Physiological Ecology of Tropical Plants

Physiological Ecology of Tropical Plants
Ulrich Lüttge
Physiological Ecology
of Tropical Plants
Second Edition
123
Professor Dr. U LRICH L ÜTTGE
Technische Hochschule Darmstadt
Institut für Botanik
Schnittsphanstraße 3–5
D-64287 Darmstadt
Germany
Cover photo showing the fringe of a rainforest on Sierrania Parú, Guayana

Highlands, Venezuela; 1250 m a.s.l., 04°25’N, 65°32’W.
ISBN-13: 978-3-540-71792-8 e-ISBN-13: 978-3-540-71793-5

DOI 10.1007/6138588
Library of Congress Control Number: 2007939890
© 2008 Springer-Verlag Berlin Heidelberg
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 permission for use must always be obtained from Springer. Violations
are liable to prosecution under the German Copyright Law.
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.
Cover photograph: Prof. U. Lüttge, Darmstadt
Cover design: WMXDesign
Typesetting and Production: LE-TEX Jelonek, Schmidt & Vöckler GbR, Leipzig
Printed on acid-free paper SPIN 11515630 31/3180
987654321
springer.com
Preface, 1st Edition
Recently, in many countries, particularly in the industrialized parts of the world,

there has been an upsurge in public opinion concerned about the alarming rate of
destruction of tropical ecosystems by man, and particularly the continuing elimina-
tion of tropical rainforest. The response to the public concern has led to a bourgeon-
ing of popular literature. More dispassionate scientific books are often devoted to
special topics of tropical ecology, e.g. biotopes such as rainforests or savannas, eco-
logically defined groups of plants such as mangroves, epiphytes or succulent plants,
and special taxa such as palms. The ecological approach of this scientific literature
is predominantly floristic (and faunal) description and analysis of the diversity in
associations, biotopes and ecosystems.
However, the development of modern experimental technology, which is increas-
ingly well adapted to the use in field work in the tropics, is also allowing more and
more detailed ecophysiological studies. Observations in the field lead to delineation
of precise problems for studies in laboratories, growth chambers and phytotrons.
The results of such work are built into hypotheses, whose ecological significance
in turn is tested again in the field. This fruitful ecophysiological interplay between
work in the field and in the laboratory leads to an increasing understanding of phys-
iological, biochemical and molecular bases of ecological adaptations. Phenotypic
physiological plasticity is important in mechanisms of ecological adaptations and
may also be involved in mechanisms of generation and maintenance of floristic and
faunal diversity in ecosystems.
I am therefore convinced that ecology must be studied at various levels of com-
plexity and integration, namely the phytogeographical, ecosystem or biotope level,
also called the synecological level, the whole organism level, also called the auteco-
logical level, and the cellular, organelle, membrane and molecular levels. These var-
ious levels may, in fact, not be so much distinguished by their respective degree of
complexity. One may consider them as fractals. With each additional magnification,
i.e. on each level, similar problems of complexity with integration of subsystems,
feedback and non-linear behaviour will be encountered. Therefore, the distinction
between levels does not appear to be basically conceptual and is more a matter of
scaling. Studies with ecological relevance must not remain isolated within the in-
vi Preface, 1st Edition
dividual levels. In addition to the interplay between work in the field and in the
laboratory, there must be continuous feedback and feedforward within and between
the levels of different scaling. Thus, although the aim is to progress towards levels
at finer scales as far as possible, I think we must also refer to the levels at larger
scales in order to put mechanisms of ecological adaptations into context.
Alexander von Humboldt was the first to recognize the relations between phys-
iognomy of plants and the environment. In this vein, it seemed appropriate to begin
the various chapters and sections of this book with descriptions of the physiognomy
of biotopes and plants and to deduce the ecophysiological problems from them. This
may also help to motivate readers, depending on their individual starting points, ei-
ther to be carried on from the experienced environment to more abstract levels of
understanding, or to consider the function of molecular, biochemical or physiologi-
cal units in relation to the performance of plants in habitats.
It is the aim of this book to cover plants of all major tropical ecosystems. In the
tropics we encounter biotopes with vast expanses. Thus, I felt that a brief record
of current trends in large scale sensing and diagnosis would be useful (Chap. 2).
The largest and most dominating tropical biomes are forests (Chap. 3) and savan-
nas (Chap. 7). Ecophysiology of tropical plants is, in general, still a limited field,
which is in its early stages of development, and in some areas knowledge is still
poor. In limited areas, however, progress is already quite advanced. For example
much work is available on epiphytes, which therefore have been given their own
chapter (Chap. 4) although they are important parts of tropical forests. Mangroves
are very specific tropical forests and also are treated in a separate chapter (Chap. 5).
Salinas, inselbergs and páramos are very characteristic tropical environments and
are covered each in a short chapter (Chaps. 6, 8 and 9). Although plants of dry and
arid habitats are discussed in various chapters, genuine deserts are excluded. First,
the major deserts lie outside the tropics. Second, very much ecophysiological work
has already been performed on desert plants, and this would go much beyond the
scope of this book.
Physiognomy is depicted with photographs, most of which were taken during
my own excursions in the tropics. Ecophysiological work on tropical plants was
compiled from the literature wherever available and I also refer to studies of my
own group. Some original publications are cited when necessary, but when possible
and appropriate preference was given to quoting summarizing and reviewing works,
because it is the aim of this book to give a general overview rather than specific
interpretations of specialized research contributions.
Abundant illustrations with drawings and tables are used to elucidate ecophysi-
ological relations. It was, moreover intended to write a simple and readily flowing
text, which is easy to read. In fact, it was aimed to make this book useful for a wide
audience interested in the tropical environment. Thus, while writing I found that,
for some readers, basic knowledge behind ideas, results and hypotheses presented
would need at least brief repetition and explanation. In order to avoid interruption
of the text, this was separated into boxes, which the reader may consult or overlook,
depending on the individual level of understanding.
Preface, 1st Edition vii
I am indebted to former students and postdoctoral fellows in my laboratory and

friends and colleagues who work in various tropical countries and made it possible
for me to become acquainted with tropical environments and to perform ecophys-
iological field research in the tropics. Most of this occurred in the Caribbean, in
Venezuela and Brazil. Unavoidably, this led to a noticeable South American bias of
the book. This may be of some disadvantage in the selection of the concrete phys-
iognomic descriptions, but in the more abstract ecophysiological context it may be
less important.
There is much goodwill in industrialized countries for an understanding of the
tropics, but much more knowledge is needed. Many efforts are being made, and
there is much research in the developing tropical countries, but more encouragement
is needed. I hope this book, may make some contribution, if humble, towards both. I
have written this book in English, which is not my mother tongue, but the scientific
lingua franca to address people in the different parts of the world. In striving to build
bridges, Alexander von Humboldt is chosen as a mentor for this book, and citations
from his lucid descriptions of his journey to South America are used here and there
as a guide.
I owe particular thanks to Professor Dr. Dr. h.c. mult. Hubert Ziegler for his en-
couragement in publishing this text, and also to him and Professor Dr. Erwin Beck
for reading some of the chapters. With great empathy, Professor Dr. Howard Grif-
fiths went through the painstaking work of correcting my English, and with a great
deal of sensitivity, being an expert in tropical plant ecophysiology himself, he made
invaluable contributions far beyond establishing linguistic discipline. I thank the
students who attended my courses and lectures on tropical plant ecophysiology and
served as guinea pigs for the development of this text. My own research in the
tropics and exchange with colleagues in tropical countries were supported by the
Deutsche Forschungsgemeinschaft (DFG), the Alexander-von-Humboldt-Stiftung,
the Deutscher Akademischer Austauschdienst (DAAD), the Volkswagenstiftung and
the Körber European Science Award, which are most gratefully acknowledged.
A great many thanks are due to Ms. Barbara Reinhards, who took care of an almost
endless succession of different versions of the manuscript with never-exhausted pa-
tience, and Ms. Doris Schäfer and Ms. Rosel Heger who handled the line drawings
and the halftone photographs, respectively.
Darmstadt, February 1997 Ulrich Lüttge

Preface, 2nd Edition
Now, after 10 years, as I present the second edition of this book, there are no reasons
to change the aims and the scope. However, during these ten years, international re-
search efforts in physiological ecology of plants in the tropics have increased enor-
mously in quantity and quality. In some fields advances were more substantial than
in others. New approaches were made in remote sensing and, at the other end of
the spectrum, in some areas molecular biology saw many developments regarding
ecological performance of tropical plants, e.g. in understanding the adaptation of
resurrection plants to the extreme habitat of inselbergs. This progress had to be cov-
ered in the second edition without too much of an increase in volume. Thus, it was
again important to strive for a balance between detail and generality as advocated by
Alexander von Humboldt who remains the spiritual mentor of the book. In this vein
his suggestion in modifying the structure of the chapters was also followed when he
writes:
In an extended work where one strives for ease of understanding and overall clarity, the
composition and layout in the arrangement of the whole are almost more important than
the richness of the contents.1
Therefore, to provide more convenient units for the reader, large chapters were
broken down into smaller ones. Material on tropical forests, occupying about half
of the entire volume of the book, is now arranged in five chapters covering structure
and function under the influence of environmental cues and including epiphytes and
mangroves as part of the tropical forest complex. Savannas are now treated in two
chapters. Coastal Salinas have been combined with a new section on the Brazilian
restingas in a chapter on coastal sand plains.
In spite of international agreements at the political level, not much has changed
since the late 1980s regarding reducing the speed of the destruction of the original
tropical environments. When Alexander von Humboldt wrote that
1 In einem vielumfassenden Werke, in dem Leichtigkeit des Verständnisses und Klarheit des To-
taleindrucks erstrebt werden, sind Composition und Gliederung in der Anordnung des Ganzen fast
noch wichtiger als die Reichhaltigkeit des Inhalts. Kosmos, vol. IV, 1858
x Preface, 2nd Edition
nature then appears greatest when, as well as the sensual impressions it is also reflected in
the depth of thought2
he argued with a plea for stringent study and mathematical understanding. Now,
as we are increasingly seeing the loss of the illusion that rationality prevents the
destruction of nature, we may again have to think more about the former, “sensual
impressions”. The description of the physiognomy of biotopes, again associated
with the discussion of ecophysiological performance of plants in this book, may
allude to this. In the preface ten years ago I commented on the fact that I present
this text in English, not in my mother tongue. Besides linguistic unevenness, this
may bear out different attitudes in thinking, as masterly expressed by Alexander
von Humboldt, who himself was able to write in many different languages:
The tone of expression in its increased vividness will in no way be the same when involving
either simple pure objectivity in the description of nature or the reflection of the surrounding
nature embracing the feelings and innate character of man. In all literature these limitations
are markedly different in accordance with the character of the language and the spirit of
the people. . . . Only at home – i.e. in the self assurance of the native mother tongue – can
the correct balance of the timbre be unconsciously found.3
I hope I have, in the main, found the balance in this text and ask the reader to
forgive any failures.
Again, I have to thank my students for much feedback, giving the book a realistic
basis for coverage of the material. I am grateful that they have supported my lectures
by their attendance, although I am now professor emeritus.
I am grateful to Dr. Dieter Czeschlik of Springer for his sustained confidence in
the book and I warmly thank Ursula Gramm for her fine cooperation in its produc-
tion.
Darmstadt, April 2007 Ulrich Lüttge
2 Die Natur erscheint da am größten, wo neben dem sinnlichen Eindruck sie sich auch in der Tiefe
des Gedankens reflectiert. Kosmos, vol. IV, 1858

3 . . . , weil die Färbung des Ausdrucks in seiner erhöhten Lebendigkeit keineswegs dieselbe sein
darf in der einfachen, in reiner Objectivität aufgefaßten Naturbeschreibung, und in dem Reflex
der äußeren Natur auf das Gefühl und die innere Natur des Menschen. In jeder Litteratur aber
sind diese Grenzen nach dem Wesen der Sprache und dem Volksgeiste anders gezogen . . . Nur
heimisch, in der angeborenen, vaterländischen Sprache kann durch Selbstgefühl das richtige Maaß
der Färbung wie bewusstlos bestimmt werden. Kosmos, vol. IV, 1858
Translated by myself from Alexander von Humboldt, Kosmos. Entwurf einer physischen
Weltbeschreibung. Ottmar Ette and Oliver Lubrich, eds., Eichborn Verlag, Frankfurt am Main,
2004
Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Historical Background of Ecophysiology . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 The Tropics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Tropical Forests and Savannas: Their Emotional, Commercial,
Ecological and Scientific Importance . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4 The Destruction of Tropical Forest . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2 Large-Scale Sensing and Diagnosis in Relation

to the Tropical Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1 Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2 Climatic Relations and Vegetation Modeling . . . . . . . . . . . . . . . . . . . . 17
2.2.1 The Klimadiagramm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2.2 Vegetation Modeling
Based on Irradiance and Water Budgets . . . . . . . . . . . . . . . . . . 22
2.3 Remote Sensing Using Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3.1 Reflection and Absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3.2 Fluorescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.4 Gas Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.5 Stable Isotope Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.6 Mathematical Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3 Tropical Forests. I. Physiognomy and Functional Structure . . . . . . . . . 51

3.1 Separation of Different Types of Tropical Forests . . . . . . . . . . . . . . . . 51
3.2 Physiognomy of Different Types of Tropical Forests . . . . . . . . . . . . . 55
3.2.1 Tropical Rain Forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.2.2 Tropical Cloud and Elfin Forests . . . . . . . . . . . . . . . . . . . . . . . 59
3.2.3 Floodplain Forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.2.4 Thorn Scrub and Cactus Forests . . . . . . . . . . . . . . . . . . . . . . . . 62
3.2.5 Mangroves and Gallery Forests . . . . . . . . . . . . . . . . . . . . . . . . 67
xii Contents
3.3 Horizontal Structure and Diversity of Tropical Forests . . . . . . . . . . . . 67

3.3.1 Diversity and the Spatial Structure of the Environment . . . . . 67
3.3.2 Diversity and Plasticity and the Biological Stress Concept . . 69
3.3.3 Diversity and the Chaos of Oscillating Mosaics . . . . . . . . . . . 73
3.3.4 Diversity and Life Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
3.4 Vertical Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.4.1 Irradiance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.4.2 Temperature and Air Humidity . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.4.3 Carbon Dioxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
3.4.4 Mineral Nutrients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4 Tropical Forests. II. Ecophysiological Responses to Light . . . . . . . . . . . 103

4.1 Light Responses of Photosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.1.1 Light-Response Characteristics of Sun and Shade Plants . . . 103
4.1.2 The Photosynthetic Apparatus:
Pigments, Enzymes and Nitrogen . . . . . . . . . . . . . . . . . . . . . . . 107
4.1.3 The Origin of High-Irradiance Stress
and General Plant Responses . . . . . . . . . . . . . . . . . . . . . . . . . . 118
4.1.4 Dissipation of Excitation Energy in the Form of Heat:
The Role of Xanthophylls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
4.1.5 Damage and Repair of Reaction Centres of Photosystem II:
The D1 -Protein . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
4.1.6 Conclusion: Summarizing Mechanisms of Dissipation
of Photosynthetic Excitation Energy . . . . . . . . . . . . . . . . . . . . 125
4.1.7 Dissipation of Excitation Energy in the Form
of Fluorescence: A Tool in Plant Ecophysiology . . . . . . . . . . 126
4.2 Varying Irradiance on the Forest Floor and in Gap Dynamics . . . . . . 131
4.2.1 The Response to Light Flecks . . . . . . . . . . . . . . . . . . . . . . . . . . 131
4.2.2 Light Quality: Signalling Functions of Light . . . . . . . . . . . . . 136
4.3 Seedlings: Germination, Establishment and Growth . . . . . . . . . . . . . . 138
4.3.1 Regulation of Seed Dormancy and Germination . . . . . . . . . . . 138
4.3.2 Growth of Seedlings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
5 Tropical Forests. III. Ecophysiological Responses to Drought . . . . . . . . 149

5.1 Drought in Moist Tropical Forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
5.2 Drought in Dry Tropical Forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
5.2.1 Leaf Shedding and Hydraulic Architecture . . . . . . . . . . . . . . . 151
5.2.2 Ecophysiological Responses of Plants
with C3 -Photosynthesis
and Crassulacean Acid Metabolism (CAM) . . . . . . . . . . . . . . 153
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
Contents xiii
6 Tropical Forests.
IV. Lianas, Hemi-Epiphytes, Epiphytes and Mistletoes . . . . . . . . . . . . . . 165
6.1 The Conquest of Space:
Cryptogams and a Diversity of Life Forms of Vascular Plants . . . . . . 165
6.2 Cryptogams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
6.2.1 Bacteria and Cyanobacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
6.2.2 Bryophytes and Lichens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
6.3 Lianas, Climbers, Vines and Hemi-Epiphytes . . . . . . . . . . . . . . . . . . . 175
6.4 Epiphytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
6.5 Mistletoes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
6.6 Stressors Driving Ecophysiological Adaptation
of Epiphytes and Hemi-Epiphytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
6.6.1 Light and the Evolution of Plants to Epiphytism . . . . . . . . . . 191
6.6.2 Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
6.6.3 Mineral Nutrients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
7 Tropical Forests. V. Mangroves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

7.1 Phytogeography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
7.2 Site Characteristics and Contrasts in Salinity . . . . . . . . . . . . . . . . . . . . 232
7.3 Morphological Characteristics of the Mangrove Tree Life Form . . . . 233
7.3.1 Hypoxia in Inundated Swampy Soils,
Root Morphology and Aeration . . . . . . . . . . . . . . . . . . . . . . . . 233
7.3.2 Hydraulic Architecture and Xylem Sap Flow . . . . . . . . . . . . . 236
7.3.3 Vivipary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
7.4 Exclusion, Inclusion and Excretion of Salt . . . . . . . . . . . . . . . . . . . . . . 237
7.5 Photosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
7.5.1 CO2 -Exchange and Stomatal Conductance . . . . . . . . . . . . . . . 246
7.5.2 Water Use Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
7.5.3 High Irradiance, Photoinhibition and Oxidative Stress . . . . . 253
7.5.4 Interacting Factors: Salinity, Irradiance, Elevated CO2 . . . . . 256
7.6 Nutrition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
7.7 Aquatic Communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
7.7.1 Macroalgae in Mangroves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
7.7.2 Microbial Mats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
7.8 Mangroves as Endangered Ecosystems with Numerous Benefits
for Man and the Need for their Conservation . . . . . . . . . . . . . . . . . . . . 259
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
8 Ecosystems of Coastal Sand Plains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265

8.1 Restingas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
8.1.1 Geological History and Vegetation Physiognomy . . . . . . . . . . 265
8.1.2 The Nurse Plant Syndrome
and Dynamics of Vegetation Islands . . . . . . . . . . . . . . . . . . . . 266
8.1.3 Ecophysiology of Photosynthesis of Restinga Plants . . . . . . . 268
xiv Contents
8.2 Salinas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269

8.2.1 Formation of Coastal Salt Marshes
and Vegetation Physiognomy . . . . . . . . . . . . . . . . . . . . . . . . . . 269
8.2.2 Dynamics of Vegetation Islands . . . . . . . . . . . . . . . . . . . . . . . . 275
8.2.3 Strategies of Adaptation of Plants
in the Different Vegetation Units . . . . . . . . . . . . . . . . . . . . . . . 277
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
9 Savannas. I. Physiognomy, Terminology and Ecotones:

Why Do Savannas Exist? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
9.1 Physiognomy and Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
9.2 Seasonality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
9.3 The Savanna Problem: Why Do Savannas Exist? . . . . . . . . . . . . . . . . 305
9.4 Ecotones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
9.4.1 Savanna-Forest Ecotone Dynamics . . . . . . . . . . . . . . . . . . . . . 306
9.4.2 Savanna-Desert Ecotone Dynamics:
The Sahel Problem as a Case Story . . . . . . . . . . . . . . . . . . . . . 308
9.5 Productivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
10 Savannas. II.
The Environmental Factors Water, Mineral Nutrients and Fire . . . . . . 313
10.1 The Water Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
10.1.1 Grasses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
10.1.2 Trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
10.2 The Nutrient Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
10.2.1 Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
10.2.2 Phosphorus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
10.2.3 Biotic Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345
10.2.4 The Aluminium Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
10.3 The Fire Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361
10.3.1 The Causes of Fire: Anthropogenic and Natural . . . . . . . . . . . 361
10.3.2 Pyrophytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364
10.3.3 Burning by Man: Losses and Gains . . . . . . . . . . . . . . . . . . . . . 365
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
11 Inselbergs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
11.1 Physiognomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
11.2 Cryptogams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382
11.2.1 Cyanobacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382
11.2.2 Lichens and Mosses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400
11.3 Vascular Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401
11.3.1 Diversity and Life Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401
11.3.2 Physiological Ecology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401
11.4 Desiccation Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404
Contents xv
11.4.1 Cyanobacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404

11.4.2 Lichens and Bryophytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404
11.4.3 Vascular Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413
12 Páramos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419
12.1 Summer Every Day, Winter Every Night . . . . . . . . . . . . . . . . . . . . . . . 419
12.2 The Stress Factor Frost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422
12.3 Life Forms of Páramo Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
12.3.1 Giant-Rosette Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
12.3.2 Other Life Forms:
Tussocks, Cushions, Acaulescent Rosettes, Sclerophylls . . . . 429
12.3.3 Cacti . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
12.4 Frost Resistance in Giant-Rosette Plants . . . . . . . . . . . . . . . . . . . . . . . 433
12.4.1 Afro-Alpine Plants: Freezing Tolerance . . . . . . . . . . . . . . . . . . 433
12.4.2 Andean Plants: Freezing Avoidance . . . . . . . . . . . . . . . . . . . . . 435
12.4.3 Comparison of the Strategies
of Freezing Tolerance and Avoidance . . . . . . . . . . . . . . . . . . . 435
12.5 Other Stress Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438
12.5.1 Water Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438
12.5.2 Mineral Nutrition and Carbon . . . . . . . . . . . . . . . . . . . . . . . . . . 438
12.5.3 Irradiance and Heat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
Scientific Name Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449

Chapter 1
Introduction
1.1 Historical Background of Ecophysiology
We aim
• to start by depicting habitats and plants physiognomically,
• to deduce problems from such observations in the field as are suited for physi-
ological, biochemical and biophysical and perhaps even molecular experimenta-
tion in the laboratory, and
• to return from the laboratory to the field with increasingly sophisticated tech-
nologies for measurements and analyses applicable to field conditions.
With this we follow a great tradition, which was begun early in the last century, and
which we may retrace from Mägdefrau’s History of Botany (1992) as follows.
The title of one of the best-known essays (1806) by Alexander von Humboldt is
Ideas for a Physiognomy of Plants (“Ideen zu einer Physiognomik der Gewächse”).
He realized that the physiognomy of
vegetation is determined by environmental conditions and that the distribution of
plants depends on the climate and, thus he became the founder of plant geography
(von Humboldt ed 1989). The selective pressure exerted by variation in environ-
mental factors then also became the most essential aspect for explanation of natural
selection in
Charles Darwin’s theory of evolution (Darwin 1859). However, it was Ernst
Haeckel who coined the term ecology in 1866. Stephen J. Gould (1977), the sharp
American essayist in phylogeny, caricatured it as follows:
“Ernst Haeckel, the great popularizer of evolutionary theory in Germany, loved to coin
words. The vast majority of his creations died with him half a century ago, but among the
survivors are ‘ontogeny’, ‘phylogeny’, and ‘ecology’. The last is now facing an opposite fate
– loss of meaning by extension and vastly inflated currency. Common usage now threatens
to make ‘ecology’ a label for anything good that happens far from cities or anything that
does not have synthetic chemicals in it”.
However, Ernst Haeckel’s own original definition of ecology was already wide
and may, in fact, encompass much of the current application of the term, since he
2 1 Introduction
Fig. 1.1A,B The hydrothermal partitioning of the tropics after Lauer (1975). A The warm tropics
(hatched) and the cold tropics (black). B The wet tropics (hatched) and the dry tropics (dotted)
wrote that ecology is “. . . the entire science of the relations of the organism to its
surrounding environment, comprising in a broader sense all conditions of its exis-
tence”.1
Andreas Franz Wilhelm Schimper (1856–1901) founded Plant Geography on
a Physiological Basis with this title of his famous text published in 1898, and he
coined the term tropical rainforest. Simon Schwendener (1829–1919) suggested that
the relations between the environment and the morphological traits of plants
are best studied in areas providing extreme conditions. His advice has found many
followers to date. It was, however, Ernst Stahl (1848–1919) who introduced exper-
imentation to ecological research and thus founded physiological ecology. He also
discovered the role of stomata in transpiration and photosynthesis. Among the late
scientists of the twentieth century Arthur George Tansley (1871–1955), Otto Stocker
(1888–1979), Arthur Pisek (1894–1975), Heinrich Walter (1898–1989), Bruno Hu-
ber (1899–1969) and Michael Evenari (1904–1989) all stimulated the development
of physiological ecology. Physiological ecology primarily addresses itself to aute-
cology, i.e. ecological relations of properties and performance of individual plants
or species, but must also develop with synecology, i.e. consideration of integration
of plant functions at the community level (Lüttge and Scarano 2004), which was
already made clear by the early pioneers such as Arthur George Tansley who
1“. . . die gesamte Wissenschaft von den Beziehungen des Organismus zur umgebenden Außen-
welt, wohin wir im weiteren Sinne alle Existenzbedingungen rechnen können.”
1.2 The Tropics 3
“made it clear that synecology and autecology are subsumed in the study of the community
by methods firstly descriptive and appreciative, and secondly, analytic and experimental”
(quoted from Godwin 1977).
1.2 The Tropics
Applying the approach of physiological ecology to understanding ecological func-

tions and relationships in the tropics requires that we first must define what we
mean by tropics.
Fig. 1.2A,B Diurnal (A) and

annual temperature cycles (B)
at a station in the warm trop-
ics (Manaos at the Amazonas,
Brazil) and in the cold tropics
(Quito at 2,660 m a.s.l. in
the Andes, Ecuador) and at
stations in the temperate zone
(Paris and Milan, respec-
tively). The diurnal cycles in
A are given for the warmest
months as indicated by solid
lines (July, September and
July for Manaos, Quito and
Paris) and the coldest months
are depicted by dotted lines
(June, November and January
for Manaos, Quito and Paris).
(After Lauer 1975)
4 1 Introduction
Volkmar Vareschi (1980) has listed several possible definitions of the tropics,
which partially overlap but give different emphasis:
• geodetically the tropics are limited by the lines of latitude 23◦ 27 north and south
of the equator, i.e. the Tropic of Cancer and the Tropic of Capricorn respectively;
• climatologically the tropics are the zones of equal day and night length; they
are not basically characterized by high temperature and moisture; depending on
altitude, warm and cold tropics can be distinguished, and depending on the
precipitation regimes in the region between the equator and the two lines of lati-
tude at 23◦ 27 north and south, wet and dry tropics can be distinguished (Lauer
1975; Fig. 1.1);
Fig. 1.3A–C Global distribution of savannas (A) and tropical rainforest (B) and optimum carbon
fixation (C). (After Vareschi 1980, A, C with permission of R. Ulmer; Walter and Breckle 1984,
B with permission of S.-W. Breckle and G. Fischer-Verlag)
1.3 Tropical Forests and Savannas: General Importance 5
• phyto-geographically the tropics are indicated by the distribution of palms;

• eco-geographically the tropics are the zone in which the climatic effects of day-
night cycles are far more important than those of seasonal cycles; day-night cy-
cles of temperature are much larger than in the temperate zone in both the coldest
and the warmest months, but annual cycles of mean monthly temperature are al-
most absent in the tropics (Fig. 1.2);
• botanically the tropics are contained in a well-separated pantropical floristic
province;
• in terms of the biology of productivity, the tropics are the zones of optimal car-
bon fixation and photosynthetic capacity with > 600 g m−2 year−1 , which glob-
ally corresponds well to the occurrence of tropical rainforest in a broader sense
(Fig. 1.3).
1.3 Tropical Forests and Savannas: Their Emotional,

