai, 188 Swamp deer, 4
O T
Onager, 2 Tapirs, 33, 39–41, 61, 316
Oryx, 280 Tarpan, 265
Thar, 3
Topi, 122, 163, 188, 280
P
Père david’s deer, 4, 40, 316
Perissodactyls, 22, 24, 26, 28, 32–34, 57, 64, W
65, 67 Wapiti, 149, 161, 162, 281
Pronghorn, 29, 36, 39, 40, 186, 205 Warthog, 27, 28, 41, 188, 315, 316
Przewalski horse, 4 Water buffalo, 2, 3, 5
Waterbuck, 164, 166, 188, 280
White rhinoceros, 4, 108, 312, 315
Q White-tailed deer, 7, 16, 161, 163, 165, 166,
Quagga, 4 169, 186, 187, 316
Wildebeest, 4, 40, 122, 161–163, 166, 167,
169, 188, 235, 239, 280, 316
R Wisent, 16, 316
Red deer, 15, 132, 137, 149, 158, 161, 165,
166, 169, 186, 187, 281, 283,
301, 316 Y
Reindeer, 2, 3, 40, 105, 106, 135, 187, 205, Yak, 2, 188, 281
206, 208, 209, 212
Rhinoceroses, 67, 77, 108, 315, 316
Roe deer, 15, 16, 78, 123, 132, 137, 162, Z
163, 169, 187, 293, 295, Zebra, 42, 149, 162–164, 166, 167, 169, 188,
299–301, 316 239, 280, 315, 316
Ecological Studies
Volumes published since 2003
Volume 163 Volume 171

Fluxes of Carbon, Water and Energy Coastal Dunes: Ecology
of European Forests (2003) and Conservation (2004)
R. Valentini (Ed.) M.L. Martínez and N. Psuty (Eds.)

Herbivory of Leaf-Cutting Ants: Biogeochemistry of Forested
A Case Study on Atta colombica in the Catchments in a Changing
Tropical Rainforest Environment:
of Panama (2003) A German Case Study (2004)
R. Wirth, H. Herz, R.J. Ryel, W. Beyschlag, E. Matzner (Ed.)
B. Hölldobler
Volume 173
Volume 165 Insects and Ecosystem Function (2004)
Population Viability in Plants: W.W. Weisser and E. Siemann (Eds.)
Conservation, Management,
and Modeling of Rare Plants (2003) Volume 174
C.A. Brigham, M.W. Schwartz (Eds.) Pollination Ecology and the Rain Forest:
Sarawak Studies (2005)
Volume 166 D. Roubik, S. Sakai, and A.A. Hamid (Eds.)
North American Temperate
Deciduous Forest Responses to Volume 175
Changing Precipitation Antarctic Ecosystems: Environmental
Regimes (2003) Contamination, Climate Change,
P. Hanson and S.D. Wullschleger (Eds.) and Human Impact (2005)
R. Bargagli
Volume 167
Alpine Biodiversity in Europe (2003) Volume 176
L. Nagy, G. Grabherr, Ch. Körner, Forest Diversity and Function: Temperate
D. Thompson (Eds.) and Boreal Systems (2005)
M. Scherer-Lorenzen, Ch. Körner,
Volume 168 and E.-D. Schulze (Eds.)
Root Ecology (2003)
H. de Kroon and E.J.W. Visser (Eds.) Volume 177
A History of Atmospheric CO2
Volume 169 and its Effects on Plants, Animals,
Fire in Tropical Savannas: and Ecosystems (2005)
The Kapalga Experiment (2003) J.R. Ehleringer, T.E. Cerling, and
A.N. Andersen, G.D. Cook, M.D. Dearing (Eds.)
and R.J. Williams (Eds.)
Volume 178
Volume 170 Photosynthetic Adaptation: Chloroplast
Molecular Ecotoxicology to Landscape (2005)
of Plants (2004) W.K. Smith, T.C. Vogelmann,
H. Sandermann (Ed.) and C. Chritchley (Eds.)
Lamto: Structure, Functioning, Managed Ecosystems and CO2:
and Dynamics of a Savanna Case Studies, Processes, and
Ecosystem (2006) Perspectives (2006)
L. Abbadie et al. (Eds.) J. Nösberger, S.P. Long, R. J. Norby,
M. Stitt, G.R. Hendrey,
Volume 180 and H. Blum (Eds.)
Plant Ecology, Herbivory, and Human
Impact in Nordic Mountain Birch Volume 188
Forests (2005) Boreal Peatland Ecosystem (2006)
F.E. Wielgolaski (Ed.) and P.S. Karlsson, R.K. Wieder and D.H. Vitt (Eds.)
S. Neuvonen, D. Thannheiser (Ed. Board)
Volume 189
Volume 181 Ecology of Harmful Algae (2006)
Nutrient Acquisition by Plants: E. Granéli and J.T. Turner (Eds.)
An Ecological Perspective (2005)
H. BassiriRad (Ed.) Volume 190
Wetlands and Natural Resource
Volume 182 Management (2006)
Human Ecology: Biocultural Adaptations J.T.A. Verhoeven, B. Beltman,
in Human Cummunities (2006) R. Bobbink, and D.F. Whigham (Eds.)
H. Schutkowski
Volume 191
Volume 183
Wetlands: Functioning,
Growth Dynamics of Conifer Tree
Biodiversity Conservation,
Rings: Images of Past and Future
and Restoration (2006)
Environments (2006)
R. Bobbink, B. Beltman, J.T.A. Verhoeven,
E.A. Vaganov, M.K. Hughes,
and D.F. Whigham (Eds.)
and A.V. Shashkin