Commercial, Ecological and Scientific Importance
Forests and savannas are the ecosystems which cover the largest areas in the tropics
(Fig. 1.3). Notwithstanding the wide public concern about destruction and decline
of tropical forests, it appears that it is deeply ingrained in the nature of man to be
frightened by the unknown in the darkness of dense and repellent forests. There
were incidents of natural catastrophes in which falling trees and landslides with
forests threatened man, as a conequence of which a call arose to remove the forests
altogether. The contrast in our emotional reactions towards forest and savanna, re-
spectively, is vividly expressed by Alexander von Humboldt (1982) in his Journey
to South America:
“If one has spent many months in the dense forests along the Orinoco, if one got used there
to seeing the stars only near the zenith like looking upwards from the bottom of a well as
soon as one leaves the bed of the river, then wandering over the steppes2 has something
pleasant and attractive in it. The new pictures which one perceives give deep impression;
like the Llanero one enjoys the feeling of being able to look around so well. However, this
comfort does not last long. If, wandering for eight to ten days, one gets used to the games
of mirages and the brilliant green of Mauritia bushes, which appear mile after mile, one
feels the desire for more variable impressions, one longs for the sight of the huge trees of
the tropics . . . ”3
2 Note that A. von Humboldt did not yet distinguish between “steppes” and “savannas”, which we
now consider as the grasslands of temperate and tropical zones respectively (Vareschi 1980; Walter
and Breckle 1984).
3 “Hat man mehrere Monate in den dichten Wäldern am Orinoko zugebracht, hat man sich dort
daran gewöhnt, daß man, sobald man vom Strome abgeht, die Sterne nur in der Nähe des Zenit
und wie aus einem Brunnen heraus sehen kann, so hat eine Wanderung über die Steppen etwas
Angenehmes, Anziehendes. Die neuen Bilder, die man aufnimmt, machen großen Eindruck, wie
dem Llanero ist einem ganz wohl, ‘daß man so gut um sich sehen kann’. Aber dieses Behagen
ist nicht von langer Dauer. Ist man nach acht- oder zehntägigem Marsch gewöhnt an das Spiel
der Luftspiegelung und an das glänzende Grün der Mauritiabüsche, die von Meile zu Meile zum
Vorschein kommen, so fühlt man in sich das Bedürfnis mannigfaltigerer Eindrücke, man sehnt sich
nach dem Anblick der gewaltigen Bäume der Tropen . . . ”.
6 1 Introduction
Commercially it is important to remember that most tropical countries have to

sustain increasingly large populations, savannas serving agriculture and forests pro-
viding resources, which are, however, renewable only to a limited extent. Although
original CO2 sequestration and hence gross primary productivity of tropical forests
is high, due to high rates of respiration and the rapid degradation of litter in the
tropical forests, net CO2 uptake and net productivity is much reduced. It may even
be lower than in a beech forest of the temperate zone (Fig. 1.4).
Large-scale deforestation proves to be irreversible (Medina 1991). Previous land
use determines successions and recovery. In the Central Amazon it was found that
sites which had been clear cut without subsequent use were dominated 6–10 years
later by the pioneer genus Cecropia and had a high diversity, while sites used for
pasture before abandonment were dominated by the pioneer genus Vismia and had
a lower diversity. Seed source, effects on soil and mineral depletion lead to a more
rapid return of primary forest species if deforestation is not followed by the use
as pastures before abandonment (Mesquita et al. 2001). Therefore, it is possible to
reconcile utilization and preservation of tropical ecosystems if human activities are
directed in the right way, as vividly summarized by Whitmore (1990). For exam-
ple, there are two types of shifting agriculture (slash-and-burn agriculture; Fig. 1.5).
One of them is destructive and unsustainable. It is an invasive system, where fields
are used until they are exhausted even for regrowth of secondary forest. The other
is a cyclic system, where clearings are used for cultivation for 1–2 years and then
left to recover so that they can be reused in due course during the cycle, without the
need for further clearing of forest. This is a sustainable mode of shifting agriculture.
Figure 1.6 shows the recovery of above-ground biomass of a wet tropical forest after
a slash and burn activity. Clearly, full recovery to a level comparable to mature orig-
inal forest may take 100–200 years if the magnitude of disturbance has not been too
Fig. 1.4 Comparison of gross and net productivity of a tropical rainforest in Thailand and a 60-
year-old beech forest in Denmark. (After data from Larcher 1980)
Fig. 1.5A,B Two types of shifting agriculture. A Sustainable cyclic system. B Destructive and
unsustainable invasive system. (After Whitmore 1990)
Fig. 1.6 Recovery of biomass after slash-and-burn activity in the region around San Carlos de
Rio Negro in Colombia and Venezuela. Data give composite biomass accumulation in secondary
forests. (Medina 1991)
large. However, if short-term cultivation after slash-and-burn action is followed by

fallow periods of 20 years or longer, recovery mechanisms of forest ecosystems re-
main intact, and long-term cyclic utilization under low population pressure remains
possible.
Scientifically, sustainable schemes of silviculture have been developed for tropi-
cal areas and should be extended and enforced, including appropriate methods of
logging and timber removal (Whitmore 1990). Afforestation and timber planta-
tions are established on degraded sites and may reduce the pressure on good nat-
ural forest. Trees exotic to the native tropical environments are frequently used for
such aforestations. In Ethiopia the introduction of Eucalyptus historically has been
8 1 Introduction
praised for having saved the country economically (Zewde 1992). In Brazil the total
area covered by plantations of Eucalyptus has increased 8-fold during 30 years in
the last century (Fig. 1.7). Ecologists have underlined the detrimental effects of ex-
otic tree plantations on the indigenous environment. The tremendous difference in
ecological quality between a more or less undisturbed woodland and an exotic tim-
ber plantation may become immediately clear at a glance (Fig. 1.8). However, there
are also great advantages where degradation of ecosystems leaves no other choice.
Some of the disadvantages and advantages are compared in Table 1.1: (i) potential
harmful effects on soil properties are juxtaposed by experience with propagation of
exotic tree species and silviculture, (ii) displacement of local native vegetation and
fauna must be compared with improved productivity, (iii) susceptibility to epidemic
diseases and pests can be counterbalanced by various nurse effects of exotic tree
plantations.
An appropriate forestry management can back up the advantages and an exotic
Eucalyptus plantation does not need to look as sterile as that of Fig. 1.8B (see
Fig. 1.9C,D). This is supported by the case story of a Eucalyptus saligna plantation
Area x106 ( ha ) 4
Fig. 1.7 Area covered by

plantations of Eucalyptus in 0
1965 ´75 ´85 ´95
Brazil from 1965 to 1995
(DaSilva et al. 1995) Year
Table 1.1 Comparison of disadvantages and advantages of exotic tree plantations
DISADVANTAGES ADVANTAGES
MANAGEMENT
Harmful effects on physical, chemical, biolog- Experience with propagation and silviculture
ical soil properties
PRODUCTIVITY AND DIVERSITY
Displacement of local native vegetation Initial fast growth and wood production
COMMUNITY RELATIONS
Susceptibility to epidemic diseases and pests Nurse effects
• microclimate
• reduction of erosion
• enhancement of litter and humus production
Fig. 1.8A,B Acacia woodland (A) and Eucalyptus plantations (B) in the Rift Valley, Ethiopia
in Ethiopia in a region where some remnants of original native forest of Podocarpus

falcatus are still preserved, but where large areas are deforested and deteriorated
(Feyera et al. 2002; Lüttge et al. 2003; Fetene and Beck 2004; Lemenih and Teketay
2004; Fritzsche et al. 2006). Figure 1.9A shows the Eucalyptus forest in the back-
ground of an old solitary female tree of P. falcatus, the seeds (Fig. 1.9B) of which
are dispersed by birds. As given by the data on ground cover, light penetration and
10 1 Introduction
Fig. 1.9A–D Forest of Eucalyptus saligna and Podocarpus falcatus, Shashemene-Munessa State
Forest, Ethiopia. A Female tree of P. falcatus with Eucalyptus forest in the background. B Seeds
on P. falcatus. C,D Various stages of young growth of P. falcatus inside the Eucalyptus forest
the number and density of naturally regenerated native woody species summarized
in Table 1.2, the E. saligna plantation can be favourably compared with the adjacent
native forest and the nurse effect of E. saligna is evident. The Eucalyptus is coppiced
regularly about every 7 years which gives a certain advantage to regenerating native
woody species. Moreover, physiological data comparing E. saligna and P. falcatus,
the major native tree the regeneration of which is of greatest concern, also show that
the Podocarpus can compete well with the Eucalyptus within the plantation. Pho-
tosynthetic electron transport rates are not pronouncedly lower in P. falcatus than
in E. saligna in the plantation, but the Eucalyptus needs very much more water as
shown by its higher transpiration rates, JH2 O , and leaf conductance for water vapour,
gH2 O , and its lower water use efficiency given by the higher 13 C value of its leaves
inside the plantation (for explanation of the latter see Sect. 2.5). Indeed, P. falcatus
can readily regrow from seeds inside the Eucalyptus plantation (Fig. 1.9C,D). This
suggests how an original native forest can be regenerated making use of the nurse
function (Table 1.1) of an exotic tree plantation. In the case described here, P. falca-
1.4 The Destruction of Tropical Forest 11
Table 1.2 Comparison of a plantation of Eucalyptus saligna and an adjacent native forest (A) and
of some ecophysiological data of the Eucalyptus and of the major and most conspicuous native
tree, the dioecious gymnosperm Podocarpus falcatus (B) in the Shashemene-Munessa State For-
est, eastern escarpment of the Great Rift Valley, Ethiopia (7◦ 13 N, 8◦ 37 E). ETR1000 = apparent
photosynthetic electron transport rate at an irradiance of 1,000 µmol m−2 s−1 obtained from in-
stant measurements at ambient irradiation, ETRmax = maximum apparent photosynthetic electron
transport rate obtained from measurements of light dependence curves (see Sect. 4.1.1, Box 4.6);
for JH2 O and gH2 O of E. saligna the first number is of the adaxial and the second number of the
abaxial leaf surface. (Data from Feyera et al. 2002, Lüttge et al. 2003, Lemenih and Teketay 2004)
A E. saligna plantation Native forest

Range of ground cover (%) 11–100 25–100
Naturally regenerating native woody species 9,658
(number/ha) in relation to age of plantation
11 years 3,575
22 years 10,100
27 years 18,650
Light penetration (% of full sunlight) 12–51 1–77
B E. saligna P. falcatus
ETR1000 (µmol m−2 s−1 ) 97 77
ETRmax (µmol m−2 s−1 ) 87 76
JH2 O (mmol m−2 s−1 ) 2/6 1
gH2 O (mmol m−2 s−1 ) 67/200 37
13 C (‰) 23.1 20.7
tus was never seen to germinate in the open secondary grassland but many seedlings
and young trees were found in the plantation (Fig. 1.9C,D).
The ecological and scientific importance of savannas and forests in the tropics
will be major topics of this book. There is a large diversity of unique grassland and
forest ecotypes, which require physiognomic classification and understanding at the
ecophysiological level. This is a prerequisite for maintaining or, perhaps more pes-
simistically, for reattaining an equilibrium between commercial necessities includ-
ing agriculture and forestry and the preservation of unique natural environments. It
must be noted, though, that no measures, even the most sophisticated schemes of
sustainable land use based on scientific understanding of tropical biotopes and their
components will be effective if man does not succeed in mastering the major domi-
nating problem which is his own unlimited reproduction. Particularly in the tropics,
increasing human population is the one major factor continuing to put pressure on
the environment.
1.4 The Destruction of Tropical Forest
Of particularly wide public concern is the ongoing destruction of tropical rainforest.

Estimations of the current destruction vary to some extent due to the application of
different definitions of what actually is meant by tropical rainforest. This applies
12 1 Introduction
both to the question of what are the tropics (see above, Sect. 1.2), and to the ques-
tion of what is rainforest. (The latter is discussed in Sects. 3.1. and 3.2.) Thus, in
analyzing the problem, one may find oneself confronted with different quantitative
estimates.
For example, by taking evergreen and semi-evergreen forests with:
• no less than 100 mm rain in any month during 2 out of 3 years;
• an average yearly temperature of 24 ◦ C without any occurrence of frost;
• an altitude of < 1,300 m above sea level, excepting Amazonia with < 1,800 m
and SE-Asia with < 750 m,
and combining 13 critical and 3 very critical regions (as described in Fig. 1.10),
respectively Myers (1988), arrives at the data given in Fig. 1.10, or globally at the
following numbers:
• forest preserved in 1980: 10 ×106 km2, covering 6 – 7% of the total land surface
of the earth;
• disturbed: 0.1 ×106 km2 per year;
• destroyed: 0.08 to 0.09 ×106 km2 per year.
Fig. 1.10A, B Original and remaining forest in 13 critical regions. A Madagascar, Atlantic coast of
Brazil, West Ecuador, Columbian Chocó, West Amazonian Highlands, Rondonia/Acre in Brasilian
Amazonia, montane forests in Tanzania/Kenya, Eastern Himalaya, Sinharaja Forest in Sri Lanca,
Malaysian peninsula, NW-Borneo, Philippines and New-Caledonia, and in three particularly criti-
cal regions: B Madagascar, Atlantic coast of Brazil and West Ecuador. (After data of Myers 1988)
1.4 The Destruction of Tropical Forest 13
Using somewhat different definitions of tropical forests, Jacobs (1988) arrives at the
following figures for the annual destruction:
• tropical rainforest in a narrow sense: 0.15 ×106 km2 per year;
• wet tropical forest: 0.24 ×106 km2 per year.
Unfortunately no significant improvements have been achieved to date on this tragic
situation documented in the late 1980s. For example, in Brazil the Atlantic rain
forest is now reduced to 7.5% of its original area (Myers et al. 2000), the Amazon
area in the year 2004 experienced the highest deforestation rate ever and only 16%
is still unharmed (Fearnside 2005). The major remaining areas covered with wet
tropical forest are in the Zaire basin, in West Brazil and Amazonia, in the Guayana
highlands and in New Guinea.
Some important global problems relate to the destruction of tropical forest with
regard to
• the CO2 budget of the atmosphere;
• the water balance;
• the nutrient balance;
• biodiversity,
the first two and the last of which are also causing considerable public anxiety.
Scientifically, the effects on CO2 budgets remain a subject of debate because it
is not clear whether alternative CO2 fixation processes in terrestrial ecosystems will
offset or even over compensate reductions due to loss of forest. Secondary vegeta-
tion may prove to be an increasingly strong CO2 sink, and increasing CO2 in the
atmosphere may be coupled to higher ecosystem productivity (Medina 1991; Plant,
Cell and Environment 1991). The relations of CO2 with mineral nutrition, especially
nitrogen, with guard cell sensing and transpiration and respiration, with tempera-
ture, with plant acclimation, and with the respective functions of forests and oceans
as CO2 sinks are highly complex non-linear interactions in feedback networks and
simple conclusions are not possible (Plant, Cell and Environment 1999).
The water balance of large areas may be severely impaired by deforestation.
For equatorial forests in Amazonia it has been shown by stable-isotope techniques
(see Sect. 2.5) that 50% of total incoming rainfall was lost again by evapotranspira-
tion from the forest. Thus, deforestation not only increases total runoff of water but
also disturbs recirculation, as observed in the Amazon basin, leading to lower total
rainfall and more pronounced seasonality (Medina 1991).
These observations also have implications for nutrient supply: due to the rapid
turnover of nutrients in tropical forests (see Sect. 3.4.4) and problems of erosion,
deforestation causes major destruction of soil systems and affects nutrient budgets.
Tropical humid forests are known to be the most diverse ecosystems in the world
(see Sects. 3.2.1 and 3.3.1). They are thought to support more than 50% of all plant
and animal species. Deforestation leads to loss of diversity, which has not been fully
assessed by census to date.
In conclusion, we are not able to predict the actual nature of all the changes
that may result from the more or less complete destruction of these forests (Whit-
more 1990). The theory of deterministic chaos (Sect. 3.3.3) suggests that long-term
14 1 Introduction
predictions about the behaviour of complex systems with feedback relations show-
ing non-linear behaviour are intrinsically impossible (Hastings et al. 1993; Schuster
1984). However, it is equally clear that if we do not succeed in preserving these
forests, we shall lose one of our greatest treasures.
References
Darwin C (1859) On the origin of species by means of natural selection. John Murray, London
DaSilva MC Jr, Scarano FR, DeSouza Cardell F (1995) Regeneration of an Atlantic forest forma-
tion in the understory of a Eucalyptus grandis plantation in south-eastern Brazil. J Trop Ecol
11:147–152
Fearnside PM (2005) Deforestation in Brazilian Amazonia: history, rates and consequences. Con-
serv Biol 19:680–688
Fetene M, Beck E (2004) Water relations of indigenous versus exotic tree species, growing at the
same site in a tropical montane forest in southern Ethiopia. Trees 18:428–435
Feyera S, Beck E, Lüttge U (2002) Exotic trees as nurse-trees for the regeneration of natural trop-
ical forests. Trees 16:245–249
Fritzsche F, Abate A, Fetene M, Beck E, Weise S, Guggenberger G (2006) Soil-plant hydrology of
indigenous and exotic trees in an Ethiopian montane forest. Tree Physiol 26:1043–1054
Godwin H (1977) Sir Arthur Tansley: the man and the subject. J Ecol 65:1–26
Gould SJ (1977) Ever since Darwin. Reflections in natural history. Penguin Book Harmondsworth
Haeckel E (1866) Generelle Morphologie der Organismen: Allgemeine Grundzüge der organi-
schen Formenwissenschaft. 2 volumes. Georg Reimer, Berlin
Hastings A, Hom CL, Ellner S, Turchin P, Godfray HCJ (1993) Chaos in ecology: is mother nature
a strange attractor? Annu Rev Ecol System 24:1–33
Humboldt A von (1982) Südamerikanische Reise. 1808, quoted after the edition of Greno. Ver-
lagsgesellschaft mbH, Nördlingen
Humboldt A von (1989) Schriften zur Geographie der Pflanzen. In: Alexander von Humboldt Stu-
dienausgabe Band I, Beck H (ed) Wissenschaftliche Buchgesellschaft, Darmstadt
Jacobs M (1988) The tropical rain forest. Springer, Berlin Heidelberg New York
Larcher W (1980) Ökologie der Pflanzen auf physiologischer Grundlage. Ulmer, Stuttgart
Lauer W (1975) Vom Wesen der Tropen. Klimaökologische Studien zum Inhalt und zur Ab-
grenzung eines irdischen Landschaftsgürtels. Akad Wiss Lit Mainz Abh Math Naturwiss Kl
1975/3:5–52
Lemenih M, Teketay D (2004) Restoration of native forest flora in the degraded highlands of
Ethiopia: constraints and opportunities. Sinet Ethiop J Sci 27: 75–90
Lüttge U, Scarano FR (2004) Ecophysiology. Rev Bras Bôt 27:1–10
Lüttge U, Berg A, Fetene M, Nauke P, Peter D, Beck E (2003) Comparative characterization of
photosynthetic performance and water relations of native trees and exotic plantation trees in
an Ethiopian forest. Trees 17:40–50
Mägdefrau K (1992) Geschichte der Botanik. Leben und Leistung großer Forscher. 2. Aufl. G Fis-
cher, Stuttgart
Medina E (1991) Deforestation in the tropics. Evaluation of experiences in the Amazon basin fo-
cussing on atmosphere-forest interactions. In: Mooney HA et al. (eds) Ecosystem experiments.
John Wiley, New York, pp 23–43
Mesquita RCG, Ickes K, Ganade G, Williamson GB (2001) Alternative successional pathways in
the Amazon Basin. J Ecol 89:528–537
Myers N (1988) Tropical forests and the botanists’ community. In: Greuter W, Zimmer B (eds)
Proc XIV Int Botanical Congr. Koeltz, Königstein, pp 291–300
Myers N, Mittermeier RA, Mittermeier CG, Fonseca GAB, Kent J (2000) Biodiversity hotspots
for conservation priorities. Nature 403:853–858
References 15
Plant Cell and Environment (1991) Special issue, elevated CO2 -levels, vol 14, no 8. Blackwell
Science Ltd., Oxford
Plant Cell and Environment (1999) Special issue, plant responses to rising CO2 , vol 22, no 6.
Blackwell Science Ltd., Oxford
Schimper AFW (1898) Pflanzengeographie auf physiologischer Grundlage. Jena
Schuster HG (1984) Deterministic chaos. Physik Verlag, Weinheim
Vareschi V (1980) Vegetationsökologie der Tropen. Ulmer, Stuttgart
Walter H, Breckle S-W (1984) Ökologie der Erde, vol 2. Spezielle Ökologie der tropischen und
subtropischen Zonen. G Fischer, Stuttgart
Whitmore TC (1990) An introduction to tropical rain forests. Oxford University Press, Oxford
Zewde B (1992) A history of modern Ethiopia 1855 – 1974. Addis Ababa University Press, Addis
Ababa
Chapter 2
Large-Scale Sensing and Diagnosis in Relation
to the Tropical Environment
2.1 Approaches
In the discussion of “global change” it has become increasingly important to develop

means allowing large-scale conclusions about the conditions and the behaviour of
ecosystems or biomes. Indeed, techniques for examination and detailed structural
analysis of the surface of our globe are continuously advanced, allowing the inte-
gration of observations in space and time. This is, of course, applicable throughout
the globe. However, it is particularly relevant for the tropical environment, with ex-
tended tracts of ecosystems like savannas and forests (see Sect. 1.3), which are often
difficult to penetrate on the ground.
Among the techniques applicable to the study of large-scale effects of plant life
we must distinguish between those where the analytical method itself directly cov-
ers wide spaces and areas, such as remote sensing (Sect. 2.3), and those where it
is the method of sampling, which can be developed to cover extended areas and
thus, indirectly, affords comprehension at large-scale levels. In the former case only
a smaller choice of analytical techniques mainly based on the use of radiation will
be available, while in the latter case almost any analytical method may prove useful
depending on availability of suitable samples. Clearly, there are many approaches
for large-scale sensing and diagnosis of the environment, some of which shall be
discussed below.
2.2 Climatic Relations and Vegetation Modeling
In the first volume of his “Kosmos” published in 1845 Alexander von Humboldt
(Humboldt 2004) vividly defined climate as follows:
“The term climate in its most general meaning defines all changes in the atmosphere, which
perceptibly affect our organs: temperature, humidity, changes of barometric pressure, calm
state of the air or effects of changeable winds, the magnitude of electrical charge, the clean-
ness of the atmosphere or its mixing with gaseous exhalations adverse to a smaller or larger
18 2 Large-Scale Sensing and Diagnosis in Relation to the Tropical Environment
extent, and finally the degree of habitual transparency and clarity of the sky; which is not
only important for an increased heat radiation of the soil, the natural development of plants
and the ripening of fruits, but also for the sensations and the entire spiritual temper of man.”1
Consequently, Walter and Breckle (1983) argue that “the climate . . . is the only
primary factor, which affects the other factors like the soil and the vegetation and
to a lesser degree also the fauna, but which in turn is influenced by them only to
a limited extent in the range of micro-climate”. Thus, Heinrich Walter developed
the concept of biomes and the technique of the Klimadiagramm (Sect. 2.2.1) and
more recently global vegetation modeling is based on climatic factors (Sect. 2.2.2).
2.2.1 The Klimadiagramm
As the primary producers of biomass, plants determine the physiognomy and char-
acter of large ecological units and provide the basis for all other life. This has con-
tributed greatly to the practice of using patterns of vegetation for a coarse global
division of the geo-biosphere in ecological terms. On the other hand, Walter and
Breckle (1983) argue that the large naturally occurring communities of flora and
fauna, or “biomes”, are best delineated on the basis of the climatic conditions under
which they occur, since in global terms a coarse division of geographical regions
is given by the large climatic zones, from which Walter and Breckle (1983) then
derived their term zono-biomes. Thus, we also obtain the large zones of vegetation
in a three-dimensional climatic gradation (Ehrendorfer 1991):
• from the equator to the poles (temperature gradient);
• from the oceans to the continents (oceanity or continentality according to the
degree of annual balance of temperature);
• in the altitudinal zones in high mountains.
This possibility of separating large-scale vegetation units according to the climate
leads to the practical question of whether one can make predictions on plant distri-
bution in extended areas based on simple models.
One simple technique, which has proven extraordinarily successful, is the Klima-
diagramm of Walter (1973). In his autobiography Walter (1982) describes vividly
how the idea developed when he was first confronted with the problem of a large-
scale interpretation of vegetation in Anatolia (Turkey) in 1954. The Klimadiagramm
(Box 2.1) essentially uses simple data which are readily available from all weather
1 “Der Ausdruck Klima bezeichnet in seinem allgemeinsten Sinne alle Veränderungen in der At-
mosphäre, die unsere Organe merklich afficiren: die Temperatur, die Feuchtigkeit, die Verändrun-
gen des barometrischen Druckes, den ruhigen Luftzustand oder die Wirkungen ungleichnamiger
Winde, die Größe der electrischen Spannung, die Reinheit der Atmosphäre oder die Vermengung
mit mehr oder minder schädlichen gasförmigen Exhalationen, endlich den Grad habitueller Durch-
sichtigkeit und Heiterkeit des Himmels; welcher nicht bloß wichtig ist für die vermehrte Wärme-
strahlung des Bodens, die organische Entwicklung der Gewächse und die Reifung der Früchte,
sondern auch für die Gefühle und ganze Seelenstimmung des Menschen.”
2.2 Climatic Relations and Vegetation Modeling 19
stations, i.e. mean monthly temperatures and precipitation. They are plotted accord-
ing to a precise scheme of scaling on the ordinate vs months on the abscissa: humid
periods are indicated by areas on the graph, where the temperature curve is below
the curve of precipitation; arid periods are delineated by the precipitation curve
being lower than that of temperature. According to a precisely defined scheme, ad-
ditional information can be built into the Klimadiagramm, so that depending on data
availability each diagram may give a complete description of the climate at a given
station.
Box 2.1 Klimadiagramm after Walter (1973)

• The months of the year are plotted on the abscissa.
• The mean monthly temperatures and precipitation are plotted on the ordinate,
so that
– at mean monthly precipitation between 0 and 100 mm one unit of scale
corresponding to 10 ◦ C gives 20 mm precipitation or the ratio of scalation
is 1 ◦ C:2 mm precipitation;
– at mean monthly precipitation above 100 mm the precipitation scale is
reduced to 1/10, and the ratio is 1 ◦ C:20 mm precipitation.
• Humid periods are indicated by precipitation curves above temperature
curves; they are marked by vertical hatching up to 100 mm and by black
colour above 100 mm precipitation.
• Arid periods are indicated by precipitation curves below temperature curves,
they are marked by dotting.
• According to a well-defined scheme, other details may be added to the Kli-
madiagramm as indicated in the examples given below.
Box 2.1 (Continued)
Examples presented are for tropical stations in Africa with an arid, a perhumid
and a seasonal climate: Moçâmedes at the Atlantic coast of Angola (15◦ 05 S,
12◦ 09 E), Djolu, Congo (00◦ 38 N, 22◦ 37 E) and Mpika, Muchinga Moun-
tains, Rhodesia (11◦ 52 S, 31◦ 26 E). For further details see Walter and Lieth
(1967) and Walter (1973).
For a large-scale evaluation of a certain area, or even an entire continent, one

requires the diagrams of many stations covering the respective area. One can inte-
grate them in a geographic information system, e.g. in the simplest way paste them
on a map to obtain a good survey of the climatic structure of the area. Areas with
humid, arid or seasonal climates are readily separated. An example is shown for
2.2 Climatic Relations and Vegetation Modeling 21
the predominantly tropical continent of Africa in Fig. 2.1. The humid belt along the
equator is clearly separated from the more semi-arid and arid regions. The power
of the approach can be seen by comparing the distribution of rainforest and savanna
on the African continent (Fig. 1.3) with the more humid and more seasonal regions,
respectively (Fig. 2.1).
Although we will have to draw attention to certain limitations of the Klimadia-
gramm technique later (Sect. 3.1), since arid and humid climates are strictly defined
by precipitation and evaporation and not by precipitation and temperature as in the
Klimadiagramm, we will make repeated use of the Klimadiagramm in this book.
Fig. 2.1 Klimadiagramm map of the African continent. Arid periods are dotted, humid periods are
hatched or black. Arid, humid and seasonal regions are readily differentiated. Dashed horizontal
line: equator. (Walter and Breckle 1984; with kind permission of S.-W. Breckle and G. Fischer-
Verlag)
2.2.2 Vegetation Modeling Based on Irradiance and Water Budgets
Climatic conditions like irradiance and temperature affect the water-vapour

pressure deficit of the atmosphere, and thus determine evapotranspiration or
the loss of water vapour of the vegetation to the atmosphere. Irradiance and water
Fig. 2.2 A Global prediction of physiognomic vegetation types on the basis of ecophysiological
models; compared with B actual observations. (Woodward 1987; with kind permission of the au-
thor and Cambridge University Press)
2.3 Remote Sensing Using Radiation 23
availability in turn are modulated by other climatic factors. Thus, it is possible to

make model calculations of evapotranspiration from climatic data using basic plant-
physiological principles of transpiratory water loss from leaves driven by the leaf/air
water-vapour pressure-difference. Furthermore, there are close links between evap-
otranspiration and leaf area index (LAI), which is the total leaf area related to
a unit of ground surface (see Sect. 3.4.1). The LAI is characteristically related to the
physiognomy of plants and vegetation. Therefore, from LAI one may then obtain
ecophysiological models of vegetation as dominated respectively by broad-leaved
trees, shrubs and herbs (Woodward 1987). Large-scale presentations of the models’
results may then be compared with real observations (Fig. 2.2).
2.3 Remote Sensing Using Radiation
Remote sensing of the biosphere is based on the analysis of electromagnetic radi-

ation (Hobbs and Mooney 1990). We may distinguish between measurements of
reflection or absorption of radiation and fluorescence. Radiation is also used in gas
analysis, but this is a somewhat different aspect (see Sect. 2.4).
2.3.1 Reflection and Absorption
The analysis of reflection, absorption and transmission of radiation by individ-

ual leaves, plants or by the vegetation canopy has become an important method in
ecophysiology. The principle relations are shown in Fig. 2.3. The range of photo-
synthetically active radiation (PAR: 400 – 700 mm) is largely identical to that of the
visible light. Here radiation absorption is dominating. Chlorophyll has an absorp-
tion minimum in the green range of the spectrum (550 nm), and this is identified by
reduced absorption, and increased reflection and transmission (Fig. 2.3). In the in-
frared range of the spectrum (above 800 nm) radiation reflection and transmission
are dominating. At very high wavelengths absorption increases again, although, this
is not so relevant as solar emission contains little radiation above 2,000 nm.
The contrast between absorption of the radiation in the visible range and reflec-
tion in the infrared range of the spectrum by green plants has been used to develop
a dimensionless vegetation index Q related to reflection between 580 and 680 nm
680 ) and between 725 and 1,100 nm (R 1100 ), respectively, as follows:
(R580 725
725 − R580
R 1100 680
Q= . (2.1)
725 + R580
R 1100 680
It results from this equation that at very low reflection between 580 and 680 nm and
very high reflection between 725 and 1,100 nm, vegetation is dense and Q tends
towards +1. In contrast, at very high reflection between 580 and 680 nm and very
Fig. 2.3 Relationship be-

tween reflection, absorption
and transmission of radia-
tion by a green leaf at varied
wavelengths. UV Ultraviolet;
PAR photosynthetically active
radiation; IR infrared. (After
Gates 1965; from Nobel 1983:
Biophysical Plant Physiology
and Ecology, Copyright 1983
by W.H. Freeman and Com-
pany; used with permission)
low reflection between 725 and 1,100 nm Q tends towards −1, indicting sparse
vegetation (Running 1990). The two values of R, R580 680 and R 1100 , can be measured
725
from aeroplanes or meteorological satellites equipped with two sensors for the re-
spective range of wavelengths. The results can be depicted on maps using false
colours, which provide informative images at the global level. Formations with par-
ticularly dense vegetation (e.g. the tropical forests and the extended forest regions
of the northern hemisphere) are readily distinguished from poorer areas like deserts,
steppes and savannas (Malingreau and Tucker 1987).
Thus, analysis of reflection and absorption with the rough vegetation index ob-
tained by comparing reflectance at two rather broad bands of wavelengths provides
information on a given state of vegetation and, if followed in time, also about its
dynamics. Although this type of analysis has been used successfully to predict har-
vests, it does not provide a real picture of the physiological state and vitality of
vegetation. This can be improved by high-altitude aircraft or space based imaging
spectroscopy with much higher spectral resolution of the solar radiation reflected
from the Earth’s surface in contiguous narrow bands. A National Aeronautics and
Space Administration (USA) Earth Observing (EO) device allows measurement of
reflected radiance in 242 spectral bands from 0.4 to 2.5 µm at a spatial resolution
of 30 m (Asner et al. 2004). The sampling of narrow bands of the optical spec-
trum allows deduction of more specific biochemical properties of canopies, e.g. by
giving a photochemical reflectance index (PRI), an anthocyanin reflectance index,
a spectroscopic water absorption index (Asner et al. 2004) and an index of nitrogen
concentration (Asner and Vitousek 2005) in canopies. With spectrometers of par-
ticularly high resolution gross biochemical composition of vegetation utilizing the
light absorption of quantitatively dominant organic compounds (e.g. sugars, cellu-
lose, starch, lignin, protein) can also be measured (Wessman 1990).
The PRI uses the wavelength of 570 nm, which is affected only by chlorophyll
absorbance, and the wavelength of 531 nm which is affected by both chlorophyll and
carotenoid absorbance (Nichol et al. 2006). Carotenoid absorbance is modulated by
the epoxidation-state of components of the xanthophyll cycle which is related to the

dissipation of surplus photosynthetic excitation energy as will be explained below
in the context of light in tropical forests (Sect. 4.1.4). PRI is calculated in different
ways in the literature, i.e. as
(R531 − R570 )
PRI = (2.2)
(R531 + R570 )
by Asner et al. (2004) and as
R570
PRI = R531 − + R570 (2.3)
R531
by Nichol et al. (2006). It allows an assessment of photosynthetic light-use effi-
ciency (LUE) of canopies. Using (2.3) in a study of an experimental mangrove
canopy good positive and negative correlations were obtained between PRI and ef-
fective quantum yield of photosystem II and non-photochemical quenching of pho-
tosynthetic excitation energy, respectively (Fig. 2.4), which are directly related to
LUE as will be explained in more detail in Sect. 4.1.7. The anthocyanin reflectance
index is obtained as
1 1
ARI = − (2.4)
R550 R700
and it is interesting because anthocyanin is an indicator of newly formed foliage be-
fore the full development of chlorophyll pigments. A spectroscopic water absorption
index uses the localized sensitivity of two major canopy water-absorption features
(930 – 1,040 and 1,100 – 1,230 nm).
The disadvantage in the analysis of radiation reflection is that clear skies are
needed. An alternative is the use of micro-wave radiometry, since micro-waves
are not absorbed by clouds. They give information on the heat-radiation of surfaces,
which is also much influenced by vegetation.
In principle remote sensing by detection of radiation allows analysis of many
vegetation parameters on a large scale, e.g.
• the vegetation density on the land surface according to the leaf area index and
the biomass;
• the density of plankton in the oceans;
• vegetation types and the structure and dynamics in ecosystems;
• photosynthetic activity and light-use efficiency;
• biochemical parameters of canopies;
• the productivity of agricultural and natural ecosystems;
• phenological cycles of growth;
• soil moisture;
• hydrology;
• changes by deforestation, fire, storm, erosion, pest infestations and other catas-
trophic events.
Fig. 2.4 Correlations between 0.00

the photochemical reflectance A
index (PRI) and (A) effective -0.02
quantum yield of photosystem
II (F/Fm ) and (B) non- -0.04
photochemical quenching
PRI
of photosynthetic excitation
-0.06
energy (NPQ) (from Nichol et
al. 2006)
-0.08
-0.10
0.2 0.3 0.4 0.5 0.6 0.7 0.8
∆F / F´m
0.00
B
-0.02
-0.04
PRI
-0.06
-0.08
-0.10
0.0 0.5 1.0 1.5 2.0 2.5 3.0
NPQ
In many of these examples analysis of temporal variations is essential, integrated

by repeated observation. Another important aspect is the need for “calibration”,
which may allow a more detailed interpretation of the radiation signals obtained in
remote sensing. Due to the complex structure and dynamics of ecosystems, signif-
icant efforts are required for calibration, involving measurements on a hierarchy of
levels with different resolution of area, e.g. on the ground, on measuring towers, in
aeroplanes and helicopters and in satellites (Sellers et al. 1990). With better calibra-
tion, interpretation of finer detail and more sophisticated resolution will be possible
from remote-sensing data. However, calibration or “ground truthing” itself is a great
problem, especially in inaccessible tropical areas.
2.3.2 Fluorescence
A way different from PRI (Sect. 2.3.1) for obtaining information on the photosyn-
thetic activity of vegetation is the analysis of fluorescence. The fluorescence from
chlorophyll is not only directly related to the concentration of chlorophyll but is
also inversely related to the efficiency of photosynthesis. This will be explained
in more detail later in relation to photosynthetic light use in tropical environments
(Sect. 4.1.7).
Remote fluorescence excitation is possible using powerful lasers. A laser-induced

fluorescence spectrum is given in Fig. 2.5. The fluorescence maxima at 690 nm
and 740 nm are particularly variable in response to stress as shown in Fig. 2.6 for
some conditions studied in the laboratory. As plants senesce fluorescence decreases
due to the degradation of chlorophyll (Fig. 2.6A). Conversely, when the photosyn-
thetic process is impaired, e.g. by K+ deficiency, drought stress or herbicide ac-
tion (Fig. 2.6B,C), fluorescence increases (Chapelle et al. 1984a). For a more sensi-
tive analysis it has also been suggested that the ratio of fluorescence at the peak of
690 nm and in the far-red region at 730 or 740 nm should be used (Hák et al. 1990;
Lichtenthaler et al. 1990).
Fig. 2.5 Laser-induced fluo-

rescence spectrum of maize
leaves (Chapelle et al. 1984b)
Fig. 2.6A–D Changes of laser-induced fluorescence due to various kinds of stress. A Senescence.
B Potassium deficiency. C Drought. D Action of the herbicide DCMU (dichlorophenyl-dimethyl-
urea) inhibiting photosynthesis. A, C, D soybean; B maize. (Chapelle et al. 1984a)
Fig. 2.7 Profile of laser-induced fluorescence emission (wavelength of fluorescence 685 nm) along
a flight path of 6 km above forests and fields. The bright profile gives fluorescence. The dark profile
indicates terrain elevation and at the same time the different parts of the landscape, i.e. green fields,
brown and freshly ploughed fields and trees. (After Hoge et al. 1983)
Thus, when used in remote sensing, fluorescence analysis allows large-scale di-
agnosis of stress effects by abiotic factors such as the availability of growth re-
sources (e.g. water, mineral nutrients, photosynthetically active radiation etc.) or
environmental pollutants and biotic factors, such as pests and pathogens. The res-
olution of analyses of laser-induced fluorescence of 685 nm during flights in me-
teorological aeroplanes flying at a height of 150 m and a nominal flight-speed of
100 m s−1 is between 10 and 80 m. An example is given in Fig. 2.7 for a flight path
of 6 km, resolving green and brown fields and forests.
2.4 Gas Analysis 29
2.4 Gas Analysis
Another means of analyzing the effects of plant life across a range of scales is that
of infrared gas-analysis (IRGA). It is of great importance because the method can
measure the gas exchange of plants, particularly respiratory and photosynthetic
CO2 exchange (but not O2 !) and also transpirational loss of water vapour. Many
molecules which play a role as environmental pollutants, such as sulfur dioxide,
nitrogen oxides, ammonia and carbon monoxide can also be analyzed. Therefore,
the IRGA is an important technique in ecophysiological studies encompassing pho-
tosynthesis as well as environmental control.
Box 2.2 Infrared active and non-active gases

An IRGA measures the specific absorption of infrared radiation with a char-

acteristic absorption spectrum for a particular gas. It is caused by an uneven distri-
bution of electrical charge in the gas molecules, i.e. a dipole moment, which can
be excited by infrared radiation. This is present in gas molecules which are com-
posed of two or more different atoms (Box 2.2). Moreover, gas molecules which are
built up of more than two identical atoms also develop a dipole moment due to the
oscillations of atoms against each other. For example in ozone, composed of three
Fig. 2.8 The principle of infrared gas analysis (IRGA): The gas analyzer consists of four gas-
filled chambers I–IV. The gas to be analyzed is passed through chamber II. It is often a gas stream
modified during photosynthesis by leaves in a gas-exchange cuvette (G) under photosynthetically
active radiation (PAR). The reference gas is contained in chamber I. An infrared radiation (IR)
incident on both chambers is absorbed in correlation to the gas concentrations in the chambers. The
radiation transmitted by chambers I and II heats up the gas contained in chambers III and IV below
them. Depending on the intensity of the transmitted radiation, which is influenced in chamber II
by the concentration in the measured gas, the gas in chambers III and IV is heated up to a larger
or smaller extent. This causes a pressure difference which moves the membrane of a membrane
condensor (M). The IR beams are chopped by a rotating disc (R), which interrupts the radiation
in both chambers in an equal rhythm. Thus, synchronized changes of capacity and potential are
obtained, which are amplified and rectified (V ) and given out on a recorder, into a data logger (D)
or on a digital read-out unit. With the appropriate choice of IR wave lengths corresponding to the
IR absorption spectrum of the gas species studied, different gases can be analyzed. Since chambers
III and IV also must contain the gas species to be measured, analysis of each gas species requires
a separate specific analyzer. Cross-sensitivities for other gases in some cases may also have to be
considered
2.4 Gas Analysis 31
oxygen atoms (O3 ), two of the atoms are always closer together and the distance to
the third atom is larger (induced dipole moment). Conversely, gases with two iden-
tical atoms are not sensitive to infrared radiation, because even during oscillation of
atoms the charge remains evenly distributed. (A list of important infrared active and
non-active gas molecules is given in Box 2.2.)
For the same reason which allows their detection by IRGA, the infrared-active
gases also contribute to the green-house effect, keeping the surface of the globe
warm by absorbing infrared heat radiation which would otherwise radiate out into
space. The most important gas in this respect is water vapour, without which the
earth would be unbearably cold. The recent increase of other atmospheric gases like
CO2 and methane (CH4 ) in the last hundred years is considered to threaten global
temperature balance.
Figure 2.8 shows how photosynthetic or respiratory gas exchange of individual
leaves or plants is measured in gas-exchange cuvettes. Studies at larger scales can be
made by taking samples during meteorological flights with aeroplanes or balloons,
which are later analyzed by IRGA-techniques. Such flights in the lower atmosphere
are increasingly valuable in the investigation of large-scale ecological problems. An
impressive example of the success of this approach is the study of gas exchange
above the canopy of a tropical forest by measuring CO2 -concentrations. It has been
demonstrated that high CO2 -concentrations build up above the canopy from respira-
tion during the night and are decreased in photosynthesis during the day (Fig. 2.9).
Large areas can also be investigated by ground-based gas-analysis systems span-
ning a range from meters up to kilometers. Such an instrument, for example, is the
Fourier-transformed-infrared-spectrometer (FTIR), which is equipped with a 60-cm
Fig. 2.9 Vertical profiles of

CO2 concentrations in the
atmosphere above a tropical
forest based on the infrared
analysis of gas samples taken
at three different times during
the morning. At 08.00 h
CO2 concentration is high
due to nocturnal respiration
of the plants; however, it
gradually decreases in the
course of the morning due to
photosynthesis, and at 12.20 h
it is clearly smaller than the
average CO2 concentration of
the troposphere at a little more
than 345 Pa/MPa. (After
Matson and Harriss 1988)
telescope and appropriate computer technology, and using mirrors can provide trace-
gas analyses over distances of up to 1.5 km (Fig. 2.10) (Gosz et al. 1988). Of course,
this approach is difficult in denser and taller vegetation. New laser sensors, which
operate not only in the infrared region of the spectrum but also in the visible and
ultraviolet region, allow direct gas analyses from onboard aeroplanes (Matson and
Harriss 1988; Matson and Vitousek 1990). Here, the troposphere at the surface of
our planet is taken as kind of a gas-exchange cuvette to consider gas-exchange be-
tween the biosphere and the atmosphere. The example of Fig. 2.11 shows the separa-
tion of different atmospheric CO2 concentrations above various tropical ecosystems
along a 50-km-long path of a flight in the morning.
Fig. 2.10 Infrared-radiation

source (IR) and Fourier trans-
form infrared spectrometer
(FTIR), which together with
telescopes and a mirror (or
a network of mirrors) allow
ground-based gas analysis
over longer distances. (After
Gosz et al. 1988)
Fig. 2.11 Horizontal profile of CO2 concentrations in the atmosphere above different ecosystems
in the tropics based on measurements on board of a meteorological aeroplane along a flight path
of a little more than 50 km between 08.30 and 08.43 h. The vegetation has increased the CO2
concentration due to nocturnal respiration. The values above rivers are much lower and correspond
to the average CO2 concentration in the troposphere (see also Fig. 2.9). (After Matson and Harriss
1988)
2.5 Stable Isotope Analysis 33
2.5 Stable Isotope Analysis
Many chemical elements in nature occur in the form of several isotopes, which for
a given proton content in the nuclei have slight differences in mass due to varied
neutron contents. In most elements one of the isotopes quantitatively predominates,
while the others have a much lower abundance.
Isotope techniques initially became well known through radioactive isotopes
serving as tracers to follow the path of certain elements in complex metabolic reac-
tion sequences and transport systems. Alternatively, radiocarbon dating uses the nat-
urally occurring carbon isotope 14 C, with a half-life of approximately 5,600 years,
and provides a means of dating over relatively recent geological time scales. More
recently, however, the analysis of stable isotopes has provided increasingly valuable
information on spatial and temporal variations in ecological and palaeohistorical
terms (Rundel et al. 1988; Máguas and Griffiths 2002). Table 2.1 provides informa-
tion on those stable isotopes which currently promise to provide the most important
information in ecology.
The differences in the physical and chemical properties of the various isotopes
of an element depend on the relative differences in mass and therefore they are
small. The exception is hydrogen with very light nuclei where the deuterium pro-
vides much greater isotope effects. However, the analytical methods are extraordi-
narily precise. The isotopes are analyzed using mass-spectrometry according to
their mass to charge ratio. The results of the analysis of a gaseous sample are not
expressed as absolute isotope contents but related to an internationally accepted
standard. Let x be a rare and x the most frequent isotope of an element, where
Table 2.1 A small selection of stable isotopes which promise to be important tools in ecology
(Ehleringer and Rundel 1989)
Element Isotope Abundance (%)
Hydrogen 1H 99.985
2H or D 0.015
Carbon 12 C 98.89
13 C 1.11
Nitrogen 14 N 99.63
15 N 0.37
Oxygen 16 O 99.759
18 O 0.204
Sulphur 32 S 95.00
34 S 4.22
Strontium 86 Sr 9.86
87 Sr 7.02
88 Sr 82.56
x may be heavier or lighter than x, one can obtain the isotope-ratio δx in ‰ as

follows:

(x /x)sample
δx = − 1 × 1000 . (2.5)
(x /x)standard
Thus, in fact, one is operating with isotope-ratios, which provides the high precision
in comparisons of different samples, expressed as a differential against the particular
standard.
Isotope effects are a measure of the behaviour of individual molecules contain-
ing different isotopes of an element and occur during diffusion or transport in vari-
ous media as well as in enzymatic reactions. The differences in the kinetic properties
and in the behaviour of molecules with different isotopic composition in thermody-
namic equilibria form the basis for many applications of the stable-isotope tech-
nique, including:
• geochemistry;
• hydrology;
• meteorology;
• paleoclimatology;
• biochemistry and physiology of metabolism;
• ecology and environmental research.
The technique originated from geochemistry. Its great importance in meteorology
may be illustrated by considering water. The vapour pressure of water is propor-
tional to the mass. Thus, heavy H2 O-molecules with the isotopes 2 H and 18 O in-
stead of 1 H and 16 O need higher temperatures to evaporate and are discriminated
against. However, at higher temperatures the absolute H2 O content in the gas-phase
is higher. During rain the heavier molecules then precipitate more readily than the
lighter ones, and hence, evaporation and precipitation provide a climate effect on
isotope composition of rain (Ziegler 1989):
• according to latitude;
the content of heavy isotopes 2 H and 18 O declines with increasing latitude fol-
lowing the temperature gradient;
• according to altitude;
the content of 2 H and 18 O declines with increasing altitude;
• according to seasons;
at higher latitudes (> 30◦) the rain contains more 2 H and 18 O in summer and less
in winter;
• according to continentality;
the 2 H and 18 O contents decrease with increasing distance from the coast;
• according to the amount of rain;
the 2 H and 18 O contents decrease with increasing amount of rain falling.
This explains hydrological isotope effects. The isotope composition of groundwa-
ter, flowing surface-water, and recent precipitation is different. Thus, although no
fractionation occurs during uptake by plants, organisms which take up such wa-
ter are also distinguished by their own isotope content. Although there are addi-
tional discrimination processes when water evaporates from leaves (enriching leaf
2 H and 18 O), a further fractionation occurs in favour of the heavy isotopes dur-
ing incorporation into organic material. By analysis of the most recently produced
biomass, allowing for certain exchange reactions one can conclude from which di-
rection the last rain falls or one can determine the geographical origin of plants and
food items (Smith 1975).
These climate effects have led to applications in palaeoclimatology. From the
isotope composition of fossil water in subterranean water reservoirs, which have
formed in geological periods, one can draw conclusions on the climate at that time.
Analysis of small gas inclusions in layers of ice in glaciers, or of water from the ice
itself, allow similar conclusions about the past. For example, in the ice of a glacier at
5,670 m a.s.l. in the Cordillera de Carabaya of the tropical Andes in Peru, relatively
low levels of 18 O [more negative values of δ 18 O, see (2.5)] indicate a warmer period
between 1000 and 1500 A.D. This was correlated with lower accumulation of ice,
although higher 18 O levels (less negative δ 18O), then represent a colder period from
1500 to about 1875 A.D. when ice-deposition was initially high (Fig. 2.12). Further
opportunities are provided by the incorporation of water into organic constituents of
plants, and the analysis of gross remnants of plants in peat accumulations or from
the study of isotope composition of annual rings in old trees and wood.
When δ 13 C signatures of several ecosystem components such as the air, the
plant biomass, fallen litter and the soil are studied systematically with a spatiotem-
poral resolution insights into the carbon dynamics, e.g. of tropical primary rain-
forests can be obtained (Buchmann et al. 2004). In extant current physiology of
water relations of plants the isotope effects of evaporation and diffusion of wa-
ter are useful for the study of transpiration and water use efficiency (WUE) dur-
Fig. 2.12A,B δ 18 O-values (A) [see (2.5)] and accumulation of ice (B) for the past 1000 years in
the glacier Quelcaya Cap, Cordillera de Carabaya, Peru (13◦ 56 S, 70◦ 50 W) related to values of
the year 1980. (After Jones 1990; with kind permission by La Recherche)
ing photosynthesis. Because the major constraint for transpirational loss of water
is the opening or closing state of stomatal pores 18 O-isotope enrichment in the leaf
biomass is a good measure of stomatal conductance integrated over the time of the
life span of the leaf sampled and there is a significant positive relationship with mean
transpiration rate (Sheshshayee et al. 2005). Since the opening/closing of stomata
likewise determines diffusion of atmospheric CO2 into the leaves, the isotope 13 C
has been much used as an indicator for time integrated stomatal conductance of
leaves, and in C3 -plants the variable rate of CO2 -diffusion via stomata primarily
determines overall changes in 13 C-discrimination during photosynthesis. Generally
with a small conductance, i.e. when stomata are more tightly closed, if photosyn-
thetic rate is maintained internal CO2 -concentration, pCO i , tends to be low. With
2
i
high conductance, i.e. when stomata are more opened, pCO2 approaches the value
of external CO2 -concentration, pCO o . Overall carbon isotope discrimination, , is
2
then proportional to the ratio of pCOi o
/ pCO and, hence by indirect association also
2 2
to leaf-conductance for water vapour, gH2 O , as follows:
o
pCO − pCO
i i
pCO i
pCO
=a 2
o
2
+b o
2
= a + (b − a) o
2
[‰] . (2.6)
pCO2
pCO2
pCO2
In this relationship a gives 13 C discrimination due to CO2 -diffusion in air (4.4‰)

and b the net fractionation caused by the carboxylation itself (ca. 27‰, i.e. b − a is
ca. 22.6‰) so that (2.6) becomes
i
pCO2
= 4.4 + 22.6 o [‰] . (2.7)
pCO 2
From the measurements of the carbon isotope ratios in dry plant material, δp (2.5),
and in air, δa , is calculated as follows:
δa − δp
= × 1,000 [‰] , (2.8)
1000 + δp
where δa and δp are given in ‰. Determined by the contribution from respiratory

CO2 the value for ambient air, δa , can vary between − 10.5 and − 7.5‰ especially
inside forests and depending on the height above ground. For normal bulk-air condi-
tions, one assumes a δa of −8‰ (Farquhar et al. 1989a,b; Broadmeadow et al. 1992;
Ehleringer 1993; Guehl et al. 2004). Thus, we can summarize the proportionally of
terms as follows:
i
pCO 1
∼ o
2
∼ gH2 O ∼ . (2.9)
pCO2
WUE
However, this approach is increasingly subject to critique because 13 C-isotope ef-

fects are also given by the activity of respiration as respiratory CO2 is enriched in
13 C (Ghashghaie et al. 2001), the activity of photosynthesis and the developmen-
tal state of the photosynthetic organs (Terwilliger et al. 2001). When the internal
conductivity of CO2 within leaves is different in different plants the relationships
of (2.9) are not allowing correct comparisons between species (Warren and Adams
2006). may also vary in response to light gradients within canopies (Medina et
al. 1991; Roux et al. 2001; Holtum and Winter 2005; Buchmann et al. 2004; see
Fig. 3.31). Nevertheless, in comparisons of C3 plants under critically comparable
conditions the application of these equations still provides very useful information
(Guehl et al. 2004).
Simultaneous analyses of 13 C and 18 O allow a distinction between effects of
stomatal diffusion and photosynthetic capacity. A change of pCO i
2
/ pCO
o
2
as indi-
cated by can be due to a change of gH2 O at constant maximum CO2 -assimilation,
Amax , or a change of Amax at constant gH2 O . The 18 O signature mirrors accumula-
tion of water in the photosynthetic organs and hence the contribution of stomatal
regulation (Adams and Grierson 2001). If other effects on the 18 O signature can be
excluded – e.g. habitat influences such as differences in the 18 O signature of the
water available to the plants – the simultaneous analysis of δ 13C and δ 18 O allows
a distinction between effects of gH2 O and Amax (Scheidegger et al. 2000). Positive
correlations between δ 13 C and δ 18O then indicate a dominant control by gH2 O at
relatively constant Amax where stomata operate in a broad range of opening and
closing from high transpiration rates (low δ 18O values) to low transpiration rates
(high δ 18O values). Strong variations of δ 13 C at low changes of δ 18 O, however, in-
dicate that carbon gain is mainly under assimilatory and not under stomatal control.
Far reaching diagnoses as attained by applications of the stable-isotope tech-
nique in ecology may result from isotope effects of metabolism. Several enzymes
discriminate against substrate molecules constituted of different isotopes leading
to different isotope composition of products. An important example is the fixa-
tion of CO2 for photosynthetic assimilation with discrimination occurring against
the heavier isotope 13 C as compared to 12 C. The enzyme ribulosebisphosphate
carboxylase (RuBPC), which directly leads to the formation of the C3 -compound
phosphoglyceric acid in the light (C3 -photosynthesis), discriminates against 13 CO2
more strongly than another carboxylating enzyme, phosphoenol-pyruvate carboxy-
lase (PEPC; Table 2.2), which forms the C4 -acid oxaloacetate by fixation of bicar-
bonate and leads to malate/aspartate (C4 -photosynthesis, Box 10.2). The latter en-
zyme (PEPC) also mediates dark fixation of CO2 in plants with crassulacean acid
metabolism (CAM, Box 5.1), where malic acid provides nocturnal CO2 storage. In
CAM, during the day, CO2 is remobilized from malate and refixed and assimilated
via RuBPC. Depending on the environmental conditions CAM-plants can also fix
CO2 in the light directly via RuBPC.
Due to the different extent of discrimination against 13 CO2 by the two enzymes
during primary CO2 -fixation by the C3 - and C4 -plants, and by the variable expres-
sion of the two mechanisms in CAM-plants, carbon-isotope ratios (δ 13C values)
differ in the three types of plants. They are most negative (most depleted of 13 C) in
C3 -plants, less negative in C4 -plants and intermediate in CAM-plants (Fig. 2.13).
Both C4 -photosynthesis and CAM are, in different ways, adaptations to reduced
supply of water and contribute to the efficiency of plant water use. This means
Table 2.2 13 C-discrimination (‰) in various steps of the CO fixation process (from Ziegler 1994)
2
Step 13 C isotope discrimination

Diffusion of CO2 in the gas phase 4.4
Dissolution of CO2 in water −0.9
Liquid phase diffusion of CO2 or HCO− 3 0.0
Hydration of CO2 −7.0
Carboxylation of phosphoenolpyruvate
relative to HCO−3 2.0
relative to CO2 −5.0
Carboxylation of ribulose-bisphosphate 27
Fig. 2.13 Ranges of carbon-

isotope ratios for C3 -, C4 - and
CAM species. Measurements
from 285 grasses (Poaceae)
with 47 C3 species and 238 C4
species and from 513 CAM
plants. (Unpub. data sets of
H. Ziegler)
that in C4 - and CAM-plants, for each CO2 -molecule fixed, a smaller number of
H2 O-molecules is lost via transpiration as compared to C3 -photosynthesis. Thus, in
addition to the 13 C-content, the 2 H- and 18 O-content of the plants is also affected.
Within the different modes of photosynthesis there are subtleties dependent on the
diffusive limitation imposed on evapotranspiration (see above) or in the relative uti-
lization of RuBPC and PEPC in CAM, and generally in the use of water. All this is
reflected in isotope composition. Thus, also various ecotypes of plants, like halo-
phytes and xerophytes can be differentiated. It must be noted for the latter, however,
that such comparisons are only allowed between C3 -species because these effects
are highly overridden by the carbon isotope effects of CO2 fixation by RuBPC and
PEPC, respectively.
Another example from metabolism is represented by N-nutrition of plants. For
nitrogen isotope effects the situation often is very complex and straightforward con-
clusions are difficult (Martinelli et al. 1999; Adams and Grierson 2001). Differences
in nitrogen uptake mechanisms and in the pathways of assimilation and recycling of
nitrogen in the plants can greatly affect δ 15 N values (Evans 2001). N-salts in the soil
tend to enrich the heavier isotope 15 N, as compared to atmospheric gaseous N2 , and
therefore one can recognize symbiotic N2 -fixers (e.g. legumes with root nodules) by
the lower 15 N content. Analyses of the natural abundance of 15 N in soils have also
served to document forest-to-pasture chronologies and record changes of land-use
pattern in the western Amazon Basin in Brazil (Piccolo et al. 1994).
The flow of various isotopes through the biomass of plants also affects the trans-
fer into other compartments of ecosystems. This allows the study of food webs,
habitat preferences in wandering animals, and even the analysis of eating and drink-
ing habits in human populations and individuals. For the latter it usually suffices to
analyze the organic matter of hairs or finger- and toenails.
Since the analysis of dry matter of the organisms is often sufficient, one can
cover large geographic areas with sampling even from remote regions using sim-
ple equipment. Even collections in herbaria may be used. In this way, for exam-
ple one can arrive at conclusions about the large-scale ecological distribution of
modes of photosynthesis. C4 -grasses dominate in tropical savannas, their relative
abundance declines with increasing altitude (Tieszen et al. 1979; Medina 1982; see
Sect. 10.1.1.2).
An impressive example, if not for a whole continent, is given for the rather
large tropical island of Madagascar of 590,000 km2 (Kluge et al. 1991). Combin-
ing the Klimadiagramm method with the stable-isotope technique, the distribution
of CAM among species of the genus Kalanchoë has been studied and related to
climatic zones and vegetation types of the island. There are 52 species of Kalan-
choë in Madagascar, of which all are either obligate or facultative CAM plants.
There is high flexibility among the species to obtain a variable amount of carbon
by direct CO2 fixation via RuBPC, and this is reflected by increasingly negative
δ 13 C values, whereas primary CO2 -fixation dominated by PEPC leads to less neg-
ative δ 13C values. The large scale effects deduced from the analysis present a very
close correlation of δ 13 C values in the dry matter of Kalanchoë species and cli-
mate and vegetation zones on the island. Less negative (CAM-like) values are dom-
inant in the drier zones with evergreen dry forest, deciduous woodland, savannas
and xerophilous thornbush, while more negative (C3 -like) values prevail in the wet-
ter zones with evergreen rainforest and montane forest (Fig. 2.14). The example
illustrates the close relations between climate, vegetation types and prevalence of
the water conserving CAM-mode of photosynthesis in a given genus. It even al-
lows some views into the evolutionary history of the genus Kalanchoë. Compar-
isons of the phytogeographic distribution of C3 - and CAM-species of Kalanchoë
in Madagascar with the phylogeny of different subgenera based on morphologi-
cal and molecular characteristics and the evolution of CAM in the CAM-species
show that CAM has only evolved once (monophyletically) from the C3 -species of
the moister regions and that the more drought resistant CAM-species then have
conquered the drier regions of the island (Kluge et al. 1991, 2001; Gehrig et al.
2001).
Large-scale isotope effects, however, may also result from transfer rates. Thus,
respiration and photosynthesis of organisms determine the vertical 13 CO2 -gradient
in tropical rainforests (Medina et al. 1986). Richey et al. (1990) have proposed to use
18 O-analyses to assess the large-scale consequences of the destruction of tropical
Fig. 2.14A, B Climate zones related to Klimadiagramm distribution (A) and vegetation types (B) of
Madagascar. The vegetation map (B) contains points indicating ranges of δ 13 C values as explained
in the inset. Note that the closed symbols marking the more negative δ 13 C values are concentrated
in the wetter regions, and the open symbols marking less negative δ 13 C values are accumulated
in the drier vegetation units. The inset also gives the frequency of δ 13 C values for the samples
collected on the island for three combinations of vegetation units as indicated. (After Kluge et al.
1991)
Fig. 2.14 (Continued)

rainforest in the Amazon region. Since C-, N- and S-compounds from biogenic and
anthrogenic sources have different isotope composition, the pathways of pollutant
emission can also be traced. The large global-cycles of a range of elements in the
gaseous atmosphere, as well as regional environmental effects, can be diagnosed in
this way.
Finally, interest has also been focussed on the element strontium, which is not
subject to metabolism, in terms of mineral nutrition. The 87 Sr/86 Sr ratios in the
bed-rock develop in geological times. The different geochemical mobilization of
the isotopes allows distinctions between nutrient supply of vegetation from the soil,
water or dust in the atmosphere. This facilitates understanding of nutrient sources
in ecosystems (Graustein 1989).
2.6 Mathematical Tools
Huge amounts of data become available in large-scale diagnostics. Networks of