Reindeer Management in Northernmost Geological Approaches
Europe: Linking Practical and to Coral Reef Ecology (2007)
Scientific Knowledge in Social-Ecological R.B. Aronson (Ed.)
Systems (2006)
B.C. Forbes, M. Bölter, L. Müller-Wille, Volume 193
J. Hukkinen, F. Müller, N. Gunslay, and Biological Invasions (2007)
Y. Konstantinov (Eds.) W. Nentwig (Ed.)

Ecology and Conservation Clusia: A Woody Neotropical
of Neotropical Montane Oak Genus of Remarkable Plasticity
Forests (2006) and Diversity (2007)
M. Kappelle (Ed.) U. Lüttge (Ed.)

Biological Invasions The Ecology of Browsing and
in New Zealand (2006) Grazing (2008)
R.B. Allen and W.G. Lee (Eds.) I.J. Gordon and H.H.T. Prins (Eds.)

Gordon and Prins 2008 - Ecology of Grazers and Browser PDF

Uploaded by

Copyright:

Available Formats

Gordon and Prins 2008 - Ecology of Grazers and Browser PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Gordon and Prins 2008 - Ecology of Grazers and Browser PDF

Uploaded by

Copyright:

Available Formats

Ecological Studies 195

Iain J. Gordon, Herbert H.T. Prins

The Ecology of Browsing

Library of Congress Control Number: 2007931594

Springer is a part of Springer Science+Business Media

© Springer-Verlag Berlin Heidelberg 2008

Editor: Dr. Dieter Czeschlik, Heidelberg, Germany

Printed on acid-free paper SPIN 11382201 31/3180 5 4 3 2 1 0

1 Introduction: Grazers and Browsers

2 An Evolutionary History of Browsing

3 The Morphophysiological Adaptations of Browsing

4 Nutritional Ecology of Grazing and Browsing Ruminants . . . . . . . . . . 89

5 The Comparative Feeding Bahaviour of Large Browsing

5.3 Foraging in Patches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126

6 The Comparative Population Dynamics of Browsing

7 Species Diversity of Browsing and Grazing Ungulates:

8 Impacts of Grazing and Browsing by Large Herbivores

9 Plant Traits, Browsing and Gazing Herbivores,

9.6.3 Large Herbivores and Predators . . . . . . . . . . . . . . . . . . . . . 238

10 The Impact of Browsing and Grazing Herbivores

11 Managing Large Herbivores in Theory and Practice: Is the

12 Grazers and Browsers in a Changing World: Conclusions . . . . . . . . . 309

12.6 Ways of Managing Browsers and Grazers. . . . . . . . . . . . . . . . . . . . 316

Subject Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323

Species Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327

Herbert H.T. Prins and Iain J. Gordon

1.2 Dominance of Domesticated Grazers and Browsers

1.3 The Last 30 Years and the Immediate Future

1940 1960 1980 2000 2020 2040

populated by 50 or fewer stragglers (Anon. 2006b). In Wyoming, Nebraska, or

Apart from economic drivers, sociological and demographic changes, and

1.4 Societal Relevance

even if they have no

distortions are possible wild herbivores. decreases, but local

Table 1.2 (continued)

Table 1.2 (continued)

By asking ourselves how grazers and browsers function in the landscape, we

Anonymous (2003) Europe’s Environment: The Third Assessment. Environmental assessment

in past times influenced the distribution and abundance of ungulates of different

2.1.1 Ungulate Phylogeny and Evolutionary History

Eurasia, the Miocene of North America, and the Miocene to Pleistocene

Box 2.1 (continued)

Eurasia, and South America, ranging from goat-sized to bison-sized, and

2.1.2 Determination of Feeding Adaptations

Fig. 2.2 Types of molar morphologies in herbivorous mammals. A. Bunodont (peccary).

2.1.3 Cenozoic Changes in Climate

Evolutionary transitions in ungulate diets can only be understood in the context

AFRICA NORTH SOUTH AUSTRALIA

2.2 Fossil Record Evidence of Dietary Evolution in Ungulates

2.2.1 Early Archaic Ungulates and Ungulate-like

2.2.2 The Eocene Emergence of Modern

The early Eocene saw a great diversification of artiodactyls and perissodactyls in

2.2.3 The Lull Before the Storm: Oligocene

2.2.4 The Rise of the Grasslands (20 to 10 Ma)

Fig. 2.4 Continent-wide generic diversity of ungulates in North America. H = hypsodont

2.2.5 The Rise of Grazing Ungulates (10 to 2 Ma)

2.2.6 The Late Cenozoic Dawn of the Modern Ungulate Fauna

2.3 Discussion and Conclusions

Grazing is a fairly recent evolutionary specialization among ungulates. Although