functional systems, sub-systems and supra-systems are emerging. The methods de-
scribed in Sects. 2.2 – 2.5 reveal spatiotemporal pattern formation. We deal with
multidimensional problems, which have to be summarized numerically or graphi-
cally and must be cast into a comprehensive presentation. This may possibly con-
stitute the most important bottleneck. Hobbs and Mooney (1990) conclude their
book on remote sensing of the biosphere functions with the statement, that even
without the intensive further development of new sensors, the currently available
technologies offer so much that the capacity for interpretation and application has
been surpassed. The major problems do not appear to be with collection but with
analysis and understanding of data.
The data mountains piling up require hardware and software technologies for
storage and organization. Special statistical and mathematical procedures are essen-
tial to reduce the relevant information from the mass of data to an interpretable form,
and the analysis of data needs the development of new mathematical approaches
(Hütt 2001). Special geographic information systems with multi factorial mapping
are designed to provide surveys related to space (Wallace and Campbell 1990). Im-
age analyses use the concept pf cellular automata and nearest neighbour algo-
rithms to
unravel spatiotemporal structures of patterns (Hütt and Lüttge 2002, 2005).
A particular problem of great fascination is the non-linear dynamics of all the
ecosystems studied. In phytosociology unpredictability, e.g. in occupation of new
habitats by species, is often readily interpreted as demonstrating that the distribution
of plants and the development of diversity is stochastic. Taking the term stochas-
tic in its strict mathematical meaning, this is by no means so simple. Stochastic
white noise in empirical data time series can not readily be distinguished from
the so-called deterministic chaos, which follows strict mathematical rules. A dis-
tinction between the two can only be made via sophisticated theoretical analyses
requiring very detailed sets of time series data. These are rarely available, and
2.6 Mathematical Tools 43
hence, it is hard to prove whether deterministic chaos or stochastic noise predomi-

nate.
Fig. 2.15A–C Comparisons A

of predictability in (A) ran-
dom, (B) regular and (C)
chaotic motion. (Modified
after Schuster 1995)
Locality ( x )
Time ( t )
Nevertheless, it is useful to consider briefly the possible implications of the the-

ory of deterministic chaos in ecology. In random motion initially adjacent points
are distributed with equal probability over all new allowed intervals (Fig. 2.15A). In
regular motion initially adjacent points stay adjacent (Fig. 2.15B). In chaotic mo-
tion initially adjacent points become exponentially separated (Fig. 2.15C). Thus,
regular motion and random motion show complete predictability and unpredictabil-
ity, respectively. Chaotic motion offers some short term predictability (Schuster
1995). Although chaotic systems totally lack long term predictability their short
term predictability is better than that of stochastic random processes. One can de-
termine the error of such predictions, and with the appropriate algorithms one can
use them to regulate a system via a transputer so that the system is forced to stay
in one of the chaotic paths (or trajectories) in time and space. Such fine regula-
tion of chaotic systems, which offers opportunities for much more delicate manip-
ulation than regulation of deterministic systems, is currently assessed in physics
(Hübinger et al. 1993; Schuster 1995) and even explored for practical applica-
tions in engineering. Biological systems must obey physical laws. Conversely,
we may also say that biological systems use physical laws to develop the diver-
sity of life. It would be most surprising if the wide scope of possibilities inher-
ent in deterministic chaos had not been used by life during evolution. It may be
noted that with pure stochastic randomness life would be deprived of any sig-
nificance and would be cast into meaninglessness. Conversely, pure determinis-
tic regularity, allowing one to retrace all things accurately in the past and predict
them precisely in the future for all (mathematical) infinity would cast life into te-
dious monotony. Only deterministic chaos, with its strict mathematical rules and
yet high variability, with its unpredictability and yet delicate means of fine regu-
lation, provides an opportunity for the adaptability, plasticity, diversity and beauty
of life to unfold (W. Martienssen, quoted after a public lecture; Lloyd and Lloyd
1995).
Due to the pioneering contributions of R. May (May 1976) population dynamics
has become one of the roots of the development of the chaos theory. Possibly the
apparently simple logistic equation discussed by May (1976) may allow biologists
a ready access to an intuitive comprehension of the theory of deterministic chaos.
Let x t be the size of a population at a certain state at time t. We then want to know
the size of the population at the next possible state in time, i.e. x t +1 . It is obvious that
the development of the population depends on its resources. These may be given by
a growth factor r or more generally an external control parameter, such that x t +1 is
proportional to r · x t . However, it is not only evident from the exorbitant increase of
the human population on the globe, but a general experience of population ecology,
that increasing population densities also bear inhibitory mechanisms in themselves,
i.e. x t +1 is also proportional to x t · (1 − x t ). Hence, the logistic equation for the
development of the population x is
x t +1 = r · x t · (1 − x t ) . (2.10)
Subsequent population sizes can be calculated by recursions, where x t +1 is used in
the place of x t to obtain x t +2 and so forth.
However, the equation is only apparently simple. It describes one of several pos-
sible routes from order or regularity into chaos (Schuster 1995; Fig. 2.16). It shows
that ordered predictability only occurs for a narrow range of the value of the control
parameter r . At low values of r there is a steady state, while at larger values of r a bi-
furcation (i.e. a branching or dichotomy) leads first to phase doubling and ordered
oscillations between two states and then further bifurcations give four states. How-
ever, very tiny additional changes of r lead the way into the non-predictability of de-
terministic chaos. This is seen in computer simulations of (2.10), shown in Fig. 2.16.
The lower diagram shows the initial steady state, effective for a large range of values
of r , then the first and second bifurcation leading to periods of 2 and 4 states, and
with increasingly smaller increments of r there are then chaotic responses to tiny
2.6 Mathematical Tools 45
Fig. 2.16 The route from order to chaos via increasing periods by augmenting r in the logistic
recursion equation (2.10). The lower diagram shows the route from steady state via bifurcations
(period 2, period 4) into chaos given by increasing r . The upper four diagrams give the calculated
population sizes by iteration of (2.10) for the steady state, period 2 (two states), period 4 (four
states) and chaos. (May 1976; Hastings et al. 1993)
changes of r . The top four graphs show the results of iterative calculations of popu-
lation sizes x t . At low r in the steady state the population is stable, at period 2 there
are two and at period 4 there are 4 predictable states, while in chaos, prediction of
subsequent population sizes from existing ones has become impossible.
This little excursion to population theory appeared useful to explain some basic
implications of the chaos theory. Chaos is a property of non-linear dynamic systems
and these are the rule and not the exception both in the living and non-living world.
However, while the theory of deterministic chaos has already had some impact in
population biology (May 1976; Hastings et al. 1993) it is intriguing that in ecol-
ogy in general it has only been accepted very reluctantly (Linsenmair 1995; Stone
and Ezrati 1996), or even been rejected in the exclusive distinction of deterministic
and stochastic development of diversity. It is intriguing because population biology
is a field so close to or even part of ecology. For example, we may consider the
control parameter r in (2.10) as an indicator of general resources or even as stress.
If then we envisage that a high state of order may be given by complexity, which
integrates functioning of diversity (Cramer 1993), we realize that this must occur
only within a rather narrow window of stress conditions, as it is in fact borne out
in experiments like those of Grime et al. (1987) and Tilman (1982) described in
Sect. 3.3.2. Likewise, we are right back in the realm of deterministic chaos when
we reject the climax theory of formation of stable steady state vegetation types and
adopt instead the oscillating mosaic model with continuous dieback and renewal
of “unpredictable irregularity” for tropical forests (Sect. 3.3.3). It is not unlikely
that deterministic chaos, which certainly governs the ecology of populations, also
determines the structure of tropical forests (as of other environments) with their
oscillatory and non-linear behaviour.
References
Adams MA, Grierson PF (2001) Stable isotopes at natural abundance in terrestrial plant ecology
and ecophysiology: an update. Plant Biol 3:299–310
Asner GP, Vitousek PM (2005) Remote analysis of biological invasion and biogeochemical change.
Proc Nat Acad Sci USA 102:4383–4386
Asner GP, Nepstad D, Cardinot G, Ray D (2004) Drought stress and carbon uptake in an Amazon
forest measured with spaceborne imaging spectroscopy. Proc Nat Acad Sci USA 101:6039–
6044
Broadmeadow MSJ, Griffiths H, Maxwell C, Borland A (1992) The carbon isotope ratio of plant
organic material reflects temporal and spatial variations in CO2 within tropical forest forma-
tions in Trinidad. Oecologia 89:435–441
Buchmann N, Bonal D, Barigah TS, Guehl J-M, Ehleringer JR (2004) Insights into the carbon
dynamics of tropical primary rainforests using stable carbon isotope analyses. In Gourlet-
Fleury S, Guehl J-M, Laroussinie O, eds, Ecology and management of a neotropical rainforest.
pp 95–113, Elsevier, Amsterdam
Chapelle EW, Wood FM, McMurtrey JE, Newcomb WW (1984a) Laser-induced fluorescence in
green plants: 1. A technique for the remote detection of plant stress and species differentiation.
Appl Opt 23:134–138
Chapelle EW, McMurtrey JE, Wood FM, Newcomb WW (1984b) Laser-induced fluorescence of
green plants. 2. LIF caused by nutrient deficiencies in corn. App Opt 23:139–142
Cramer F (1993) Chaos and order. The complex structure of living systems. VCH, Weinheim
Ehleringer JR (1993) Variation in leaf carbon isotope discrimination in Encelia farinosa: implica-
tions for growth, competition, and drought survival. Oecologia 95:340–346
Ehleringer JR, Rundel PW (1989) Stable isotopes: history, units, and intrumentation. In: Rundel
PW, Ehleringer JR, Nagy KA (eds), Stable isotopes in ecological research. Ecological studies,
vol 68. Springer, Berlin Heidelberg New York, pp 1–19
Ehrendorfer F (1991) Geobotanik. In: Sitte P, Ziegler H, Ehrendorfer F, Bresinsky A (eds) Stras-
burger Lehrbuch der Botanik. G Fischer, Stuttgart
Evans RD (2001) Physiological mechanisms influencing plant nitrogen isotope composition.
Trends Plant Sci 16:121–126
Farquhar GD, Ehleringer JR, Hubick KT (1989a) Carbon isotope discrimination and photosynthe-
sis. Annu Rev Plant Physiol Plant Mol Biol 40:503–537
Farquhar GD, Hubick KT, Condon AG, Richards RA (1989b) Carbon isotope fractionation and
plant water-use efficiency. In: Rundel PW, Ehleringer JR, Nagy KA (eds) Stable isotopes in
ecological research. Ecological studies, vol 68. Springer, Berlin Heidelberg New York, pp
21–40
Gates M (1965) Energy, plants and ecology. Ecology 46:1–14
References 47
Gehrig HH, Gaußmann O, Marx H, Schwarzott D, Kluge M (2001) Molecular phylogeny of the
genus Kalanchoë (Crassulaceae) inferred from nucleotide sequences of the IST-1 and IST-2
regions. Plant Sci 160:827–835
Ghashghaie J, Duranceau M, Badeck F-W, Cornic C, Adeline M-T, Deleens E (2001) δ 13 C of CO2
respired in the dark in relation to δ 13 C of leaf metabolites: comparison between Nicotiana
sylvestris and Helianthus annuus under drought. Plant Cell Environ 24:505–515
Gosz JR, Dahm CN, Risser PG (1988) Long-path FTIR measurement of atmospheric trace gas
concentrations. Ecology 69:1326–1330
Graustein WC (1989) 87 Sr/86 Sr ratios measure the sources and flow of strontium in terrestrial
ecosystems. In: Rundel PW, Ehleringer JR, Nagy KA (eds) Stable isotopes in ecological re-
search. Ecological studies, vol 68. Springer, Berlin Heidelberg New York, pp 491–512
Grime JP, Mackey JML, Hillier SH, Read DJ (1987) Floristic diversity in a model system using
experimental microsoms. Nature 328:420–422
Guehl J-M, Bonal D, Ferhi A, Barigah TS, Farquhar G, Granier A (2004) Community-level diversity
of carbon-water relations in rainforest trees. In Gourlet-Fleury S, Guehl J-M, Laroussinie O, eds,
Ecology and management of a neotropical rainforest. pp 75–94. Elsevier-Amsterdam
Hák R, Lichtenthaler HK, Rinderle U (1990) Decrease of the chlorophyll fluorescence ratio
F690/F730 during greening and development of leaves. Radiat Environ Biophys 29:329–336
Hastings A, Hom CL, Ellner S, Turchin P, Godfray HCJ (1993) Chaos in ecology: is mother nature
a strange attractor? Annu Rev Ecol Syst 24:1–33
Hobbs RJ, Mooney HA (eds) (1990) Remote sensing of biosphere functioning. Ecological studies,
vol 79. Springer, Berlin Heidelberg New York
Hoge FE, Swift RN, Yungel JK (1983) Feasibility of airborne detection of laser-induced fluores-
cence emissions from green terrestrial plants. Appl Opt 22:2991–3000
Holtum JAM, Winter K (2005) Carbon isotope composition of canopy leaves in a tropical forest in
Panama throughout a seasonal cycle. Trees 19:545–551
Hübinger B, Doerner R, Martienssen W (1993) Local control of chaotic motion. Z Physik B
90:103–106
Humboldt, A von (2004) Kosmos. Entwurf einer physischen Weltbeschreibung. Ette O, Lubrich O,
eds, Eichborn Verlag, Frankfurt M
Hütt M-T (2001) Datenanalyse in der Biologie. Springer, Berlin Heidelberg New York
Hütt M-T, Lüttge U (2002) Non-linear dynamics as a tool for data analysis and modeling in plant
physiology. Plant Biol 4:281–297
Hütt M-T, Lüttge U (2005) Network dynamics in plant biology: current progress in historical per-
spective. Progr Bot 66:277–310
Jones PD (1990) Le climat des mille dernières années. Recherche 21:304–312
Kluge M, Brulfert J, Ravelomana D, Lipp J, Ziegler H (1991) Crassulacean acid metabolism in
Kalanchoë species collected in various climatic zones of Madagascar: a survey by δ 13 C anal-
ysis. Oecologia 88:407–414
Kluge M, Razanoelisoa B, Brulfert J (2001) Implications of genotypic diversity and phenotypic
plasticity in the ecophysiological success of CAM plants, examined by studies on the vegeta-
tion of Madagascar. Plant Biol 3:214–222
Lichtenthaler HK, Hák R, Rinderle U (1990) The chlorophyll fluorescence ratio F690/F730 in
leaves of different chlorophyll content. Photosynth Res 25:295–298
Linsenmair KE (1995) Biologische Vielfalt und ökologische Stabilität. In: Markl H, Geiler G,
Großmann S, Oesterhelt D, Schmidbaur H, Quadbeck-Seeger HJ, Truscheit E (eds) Wis-
senschaft in der globalen Herausforderung. Verh Ges Dtsch Naturforsch Ärzte, 118 Vers Ham-
burg, Wiss Verlagsgesellschaft, Stuttgart, pp 267–295
Lloyd AL, Lloyd D (1995) Chaos. Its significance and detection in biology. Biological Rhythm
Res 26:233–252
Máguas C, Griffiths H (2002) Applications of stable isotopes in plant ecology. Progr Bot 64:472–
505
Malingreau J-P, Tucker CJ (1987) La végétation vue de l’espace. Recherche 18:180–189
Martinelli LK, Piccolo MC, Townsend AR, Vitousek PM, Cuevas E, McDowell W, Robertson GP,
Santos OC, Treseder K (1999) Nitrogen stable isotope composition of leaves and soil: tropical
versus temperate forests. Biogeochemistry 46:45–65
Matson PA, Harriss RC (1988) Prospects for aircraft-based gas exchange measurements in ecosys-
tem studies. Ecology 69. Ecol Soc of Am, Washington, DC (5), 1318–1325
Matson PA, Vitousek PM (1990) Remote sensing and trace gas fluxes. In: Mooney HA, Hobbs
RJ (eds) Remote sensing of biosphere functioning. Springer, Berlin Heidelberg New York, pp
157–167
May RM (1976) Simple mathematical models with very complicated dynamics. Nature 261:459–
467
Medina E (1982) Physiological ecology of neotropical savanna plants. In: Huntley BJ, Walker BH
(eds) Ecology of Tropical Savannas, Ecological Studies, vol 42. Springer, Berlin Heidelberg
New York, pp 308–335
Medina E, Montest G, Cuevas E, Roksandic Z (1986) Profiles of CO2 concentration and δ 13 C
values in tropical rain forests of the upper Rio Negro basin, Venezuela. J Trop Ecol 2:207–217
Medina E, Sternberg L, Cuevas E (1991) Vertical stratification of δ 13 C values in closed natural and
plantation forests in the Luquillo mountains, Puerto Rico. Oecologia 87:369–372
Nichol CJ, Rascher U, Matsubara S, Osmond B (2006) Assessing photosynthetic efficiency in
an experimental mangrove canopy using remote sensing and chlorophyll fluorescence. Trees
20:9–15
Nobel PS (1983) Biophysical plant physiology and ecology. WH Freeman, San Francisco
Piccolo MC, Neill C, Cerri CC (1994) Natural abundance of 15 N in soils along forest-to-pasture
chronosequences in the western Brazilian Amazon Basin. Oecologia 99:112–117
Richey JE, Adams JB, Victoria RL (1990) Synoptic-scale hydrological and biogeochemical cycles
in the Amazon river basin: a modeling and remote sensing perspective. In: Hobbs RJ, Mooney
HA (eds) Remote sensing of biosphere functioning. Ecological studies, vol 79. Springer, Berlin
Heidelberg New York, pp 249–268
Roux X le, Bariac T, Sinoquet H, Genty B, Piel C, Mariotti A, Girardin C, Richard P (2001) Spatial
distribution of leaf water-use efficiency and carbon isotope discrimination within an isolated
tree crown. Plant Cell Environ 24:1021–1032
Rundel PW, Ehleringer JR, Nagy KA (eds) (1988) Stable isotopes in ecological research. Ecologi-
cal studies, vol 68. Springer, Berlin Heidelberg New York
Running SW (1990) Estimating terrestrial primary productivity by combining remote sensing and
ecosystem simulation. In: Hobbs RJ, Mooney HA (eds) Remote sensing of biosphere func-
tioning. Ecological studies, vol 79. Springer, Berlin Heidelberg New York, pp 65–86
Scheidegger Y, Saurer M, Bahn M, Siegwolf R (2000) Linking stable oxygen and carbon iso-
topes with stomatal conductance and photosynthetic capacity: a conceptual model. Oecologia
125:350–357
Schuster HG (1995) Deterministic chaos. VCH, Weinheim
Sellers PJ, Hall FG, Strebel DE, Asrar G, Murphy RE (1990) Satellite remote sensing and field
experiments. In: Hobbs RJ, Mooney HA (eds) Remote sensing of biosphere functioning. Eco-
logical studies, vol. 79. Springer, Berlin Heidelberg New York, pp 169–201
Sheshshayee MS, Bindumadhava H, Ramesh R, Prasad TG, Lakshminarayana MR, Udayakumar
M (2005) Oxygen isotope enrichment (18 O) as a measure of time averaged transpiration rate.
J exp Bot 56:3033–3039
Smith BN (1975) Carbon and hydrogen isotopes of sucrose from various sources. Naturwis-
senschaften 62:390
Stone L, Ezrati S (1996) Chaos, cycles and spatiotemporal dynamics in plant ecology. J Ecol
84:279–291
Terwilliger VJ, Kitajima K, Roux-Swarthout DJ le, Mulkey S, Wright SJ (2001) Intrinsic water-
use efficiency and heterotrophic investment in tropical leaf growth of two neotropical pioneer
species as estimated from δ 13 C values. New Phytol 152:267–281
References 49
Tieszen LL, Senyimba MM, Imbamba SK, Troughton JH (1979) The distribution of C3 and C4
grasses and carbon isotope discrimination along an altitudinal and moisture gradient in Kenya.
Oecologia 37:337–350
Tilman D (1982) Resource competition and community structure. Princeton Univ. Press, Princeton
Wallace JF, Campbell N (1990) Analysis of remotely sensed data. In: Hobbs RJ, Mooney HA
(eds) Remote sensing of biosphere functioning. Ecological studies, vol 79. Springer, Berlin
Walter H (1973) Vegetationszonen und Klima. Ulmer, Stuttgart
Walter H (1982) Bekenntnisse eines Ökologen. 3. Auflage. G Fischer, Stuttgart
Walter H, Breckle S-W (1983) Ökologie der Erde, Bd 1, Ökologische Grundlagen in globaler Sicht.
G Fischer, Stuttgart
Walter H, Breckle S-W (1984) Ökologie der Erde, Bd 2, Spezielle Ökologie der tropischen und
Walter H, Lieth H (1967) Klimadiagramm – Weltatlas. G Fischer, Jena
Warren CR, Adams MA (2006) Internal conductance does not scale with photosynthetic capacity:
implications for carbon isotope discrimination and the economics of water and nitrogen use in
photosynthesis. Plant Cell Environ 29:192–201
Wessman CA (1990) Evaluation of canopy biochemistry. In: Hobbs RJ, Mooney HA (eds) Remote
sensing of biosphere functioning. Ecological studies, vol 79. Springer, Berlin Heidelberg New
York, pp 135–156
Woodward FI (1987) Climate and plant distribution. Cambridge University Press, Cambridge
Ziegler H (1989) Hydrogen isotope fractionation in plant tissues. In: Rundel PW, Ehleringer JR,
Nagy KA (eds) Stable isotopes in ecological research. Ecological studies, vol 68. Springer,
Berlin Heidelberg New York, pp 105–123
Ziegler H (1994) Stable isotopes in plant physiology and ecology. Progr Bot 56:1–24
Chapter 3
Tropical Forests.
I. Physiognomy and Functional Structure
3.1 Separation of Different Types of Tropical Forests
In the tropics seasonality is determined by rainfall. Seasonality of rainfall is absent

over only a very narrow zone 1◦ north and south of the equator (Fig. 3.1). Season-
ality of rainfall is the most important factor determining the definition of different
types of tropical forest. Hence, a detailed separation of types of tropical forests is
obtained when they are related to Klimadiagramm graphs (after Walter 1973, see
Sect. 2.2.1, Box 2.1), which depict the duration and severity of dry and wet seasons.
Humid and arid climates, respectively, are defined by the difference between rain-
fall and evaporation. Positive values indicate a humid climate, where precipitation
is larger than evaporation, and negative values mark arid climates with evaporation
larger than precipitation. As we have demonstrated above (Sect. 2.2.1), the Klima-
diagramm technique based on temperature and precipitation is highly successful in
demonstrating the degree of aridity and humidity of individual stations as well as
large areas. In contrast to data on temperature and precipitation, which usually are
readily available from many weather stations, measurements of free evaporation are
scarce. Other factors, especially wind, may significantly affect evaporation. Thus,
where evaporation data is available, it can add much to the information provided
by Klimadiagramm graphs. As seen in the examples of seven stations in Venezuela
(Fig. 3.2) the results of Klimadiagramm graphs regarding the occurrence of humid
seasons is roughly confirmed by the rainfall minus evaporation (R − E) index.
However, the wet seasons generally appear somewhat shorter with the R − E index,
and on the basis of this criterion small apparent humid seasons of the Klimadia-
gramm graphs disappear entirely (Fig. 3.2 diagrams 1 and 2). These seven stations
carry different types of forest and represent a latitudinal transect from the wet conti-
nental areas close to the equator to the drier coastal regions in the north of Venezuela
(Fig. 3.3). A very strong effect of latitude on the expression of tropical forest types
is highly evident, which agrees with the predictions of Fig. 3.1.
Using India and Venezuela as examples comparison of forest types is made with
a diagram of precipitation versus extension of dry periods (Fig. 3.4). The corres-
ponding Klimadiagramm graphs for Venezuela are shown in Fig. 3.5. The major
52 3 Tropical Forests. I. Physiognomy and Functional Structure
Fig. 3.1 Dry and wet seasons

in the subtropical and tropical
zone with zenithal precipi-
tation. Wet seasons hatched,
dry seasons dotted. (Walter
and Breckle 1984, with kind
permission of S.-W. Breckle
and G. Fischer-Verlag)
types of forests distinguished in relation to the degree of seasonality which increases

with increasing numbers of dry months per year and decreasing annual rainfall are:
• evergreen rainforest,
• seasonal rainforest,
• semi-evergreen moist and dry monsoon or trade-wind forest,
• drought-deciduous forest,
• thorn scrub and cactus forest.
Fig. 3.2 Diagrams for seven stations of a transect across Venezuela as shown in Fig. 3.3. The
diagrams indicate the extensions of wet seasons according to the Klimadiagramm concept (thinner
plus thicker vertical bars, i.e. where the rainfall curve is above the temperature curve; see Box 2.1),
and the dry seasons according to the Klimadiagramm concept are dotted as usual. The thick line
gives free evaporation. The R − E index is positive where the rainfall curve of the Klimadiagramm
is above the evaporation curve, and this indicates the extensions of the wet seasons according to
the R − E index (thicker vertical bars). The R − E index is negative where the rainfall curve of
the Klimadiagramm is below the evaporation curve and this indicates the extensions of the dry
seasons according to the R − E index (left white in the diagrams). Annual precipitation P; annual
free evaporation E. (After Medina 1983, with kind permission from Elsevier Science-NL, Sara
Burgerhartstraat 25, NL-1055 KV Amsterdam, The Netherlands)
3.1 Separation of Different Types of Tropical Forests 53
Fig. 3.3 Transect across

Venezuela with rainfall minus
evaporation (R − E index) at
different stations with var-
ious forest types. Numbers
in parentheses indicate dry
months per year after the
R − E index. See also Kli-
madiagramms in Fig. 3.2.
(After Medina 1983, with
kind permission from Elsevier
Science-NL, Sara Burger-
hartstraat 25, NL-1055 KV
Amsterdam, The Netherlands)
Gradation of rainfall is also inherent in Beard’s distinction (Beard 1946, 1955)

of forests according to altitude (Fig. 3.6):
• low-land rainforest,
• lower montane rainforest,
• upper montane rainforest,
• elfin forest.
Some of the forest types will be discussed separately below.
3.2 Physiognomy of Different Types of Tropical Forests 55
Fig. 3.4A,B Forest types in India (A) and in Venezuela (B) related to annual amount of precipitation
and the duration of drought periods (number of dry months per year). (After Walter and Breckle
1984 and Vareschi 1980, with kind permission of R. Ulmer)
3.2 Physiognomy of Different Types of Tropical Forests
3.2.1 Tropical Rain Forests
Tropical evergreen rainforests are said to contribute about 35% of global net primary
production (Löscher et al. 2003). However, there is a problem of deciding what we
define as “tropical rainforest”. This is best illustrated by eight maps of Venezuela
presented by Vareschi (1980), where he depicts the distribution of rainforest in this
tropical country according to the views of different authors (Fig. 3.7). In the two
Fig. 3.5A–F Klimadiagramm graphs for the different forest types distinguished in the diagram of
Fig. 3.4B
Fig. 3.6A–D Types of tropical forest at different altitudes. A Semi-evergreen lowland rainforest
(East Venezuela). B Montane rainforest (northern range Trinidad). C Upper montane rainforest
(cloud or fog forest; Rancho Grande, northern coastal range Venezuela). D Elfin forest (Serro
Santa Ana, Paraguana Peninsula Venezuela)
extreme cases either almost 2/3 of the whole country is covered by rainforest (up-
per left in Fig. 3.7) or there is no rainforest at all (lower right in Fig. 3.7); and there
are gradations in between these extremes. In Brazil much of the Amazon basin is
covered by rainforest. The Brazilian Atlantic forest also is a typical tropical rain
forest. It belongs to the 25 biodiversity hot spots of the world. Of an original area of
Fig. 3.7 Distribution of “rainforest” (black) in the tropical country Venezuela according to differ-
ent authors. (Vareschi 1980, with kind permission of R. Ulmer)
1,227,600 km2 only 91,930 km2 (7.5%) are remaining to date of which 33,084 km2
(35.9%) are protected, and there are 20,000 plant species, 8,000 of which are en-
demic (Myers et al. 2000).
The purist’s definition of tropical rainforest requires that there should be no sea-
sonality of rainfall whatsoever. Hence, the lack of any rainforest in Venezuela ac-
cords with the purist’s definition (Fig. 3.7, lower right). If one begins to broaden the
definition, of course, it becomes a matter of taste how far this term may be extended.
How long should rainless periods be and in how many successive years should they
occur in order to retain the term “rainforest”? Accordingly, we then derive the gra-
dations seen in the maps of Venezuela, shown in Fig. 3.7.
In addition to seasonality there are other features which characterize tropical rain-
forests. One of the most conspicuous is the extraordinary diversity of tree species.
In contrast to the temperate and boreal zones, where forests can be named after dom-
inant tree species, e.g. spruce, fir, pine, beech, oak, birch etc., there is no dominance
of anyone particular tree species in tropical rainforests. One may encounter up to
300 tree species per ha, which represent about 1/3 of all plant species present. The
most frequent tree species rarely represent more than 15% of all species of trees
present (Jacobs 1988; Whitmore 1990). This phenomenon also makes it difficult
to define the minimal quadrat of sampling plots in plant sociology. The minimal
quadrat is given by the size of a plot in a system being studied above which the
total number of species observed does not increase (see Sect. 3.3.1 and Fig. 3.15).
Such minimal quadrats in tropical rainforests may be quite large and may never be
attained.
3.2.2 Tropical Cloud and Elfin Forests
One of the best preserved tropical cloud forests is the forest of Rancho Grande
on the northern coastal range of Venezuela (Fig. 3.6C). It is determined by the
trade winds coming from a north-eastern direction (Fig. 3.8). Clouds build up over
the Caribbean islands and dissolve as the wind moves south, where it then hits
the coastal range of mountains on the South American continent. The hot wind
cools down as it climbs upwards and on top clouds are formed almost continuously
(Figs. 3.6C and 3.8). We find an altitudinal gradation of vegetation from cactus-
forest, thorn scrub and dry forest to evergreen cloud forest. Clouds break up as
the wind drives down the southern slopes and also move further south with the
trade winds. In places, the constant exposure to the wind may lead to formation
of elfin forest with dwarf trees similar to those on the top of Santa Ana Moun-
tain on the Paraguana Peninsula of Venezuela, which are here dominated by Clusia
multiflora and the palm Geonema paraguanensis about 1 m tall (Fig. 3.6D). In ad-
dition the reduction of growth and formation of dwarf forms may also be caused
by reduced availability of nutrients, especially phosphorus, due to the lower min-
eralization rates at lower temperatures when altitude increases (Kitayama and Aiba
2002).
Fig. 3.8 Profile in a NW to SE direction (from right to left) explaining the formation of cloud
forest on the northern coastal range of Venezuela. (After Vareschi 1980, with kind permission of
R. Ulmer)
3.2.3 Floodplain Forests
Wetland ecosystems are found to be related to hyperseasonal and marsh savannas

(Sect. 9.2, Fig. 9.10), but in addition we mention them here because we also find
fresh water flooded sites in moist forests all over the world’s tropics (Esteves 1998)
where all stages from open water to high forest can often be traced. We may dis-
tinguish between swamp forest which is permanently waterlogged and marsh forest
which is flooded only seasonally (Richards 1996; Fig. 3.9). In Brazil the Central
Amazon floodplains cover an area of more than 300,000 km2 (Junk 1997) where
we may distinguish nutrient poor bottomland floodplains along black water rivers
(várzea) and nutrient rich periodically inundated riverine wetlands fed by white
water rivers (igapó) (Esteves 1998). In Central Amazon floodplains in the pioneer
species Salix maritima even the canopy can remain under water for months, while
the canopy of the shade tolerant late successional species Tabernaemontana juruana
always remains out of the water (de Simone et al. 2002).
Flooding leads to hypoxic and even anaerobic conditions (Richards 1996; Lobo
and Joly 1998), which require morphological and anatomical as well as physiolog-
ical adaptations. Phytohormones such as ethylene and auxin are involved in their
regulation. At the morphological/anatomical level adventitious roots with lenticels
above the water (Sect. 3.3.4.1) and root aerenchymas facilitate aeration of root
tissues because plants subject to flooding have to rely on supply of oxygen from
parts of the plants which are not flooded (Pimenta et al. 1998). There is adaptive
hypertrophy of lenticels under the control of ethylene. In Sesbania commersoniana
the height × length of the main root lenticels was 0.31 × 1.56 (mm × mm) in non-
flooded controls and 0.44 × 2.06 (mm × mm) when flooded (Pimenta et al. 1998).
The submerged roots may produce suberized and lignified barriers in the exodermis
to reduce diffusion of oxygen from the root tissue to the ambient soil (de Simone et
al. 2003a) although some loss of oxygen from the roots will also help to build up
Fig. 3.9 A, B Flooded forest Reserva Biologica Poço dos Antos, Rio de Janeiro State, Brazil.
C Euterpe oleracea palm swamp, French Guyana
a partially oxidant atmosphere at the root periphery important for acquisition of nu-
trients and control of harmful reduced compounds at the root periphery (Lobo and
Joly 1998). White water fed flood plains are rich and black water fed flood plains
are poor in nutrients. Symbiotic dinitrogen fixation is frequent but is discussed in
Sect. 10.2.3.2 in the context of savanna floodplains. The reducing root environ-
ment during long periods of flooding also may lead to formation of several harmful
reduced compounds and especially dangerously high levels of reduced iron, Fe2+ ,
and manganese, Mn2+ , and stress due to uptake and storage of these metals in the
plants (Lobo and Joly 1998; de Simone et al. 2003b). At the metabolic level glycol-
ysis and fermentation with the production of ethanol or lactic acid serve anaerobic
metabolic energy turnover (Lobo and Joly 1998; Pimenta et al. 1998; de Simone et
al. 2002).
Trees and herbs may respond differently to flooding. There are basically two dif-
ferent response patterns among trees, deciduous tree species respond to the flooded
period by complete defoliation and evergreen tree species are maintaining their fo-
liage. Thus, Fernandez et al. (1999) found two different physiological reactions,
(i) a decreasing rate of photosynthesis, stomatal closure and reduced leaf conduc-
tance for water vapour with flooding, and (ii) flood tolerance, where both photosyn-
thesis and leaf conductance were independent of flooding. The C4 -photosynthesis
grass Echinochloa polystachya on submerged Central Amazon floodplains showed
the very high rates of photosynthesis of 30 – 40 µmol m−2 s−1 typical for C4 -
photosynthesis (Sect. 10.1.1.2) but much lower rates of 17 µmol m−2 s−1 during
shorter dry periods (Piedade et al. 1994).
3.2.4 Thorn Scrub and Cactus Forests
Examples comparing the physiognomy of a thorn scrub and a cactus forests are
shown in Figs. 3.10 and 3.11. The term cactus forest was coined by Vareschi (1980).
Cactus-forests represent an example of thornbush-succulent forests. This type of for-
est in Madagascar covers a small area in the south-west of the island (see Fig. 2.14)
dominated by Didieraceae (11 species in 4 genera), Euphorbiaceae and other suc-
culent and deciduous woody species (Fig. 3.12). It is also characteristic of the
“Caatinga” formation in NE Brazil (Fig. 3.13A). In Venezuela there is the “Espinar”
in the area around Carora, where Vareschi (1980) distinguishes a thornbush-forest,
with Mimosaceae (Haematoxylon praecox) and Caesalpiniaceae (Cercidium prae-
cox) which has only one species of cactus, and a cactus-forest with 10 or more
different species of cacti (Figs. 3.11 and 3.13B,C). However, both types of forest
mutually intermingle and form a mosaic-like pattern. The term cactus forest relies
on the acceptance of large columnar cacti as tree-like plants, and depends on the
size of the area occupied by them. Alternatively or additionally, one may require the
most important physiognomical aspect of a forest to be the formation of a closed
canopy. The area around Carora in Venezuela, for which Vareschi originally coined
the term, is dominated by columnar cacti of Cereus lemairei, Ritterocereus griseus
and Cephalocereus moritzianus. However, it appears, that woody Mimosaceae, Cap-
Fig. 3.10A,B Types of dry tropical forest. A Drought deciduous forest (Falcon, Venezuela).
B Cactus-forest (Carora, Venezuela)
paridaceae and Caesalpiniaceae like Cercidium praecox very much add to the im-
pression of a closed canopy, which is certainly given when one is walking around
in these forests. The floor of these forests is bare of vegetation or covered by a for-
Fig. 3.11A,B Transects of a thornbush forest (A) and a cactus forest (B) in the area of Carora,
Venezuela. In A the trees are Mimosaceae (1), the Apocynaceae Plumeria alba (2), and the Cac-
taceae Pereskia guamacho (3). The only cactus is Cereus jamacaru (4). In B Cactaceae dominate
the higher and lower strata. (Vareschi 1980, with kind permission of R. Ulmer)
bidding muddle of thorny cacti, particularly Opuntia wentiana and O. caribea. This
may be due to a large extent to overgrazing by goats. At places where the access
of goats is prevented, grasses are entering these cactus-forests, which then obtain
a physiognomy more comparable to the equivalent thornbush-savanna in Africa.
The climate at the sites of these thornbush-succulent forests is strongly seasonal
and very dry during most of the year. The Klimadiagramm of Carora (Fig. 3.14)
for example only shows a pronounced wet period from September to December and
a very short rainy season in May to June. The open canopy of these deciduous forests
allows much penetration of full sunlight, and the plants are subject to stress by:
• high irradiance (hν),
• high temperature (T ),
• scarcity of water or drought (H2 O),
• limited nutrient availability especially nitrogen (N).
The ecophysiological responses of plants in dry tropical forests will be discussed in
Sect. 5.2).
Fig. 3.12 Thornbush-succulent forest in SW Madagascar with Didieraceae and Euphorbiaceae.

(Photographs courtesy M. Kluge)
Fig. 3.13A–C Caatinga in the state of Ceará, Brazil (A), and thornbush-succulent forests near
Carora, Venezuela (B, C)
3.3 Horizontal Structure and Diversity of Tropical Forests 67
Fig. 3.14 Klimadiagramm of

the thornbush-succulent forest
site at Carora. (Vareschi 1980,
with kind permission of R.
Ulmer)
3.2.5 Mangroves and Gallery Forests
Important and typical tropical forests are also mangroves and gallery forests which
deserve their own sections in this book (Chap. 7 and Sect. 9.1, respectively).
3.3 Horizontal Structure and Diversity of Tropical Forests
Biodiversity has many facets. The understanding of mechanisms of biodiversity is

emerging only gradually, especially for the tropical environment, and there is a va-
riety of hypotheses on diversity. Kratochwil (1998) has formulated and is assessing
a set of 30 hypotheses as a basis for the development of a general theory of biodiver-
sity. It is not the place here to develop a systematic account although this would be
an attractive task in view of the particularly high diversity in many tropical forests
constituting so called diversity hot spots. However, I shall allude to questions of
diversity as I move through various aspects of the horizontal structure of tropical
forests.
3.3.1 Diversity and the Spatial Structure of the Environment
Considering the various types of forests in Sects. 3.1 and 3.2, diversity of tropical
forests has already been alluded to and the high floristic diversity in tropical rain-
forests has been mentioned (Sect. 3.2.1). This leads to the problem that the mini-
mum quadrats of plant sociology become very large. Minimum quadrats give the
smallest possible area in which all species occurring in a habitat are present, as il-
lustrated in Fig. 3.15. Whitmore (1990) has used such species to area diagrams to
illustrate the concept of floristic α-, β- and γ-diversity referring to different lev-
els in landscapes (Noss 1983) (Fig. 3.16) as first defined by Whittaker (1975), i.e.
by taking the richness of species as given by the number of different species per
minimum quadrat:
Fig. 3.15 Species/area curves for four tropical forests in Venezuela. A Cloud forest La Carbonera
(minimal quadrat 2,500 m2 ); B moist trade-wind forest Rio Caura (minimal quadrat 1,100 m2 );
C and D cloud forest of Rancho Grande, C all species, D only trees with a stem diameter > 10 cm,
for which the minimal quadrat has not been attained at 7,500 m2 . (After data of Vareschi 1980,
from Lüttge 1995)
• at the level of communities or coenoses, α-diversity,

• at the level of ecosystems with different vegetation types, for example at ridges,
on hillsides, in valleys etc., β-diversity,
• at the level of landscapes comprising several ecosystems, γ-diversity.
Another approach is the evaluation of diversity in the spatial structure of the physical
environment in relation to environmental factors. Bell et al. (1993) have suggested
the use of log–log regressions of environmental factors, such as edaphic variables,
water chemistry and climate, vs distance over large spatial scales (106 m). Plotting
the log variance of environmental factors, e.g. the soil nitrogen, soil phosphorus
etc., on log distance allows comparisons of both the heterogeneity of different en-
vironments with respect to that given factor and the different factors within a par-
ticular environment. With consideration of environmental factors, of course, eco-
physiological aspects are introduced into the discussion of diversity. Varying
ecophysiological behaviour of given genotypes is expressed as phenotypic plastic-
ity. In the log–log analysis of regression of variance of environmental factors on
distance by Bell et al. (1993) there was no indication that the variance of the physi-
cal environment tended to approach some maximal value as the distance increased.
In contrast, it increased continuously with distance. Thus, the slopes of the log–log
regressions provide a means to compare heterogeneity of environments and factors
in relation to each other, i.e. slopes of the log–log regressions indicate the correla-
tions of nearby sites between each other. Where nearby sites are highly correlated,
selection will tend to favour specialization because dispersing offspring find condi-
tions for growth similar to their parents. This leads to diversity of genotypes. On the
other hand, where nearby sites show little correlation between each other, offspring
tend to find conditions different to their parents. This favours plasticity.
Fig. 3.16 Floristic diversity related to species-area diagrams. (After Whitmore 1990, from Lüttge
1995). αβ. . . Species-poor community (low α-diversity) with low β-diversity, viz. few species,
small minimal quadrat; αβ. . . species-poor community (low α-diversity) with high β-diversity,
viz. few species, large minimal quadrat; αβ. . . species-rich community (high α-diversity) with
low β-diversity, viz. many species, small minimal quadrat; αβ. . . species-rich community (high α-
diversity) with high β-diversity, viz. many species, large minimal quadrat; γ . . . . bi- (or multiphasic)
curves indicating γ -diversity
3.3.2 Diversity and Plasticity and the Biological Stress Concept
Environmental factors direct us to the consideration of the role of stress, because

any environmental factor can become a stress factor or stressor if its dosage is too
high or too low. This is explained by the biological stress concept as described in
Box 3.1. We may ask the questions as to how stress may be involved in regulating
plasticity and diversity and whether in fact, plasticity and diversity are related.
In an experiment applying different degrees of stress to experimental micro-
cosms, Grime et al. (1987) have demonstrated that high diversity is given only
within a rather narrow range of stressed conditions. For conditions in the British
Isles high species diversity occurred at stress intensities which allowed no less than
350 g m−2 dry biomass, but no more than 750 g m−2 . At lower stress there is dom-
inance of one or only a few robust and competitive species. At higher stress only
a few highly adapted “specialists” survive as competition peaks in circumstances of
abundance. In relation to plant nutrition, despite the abundance of resources avail-
able to vegetation established on fertile soil, plant growth results in the simultaneous
development of depletion zones which creates circumstances of unequal access and
traps subordinates in the depletion zones while supporting the monopoly by growth
forms with the genetic potential for coarse-grained foraging (Campbell et al. 1991).
In a Malaysian rainforest it was shown that the species diversity of trees and vines
strongly responded to combined phosphorus and potassium concentrations in the
soil. Diversity was highest on soils with a medium P + K index, it declined towards
both very poor soils and soils with more ample nutrient supply (Fig. 3.17). The re-
lationship between diversity and environmental richness can also be demonstrated
for tropical birds and mammals (Reichholf 1994).
Box 3.1 The biological stress concept
A good physical analogue for developing the terminology is a spring. Stress is

put on the spring by strain. Reversible stress is brought about by strain in the
elastic range of the spring material: elastic strain. Irreversible stress is due to
strain beyond the elastic range of the spring material: plastic strain.
The biological stress concept was developed by Selye (1973), Levitt (1980)
and Larcher (1987). Any external factor (biotic or abiotic) and internal factor
can induce stress, i.e. become a stressor, if its dosage is too high or too low.
The terminology of the biological stress concept is explained in the diagram
giving four different possible cases for the development of a biological system
with time (abscissa).
1. Strong stress
Strong stress out of an alarm phase more or less rapidly leads into a phase
of exhaustion followed by acute damage and death. The stress has negative
effects, it is a distress.
Box 3.1 (Continued)
2. Low stress followed by stress removal

Low stress leads into an alarm phase generating recovery mechanisms.
In a recovery phase the system develops out of the conditions in which
stress has negative effects, to conditions in which stress has positive effects
and stimulates the system; stress is a eustress. The system stabilizes during
a hardening phase and attains a resistance phase, in which it may remain,
unless the degree of stress is changed or external or internal reserves re-
quired for resistance are exhausted. With respect to the latter, it is clear that
time, i.e. the duration of stress application may be important. Upon stress
removal the system enters a dehardening phase and returns to the normal
level.
3. Low stress followed by additional stress
The system first develops like that of case (2), but then additional stress is
applied either by the original stress becoming stronger or by additional dif-
ferent stressor(s). The system now goes into the condition of distress, an
exhaustion phase and chronic damage.
4. Strong stress with acute damage followed by repair after stress removal
The system first develops like that of case (1), but then a stress-free period
follows, and during a repair phase the system is restored and returns to the
“normal” level.
(After Beck and Lüttge 1990.)
Phenotypes are the receivers and modulators of environmental input and produc-
ers of output performance at the community level. The development of phenotypes
from genotypes is the real origin of complexity in biology (Schuster 1998; see also
West-Eberhard 1986, 1989, 2003; Sultan and Bazzaz 1993; Solbrig 1994; Gehrig et
al. 2001). Booth and Grime (2003) tried to test this using microcosms planted with
different species, where each species was present in different microcosms at a dif-
ferent degree of genetic diversity. The genetic diversity was generated by planting
mixtures of varying numbers of different clones of plants obtained from vegetative
propagation. Stress was applied in the form of trampling and artificial grazing. In
Fig. 3.17 Relationship be-

tween the diversity of trees
and vines in a Malaysian
rainforest (ordinate) and
combined phosphorus and
potassium concentrations in
the soil (P + N index on
the abscissa). (After Tilman
1982)
this way it was attempted to relate genetic diversity to community diversity. Di-
versity declined in all of the microcosms, and there were no dramatic differences in
the first 3 years of the 5 years experiment. However, in the last 2 years the decline
of diversity was lower in the microcosms having the largest genetic diversity.
With regard to plasticity it appears that both high stress and low stress do not
favour traits, which support such phenotypic variability. High stress favours spe-
cialized adaptations to the prevailing specific and strong stressor (e.g. frost near
the poles or drought in deserts). Low stress allows the success of few species,
which can competitively procure resources for development of their own biomass
(e.g. nitrophilous plants in sites rich in nitrate and other nutrients). Only medium
stress advances the unfolding of variability. Stress of medium intensity and high
variability in time is typical for the environment of the tropical forests, where the
important factors are:
• nutrients,
• water,
• CO2 ,
• light,
• temperature.
None of these factors ever really becomes extreme, but their dynamic spatiotem-
poral variations and interactions cause stress in tropical forests and determine the
struggle for existence between species (Richards 1996). This makes it important to
study functional diversity and in this respect the observation of cross diversity
by Guehl et al. (2004) is highly interesting. These authors have used the relation
between δ 13 C-data and leaf conductance for water vapour, gH2 O (see Sect. 2.5) to
assess water relations of functional types of rain forest trees, i.e. shade tolerant, hemi
tolerant and heliophilic species. Differences in δ 13 C among species were primarily
driven by gH2 O , and there was some indication for the expected higher water use
efficiency of heliophilic species. However, there was a non-linear pattern in the re-
lations of δ 13 C and the gradient of shade tolerance and the results were not in full
agreement with a simple concept of functional types, so that cross diversity among
main functions or traits may reflect an important aspect in functional diversity.
Phenotypic plasticity must be considered in relation to co-occurrence of differ-
ent genotypes within a population which are each adapted to a slightly different en-
vironment. Genetic variation is reflected in phenotypic plasticity (Booy et al. 2000).
Plasticity itself can be considered as a trait, which is subject to selection (West-
Eberhard 1986, 1989, 2003). However, plasticity per se is not adaptive. This much
depends on the physiological costs of plasticity (van Kleunen and Fischer 2005).
In any case, phenotypic plasticity offers material for selection, since selection is
acting on the phenotypes. This is particularly important in systems, which are not
strictly homeostatic (see Sect. 3.3.3), and where radiative movement of phenotypes
followed by separation may be one of the bases for the development of genetic
diversity (West-Eberhard 1986, 1989, 2003). Thus, the promotion of phenotypic
plasticity by variable and medium stress may be one of the reasons for the extraor-
dinarily high biodiversity of tropical forests (Lüttge 1995, 2005, 2007).
3.3.3 Diversity and the Chaos of Oscillating Mosaics
After assessing diversity and plasticity, this then leads us to the question: Is there
homeostasis? The tropical rainforest is frequently considered to be a climax asso-
ciation. Ideally, climax associations are steady-states for vegetation, representing
stable or homeostatic ecological equilibria, determined by the natural environmen-
tal factors in a given climatic zone. According to the climax theory (Clements 1936)
independent of the starting conditions at a given location, progressive successions
should always lead to the same final equilibrium or climax association. Only when
there is an effect of external influences, such as natural or man-made catastrophes,
are regressive successions elicited, which cause a deviation from the climax asso-
ciation. Of course, the latter involves changes of environmental conditions.
The question arises, however, as to whether the final steady state is really in-
dependent of the starting conditions. Can predictions be made? Is there only one
possible final equilibrium? Or are there several different possible ecological states?
Is it then possible that small initial deviations from the mean climatic conditions
(e.g. mean annual precipitation and temperatures) might divert the development of
succession towards very different end points?
In fact, the straightforward application of the climax theory to large ecosystems
has been challenged by Remmert (1985, 1991), using the very example of the trop-
ical rainforest. He strongly underlines the dynamic nature of rainforests (see also
Whitmore 1990) and in his view the tropical rainforest is subject to a continuous
cycle of series of successional states. It represents a diverse cyclically changing
mosaic pattern (see also Watt 1947). This can be illustrated by considering the dy-
namics of gaps or chablis in the tropical rainforest (Orians 1982; van der Meer and
Bongers 1996). In the original French meaning chablis are clearings in forests due
to storms. In wet tropical forests tall falling trees with their large crowns cause two
adjacent gaps, one beneath the original location of the crown and the other one at
the site of impact on the ground (Fig. 3.18).
When this is set in the context of floristic diversity, destruction such as the forma-
tion of chablis, as well as the introduction of roads may increase diversity because
β-diversity is introduced (Sect. 3.3.1). Gaps and chablis are reinvaded by vegeta-
tion, and with various successional stages the forest is restored (Fig. 3.19). Thus,
such chablis are sites of destruction and renewal, which at any given time may com-
prise 3 – 10% of the total forest area (Jacobs 1988). Larger clearings also result from
shifting agriculture (Fig. 3.20) and other human activities.
In the renewal of forest in natural clearings resulting from falling trees, hurri-
canes, earthquakes, volcanic eruptions, fires and landslides, or in farms abandoned
due to exhaustion of nutrients or the take-over of weeds and pests there is no pre-
determined pattern. As Jacobs (1988) comments “. . . the selection is one of un-
predictable irregularity”. There is a plethora of environmental factors affecting
regeneration and to which the unpredictability of proximal species regeneration will
be related to. It depends among others on the extent of diversity in adjacent veg-
etation and the age of the surrounding communities, the light climate (Torquebiau
1988), the availability and viability of propagules, seed and seedling bank compo-
Fig. 3.18 A,B Formation of gaps or chablis by falling trees. Subsequently the crown gap fills
more rapidly than the gap created by the impact. (After Jacobs 1988). C Fallen tree in a forest of
Sierrania Páru, Venezuela
Fig. 3.19 Reinvasion of a chablis. (After Jacobs 1988)

Fig. 3.20 Slash and burn agriculture (see also Fig. 1.5)
sition (Dalling et al. 1998; Massey et al. 2006), dispersal limitation and dormancy
parameters (Dalling et al. 1998), differences in growth and mortality rates among
species with different carbon allocation patterns (Newell et al. 1993; Dalling et al.
1998), edaphic specialization and soil resources (Phillips et al. 2003; Palmiotto et al.
2004; Valencia et al. 2004). With such unpredictability we are not naturally facing
a stochastic development of diversity. This would be an a priori unjustified sim-
plification. However, we are right back in the realm of non-linear dynamics and
deterministic chaos (Sect. 2.6; Solé et al. 1994; Solé and Manrubia 1995a,b; Man-
rubia and Solé 1996; Levin and Muller-Landau 2001), where we may recall one
of the most characteristic implications of the chaos theory, namely that even the
slightest differences in the starting conditions – or the many factors affecting gap
dynamics – may lead to the most dramatic differences in subsequent development
(see also Fig. 2.15C).
3.3.4 Diversity and Life Forms
Diversity is also given by the variability of life-forms. The life-form concept is de-
fined rather loosely. Many attempts have been made to categorize the diversity of
plant types by distinction of life forms which are morphologically more or less
conspicuous. Life forms represent morphological adaptations to environments or
towards a given stress factor or set of stress factors. An original example illustrating
an approach to such a classification is R AUNKIAER’s crisp distinction of five life
forms of vascular plants according to the life time of shoots and the position and
protection of regenerating buds, namely:
• the phanerophytes with regenerative buds higher than 50 cm above ground,
• the chamaephytes with buds closer to the ground (10 – 50 cm above ground),
• the hemicryptophytes having a close contact of the buds with the ground,
• the cryptophytes with below ground regenerative organs (rhizomes, onions,
bulbs, storage roots etc.),
• the therophytes or annuals.
These life forms are particularly conceived for analysis of vegetation in mesic and
temperate climates as they describe strategies for over-wintering. However, they
may also be used to describe adaptation to regular seasonal drought periods and,
thus, are useful with respect to the tropical environments. Moreover, of course, the
distinction between trees (macrophanerophytes), shrubs or bushes (microphanero-
phytes) and herbs of different forms inherent in these definitions is always appli-
cable. However, this scheme is hardly sufficient to come to grasp of diversity in
a tropical forest.
Vareschi (1980) has noted that as a basis for schemes defining different life forms,
morphological modifications of any of the major plant organs could be chosen, e.g.
life forms based on roots, shoots, leaves, flowers or propagules. Another approach
would be to derive life forms which use morphological modifications at the whole
plant level with distinct mechanistic relations. Both shall be illustrated here by giv-
ing a few specific examples.
3.3.4.1 Root Categories
Root categories of trees frequent in tropical forests are stilt roots and buttress roots
(Fig. 3.21). It has been debated whether such roots have mainly mechanical func-
tions or serve aeration and O2 -supply to below ground root tissues. Presumably
both are important. Buttress roots, in particular, may function like ropes with ef-
fective anchoring of the trunk to the ground (Mattheck 1992) and it was shown
experimentally that they indeed have a clear anchorage function (Crook et al. 1997).
In addition, the increase in above-ground root surface brought about by these root
types may facilitate aeration. In wet tropical soils, where gas diffusion is limited and
where vigorous soil-respiration will lower O2 -concentration, this may be a partic-
ularly important aspect. Indeed, immediately below ground the buttress roots show
much branching and produce many fine absorptive roots, which can be supplied
with O2 via pores in the bark of the above ground buttress (Fig. 3.21E,F). Stilt roots
also are often covered with lenticels facilitating gas exchange with the atmosphere
(Fig. 3.21G,H).
3.3.4.2 Leaf Categories
The most variable plant organ in form is the leaf. Vareschi’s (1980) leaf analyses
of plants in the cloud forest of Rancho Grande in Venezuela reveal more than 300
forms, most of which are reproduced in Fig. 3.22 for the sake of their graphic attrac-
tiveness. R AUNKIAER’s size classes of leaves offer a more systematic approach,
where leaves are distinguished by their area:
• megaphyll > 1,500 cm2 ,
• macrophyll 1,500 – 180 cm2 ,
• mesophyll 180 – 20 cm2 ,
• microphyll 20 – 2 cm2 ,
• nanophyll 2 – 0.2 cm2 ,
• leptophyll < 0.2 cm2 .
Table 3.1 gives an idea of the percentage distribution of leaf sizes in a rainforest
and an evergreen bushland. It shows that larger leaves predominate in the rainforest
whilst smaller leaves are found in the bushland. Naturally, the leaf shape is an es-
sential additional feature of diversity (Fig. 3.22). It is difficult, however, to delineate
distinct categories, and without any mechanistic basis this approach may devalue the
use of life form classifications. Vareschi (1980) lists about 18 leaf forms according
to shape and special surface features, in addition to R AUNKIAER’s size classes, and
then derives a diversity coefficient, cd as follows:
cd = n · f , (3.1)
Fig. 3.21A–H Root types of trees in tropical forests. Giant buttress roots in a rain forest in French
Guyana (A,B). Buttressed tree in the cloud forest of Rancho Grande, northern coastal range of
Venezuela (C). Stilt roots of Pandanus (D) in Queensland, Australia. Buttresses with absorptive
roots in a rain forest in French Guyana (E,F). Palm in a forest of the Gran Sabana, Venezuela
(G,H), with lenticels clearly seen in H

Fig. 3.22 Leaf-forms obtained from leaf analyses in the cloud forest of Rancho Grande, Venezuela
(Vareschi 1980, with kind permission of R. Ulmer)
where n is the number of species occurring and f the number of leaf categories. The
lowest possible value of cd is 1 (one species only occurring). In mesic environments
with few species and low leaf form diversity it may be several tens or hundreds,
whereas in tropical environments under favourable conditions it may reach up to
20,000.
Table 3.1 Percentage distribution of leaf sizes in an evergreen rainforest in Brazil and an evergreen
bushland at Port Henderson Hill, Jamaica (Medina 1983)
Rainforest macrophanerophytes Bushland microphanerophytes

Number of species: 49 43
Macrophyll 2 –
Mesophyll 75 16
Microphyll 16 74
Nanophyll 2 7
Leptophyll 4 2
3.3.4.3 Stem Characteristics
Conspicuous life forms based on stem characteristics are the fleshy and woody
stem succulents (Sects. 8.2.3.2.1 and 10.1.2.2, respectively) and the xylohemicryp-
tophytes with their xylopodia or lignotubers (Sect. 10.1.2.2).
3.3.4.4 Whole Plant Modifications
Very typical special life forms of vascular plants in tropical forests are epiphytes,
hemi-epiphytes and lianas to which a special chapter (Chap. 6) is devoted.
Another example are the myrmecophytes (ant plants). The symbiosis between
plants and ants plays a particularly important role in the tropical environment. It
is frequently found in epiphytes (see Sect. 6.6.3) but also in terrestrial plants, where
the genera Tococa (Melastomataceae), Cecropia (Moraceae) and woody Legumi-
nosae are the best known ant plants. In the Sira-mountains of Peruvian Amazo-
nia, Morawetz and Wallnöfer (1992) counted that 4.4% of all species were genuine
myrmecophytes. The ant plants prefer disturbed sites, which may be naturally due
to land-slides; 86% of all genuine myrmecophytes (38 species) were found in such
sites, while only 14% (6 species) regularly occurred in the primary forest.
The plants provide hollows, so-called domatia, where the ants find protected
spaces for nests, e.g. in the inflated leaf bases of Tococa (Fig. 3.23A), the hollow
stems of Cecropia (Fig. 3.23B) and thorns and leaf petioles of many Leguminosae
(Caesalpiniaceae, Mimosaceae). In addition the ants may receive nutrition in the
form of nectar from extra floral nectaries or various food bodies (Webber et al.
2006) and special nutritive appendices containing fat and oils, protein and carbohy-
drates, e.g. the elaiosomes (appendices of seeds) or the Müller-bodies of Cecropia
which are made up of glycogen and not the usual plant-storage carbohydrate starch.
It is thought that in return the ants provide to the plants protection from phy-
tophagous animals especially insects (Davidson and Epstein 1989; Duarte Rocha
and Godoy Bergallo 1992). It has also been observed that ants keep their host plants
free of epiphytes. A most astonishing story has been reported by Morawetz et al.
(1992). The Myrmecochista ants of Tococa occidentalis systematically and rapidly
kill all angiosperms coming closer than 4 m to their host plants. T. occidentalis is
Fig. 3.23A, B Ant-nest plants. A Tococa sp. with ant nests in the inflated leaf bases (arrows).
B Cecropia sp. with ant nests in the hollow stem
a light demanding species and its growth and proliferation is stimulated by the ants’
clearing of the surrounding competitors. Thus, after an initial T. occidentalis-plant
has been colonized by the ants, pure T. occidentalis stands with a diameter of several
Fig. 3.24 Stand of Tococa

occidentalis (To) with a sur-
rounding safety corridor (s)
and adjacent vegetation. (Af-
ter Fig. 2 in Morawetz et al.
1992, with kind permission of
the author and Chapman and
Hall)
meters (10 – 30 m) may then develop, around which the ants even maintain a “safety
corridor” (Fig. 3.24). Using their mandibles the ants cut the veins of the leaves of
the competing plants. In the case of palmate leaves they attack the point at the base,
where all veins join; pinnate venation is destroyed by cutting the first and second
order veins at the base; the veins of moncotyledons (e.g. palms) are cut one by one
along the entire leaves. After cutting the veins the ants inject a poisonous excretion
from their abdomen. Apical meristems are also attacked. In this way the plants die
rather rapidly. Although the ants can effectively kill 10 – 50 m tall trees, they do not
usually attack such emergent trees at some distance of their host plants which may
then form a closed canopy 10 – 25 m above the stand of T. occidentalis. As the light
demanding T. occidentalis-plants die away back in the shade, the stand deteriorates
and the ants emigrate to start a new cycle elsewhere.
3.4 Vertical Structure
Stratification, meaning the vertical structure of tropical forests, is directly linked to

local action of specific environmental factors, such as light, temperature, humidity,
CO2 and minerals, the vertical distribution of which can be described. The vertical
structure of tropical forests is determined by several more or less distinct and typical
canopy layers. In simplified terms one may distinguish three major layers:
• a layer of emerging giant trees up to 60 – 80 m tall,
• an intermediate main canopy layer up to 24 – 26 m,
• a lower canopy layer,
(Whitmore 1990), as shown in the schematic transect of Fig. 3.25. More realistic
transects of actual forests often show a larger complexity, and there is also much
diversity of vertical structures among forest types.
3.4 Vertical Structure 85
Fig. 3.25 Schematic representation of the strata structure of a tropical forest
The abundant plant life in these various strata determines vertical gradients of
many important environmental factors such as:
• light intensity and spectral composition,
• temperature,
• air humidity,
• CO2 -concentration,
• mineral nutrients.
3.4.1 Irradiance
Due to absorption by the foliage intensity of irradiance may decrease exponentially

from the main canopy layer down to the forest floor, which often obtains only a few
per cent of the intensity received by the upper canopy or by a large forest clearing
(Fig. 3.26). One consequence of such light gradients in tropical forests is that the
maximum heights individual trees may reach are negatively correlated with shade
tolerance and that there is a positive correlation of maximum height with light sat-
urated rates of photosynthesis which are usually higher in sun plants (Sect. 4.1.1)
(Davies et al. 1998).
Light-absorption by the canopy of forests may be treated according to Lambert–
Beer’s law, which is well known from photometry. The ratio of the light intensity I
Fig. 3.26 Light penetration through the canopy of a tropical forest. (After Jacobs 1988)
of a beam passing through a sample solution of a thickness d and the light intensity
of the incident light beam Io is given by
I
= 10−εdc , (3.2)
Io
where ε is the molar extinction coefficient and c the concentration of the sample. In
analogy, the situation for canopies may be written as
Il
= e−kLAIo−e . (3.3)
Io
Here Io is the light intensity outside the canopy and Il the intensity at level l,
LAIo−e is the leaf area index between the top of the canopy, o, and level l, and k is
a constant. The leaf area index is dimensionless and is given by a projection of all
foliage onto a certain level l or onto the ground; it thus is the ratio of
[total foliage area above a unit area at layer l ] : [unit area] ,
and a typical value for tropical rainforests is 8.

Light-absorption by the canopy not only reduces light intensity but also changes
the light quality or spectral composition. In the red region of the spectrum absorp-
tion by photosynthetic pigments changes the red:far red ratio (R:FR) so that phy-
tochrome regulated processes, which respond to this ratio are affected (Sect. 4.2.2).
In a low-land rainforest in Costa Rica R:FR was found to be 1.23 in a large clear-
ing but only 0.42 on the forest floor (Chazdon and Fetcher 1984) and in rain
forests in French Guiana R:FR values of 0.10 to 0.15 were recorded (Bongers et
al. 2001).
The degree to which the canopy is closed above the forest floor or leaves open-
ings for light penetration can be determined by quantitative image analysis of pho-
tographs taken around noon with wide-angle or fish-eye lenses pointing upwards
from the ground (Fig. 3.27). In addition to the diffuse light filtering through the
canopy foliage, the forest floor may also obtain light in the form of light flecks.
Light flecks occur when movements of leaves in the wind or the changing angle of
the sun allow direct light penetration for intermittent periods of time. These light
flecks may provide up to 80% of the total irradiation received by the forest floor,
and their intensity ranges from 10 to 70% of full sunlight. They are important for
photosynthetic productivity (see Sect. 4.2.1). Figure 3.28 shows that for short peri-
ods lower strata in tropical forests may obtain quite high irradiance, which may at
times exceed that received by strata higher up. Thus, the gradual decline of irradi-
ance from the top of the canopy to the ground shown in Fig. 3.26 does not always
correspond to the actual situation.
Fig. 3.27 Vertical fish-eye

camera view from the floor to
the canopy of a rain forest in
Panamá. (Photograph Bettina
Engelbrecht)
Fig. 3.28 Daily course of irradiance at different strata of a wet tropical forest in northern Australia
on an almost cloudless day; i.e. 36 m = above the canopy, 22 m and 10 m = increasingly lower
levels inside the forest. (After Doley et al. 1987)
3.4.2 Temperature and Air Humidity
Diurnal variations of soil temperature and air temperature 1.5 m above the ground
within a forest in Surinam in comparison to small and large clearings are shown in
Fig. 3.29. There is strong dampening of daily changes of temperature inside the
forest, where it remains much cooler than in the gaps and clearings.
Relative air humidity (RH) is related to temperature. Hence, the water-vapour
pressure saturation deficit of the atmosphere at the top of the canopy and at var-
ious levels inside the forest shows a similar pattern (Fig. 3.30). It is close to zero
at all levels at sunrise, shows a maximum at 14.00 h and decreases inside the for-
est from the higher to the lower strata. Since water-vapour pressure deficit of the
atmosphere determines the driving force for transpiration, it constitutes a highly
important ecophysiological factor. High RH also increases the CO2 -sensitivity of
stomata and there is acclimation of stomatal responses which may play a role for
photosynthesis in the lower strata of tropical forests (Talbott et al. 2003).
Fig. 3.29A,B Daily course of soil temperature at 2 cm depth (A) and air temperature 1.5 m above
the floor (B) in a forest in Surinam. Comparisons between the closed forest and clearings are made
for the hot and dry season. (After Jacobs 1988)
Fig. 3.30 Daily course of

water-vapour pressure satura-
tion deficit of the atmosphere
at different strata of a forest in
Surinam during the dry sea-
son; i.e. above the canopy and
at increasingly lower levels
inside the forest as indicated.
(After Jacobs 1988)
3.4.3 Carbon Dioxide
Carbon-dioxide concentration in the atmosphere inside forests is influenced by pho-

tosynthesis and respiration of the organisms living in the forests including soil res-
piration (Buchmann et al. 1996). Daily averages of CO2 -concentration at the soil
surface may be quite large, i.e. up to 1,000 ppm, due to the respiration of plant roots
and soil organisms. One meter above the floor of two forests of the upper Rio Negro
Basin in Venezuela, daily average CO2 -concentration was still 508 and 541 ppm re-
spectively, and then showed little decline for up to 20 m (Medina et al. 1986). More
detailed analyses of the diurnal and seasonal dynamics of the vertical profile of CO2
concentration in a tropical rain forest are shown in Fig. 3.31A. Clearly at times
plants within the canopy may photosynthesize at CO2 -concentrations well above
40 A B
Canopy height ( m )
30
20
10
0
300 400 500 600 -17 -14 -11 -8
CO2 ( mmol mol -1 ) δ13Cleaf ( ‰ )
Canopy height ( m )
C Brazil Trinidad French Guiana

30
20
10
0
-4 -2 0 - 4 -2 0 -4 -2 0
Depletion in δ13Cleaf ( ‰ )
Fig. 3.31A–C Vertical gradients of CO2 and carbon isotope signatures. A CO2 concentrations of
the air at different levels above the soil surface in a tropical rain forest of French Guiana, where
closed symbols are for the early morning and open symbols for midday, when the vegetation has
already reduced the CO2 levels by photosynthesis, and circles and triangles are for the wet and
the dry season, respectively. B δ 13 C values of the air (symbols as for A). C Depletion of δ 13 C in
the leaf biomass in three lowland tropical forests, where the open and closed symbols for French
Guiana are for the dry and the wet season, respectively. The δ 13 C values of the leaves at the top of
the vertical profile were taken as a reference (zero depletion) to calculate the intra-canopy depletion
of δ 13 C with decreasing height above the ground. (Modified after Buchmann et al. 2004)
the average CO2 -concentration in the atmosphere outside the forests. It is seen in
Fig. 3.31B,C that there are also vertical gradients of δ 13C of the air and of the leaf
biomass (Buchmann et al. 2004; see also Sect. 2.5).
3.4.4 Mineral Nutrients
3.4.4.1 Inorganic Nutrient Cycling
When we consider inorganic nutrient cycling in tropical forests we must include, of

course, the soil. It is frequently assumed that the major portion of minerals in wet
tropical forests is bound in the living biomass. This is not always true and the soil
may contain a considerable fraction of the minerals in the ecosystem. It is often the
soil which is the most vulnerable part in tropical forests. Exposed by unbalanced
logging systems or methods of shifting agriculture it may rapidly become oxidized
and eroded. With vegetation and soil we obtain a conspicuous vertical structure
(Figs. 3.32 and 3.33). Mineral nutrients show variability over the horizontal strata
of forests, because interaction with leaves and stems causes precipitation to become
enriched in nutrients and the throughfall of rain and the stemflow in the forests
show much higher concentrations of most mineral ions than the rain water itself
(Table 3.2).
An example of inorganic nutrient cycling in a wet tropical forest is given in
Fig. 3.32. Input of minerals to the soil is via rain, canopy leaching with throughfall
and stemflow (see also Table 3.2) and via litter fall (see also tables on page 232 and
Fig. 10.2, page 277 in Richards 1996).
Table 3.2 Mineral nutrients in rainwater, throughfall and stemflow. (A) Average nutrient concen-
tration in rainwater and annual input in several tropical locations throughout the world. (Data from
Medina and Cuevas 1994) (B) Values measured in a forest in Central-America. (Data from Junk
and Furch 1985)
A B
Nutrient
Nutrient concen- Annual input Nutrient concentration
tration in via rainwater (µmol/l)
rainwater (µmol/l) (mmol/m2 )
Rainwater Throughfall Stemflow
Na n.d. n.d. 5.2 11.7 91.7
K 1.0 20 2.7 31.8 168.7
Ca 11.1 200 1.8 6.3 43.0
Mg 8.2 152 0.8 7.8 39.9
NH+4 -N 25.9 493 12.1 < 3.6 657.1
NO−3 -N 9.3 196 7.9 40.0 19.3
PO4 -P 0.7 15 0.1 4.9 3.1
SO4 -S 21.8 424 n.d. n.d. n.d.
Fig. 3.32 Diagram of inorganic nutrient cycling in the lower montane rainforest at Kerigoma, New
Guinea. Numbers are in kg ha−1 year−1 for flows (arrows) and in kg ha−1 for pools (boxes). (After
Whitmore 1990 by permission of Oxford University Press)
Tropical forests frequently have a nutrient limitation of some sort. Although

the soil is often poor in mineralized nutrients the vegetation is luxuriant and so
highly rich in species. As in savannas (Sect. 10.2.2), phosphorus is often the most
problematic element. Mycorrhizal symbioses between plant roots and fungi are im-
portant (Medina and Cuevas 1994; Béreau et al. 2004) and there is feedback be-
tween the essential controlling steps of retranslocation and mineralization in trop-
ical rain forest ecosystems dominated by ectomycorrhizal trees (Chuyong et al.
2000). P/N-ratios in canopy leaves of tropical humid forests range from 15 to 35

(mol/mol) ×103 (Medina and Cuevas 1994) similar to those in plants of savan-
nas. Nitrogen cycles for a semideciduous forest are shown in Fig. 3.33. Nodu-
lation and atmospheric dinitrogen fixation is nutrient limited in tropical forests
(Souza Moreira et al. 1992). In a rainforest in French Guyana, however, 43% of
the Leguminosae were fixing N2 , the contribution of N2 -fixation to their nitro-
gen nutrition averaged 54%, and overall the N2 -fixing species are important for
the performance of the ecosystem increasing its nitrogen-biomass by 10% (Dom-
enach et al. 2004). N2 -fixation capacity of the rain forest trees is controlled and
Fig. 3.33 Compartmentation and annual turnover of nitrogen in a semideciduous forest in Ghana.
(From the Ecology of Neotropical Savannas by Guillermo Sarmiento. Copyright © 1984 by the
President and Fellows of Harvard College. Reprinted by permission of Harvard University Press
(Sarmiento 1984))
limited by low phosphorus and high nitrogen levels in the soil (Pons et al. 2007;
Sect. 10.2.3.2.2).
Roots, which often form very dense mats, are mainly restricted to the upper 0.1 –
0.3 m of the soil, and frequently the soil layer itself on top of the bed rock is rather
thin. Soil respiration, which is very high in wet tropical forests has a typical rate of
4 µmol CO2 m−2 s−1 (Buchmann et al. 2004) and corresponds to 600 – 670 g organic
matter m−2 year−1 , of which 67 – 82% is due to respiration within root mats. Rates
of mineralization of organic litter are high and recirculation of minerals is rapid.
Phosphorus in the soils of humid tropical forests is correlated with the litter fall
mass (Silver 1994). Rapid recirculation is important for avoiding nutrient leaching
(Chuyong et al. 2000). Litter turnover may support seedling growth (Brearley et al.
2003) and is essential for the development of photosynthetic capacity (Santiago and
Mulkey 2005).
Ants play a role in biomass turnover and mineralization. In myrmecophytes the
ants may contribute to nutrition of their hosts (see Sect. 6.6.3). In Tococa guianen-
sis it was observed that one of the two adjacent domatia at the base of each leaf (see
Fig. 3.23A) is used for nesting and the other for dumping excrement and debris.
By radioactive tracer studies the inner surface of the domatia has been shown to be
absorptive of low molecular substrates like amino acids and phosphate, in contrast
to the surfaces of the leaf lamina. Thus, it is highly likely that nutrients are absorbed
from the rotting material in the trash-domatium (Nickol 1992). On the other hand,
in contrast to other reports in the literature that in Cecropia (Fig. 3.23B) 93% of its
nitrogen is supplied by ants, a recent study suggested that it is only ≤ 1% (Fischer
et al. 2003).
Leaf-cutter ants (Fig. 3.34) are a special case. (In the following text I follow the
monograph of Wirth et al. 2003.) Herbivory in general can enhance nutrient cycling
by, e.g.
• enhancing the leaching rate of nutrients from foliage,
• increasing the rate of litter fall,
• stimulating nutrient cycling and turnover within plants,
• promoting activities of decomposer organisms.
Table 3.3 Selection of some quantitative data on the activity of leaf-cutter ants, Atta colombica,
extracted from Wirth et al. (2003) and Herz et al. (2006)
Nest size (m2 ) A few to about 100

Dry leaf mass collected (kg per year)
average 273 ± 161 (SD)
range 85 – 470
Leaf area collected (m2 per year)
average 2700
range 835 – 4550
Leaf area (m2 ) collected per ground area (ha) per year 1217
Dry biomass (kg) collected per ground area (ha) per year 132
Foliage loss of host plants (%) 8 – 40
Fig. 3.34 Leaf-cutter ants
Leaf-cutter ants are a special case because they collect an enormous amount of plant
biomass and carry it into complicated underground chamber systems where they cul-
tivate fungi. Some quantitative data are compiled in Table 3.3. The loss of foliage by
the host plants of up to 40% can reduce their fitness. However, there is also compen-
satory feedback. Leaf-cutter ants prefer to collect leaf pieces from the upper canopy.
Partial defoliation causes an increase in the frequency and variability of light flecks
and may result in higher rates of photosynthesis due to increased light penetration
and incident irradiance on the remaining parts of tree crowns. Thus, herbivory by
leaf-cutter ants appears to have little effect on whole forest canopy photosynthesis
although it may considerably reduce photosynthesis locally. The effects on nutri-
ent flow in the forests are patchy because the ants, Atta colombica as the species
of the case study of Wirth et al. (2003), bury the nitrogen rich exhausted fungal
substrate or refuse from their fungi gardens in large refuse chambers below the fun-
gus chambers of their nests at a depth of 7 m below the surface and only partially
on the soil surface. Only deep rooting trees may have access to the former but fine
roots of nearby plants may penetrate the latter. The refuse is enriched in nitrogen
because the ants prefer to collect the more N-rich young leaves and use more the
carbon than the nitrogen. The carbon/nitrogen ratios are 22 in canopy foliage and
36 in normal leaf litter, 21 in the leaf harvest of the ants and only 16 in the refuse
dump from the ant’s fungi gardens. In the study area of Wirth et al. (2003) on Barro
Colorado Island, Panamá, in locations covering less than 0.5% of the area popu-
lated by leaf-cutting ants nitrogen flux is therefore about 20 – 30 times higher than
in the rest of the forest. Hence, while a large scale benefit of plant nutrition from
leaf-cutter ants is debatable local positive effects on plant growth and fitness are
evident.
3.4.4.2 Nutrient Partitioning
Different partitioning of inorganic nitrogen assimilation between the roots and

shoots of trees is observed in pioneer and mature phase trees of tropical forests.
In gaps the mineralization of a large mass of fresh litter, e.g. from fallen trees, may
lead to higher availability of NO− +
3 -N (not NH4 -N) and PO4 -P (Denslow et al. 1998).
Thus, rapidly growing pioneer or colonizing tree species, which are exposed to high
irradiation, exhibit a large capacity to assimilate nitrate in their leaves, where light
energy can be directly used in photosynthetic nitrate reduction (Aidar et al. 2003).
Partitioning of NO− 3 -assimilation between roots and shoots is strongly related with
average daily photosynthetically active radiation rather than the availability of NO− 3
in the soil (Stewart et al. 1992). Leaves of shaded species have low levels of nitrate
reductase and show little capacity to utilize nitrate, even when it was readily avail-
able, and primarily assimilate ammonia (Stewart et al. 1988, 1990; Fredeen et al.
1991; Fredeen and Field 1992; Denslow et al. 1998).
3.4.4.3 Flushing of New Leaves and Longevity of Mature Leaves

Related to Nutrient Budgets
An interesting phenomenon, which may also be related to nutrient budgets is leaf

flushing. New leaves and shoots expand from their buds very rapidly to attain
a size close to that of mature leaves, much before they reach their final rigidity
and pigmentation. In fact they hang down from the branches as if wilted, and often
are coloured brightly yellow or red (Fig. 3.35). The development of chloroplasts
and the photosynthetic apparatus is delayed which are both particularly nitrogen-
demanding. This can be considered to be an adaptation to conditions of high fungal
and herbivore damage to the expanding leaves. Damage may be 100 times higher
to young than to mature leaves. Mature leaves are better protected (Kursar and Co-
ley 1992a,b; Schlindwein et al. 2006). Costs of damage to the newly flushed leaves
remains low since not so many resources have been invested in them. Resource allo-
cation to leaves becomes beneficial when they mature and establish photosynthetic
productivity in return. Delayed greening is observed in many species (Kursar and
Coley 1992b; Miyazawa and Terashima 2001) and occurs mainly in shade tolerant
species and not in gap-requiring species. In the shade a late development of pho-
tosynthesis is less disadvantageous than in high light (Kursar and Coley 1992b).
It would be interesting to know if the bright colour of freshly flushed leaves even
functions in attracting herbivores to these “cheap” leaves, thus protecting the “ex-
pensive” mature leaves. In a tropical dry-deciduous forest and a dry-thorn forest in
India, phenological strategies have also been observed in relation to leaf flushing.
Flushing occurs in the dry season and reaches a peak before the onset of the rains.
Herbivorous insects emerge with the rains and attain a peak biomass during the wet
months, so that early leaf flushing and maturation provides protection (Murali and
Sukumar 1993).
Nutrient availability also affects the structure and longevity of leaves of forest
trees. Leaf longevity may vary in different tropical forest tree species from about
Fig. 3.35A,B Leaf-flushing. A in a mango tree, B in Brownea sp.
18 months to several years (Richards 1996). It is highly plastic and can respond to
light (Osada et al. 2001). Small leathery leaves (“scleromorphic microphylls”) are
developed on infertile soils due to N- but mainly P-deficiency (Medina and Cuevas
1989; Medina et al. 1990). Such leaves are more durable and better protected from
herbivory (Choong et al. 1992) than large, thin leaves. Thus, nutrient investment in
leaf structure provides a return in the form of photosynthetic products for a longer
period of time. Deciduous and evergreen species coexist in tropical dry forests. They
differ greatly in their investments of resources for leaf construction and mainte-
nance. In deciduous species, with roots occurring under relatively nutrient-rich con-
ditions, leaves can have a potentially high nitrogen-use efficiency (CO2 -assimilation
related to leaf N-content; see Sect. 4.1.2). Conversely, in evergreen species with
lower nitrogen-use efficiency, the long residence time of nitrogen is favourable be-
cause roots occur in nutrient-poor soil microhabitats (Sobrado 1991). Both decid-
uous and evergreen species preserve nitrogen resources. Reserves of nitrogen are
maintained in the twigs in drought-deciduous species and in the older leaves in ever-
green species, providing some nitrogen for the reconstruction of new leaves follow-
ing drought and during leaf exchange respectively (Sobrado 1995). In conclusion,
plant species obviously allocate resources either to obtain a high photosynthetic as-
similation rate from large and fragile leaves for a brief time or to provide a resistant
physical structure which results in a lower rate of CO2 assimilation over a longer
time (Reich et al. 1991). Thus, mineral nutrition influences the lifespan of leaves.
References 97
References
Aidar MPM, Schmidt S, Moss G, Stewart GR, Joly CA (2003) Nitrogen use strategies of neotrop-
ical rainforest trees in threatened Atlantic Forest. Plant Cell Environ 26:389–399
Beard JS (1946) The natural vegetation of Trinidad. Oxford Forestry Memoirs, No 20. Oxford
University Press, Oxford
Beard JS (1955) The classification of tropical American vegetation types. Ecology 36:89–100
Beck E, Lüttge U (1990) Streß bei Pflanzen. Biol Unserer Zeit 20:237–244
Bell G, Lechowicz MJ, Appenzeller A, Chandler M, DeBlois E, Jackson L, Mackenzie B, Preziosi
R, Schallenberg M, Tinker N (1993) The spatial structure of the physical environment. Oe-
cologia 96:114–121
Béreau M, Louisanna E, Garbaye J (2004) Mycorrhizal symbiosis in the tropical rainforest of
French Guiana and its potential contribution to tree regeneration and growth. In: Gourlet-
Fleury S, Guehl J-M, Laroussinie O (eds) Ecology and management of a neotropical rainforest.
Elsevier, Amsterdam, pp 114–119
Bongers F, Meer PJ van der, Théry M (2001) Scales of ambient light variation. In: Bongers F,
Charles-Dominique P, Forget P-M, Théry M (eds) Nouragues. Dynamics and plant animal
interactions in a neotropical rainforest. Kluwer Academic Publishers, Dordrecht, pp 19–29
Booth RE, Grime JP (2003) Effects of genetic impoverishment on plant community diversity.
J Ecol 91:721–730
Booy G, Hendriks RJJ, Smulders MJM, Groenendael JM van, Vosman B (2000) Genetic diversity
and the survival of populations. Plant Biol 2:379–395
Brearley FQ, Press MC, Scholes JD (2003) Nutrients obtained from leaf litter can improve the
growth of dipterocarp seedlings. New Phytol 160:101–110
Buchmann N, Kao W-Y, Ehleringer JR (1996) Carbon dioxide concentrations within forest canopies
– variation with time, stand culture, and vegetation type. Global Change Biol 2:421–432
Buchmann N, Bonal D, Barigah TS, Guehl J-M, Ehleringer JR (2004) Insights into the carbon
dynamics of tropical primary rainforests using stable carbon isotope analyses. In: Gourlet-
Fleury S, Guehl J-M, Laroussinie O (eds) Ecology and management of a neotropical rainforest.
Elsevier, Amsterdam, pp 95–113
Campbell BD, Grime JP, Mackey JML, Jalili A (1991) The quest for a mechanistic understanding
of resource competition in plant communities: the role of experiments. Funct Ecol 5:241–253
Chazdon RL, Fetcher N (1984) Light environments of tropical rainforests. In: Medina E, Mooney
HA, Vázquez-Yanes C (eds) Physiological ecology of plants in the wet tropics. Dr W Junk,
The Hague, pp 27–50
Choong MF, Lucas PW, Ong JSY, Pereira B, Tan HTW, Turner IM (1992) Leaf fracture toughness
and sclerophylly:their correlations and ecological implications. New Phytol 121:597–610
Chuyong GB, Newbery DM, Songwe NC (2000) Litter nutrients and retranslocation in a central
African rain forest dominated by ectomycorrhizal trees. New Phytol 148:493–510
Clements FE (1936) Nature and structure of the climax. J Ecol 24:252–284
Crook MJ, Ennos AR, Banks JR (1997) The function of buttress roots: a comparative study of
the anchorage systems of buttressed (Aglaia and Nephelium ramboutan species) and non-
buttressed (Mallotus wraja) tropical trees. J Exp Bot 48:1703–1716
Dalling JW, Hubbell SP, Silvera K (1998) Seed dispersal, seedling establishment and gap partition-
ing among tropical pioneer trees. J Ecol 86:674–689
Davidson DW, Epstein WW (1989) Epiphytic associations with ants. In: Lüttge U (ed) Vascular
plants as epiphytes. Evolution and ecophysiology. Ecological Studies, vol 76. Springer, Berlin
Davies SJ, Palmiotto PA, Ashton PS, Lee HS, Lafrankie JV (1998) Comparative ecology of 11
sympatric species of Macaranga in Borneo: tree distribution in relation to horizontal and ver-
tical resource heterogeneity. J Ecol 86:662–673
Denslow JS, Ellison AM, Sanford RE (1998) Treefall gap size effects on above- and below-ground
processes in a tropical wet forest. J Ecol 86:597–609
Doley D, Yates DJ, Unwin GL (1987) Photosynthesis in an Australian rainforest tree, Argyroden-
dron peralatum, during the rapid development and relief of water deficits in the dry season.
Domenach A-M, Roggy J-C, Molino J-F, Marechal J, Sabatier D, Prévost M-F (2004) Diversity of
the leguminous tree Rhizobium associations and role of the nitrogen fixation on the stability of
the rainforest in French Guiana. In: Gouret-Fleury S, Guehl JM, Laroussinie O (eds) Ecology
and management of a neotropical rainforest. Elsevier, Amsterdam, pp 120–143
Duarte Rocha CF, Godoy Bergallo H (1992) Bigger ant colonies reduce herbivory and herbivore
residence time on leaves of an ant-plant: Azteca muelleri vs. Coelomera ruficornis on Cecropia
pachystachya. Oecologia 91:249–252
Esteves FA (1998) Considerations on the ecology of wetlands, with emphasis on Brazilian flood-
plain ecosystems. In: Scarano FR, Franco AC (eds) Ecophysiological strategies of xerophytic
and amphibious plants in the neotropics. Oecologia Brasiliensis, vol IV, Universidade Federal
do Rio de Janeiro, Rio de Janeiro, pp 111–135
Fernandez MD, Pieters A, Donoso C, Herrera C, Tezara W, Rengifo E, Herrera A (1999) Seasonal
changes in photosynthesis of trees in the flooded forest of the Mapire river. Tree Physiol 19:79–
85
Fischer RC, Wanek W, Richter A, Mayer V (2003) Do ants feed plants? A 15 N labelling study of
nitrogen fluxes from ants to plants in the mutualism of Pheidole and Piper. J. Ecol 91:126–134
Fredeen AL, Field CB (1992) Ammonium and nitrate uptake in gap generalist and understory
species of the genus Piper. Oecologia 92:207–214
Fredeen AL, Griffin K, Field CB (1991) Effects of light quantity and quality and soil nitrogen status
on nitrate reductase activity in rainforest species of the genus Piper. Oecologia 86:441–446
Gehrig H, Gaussmann O, Marx H, Schwarzott D, Kluge M (2001) Molecular phylogeny of the
genus Kalanchoë (Crassulaceae) inferred from nucleotide sequences of the IST-1 and IST-2
regions. Plant Sci 160:827–835
Grime JP, Mackey JML, Hillier SH, Read DJ (1987) Floristic diversity in a model system using
experimental microsoms. Nature 328:420–422
Guehl J-M, Bonal D, Ferhi A, Barigah TS, Farquhar G, Granier A (2004) Community-level diver-
sity of carbon-water relations in rainforest trees. In: Gourlet-Fleury S, Guehl J-M, Laroussinie
O (eds) Ecology and management of a neotropical rainforest. Elsevier-Amsterdam, pp 75–94
Herz H, Beyschlag W, Hölldobler B (2006) Herbivory rate of leaf-cutting ants in a tropical moist
forest in Panama at the population and ecosystem scales. Biotropica (in press)
Jacobs M (1988) The tropical rain forest. Springer, Berlin Heidelberg New York
Junk WJ (1997) Distribution and size of neotropical floodplains. In: Junk WJ (ed) The Central
Amazon floodplain. Ecological Studies vol 126. Springer, Berlin Heidelberg New York, pp
12–16
Junk WJ, Furch K (1985) The physical and chemical properties of Amazonian waters and their
relationships with biota. In: Prance GT, Lovejoy TE (eds) Amazonia. Pergamon, Oxford, p 7
Kitayama K, Aiba S-I (2002) Ecosystem structure and productivity of tropical rainforests along al-
titudinal gradients with contrasting soil phosphorus pools on Mount Kinabalu, Borneo. J Ecol
90:37–51
Kleunen M van, Fischer M (2005) Constraints on the evolution of adaptive phenotypic plasticity
in plants. New Phytol 166:49–60
Kratochwil A (1998) Biodiversity in ecosystems. Atti dei Convegni Lincei 145:23–62
Kursar TA, Coley PD (1992a) Delayed development of the photosynthetic apparatus in tropical
rain forest species. Funct Ecol 6:411–422
Kursar TA, Coley PD (1992b) The consequences of delayed greening during leaf development for
light absorption and light use efficiency. Plant Cell Environ 15:901–909
Larcher W (1987) Streß bei Pflanzen. Naturwissenschaften 74:158–167
Levin SA, Muller-Landau HC (2001) The emergence of diversity in plant communities. CR Acad
Sci Paris Sci de la Vie 323:129–139
Levitt J (1980) Responses of plants to environmental stresses, vol I. Academic Press, New York
References 99
Lobo PC, Joly CA (1998) Tolerance to hypoxia and anoxia in neotropical tree species. In: Scarano
FR, Franco AC (eds) Ecophysiological strategies of xerophytic and amphibious plants in the
neotropics. Oecologia Brasiliensis, vol IV, Universidade Federal do Rio de Janeiro, Rio de
Janeiro, pp 111–135
Löscher HW, Oberbauer SF, Gholz HL, Clark DB (2003) Environmental controls on net ecosystem-
level carbon exchange and productivity in a Central American tropical wet forest. Global
Change Biol 9:396–412
Lüttge U (1995) Ecophysiological basis of the diversity of tropical plants: the example of the genus
Clusia. In: Heinen HD, San José JJ, Caballero-Arias H (eds) Nature and human ecology in the
neotropics. Sci Guaianae 5:23–26
Lüttge U (2005) Genotypes-phenotypes-ecotypes: relations to Crassulacean acid metabolism.
Nova Acta Leopoldina NF 92/342, pp 177–193
Lüttge U (2007) Physiological ecology. In: Lüttge U (ed) Clusia: a woody neotropical genus of re-
markable plasticity and diversity. Ecol Studies vol. 194. Springer, Berlin Heidelberg NewYork,
pp 187–234
Manrubia SC, Solé RV (1996) Self-organized criticality in rainforest dynamics. Chaos, Solitons
and Fractals 7:523–541
Massey FP, Massey K, Press MC, Hartley SE (2006) Neighbourhood composition determines
growth, architecture and herbivory in tropical rain forest tree seedlings. J Ecol 94:646–655
Mattheck C (1992) Design in der Natur. Der Baum als Lehrmeister. Rombach Freiburg i. Breisgau
Medina E (1983) Adaptations of tropical trees to moisture stress. In: Golley FB (ed) Tropical rain
forest ecosystems, A. Structure and function. Elsevier, Amsterdam, pp 225–237
Medina E (1986) Forests, savannas and montane tropical environment. In: Baker NR, Long SP
(eds) Photosynthesis in contrasting environments. Elsevier Amsterdam, pp 139–171
Medina E, Cuevas E (1989) Patterns of nutrient accumulation and release in Amazonian forests of
the upper Rio Negro basin. In: Proctor J (ed) Mineral nutrients in tropical forest and savanna
ecosystems. Blackwell Oxford, pp 217–240
Medina E, Cuevas E (1994) Mineral nutrition:humid tropical forests. Prog Bot 55:115–129
Medina E, García V, Cuevas E (1990) Sclerophylly and oligotrophic environments: relationships
between leaf structure, mineral nutrient content, and drought resistance in tropical rain forests
of the upper Rio Negro region. Biotropica 22:51–64
Meer PJ van der, Bongers F (1996) Patterns of tree-fall and branch-fall in a tropical rain forest in
French Guiana. J Ecol 84:19–29
Miyazawa S-I, Terashima I (2001) Slow development of leaf photosynthesis in an evergreen broad-
leaved tree, Castanopsis sieboldii: relationships between leaf anatomical characteristics and
photosynthesis and photosynthetic rate. Plant Cell Environ 24:279–291
Morawetz W, Wallnöfer B (1992) Die Ameisenpflanzen entlang eines Transekts durch das Sira-
Gebirge (Peruanisches Amazonien) und ihre ökologische Stellung im Regenwald. Dtsch Ges
Tropenökologie, Jahrestagung Bonn
Morawetz W, Henzl M, Wallnöfer B (1992) Tree killing by herbicide producing ants for the estab-
lishment of pure Tococa occidentalis populations in the Peruvian Amazon. Biodivers Conserv
1:19–33
Murali KS, Sukumar R (1993) Leaf flushing phenology and herbivory in a tropical dry deciduous
forest, southern India. Oecologia 94:114–119
Newell EA, McDonald EP, Strain BR, Denslow JS (1993) Photosynthetic responses of Miconia
species to canopy openings in a lowland tropical rainforest. Oecologia 94:49–56
Nickol MG (1992) Untersuchungen der Myrmekodomatien von Tococa guianensis (Melastomat-
aceae). Dtsch Ges Tropenökologie, Jahrestagung Bonn
Noss RF (1983) A regional landscape approach to maintain diversity. BioScience 33:700–706
Orians GH (1982) The influence of tree-fall in tropical forest on tree species richness. Trop Ecol
23:255–279
Osada N, Takeda H, Furukawa A, Awang M (2001) Leaf dynamics and maintenance of tree crowns
in a Malaysian rain forest stand. J Ecol 89:774–782
Palmiotto PA, Davies SJ, Vogt KA, Ashton MS, Vogt DJ, Ashton PS (2004) Soil-related habitat
specialization in dipterocarp rain forest tree species in Borneo. J Ecol 92:609–623
Phillips OL, Vargas PN, Monteagudo AL, Cruz AP, Zans MEC, Sánchez WG, Yli-Halla M, Rose S
(2003) Habitat association among Amazonian tree species: a landscape-scale approach. J Ecol
91:757–775
Piedade MTF, Long SP, Junk WJ (1994) Leaf and canopy photosynthetic CO2 uptake of a stand
of Echinochloa polystachia on the Central Amazon floodplain. Are the high potential rates
associated with the C4 syndrome realized under the near-optimal conditions provided by this
exceptional natural habitat? Oecologia 97:193–201
Pimenta JA, Bianchini E, Medri ME (1998) Adaptations to flooding by tropical trees: morpholog-
ical and anatomical modifications. In: Scarano FR, Franco AC (eds) Ecophysiological strate-
gies of xerophytic and amphibious plants in the neotropics. Oecologia Brasiliensis, vol IV,
Universidade Federal do Rio de Janeiro, Rio de Janeiro, pp 157–176
Pons TL, Perreijn K, Kessel C van, Werger MJA (2007) Symbiotic nitrogen fixation in a tropical
rainforest: 15 N natural abundance measurements supported by experimental isotopic enrich-
ment. New Phytol 173:154–167
Reich PB, Uhl C, Walters MB, Ellsworth DS (1991) Leaf life span as a determinant of leaf structure
and function among 23 amazonian tree species. Oecologia 86:16–24
Reichholf JH (1994) Biodiversity. Why are there so many different species? Universitas 1994/1:42–
51
Remmert H (1985) Was geschieht im Klimax-Stadium? Ökologisches Gleichgewicht durch Mo-
saik aus desynchronen Zyklen. Naturwissenschaften 72:505–512
Remmert H (1991) The mosaic cycle of ecosystems. Ecological Studies, vol 85. Springer, Berlin
Heidelberg New York
Richards PW (1996) The tropical rain forest. An ecological study, 2nd edn. Cambridge Univ Press,
Cambridge
Santiago LS, Mulkey SS (2005) Leaf productivity along a precipitation gradient in lowland
Panama: patterns from leaf to ecosystem Trees 19:349–356
Sarmiento G (1984) The ecology of neotropical savannas. Harvard University Press, Cambridge
Schlindwein CCD, Fett-Neto AG, Dillenburg LR (2006) Chemical and mechanical changes during
leaf expansion of four woody species of a dry restinga woodland. Plant Biol 8:430–438
Schuster P (1998) Evolution in molekularer Auflösung. Ber u Abh Berlin-Brandenb Akad Wiss
6:187–215
Selye H (1973) The evolution of the stress concept. Am Sci 61:693–699
Silver WL (1994) Is nutrient availability related to plant nutrient use in humid tropical forests?
Simone O de, Haase K, Müller E, Junk WJ, Gonsior G, Schmidt W (2002) Funct Plant Biol
29:1025–1035
Simone O de, Haase K, Müller E, Junk WJ, Hartmann K, Schreiber L, Schmidt W (2003a)
Apoplasmic barriers and oxygen transport properties by hypodermal cell walls in roots from
four Amazonian tree species. Plant Physiol 132:206–217
Simone O de, Müller E, Junk WJ, Richau K, Schmidt W (2003b) Iron distribution in three central
Amazon tree species from white water-inundation areas (várza) subjected to different iron
regimes. Trees 17:535–541
Sobrado MA (1991) Cost-benefit relationships in deciduous and evergreen leaves of tropical dry
forest species. Func Ecol 5:608–616
Sobrado MA (1995) Seasonal differences in nitrogen storage in deciduous and evergreen species
of a tropical dry forest. Biol Plant 37:291–295
Solbrig OT (1994) Plant traits and adaptive strategies: their roles in ecosystem function. In: Schulze
E-D, Mooney HA (eds) Biodiversity and ecosystem function. Ecological Studies 99. Springer,
References 101
Solé RV, Manrubia SC (1995a) Self-similarity in rain forests: evidence for a critical state. Phys
Rev E 51:6250–6253
Solé RV, Manrubia SC (1995b) Are rainforests self-organized in a critical state? J Theor Biol
173:31–40
Solé RV, Manrubia SC, Luque B (1994) Multifractals in rainforest ecosystems: modelling and sim-
ulations. In: Novak NM (ed) Fractals in the natural and applied sciences. Elsevier, Amsterdam,
pp 397–407
Souza Moreira FM de, Silva MF da, Faria SM de (1992) Occurrence of nodulation in legume
species in the Amazon region of Brazil. New Phytol 121:563–570
Stewart GR, Hegarty EE, Specht RL (1988) Inorganic nitrogen assimilation in plants of Australian
rainforest communities. Physiol Plant 74:26–33
Stewart GR, Gracia CA, Hegarty EE, Specht RL (1990) Nitrate reductase activity and chlorophyll
content in sun leaves of subtropical Australian cloud-forest (rainforest) and open-forest com-
munities. Oecologia 82:544–551
Stewart GR, Joly CA, Smirnoff N (1992) Partitioning of inorganic nitrogen assimilation between
the roots and shoots of cerrado and forest trees of contrasting plant communities of South East
Brasil. Oecologia 91:511–517
Sultan SE, Bazzaz FA (1993) Phenotypic plasticity in Polygonum persicaria. I. Diversity and uni-
formity in genotypic norms of reaction to light. Evolution 47:1009–1031
Talbott LD, Rahveh E, Zeiger E (2003) Relative humidity is a key factor in the acclimation of the
stomatal response to CO2 . J Exp Bot 54:2141–2147
Tilman D (1982) Resource competition and community structure. Princeton Univ. Press, Princeton
Torquebiau EF (1988) Phtosynthetically active radiation environment, patch dynamics and archi-
tecture in a tropical rainforest in Sumatra. Aust J Plant Physiol 15:327–342
Valencia R, Foster RB, Villa G, Condit R, Svenning JC, Hernández C, Romoleroux K, Losos
E, Magård E, Balsev H (2004) Tree species distributions and local habitat variation in the
Amazon: large forest plot in eastern Ecuador. J Ecol 92:214–229
Walter H (1973) Vegetationszonen und Klima. Ulmer, Stuttgart
Walter H, Breckle S-W (1984) Ökologie der Erde, vol 2. Spezielle Ökologie der tropischen und
Watt AS (1947) Pattern and process in the plant community. J Ecol 35:1–22
Webber BL, Abaloz BA, Woodrow IE (2006) Myrmecophilic food body production in the under-
storey tree, Ryparosa kurrangii (Achariaceae), a rare Australian rain forest taxon. New Phytol
173:250–263
West-Eberhard MJ (1986) Alternative adaptations, speciation, and phylogeny (a review). Proc Natl
Acad Sci USA 83:1388–1392
West-Eberhard MJ (1989) Phenotypic plasticity and the origins of diversity. Annu Rev Ecol Syst
20:249–278
West-Eberhard MJ (2003) Developmental plasticity and evolution. Oxford University Press, Ox-
ford
Whitmore TC (1990) An introduction to tropical rain forests. Oxford University Press, Oxford
Whittaker RH (1975) Communities and ecosystems, 2nd edn. Macmillan, New York
Wirth R, Herz H, Ryel RJ, Beyschlag W, Hölldobler B (2003) Herbivory of leaf-cutting ants.
A case study on Atta colombica in the tropical rainforest of Panamá. Ecological Studies, vol.
164. Springer, Berlin Heidelberg New York
Chapter 4
Tropical Forests.
II. Ecophysiological Responses to Light
4.1 Light Responses of Photosynthesis
Although gradients of several different environmental factors are noticeable in trop-

ical forests (Sect. 3.4), intensity of irradiance is most highly variable and appears to
play the most prominent role in determining the ecophysiological comportment of
forest plants. At the top of the canopy and in larger clearings in full sun-light inten-
sity of photosynthetically active radiation (PAR) at 400 – 700 nm wavelength may
range from well above 1,000 up to 2,500 µmol m−2 s−1 photons. On the forest floor
there may be less 5 µmol m−2 s−1 photons (see Figs. 3.26 and 3.28). Thus, light
can become a stress factor from both too much (when it causes overenergization
of the photosynthetic apparatus and hence photoinhibition or even photodestruc-
tion), or too little (when it becomes limiting as an energy source of photosynthesis).
Focussing of light by leaf epidermal cells may increase irradiance intensity in the
mesophyll of under story plants of tropical forests (Vogelmann et al. 1996). How-
ever, the major responses are those of the photosynthetic machinery.
4.1.1 Light-Response Characteristics of Sun and Shade Plants
The photosynthetic utilization of light by plants is described quantitatively by light-

response curves (Fig. 4.1), which are distinguished by several cardinal points as
follows (see also Box 4.1). In darkness (zero PAR), there is net-CO2 release due to
respiration. As light intensity increases, net-CO2 release is gradually reduced until
the light-compensation point is attained, where net-CO2 exchange is zero because
photosynthetic CO2 -uptake just balances respiratory CO2 -release. Above this point,
net-CO2 uptake increases until light saturation is reached. The light-saturation
point often is hard to determine precisely, because light saturation is approached
gradually. Hence, half saturation of photosynthesis is often quoted alternatively or
additionally. The slope of the nearly linear part of the curve below saturation gives
the apparent quantum yield (mol CO2 per mol photons) of photosynthesis.
104 4 Tropical Forests. II. Ecophysiological Responses to Light
Fig. 4.1 A Light-response

curves of a sun plant Ploiar-
ium alternifolium and a shade
plant Lycopodium cernum
measured in the field in
a tropical secondary forest
in Singapore. (After Lüttge et
al. 2005). B Light-response
curves of four species in the
gradient of understory, cli-
max and pioneer systems in
a secondary lower montane
rainforest in Costa Rica (al-
though only old units for
light intensity are available
this figure is quite illustra-
tive; Stephens and Waggoner
1970)
Plants may be genetically determined for growth at low or high light intensity.
In this case we can distinguish genuine shade and sun species. However, there are
also ontogenetic and developmental modifications, where light exerts a signalling
function rather than being only the energy source of photosynthesis, and plants may
acclimate or adapt ecophysiologically to low and high irradiance, respectively (see
Sect. 4.3.2). The potential for light acclimation is species specific and may involve
major structural and functional changes in the photosynthetic apparatus (Bailey et al.
2001). Thus, there may be shade and sun forms of given species. Plants surviving in
4.1 Light Responses of Photosynthesis 105
Box 4.1 Light-response characteristics
• Light-response curves and their cardinal points
• Distinctive characters of sun and shade plants

Shade plants Sun plants
Respiration Low High
Light compensation point Low High
Half saturation point Low High
Light saturation point Low High
Maximum rate of photosynthesis Low High
Quantum yield High Low
the understory or below canopy of tropical rainforests may experience opening and
closure of the canopy several times during their lifetime. Understory species of
Miconia (Melastomataceae) responded to canopy openings by production of new
sun-leaves rather than acclimation of old shade-leaves, which could significantly in-
crease maximum assimilation rates (Newell et al. 1993). Individual plants may also
have both shade- and sun-leaves when part of the foliage is shaded and exposed re-
spectively. Then we may define such plants as shade or light tolerant, but not shade
or light demanding.
All of these various aspects are very important for plant life in tropical forests
with their highly variable light climates. Comparisons of light response curves and
their cardinal points provide distinctive characters for shade and sun species or phe-
notypes (Box 4.1). Shade plants usually have lower rates of respiration but the activ-
ity of the alternative cyanide resistant pathway of mitochondrial electron transport
not producing ATP is low and therefore the ATP/O2 efficiency of respiration is
higher than in sun plants as required due the lower overall energy input at low irra-
diance in the shade (Noguchi et al. 2001b). Shade plants have lower rates of pho-
tosynthesis at light saturation and together with the lower rates of respiration this
leads to lower light compensation and light saturation points, but higher quantum
yields than in high light or sun plants. In Fig. 4.1 this is illustrated by the compar-
ison of sun and shade plants (Fig. 4.1A) and pioneer, climax and understory trees
and shrubs (Fig. 4.1B).
The different light-use characteristics of sun and shade plants are very important
for understanding the distinct stages in the dynamics of tropical forests (Sects. 3.3.3
and 4.3.2). They distinguish pioneer species from climax species, and from plants of
the under-growth. During growth (Sect. 4.3.2), pioneer species show the character-
istics of sun plants, understory species that of shade plants (Eschenbach et al. 1998),
and the dominant trees of later successions (climax species) show an intermediate
behaviour as they may be extremely sun exposed in the upper canopy and shaded
in the lower canopy layer (Fig. 4.1B, Tables 4.1 and 4.2). Thus, the pioneer species
Cecropia peltata needs much higher light intensities for light saturation of photosyn-
thesis and has much higher rates of maximum photosynthesis than the shade plant
Croton glabellus, and a comparison of the light-response curves of dominant cli-
max trees in the upper canopy, i.e. Cordia alliodora and Goethalsia meiantha, with
Table 4.1 Values for cardinal-points of light-response curves of sun and shade plants in general
and of plants in tropical forests. (After Lüttge 1985)
Plant type Light-compensation Light saturation Rate of CO2 -uptake
point of CO2 -uptake at light saturation
(µmol photons m−2 s−1 ) (µmol photons m−2 s−1 ) (µmol m−2 s−1 )
Sun plants 20 – 30 400 – 600 10 – 20
Shade plants 0.5 – 10 60 – 200 1–3
Tropical rainforest
Upper canopy
Sun types 12 250 – 370 13 – 19
Shade types 6 – 12 125 – 185 6 – 10
Lower canopy
Shade types 6 – 12 125 4–5
Herbs 2.6 – 6 25 – 37 1.3 – 1.9
of the undergrowth
Table 4.2 Maximum rates of photosynthesis of plants in various stages of succession dynamics,
and daily rates of photosynthesis in various strata, of an Australian rainforest (Doley et al. 1988)
Maximum rates (µmol CO2 m−2 s−1 )
Species of early stages of forest successions 15

Tree species of later stages of forest successions 4 – 15
Undergrowth species 1–3
Daily rates (mmol CO2 m−2 day−1 )
Leaves of upper canopy level 250
Leaves at the fringe of a large clearing 97
Leaves of the undergrowth 24
Fig. 4.2 Leaf-internal CO2 -partial pressures ( pCOi ) and CO2 -assimilation (A) to leaf conductance
2
to water vapour (gH2 O ) ratios of five tropical rainforest tree species in artificial stands under com-
mon conditions in French Guiana. Closed circles: Pioneer species (Jacaranda copaia, Goupia
glabra, Carapa guianensis); open circles: late successional species (Dicorynia guianensis, Eperna
falcata). (After data of Huc et al. 1994)
that of the understory shrub C. glabellus in Fig. 4.1B shows the distinct differences
between sun and shade plants with respect to all the features listed in Box 4.1.
Pioneer and late successional rainforest species also regulate their leaf gas ex-
change in different ways. Shade plants have lower leaf conductance to water vapour,
gH2 O , than sun plants which leads to lower gas exchange and growth (Bonal et al.
2000; Sack et al. 2005). Thus, pioneer species operate at lower A/gH2 O ratios (CO2 -
assimilation A), i.e. with a greater stomatal aperture and higher internal CO2 -partial
i
pressures ( pCO 2
) as compared to trees found in the later stages of succession (Huc
et al. 1994; Fig. 4.2).
4.1.2 The Photosynthetic Apparatus:

Pigments, Enzymes and Nitrogen
It is very often observed in the tropics, that individual plants of a given species
growing in deep shade inside a forest and exposed to full sun-light in an open
habitat respectively, form morphologically very different phenotypes, which are
also strongly distinguished by pigmentation. For example, this is frequently found
among rosettes of bromeliads, e.g. in the genera Bromelia and Ananas which belong
to genetically identical clones propagating vegetatively by formation of ramets. In
Bromelia humilis shade plants are much larger than sun plants; they have long and
slender leaves, whereas sun plants overall have a more stunted appearance (Fig. 4.3).
The most conspicuous difference is leaf colour, which is dark green in the shade
plants, brightly yellow in the sun plants and light green in intermediate forms.
What is behind this pronounced difference in pigmentation? Addressing this
question requires a reminder of the basic structure of the photosynthetic appa-
ratus situated in the thylakoid membranes of the chloroplasts (Box 4.2). The major
Fig. 4.3 Phenotypes of Bromelia humilis, from left to right yellow/exposed form, intermediate
light green/exposed form and dark green/shaded form
Box 4.2 Structure and function of the photosynthetic apparatus
A. Scheme of a chloroplast with outer membrane (envelope), inner membrane,

the stroma (= plastoplasm or “cytoplasm” of the chloroplast), stromal and
granal thylakoids and thylakoid interior.
Box 4.2 (Continued)

B. Scheme of the thylakoid membrane with the elements of photosynthetic elec-
tron transport.
There are four important integral proteins or protein complexes in the thy-
lakoid membrane:
• Photosystem II (PS II) with antenna and reaction-centre pigments; it oc-
curs only in the appressed regions of granal thylakoids.
• The cytochrome b6,f-complex.
• Photosystem I (PS I) with antenna and reaction-centre pigments; it oc-
curs in the stroma thylakoids.
• The NADP-reductase (NADP = nicotinamide-adenine-dinucleotide-
phosphate).
• The F0 F1 -ATPase or coupling factor.
When PS II and PS I are excited by the absorption of photons (hν), H2 O is
split into oxygen, protons and electrons (e− ), electrons flow from PS II via
membrane-bound plastoquinone (Q) and mobile plastoquinone (PQ) to the
cytochrome-b6,f-complex and plastocyanine (PCy) at the lumen side of the
membrane, and from PCy further to PS I, ferredoxin at the stroma side of
the membrane and to the NADP-reductase, which finally generates the re-
ducing equivalents needed for CO2 reduction. If they are not utilized in CO2
reduction, the systems of this electron-transport chain may become overre-
duced, and damage may result if the excitation energy cannot be dissipated
in other ways.
Electron transport is associated with charge separation and simultaneously
leads to establishment of a proton-electrochemical gradient across the thy-
lakoid membranes between the lumen (∼pH 5) and the stroma (∼pH 8). This
is the driving force for the movement of protons through the ATPase which is
coupled to ATP synthesis providing the energy needed for CO2 assimilation.
Box 4.2 (Continued)
C. Model of a light trap or light-harvesting complex (PS I or PS II) with antenna

pigments and reaction-centre pigment.
It is sufficient that one of the antenna pigments absorbs a photon (hν) and is
excited. Transfer of the excitation energy between antenna pigments always
eventually leads to excitation of the reaction center. The larger the light trap
the more probable is excitation of the reaction centre.
components are the pigments of the two photosystems (photosystem I and II), the
thylakoid proteins embedded in the membrane lipid-bilayer and the cofactors of the
photosynthetic electron-transport chain. Pigments are the chlorophyll a of the light
harvesting complex (light trap) and accessory antenna pigments such as chloro-
phyll b, carotenoids and xanthophylls. Thylakoid-proteins are the various elements
of the electron-transport chain (redox-chain), the ATP-generating coupling factor
(F0 F1 -ATPase) and chlorophyll-protein light harvesting complexes.
Since proteins as well as chlorophyll and cytochrome molecules contain much
nitrogen, this element plays a prominent role in constructing the photosynthetic
apparatus in the thylakoids. Given that the components of the photosynthetic ap-
paratus acclimate to changing light climate then nitrogen supply is important for
these processes. Thus, it is appropriate to determine the N-costs of thylakoid mem-
branes, which can be expressed in units of mol N:mol chlorophyll. For exam-
ple, in Alocasia macrorrhiza, a shade tolerant species native to tropical rainfor-
est understories in Australia, this was 45 mol N/mol chl under natural shade, and
56 mol N/mol chl in Pisum sativum grown at high irradiance (Evans 1988). A gen-
eral comparison of shade plants and sun plants reveals a number of characteristics,
as follows:
Box 4.2 (Continued)
D. Photosynthetic Pigments
Box 4.2 (Continued)
The reaction-centre pigment chlorophyll a and antenna pigments chloro-

phyll b and carotenoids. The essential feature of the light absorbing pigments
are systems of conjugated double bonds.
(Schemes after Lüttge et al. 2005)
• Shade plants contain more chlorophyll b or have smaller chlorophyll a:b

ratios.
The larger relative amount of antenna pigments assures that low photosyntheti-
cally active photon flux densities (PPFDs) or light intensities are used efficiently, i.e.
at low flux densities photons are absorbed effectively and the excitation energy can
be transferred to the light trap reaction center chlorophyll (Box 4.2C). This explains
the higher quantum yield of shade plants (Sect. 4.1.1, Fig. 4.1, Box 4.1). At a given
nitrogen availability chlorophyll a : b ratios increase with increasing irradiance (Ki-
tajima and Hogan 2003).
• Shade plants have lower rates of electron flow along the redox-chain in the
thylakoids related to chlorophyll.
This is a consequence of the larger chlorophyll content of the photosystems.
• Shade plants have less soluble protein in relation to chlorophyll but shade
plants have larger total N-contents in their biomass.
The soluble proteins of leaves include ribulose-bis-phosphate carboxylase/oxy-
genase (RuBISCO, see also below or RuBPC in Sect. 2.5). This enzyme-protein
is responsible for photosynthetic CO2 -fixation. It is the single major protein and
hence N-containing compound in plant leaves. The lower protein/chlorophyll ra-
tio in shade plants is due to the higher chlorophyll content and lower content of
RuBISCO. However, generally shade plants have larger total N-contents in their
biomass, and chlorophyll a : b ratios increase with decreasing N-availability espe-
cially under high irradiance conditions (Kitajima and Hogan 2003). An over produc-
tion of RuBISCO may occur for an N-reserve and then there is a weak correlation of
N-contents and RuBISCO and maximal rate of photosynthesis (Warren et al. 2000).
• Shade plants have larger photosystem II/photosystem I ratios.
This is related to the change of the spectral composition of light passing through the
canopy, where the shorter-wave length red-light is filtered out to a larger extent than
the longer-wave length red-light (see Sect. 4.2). The chlorophyll of photosystem II
(PS II) is excited by somewhat shorter wavelengths (P-680 for absorption at λ =
680 nm) than that of photosystem I (PS I; P-700). Since both photosystems must
co-operate in photosynthesis, shade plants need more PS II in relation to PS I.
• Shade plants have larger chloroplasts and more grana formation.
Experiments to elucidate these relationships have often been made with tropical
plants, since the contrast between deep shade in the dark rainforests and full sun
exposure in clearings and open habitats with small solar inclination throughout the
year is more pronounced in the tropical environment. Figure 4.4 shows the results of
a study, where Alocasia macrorrhiza, which has a large capacity for photosynthetic
acclimation to different light environments, was adapted to various light intensities
during growth, i.e. from very low PPFDs up to 800 µmol m−2 s−1 . The photosyn-
thetic capacity, the activity of RuBISCO, the content of cytochrome f – an important
element of the photosynthetic electron-transport chain (Box 4.2) – and the chloro-
phyll a:b ratio increased considerably with light acclimation. The amount of trap
chlorophyll of PS I showed no change, but PS II increased slightly at high growth
irradiance up to 800 µmol m−2 s−1 . Thus, in this particular experiment a consider-
ably larger PS II/PS I ratio was not seen in the shade grown plants. Total chlorophyll
content decreased. There was a small decrease of quantum yield. It may be noted ad-
ditionally, however, that a cost-benefit study which includes modelling and simula-
tion suggests that shade leaves not necessarily have a lower photosynthetic capacity
than sun leaves when leaf mass rather than area (as in Fig. 4.4) is used as a basis, al-
though the higher investment in CO2 -fixation-cycle enzymes and electron-transport
carriers per unit of leaf surface in shade plants remains evident (Sims and Pearcy
1994; Sims et al. 1994). Thus, by and large the differences between low- and high-
light grown Alocasia plants are in conformity with the general distinctions between
shade and sun plants made above, despite the general observation that A. macror-
rhiza is a typical understory rainforest plant in Australia, and still more pronounced
effects may be observed with plants showing a greater phenotypic plasticity at still
higher irradiance.
A study with the low and high-light grown phenotypes of Bromelia humilis (see
Fig. 4.3) also corroborates these basic relationships, and emphasizes the modula-
tion by nitrogen nutrition (Fetene et al. 1990; Table 4.3). In the low-light plants,
Fig. 4.4 Components and photosynthetic functions of leaves of the tropical understory plant
Alocasia macrorrhiza grown under varying light intensities up to 800 µmol m−2 s−1 photons
(λ = 400 − 700 nm). (After Chow et al. 1988; see Anderson and Thomson 1989)
Table 4.3 Effects of light intensity and nitrogen on photosynthesis and leaf parameters of Bromelia
humilis (Fetene et al. 1990)
Irradiance during growth [µ mol photons m−2 s−1 ]

High (700 – 800) Low (20 – 30)
+N −N +N −N
Light-saturated rate of photo- 14.0 ± 3.5 8.3 ± 2.0 16.7 ± 2.6 6.9 ± 2.5
synthesis (µ mol O2 m−2 s−1 )
Apparent quantum yield 0.077 ± 0.010 0.042 ± 0.002 0.085 ± 0.015 0.041 ± 0.003
(mol O2 mol−1 photons)
Light compensation point 25 ± 4 40 ± 5 10 ± 2 10 ± 5
(µ mol photons m−2 s−1 )
Total chlorophyll (µ g g−1 FW) 102 ± 21 99 ± 10 725 ± 19 364 ± 18
Chlorophyll a/b ratio 2.61 2.83 2.35 2.32
Nitrogen-use efficiency at light 1.2 0.5 1.0 0.4
saturation of photosynthesis
(mol CO2 mol−1 leaf N)
chlorophyll levels were high as compared to the high-light plants and increased
by additional N-nutrition. The chlorophyll a : b ratio was higher in the high-light
plants. The different construction of the photosynthetic apparatus was reflected in
a development of the typical cytological structure of sun- and shade-plant chloro-
plasts, respectively (Fig. 4.5). Low-light grown plants developed characteristically
large, globular, shade-acclimated chloroplasts, with extensive grana-formation and
hence appressed thylakoid membranes (Box 4.2). The number of thylakoid mem-
branes per granum was threefold larger in low-light than in high-light grown plants,
and the ratio of appressed to non-appressed thylakoid membranes was 3.5 – 5.0 and
1.0 – 1.5 respectively. Chloroplasts of high-light plants grown without N had only
poorly developed thylakoids. Light compensation points were higher in the high-
light plants of B. humilis (Table 4.3). It was most noteworthy, however, that with
extra N-nutrition the low-light plants could attain similar light saturated rates of
photosynthesis as the high-light plants and that the high-light plants almost reached
the apparent quantum yield of the low light plants. Expressed on a leaf-nitrogen ba-
sis net photosynthetic CO2 -fixation was similar in low and high-light grown plants
independent of whether additional N as supplied. This demonstrates the optimisa-
tion of nitrogen use, but it also suggests an interaction with other factors, since the
ratio of net-CO2 -fixation to nitrogen levels was not constant but rather low at low
N-levels and higher at high N-levels.
Maximum assimilation rates related to nitrogen levels of leaves give the nitrogen-
use-efficiency (NUE), which is an important parameter relating the functioning of
the photosynthetic apparatus to mineral nutrition. In general, assimilation vs leaf-
nitrogen curves are linear over certain ranges of N-levels (Fig. 4.6; Nielsen et al.
1997). Often they do not appear to extrapolate to the origin, and thus, show reduced
NUE at low N-levels. At higher N-levels there may be N-saturation of assimila-
tion (see Diplacus in Fig. 4.6B). Therefore, Evans (1988) has evaluated the general
N-relationships of sun and shade plants somewhat critically. While a correlation be-
tween assimilation and N-levels is clearly given, other factors like characteristics
of individual species (Fig. 4.6) and growth conditions must also be involved. This
may include irradiance during growth since with spinach and peas (unlike in the
experiment of Table 4.3), there was an effect of light intensity on NUE.
Fig. 4.5A–D Chloroplast ultrastructure of Bromelia humilis grown in low light with N (A) or
without N (B) and in high light with N (C) or without N (D). (Fetene et al. 1990)
Fig. 4.6A, B Correlation between photosynthetic capacity and nitrogen levels in leaves of: A Pisum
hispidum (closed symbols) and Pisum auritum (open symbols); B Diplacus aurantiacus (closed
symbols) and Solidago altissima (open symbols). r correlation coefficients. (After: A Field 1988;
B Evans 1988)
Reich et al. (1994) have presented a detailed investigation of these relations per-
formed in two open and disturbed sites and three late successional forest types in
the Amazon basin, which differed in light climate and nutritional status (Fig. 4.7).
Photosynthesis-N relations are steepest and intercepts on the x-axis (N-levels) are
highest in the disturbed open habitats with high resource acquisition and rapid plant
growth (lines a and b in Fig. 4.7A,B), which also have the highest rates of photosyn-
thesis (sun plants). Among the other sites, under N-limitation curves were steeper
(lines d and e in Fig. 4.7A) than when P plus Ca were more deficient (line c in
Fig. 4.7A). This was somewhat less clearly seen, when leaf N was expressed on
a leaf area basis (Fig. 4.7B) as compared to the leaf mass basis (Fig. 4.7A), so that
besides species and site characteristics a certain effect of the basis of data expres-
sion, i.e. mass or leaf area is also noted.
Fig. 4.7A, B Photosynthesis/leaf-nitrogen relationships in five different forest types near San Car-
los de Rio Negro, Venezuela (1◦ 56 N, 67◦ 03 W) at ∼ 100 m a.s.l. in the north central Amazon
basin. Disturbed sites, high resource regeneration sites: a cultivated, and b early secondary suc-
cessional Tierra Firme plots. Late successional forest types: c Tierra Firme, P- and Ca-limited; d
Caatinga, N-limited; e Bana, P- and N-limited. Curves were drawn from the y-intercepts and slopes
of linear regressions given by Reich et al. (1994) for the various samples of a total of 23 species,
which they studied at the five sites. The lengths of the lines in each case indicate the range in which
points were obtained. Results are expressed on a mass basis (A) as well as on an area basis (B)
4.1.3 The Origin of High-Irradiance Stress

and General Plant Responses
The biological stress concept has shown us that stress can result from low or high
dosage of any particular environmental factor (Box 3.1). In sun plants increased
chlorophyll a/b ratios and a comparatively small size of chlorophyll a and b bind-
ing antennae (Sect. 4.1.2) contribute to protection from too high irradiance (Krause
et al. 2001). Conversely shade plants, or phenotypes considered in Sects. 4.1.1 and
4.1.2, are adapted to low irradiance stress typical of the interior of dense forests.
Stress by high irradiance might generally appear to be more characteristic of open
habitats like savannas. However, it is also found in deciduous and semi-deciduous
dry forests and in the upper canopy of wet forests. Moreover, in view of the very high
light intensities of some light flecks (Sect. 4.2.1), it may even be a particular prob-
lem for the shade-adapted plants in the understory of moist forests. Finally, shade
plants of the forest floor may be suddenly exposed to high irradiance when gaps
are created by falling trees. Hence, it is necessary to address of how plants avoid
damage from excess irradiation. Strategies of avoidance include vertical leaf angles
and small lamina areas (He et al. 1996) and chloroplast movements (Augustynowicz
and Gabrỳs 1999; Gorton et al. 1999).
With respect to excess absorbed irradiance first it is necessary to recall the various
states of excitation and relaxation of chlorophyll in the photosystems (Box 4.3).
Excitation to the 2nd singlet level follows absorption of photons of blue light, while
red light causes excitation to the 1st singlet level. Relaxation from the 2nd singlet
state occurs by emission of heat, while there are several ways of relaxation from
the 1st singlet state to the ground level. The normal means of energy dissipation
in photosynthesis is photochemical work (Box 4.3), i.e. the eventual reduction of
CO2 fixed via RuBISCO. This, however, under certain circumstances may become
a limiting process, e.g. due to:
• low intercellular CO2 -concentrations in leaves due to high CO2 -fixation rates
following over saturation of the photosynthetic apparatus by PPFD,
• closure of stomata to reduce transpirational loss of water in response to high
irradiance and heat with the concomitant consequence of low intercellular CO2 -
concentrations,
• general over excitation of the photosynthetic apparatus and over reduction of the
redox-elements of the photosynthetic electron-transport chain (see Box 4.2).
Box 4.3 Light absorption, chlorophyll excitation and relaxation
A. Absorption spectra of
the reaction-centre pig-
ment chlorophyll a and
the antenna pigments
chlorophyll b and caro-
tenoids (see Box 4.2).
Box 4.3 (Continued)

B. Ground state and excitation states with substates (horizontal lines) and re-
laxation of chlorophyll:
Absorption of the more energy-rich blue-light quanta (shorter wave-lengths)

leads to the second and third excited singlet state (half-life 10−14 – 10−15 s).
Absorption of the less energy-rich red-light quanta (longer wavelengths)
leads to the first excited singlet state (half-life 10−9 – 10−11 s). Relaxation
can occur by transition between systems, and energy is dissipated as heat.
Transition from the first singlet state to the triplet state (half-life 10−4 –
10−2 s) is only probable when the whole system is overexcited. Relaxation
from the first singlet state to the ground state in addition to energy dissipa-
tion as heat can occur by emission of light as fluorescence. Relaxation from
the triplet state to the ground state is possible by the emission of light as
phosphorescence. Energy transfer from the first singlet state and the triplet
state can lead to photochemical work, i.e. CO2 -assimilation or photorespi-
ration in the case of the first singlet state and formation of oxygen radicals
and photodamage in the case of the triplet state.
Overall there is a set of several means of dissipating excitation of chlorophyll a

in photosystem II (see Boxes 4.3 and 4.4):
• PS II photochemistry, due to CO2 or O2 binding,
• relaxation by emission as heat (Sect. 4.1.4),
• relaxation by emission as fluorescence (Sect. 4.1.7),
• relaxation by energy transfer to photosystem I,
• transfer of excitation from the 1st singlet state to the triplet state, which is cor-
related with a change of electron spin from antiparallel to parallel.
One valve for dissipation of surplus energy is photochemical work. Increased con-
tents of electron-transfer-chain components and RuBISCO are, of course, protec-
tive (Ramalho et al. 1999). A photochemical work different from CO2 assimilation
is photorespiration. This is possible, since RuBISCO (ribulosebisphosphate car-
boxylase/oxygenase) not only reacts with CO2 but also with O2 and can oxygenate
ribulosebisphosphate to form phosphoglycolate, which is metabolized in the pho-
torespiratory reaction cycle. However, this still may be of limited capacity to avoid
adverse effects of overenergization, which leads to photoinhibition. The primary
site of photoinhibition is photosystem II. It can be protected by energy dissipation
as heat which is correlated with chlorophyll fluorescence. These are important func-
tions which need to be treated in separate sections (Sects. 4.1.4 and 4.1.7, respec-
tively).
Energy transfer to photosystem I is also called spill over. Intact photosystem II
in higher plants is located in the appressed thylakoid regions of chloroplast grana
(Box 4.2) but for spill over the peripheral light harvesting complexes of photosystem
II get phosphorylated by a specific kinase and are transferred to the stroma region
of the thylakoids. This is a reversible process highly regulated by irradiance and the
redox state of plastoquinone, PQH2 (see Box 4.2).
The triplet state is much more stable, i.e. it has a much longer half-life, than
the two singlet states. Therefore, it may lead to the formation of reactive oxygen
species and oxygen radicals with the subsequent destruction of pigments, lipids
and membranes. Radical scavengers or antioxidants are a number of redox sub-
stances, such as dihydroascorbate, reduced glutathione and for the chloroplast espe-
cially tocopherol (Krieger-Liszkay and Trebst 2006) and the enzymatic reactions of
superoxide dismutases. The xanthophyll cycles of chloroplasts also are mechanisms
for the scavenging of highly reactive singlet oxygen (Sect. 4.1.4).
4.1.4 Dissipation of Excitation Energy in the Form of Heat:

The Role of Xanthophylls
Dissipation of surplus excitation energy in the form of heat is mediated by cycles

of xanthophylls. The zeaxanthin cycle is one of the most important photoprotec-
tional mechanisms in photosynthesis (Demmig-Adams et al. 1996; Gilmore 1997;
Gilmore and Govindjee 1999). It binds harmful singlet activated oxygen (1 O∗2 ) and is
a futile cycle in terms of energy turnover. The basic metabolic reactions are shown in
Box 4.4. The xanthophylls of the zeaxanthin cycle are peripherally associated with
the light harvesting complex of photosystem II, LHCII (Horton and Ruban 2005).
Singlet activated oxygen is bound in the formation of the epoxides antheraxanthin
and violaxanthin from zeaxanthin, which also requires redox energy in the form of
NADPH + H+ . The zeaxanthin cycle only functions when a trans-thylakoid proton
gradient, pH, is built up by photosynthetic electron transport. The pH otherwise
used to power photophosphorylation is the primary controlling factor of the zeax-
anthin cycle. pH and the electrons are used for deepoxidation and the return of
the cycle to zeaxanthin when the originally harmful oxygen is eventually reduced to
water and the energy of redox equivalents, electrons and pH across the thylakoid
Box 4.4 Xanthophyll-cycles
The zeaxanthin-cycle scheme was originally proposed by Hager (1980) and

later forcefully propagated by Demmig-Adams (1990), Demmig-Adams and
Adams (1992) and Pfündel and Bilger (1994). The turnover in the cycle in-
volves detoxification of singlet activated oxygen (1 O∗2 ) and energy dissipation,
namely binding of activated oxygen by oxidation of zeaxanthin to violaxanthin
(epoxidation) and dissipation of the energy of photosynthetic electron transport
by rereduction (deepoxidation) to zeaxanthin. The epoxidase is located on the
stroma side and the de-epoxidase on the thylakoid-lumen side of the thylakoid
membrane, and the pH optima of the epoxidase (pH 7.5) and the deepoxidase
(pH 5.2) correspond to the prevailing conditions in the chloroplast stroma and
the thylakoid interior, respectively. The deepoxidase is mobile within the thy-
lakoid lumen at neutral pH but becomes membrane-bound when the pH drops
and a pH gradient is established across the thylakoid membrane by photosyn-
thetic electron flow (Hager and Holocher 1994); it has a narrow pH optimum at
pH 5.2 and is presumed to be activated by the photosynthetic electron transport
via acidification of the lumenal pH (Büch et al. 1994). Ascorbate is a cofactor
in the deepoxidation reaction (not drawn in the cycle itself).
An additional xanthophyll-cycle is the lutëin/lutëin-epoxide cycle. While
zeaxanthin has two β-ionon rings and can form two epoxides, lutëin has one α-
ionon ring and only one β-ionon ring, and therefore, can form only one epoxide.
The deepoxidase is the same as in the zeaxanthin-cycle (violaxanthin deepox-
idase), while the epoxidase either is the same as the zeaxanthin epoxidase or
a homologous enzyme (Matsubara et al. 2001, 2003).
membranes is effectively dissipated as heat. The zeaxanthin cycle directly protects

PSII rather than PSI but may indirectly protect also PSI by restricting electron flow
(Barth et al. 2001).
As an alternative, or in addition, to their function in the futile zeaxanthin cy-
cle, xanthophylls may be involved in directly affecting the structure of the light-
harvesting complex itself (Horton et al. 1994) and thus diverting excitation by ab-
sorbed photons from the reaction centre of photosystem II (PSII). Zeaxanthin can
bind to the internal binding site of the xanthophyll lutëin in LHCII (Horton and
Ruban 2005). Zeaxanthin binds in dependence of pH at the thylakoid membrane
(Gilmore 1997; Gilmore and Yamasaki 1998; Gilmore et al. 1998; Gilmore and
Govindjee 1999) causing conformational changes of the system, possibly involving
aggregation of the light-harvesting complex of PSII, which may be facilitated by an
absence of violaxanthin and/or a presence of zeaxanthin and converting the reac-
tion centre of PSII into a centre dissipating heat (Bilger and Björkman 1994; Horton
et al. 1994; Gilmore et al. 1996; Horton and Ruban 2005) through charge separa-
tion of a chlorophyll-zeaxanthin heterodimer (Holt et al. 2005). This mechanism
operates under prolonged and very extreme excessive light, while the zeaxanthin
cycle responds to much lower and variable levels of excess light (Gilmore et al.
1996).
Box 4.4 (Continued)
Independent of the zeaxanthin-cycle, zeaxanthin can also directly function as

an antioxidant and prevent lipid oxidation removing epoxy groups from fatty
acids, where violaxanthin formation provides protection against lipid peroxidation
(Schindler and Lichtenthaler 1996; Baroli et al. 2003).
Thus, three possible xanthophyll mechanisms are listed by Schindler and Licht-
enthaler (1996), namely:
• reaction of zeaxanthin with highly reactive oxygen species and the futile energy
dissipating zeaxanthin cycle,
• aggregation/dissociation of the light-harvesting complex,
• reaction of zeaxanthin with reactive oxygen species.
An additional xanthophyll-cycle is the lutëin/lutëin epoxide cycle (Box 4.4). Lutëin

is an intrinsic component of the light harvesting complex of PSII. It facilitates
enhanced photoprotection through its superior singlet and/or triplet chlorophyll
quenching (Matsubara et al. 2005). The lutëin cycle operates in parallel to the
zeaxanthin-cycle in some plants, e.g. in mistletoes (Sect. 6.5) and the tropical
Fabaceae Inga (Matsubara et al. 2001, 2003, 2005).
Over energization of the photosynthetic light harvesting apparatus is always a po-
tential danger and ecophysiologically the xanthophylls provide an important protec-
tive machinery for many circumstances. Plants permanently exposed to full sunlight
have effective protective mechanisms (Krause et al. 2006) and young leaves develop
them to a higher degree than mature leaves (Krause et al. 1995). Zeaxanthin-cycle
dependent energy dissipation is an extremely flexible process and can kinetically
respond within seconds to minutes. Via adaptive changes of pool sizes of the xan-
thophylls time-scales of days up to seasons can be covered (Demmig-Adams et
al. 1996). Ecophysiologically the zeaxanthin-cycle, for example, is operative in the
light-exposed phenotypes of bromeliads in the tropics, zeaxanthin was only detected
in the yellow high-light plants of Bromelia humilis (see Fig. 4.3) and not in the shade
plants (Fetene et al. 1990) and also plays a significant role in Guzmania monostachia
(Maxwell et al. 1994, 1995). A survey of several other sun and shade plants includ-
ing tropical rainforest species also showed that sun plants possessed larger xantho-
phyll pools and greater maximal zeaxanthin and antheraxanthin contents than shade
plants. Sun plants displayed a greater maximal capacity for photoprotective energy
dissipation via the pigments than plants acclimated to very low irradiance, and in
sun leaves the reduction state of PSII at full sun light remained at a much lower
level than in shade leaves (Demmig-Adams and Adams 1994; Gilmore 1997). In
the vertical light gradients in tropical forests levels of zeaxanthin-cycle pigments
increase from the forest floor to the canopy (Logan et al. 1996). Xanthophyll pool
size and zeaxanthin-cycle activity are also enhanced when photosynthesis is under
nitrogen limitation (see Figs. 4.6 and 4.7) (Tóth et al. 2002; Cheng 2003).
4.1.5 Damage and Repair of Reaction Centres of Photosystem II:

The D1 -Protein
Turnover in the xanthophyll-cycle can be very rapid. Alternatively to energy dissi-

pation as heat and especially in the absence of xanthophyll-cycle associated photo-
protection (Thiele et al. 1997) a slower process develops, which is irreversible when
protein biosynthesis is inhibited. It is due to a functional dissociation between the
reaction centres and antennae of photosystem II (see Box 4.2; Demmig-Adams and
Adams 1993). The reaction centres of photosystem II are built up of one copy each
of a D1 - and a D2 -protein. The destruction of these proteins prevents coupling of
excitation and electron transport via the light trap reaction centres because in the
light harvesting complex of PSII (LHCII) excitation is passed to plastoquinone (Q)
via the D1 -protein. Repair needs protein synthesis (Box 4.5). The D1 -protein is al-
ways destroyed by light even at rather low intensities. Repair is most effective at low
irradiance (30 µ mol m−2 s−1 ). The D1 -protein is always under turnover, but at high
irradiance and without photoprotection repair mechanisms are too slow resulting in
damage of D1 in the LHCII (Tyystjärvi and Aro 1996; He and Chow 2003), where
light stress is amplified by high temperature stress (Königer et al. 1998). However,
this damage which is slowly reversible in the turnover of D1 -protein destruction
and resynthesis (Box 4.5) is also a mechanism of photoprotection as it inhibits over
reduction of the photosynthetic electron transport chain.
Box 4.5 Damage and repair of the reaction centres of photosystem II in

light-stress and recovery cycles
Under high light intensities (HL) the D1 -protein is first modified reversibly
and recovery needs low light intensities (LL); (1). Under further stress the D1 -
protein is damaged irreversibly (2) and repair needs protein synthesis (3). The
turnover of the D2 -protein is slower although it also can be damaged irreversibly
by light stress (Schäfer and Schmid 1993; Critchley and Russel 1994).
4.1.6 Conclusion: Summarizing Mechanisms of Dissipation

of Photosynthetic Excitation Energy
In summary, the photosynthetic excitation energy not used by reduction and assimi-
lation of CO2 can be dissipated by a suite of protective and destructive mechanisms:
• the oxidase function of RuBISCO, photorespiration,
• the separation of peripheral and core parts of PSII, spill-over,
• capture of 1 O∗2 and dissipation of excitation energy as heat,
• reversible damage of the D1 -protein of LHCII,

• irreversible photodestruction.
The second to the fourth of these mechanisms imply photoinhibition of photochem-
ical work of either CO2 -assimilation or photorespiration (Box 4.3), and the fourth
one includes photodestruction. This makes us realizing difficulties and ambiguities
in our nomenclature: What is “inhibition” and what is “damage”, what is a reac-
tion for “protection” and what is “destruction”? The differences between acute and
chronic damage may be gradual and the quantitative relationships between various
ways of energy dissipation in relation to dynamics of photoprotective forces remain
a conundrum (Osmond and Grace 1995). In terms of the stress-concept (Box 3.1)
reversible and irreversible photoinhibition of CO2 -assimilation, represent elastic
and plastic strain, respectively.
4.1.7 Dissipation of Excitation Energy in the Form

of Fluorescence: A Tool in Plant Ecophysiology
From the first singlet excited state chlorophyll can relax directly to the ground state
by emission of light. The wavelength of this light is shifted somewhat towards the
far red range of the spectrum as compared to the wavelength optimum of 680 nm
of the excitation of chlorophyll a of PSII (P680). This is called chlorophyll fluo-
rescence. The intensity of this energy-dissipation is low and fluorescence does not
contribute to photoprotection under high irradiance stress. However, as a pathway
alternative to photochemical work and the non-photochemical processes of harmless
thermal dissipation of energy via the xanthophylls and oxidative photodestruction
(Sect. 4.1.6) fluorescence of chlorophyll a of PSII is an excellent indicator of the
state of PSII and its operation. Emission of fluorescence is most readily measured
in a non-invasive way by photometric techniques, where the yield and quenching,
respectively, of fluorescence are analyzed. As fluorescence is a reaction compet-
ing with photochemistry and with non-photochemical processes the magnitude of
fluorescence quenching quantitatively represents:
• photochemical work, i.e. transfer of electrons from excited photosystem II to
plastoquinone (“photochemical quenching”) and effective functioning of the
mechanism leading to photosynthetic CO2 -assimilation,
• energization of thylakoid membranes with xanthophyll mediated thermal energy
dissipation ( “non-photochemical quenching”),
• reversible and irreversible photoinhibition.
This has led to the development of sophisticated analyses of fluorescence signals,
which allow to distinguish between photochemical competence and the balance
between reversible and irreversible photoinhibition, as explained in Box 4.6. The
method has become a powerful tool in plant-ecophysiology and as an essentially
non-invasive and a non-destructive method it is also potentially suitable for re-
mote sensing, and hence, for the ecophysiological survey of large tropical forest
and savanna areas (see Sect. 1.3). Fluorescence analysis is based on the K AUT-
SKY -effect explained in Box 4.6. H. K AUTSKY worked in the early 1930s and
had to use elaborate optical equipment at 73 K. Currently we use pulse ampli-
tude modulated spectrophotometers, we can work at ambient solar radiation and
1.0
0.8
∆F / Fm´
0.6
0.4
0.2
0.0
ETR ( μmol m-2 s-1 )
150
100
50
8
6
NPQ
2
0
0 500 1000 1500 0 500 1000 1500

PAR ( μmol m-2 s-1 )
Fig. 4.8 Light response curves of F/Fm , ETR and NPQ (for explanation see Box 4.6) given
by plotting momentary measurements of chlorophyll fluorescence of Podocarpus falcatus trees
obtained during day-courses at two different sites, i.e. in a natural forest (left) and in a plantation of
Eucalyptus saligna (right) of a montane forest in Ethiopia (see also Sect. 1.3) vs incident irradiance
(PAR). (From Lüttge et al. 2003)
Box 4.6 Fluorescence analysis

A. Fluorescence induction kinetics (K AUTSKY effect)
In the first second after excitation fluorescence of photosystem II is induced and
rises from the fluorescence (O) of the dark-adapted leaf obtained at very weak
excitation energy to I , where all primary electron acceptors (Q in Box 4.2B)
are reduced, and after a small shoulder further to P, where the plastoquinone
pool (PQ in Box 4.2B) is also reduced.
Box 4.6 (Continued)
The high fluorescence obtained at P after the first second is quenched as the
electron-transport capacity at the acceptor side of photosystem I and other
fluorescence-quenching processes are activated.
B. Fluorescence analysis
In a pulse modulated-fluorescence-analysis system the following kinetics are
obtained during induction after a period of darkness (the example given was
obtained from a leaf of Clusia multiflora):
Arrowheads pointing upwards and downwards respectively, indicate switching

on and off of light, namely:
• m weak measuring light,
• p a single pulse of saturating actinic light,
• a actinic light with regular light-saturating pulses.
Box 4.6 (Continued)
Symbols in the graph have the following meaning:

• Fo minimal fluorescence yield of dark-adapted sample in weak measuring
light,
• Fm maximum fluorescence yield of the dark-adapted sample,
• Fv maximum variable fluorescence,
• Fo minimal fluorescence yield of the light-adapted sample,
• Fm maximum fluorescence yield of the light-adapted sample.
Calculations which can be made include the following:
1. Fv /Fm as a measure of potential quantum yield of photosystem II after dark
adaptation, which is lowered by photoinhibition,
Fv /Fm = (Fm − Fo )/Fm .
2. F/Fm as a measure of effective quantum yield,
F/Fm = (Fm − F)/Fm .
3. ETR = 0.86 × 0.5 × (F/Fm ) × PPFD as an empirical approximation of

the relative electron transport rates, where PPFD is incident photosynthetic
photon flux density, the factor 0.86 accounts for an average light absorption
of leaves of 86% (unless measured specifically) and the factor 0.5 for equal
distribution of absorbed photons to PSII and PSI.
4. The quenching coefficient for photochemical quenching of fluorescence, qP
qP = (Fm − F)/(Fm − Fo ) .
5. The quenching coefficient for non-photochemical quenching of fluorescence,

qN
qN = 1 − (Fm − Fo )/(Fm − Fo ) .
6. The extent of non-photochemical quenching, NPQ,
NPQ = (Fm − Fm )/Fm .
(Refs.: Genty et al. 1989; van Koten and Snel 1990; Schreiber and Bilger 1993;
Bilger et al. 1995; Maxwell and Johnson 2000)
Fig. 4.9A, B Light dependence curves of net CO2 uptake (A) and photochemical quenching (B) of
sun and shade leaves of Arbutus unedo. (Schreiber and Bilger 1987)
we benefit from instrument miniaturization for large scale operations in the field
(Sect. 2.3.2).
Among the parameters explained in Box 4.6 potential quantum yield of PSII,
Fv /Fm , is extremely useful in ecophysiology as it allows to distinguish between
acute and chronic photoinhibition. A dark adapted sample where all elements of
the electron transport chain are oxidized (“open”) shows maximum Fv /Fm upon
a saturating light pulse. This is close to 0.83, because maximally 83% of the light
are used (Björkman and Demmig 1987). If Fv /Fm is lower than that, the sample is
under photoinhibition. Then one can study the time of darkening needed to restore
it to close to 0.8 (Thiele et al. 1998). If the photoinhibition indicated by Fv /Fm
was reversible within several tens of minutes one was observing acute photoinhibi-
tion due to built up of an electrochemical gradient at the thylakoid membranes and
xanthophyll type energy dissipation. If it was not reversible before several hours,
destruction of D1 -protein must have been involved. If it was not reversible for an
extended period, e.g. overnight, there has been irreversible photodamage and pho-
toinhibition was chronic.
Effective quantum yield of PSII, F/Fm , and apparent electron transport rate,
ETR, indicate the activity of photochemical work. F/Fm decreases with increas-
ing light intensity as an increasingly less proportion of the incident irradiance is
used for photosynthesis, and ETR increases up to light saturation. In this way for
example from actual momentary measurements at varying irradiance, e.g. during
the course of a day, light saturation curves can be obtained and non-photochemical
fluorescence quenching can also be related to irradiance (Fig. 4.8). In Fig. 4.9 light-
response curves of net CO2 -uptake and photochemical quenching of a sun and
a shade leaf are compared. The CO2 -uptake curves show the typical characteris-
tics of sun and shade types (see Fig. 4.1). Photochemical quenching is high in both
cases at low PPFD and decreases much more rapidly with increasing light intensity
in the shade leaf than in the sun leaf, suggesting increased over reduction of the
photosynthetic electron transport chain.
4.2 Varying Irradiance on the Forest Floor and in Gap Dynamics 131
4.2 Varying Irradiance on the Forest Floor and in Gap Dynamics
4.2.1 The Response to Light Flecks
Light flecks were already mentioned in relation to the vertical structure of forests
(Sect. 3.4.1) describing the dynamics of light penetrating through the forest canopy.
The importance of such dynamics is illustrated by modelling canopy photosynthesis
with steady state and dynamic models, respectively, where the former overestimate
carbon gain by 13.4% at open sites and even by 86.5% at low light environments
of the understory (Stegemann et al. 1999; Timm et al. 2002, 2004). Clearly, light
flecks must be important in any type of forest. However, in the very dark, moist
tropical forests the dynamics of the responses of photosynthesis to light flecks play
an essential role in fulfilling the energy demands of photosynthesis in lower canopy
layers and particularly on the forest floor.
A prerequisite for photosynthetic utilization of the irradiance of rapidly formed
and transient light flecks are swift and co-ordinated reactions of stomata as
well as the biophysical and biochemical machineries of CO2 -assimilation. Fig-
ure 4.10 shows that CO2 -uptake, stomata-limited leaf conductance for water vapour
(gH2 O ) and intercellular CO2 -concentration ( pCO
i
2
) in the leaves of Claoxylon sand-
wicense and Euphorbia forbesii respond within minutes to a stepped increase of irra-
diance from low to high intensity. Both species occur in the understory of a Hawai-
ian forest (Pearcy 1983). In C. sandwicense, a tree with C3 -photosynthesis, gH2 O
Fig. 4.10A, B Response of net CO2 uptake, leaf conductance for water vapour (gH2 O ) and
i
intercellular CO2 concentration ( pCO 2
) to a stepped increase of light intensity to a PPFD of
500 µmol m s photons (arrow) after the leaves had been at 22 µmol m−2 s−1 photons for 2 h;
−2 −1
A Claoxylon sandwicense; B Euphorbia forbesii. (Pearcy et al. 1985)

showed an immediate linear increase, which continued for more than 50 min; net
CO2 -uptake first increased very rapidly and then more gradually in correlation with
i
gH2 O ; pCO decreased initially in response to increased availability of light energy
2
for photosynthesis and then increased again slightly as stomata opened more widely
(increased gH2 O ) allowing CO2 -uptake from the atmosphere. E. forbesii, perform-
ing C4 -photosynthesis (see Box 10.2, Sect. 10.1.2), responded in a somewhat dif-
ferent fashion. There was a slight delay in stomatal opening, but then maximal
stomatal conductance was attained within 15 min. After an initial decline, pCO i
2
was stabilised at an intermediate level while CO2 -uptake reached a high, constant
rate.
Are these responses sufficient to allow the plants to make efficient use of short
light flecks? Interestingly there is an effect of accelerating CO2 -uptake with time,
when very short light flecks are imposed repeatedly over short periods. Figure 4.11
shows, for the two species discussed above, that CO2 -uptake during artificial light
flecks increases gradually when subsequent light flecks of a duration of 1 min are
alternated with low background intensity for about 90 s. Conditioned or induced
leaves have a considerably higher light use efficiency in light flecks than uninduced
ones which is decreasing as light fleck length increases (Fig. 4.12; Valladares et al.
1997). The intensity of this background light is also important in maintaining the
conditioning effect, which leads to increasing efficiency in the use of light flecks. As
shown in Fig. 4.13, the efficient use of 5-s light flecks at 500 µ mol photons m−2 s−1
increases considerably when the background light intensity following each light
fleck is increased from 0 to 10 µ mol photons m−2 s−1 . At very high intensities the
irradiance of light flecks is used more effectively when there are interruptions by
low intensity background irradiation. This is explained by after effects of the high
Fig. 4.11A, B Response of CO2 -uptake of: A C. sandwicense; B E. forbesii during 1-min light
flecks (PPFD = 510 µ mol photons m−2 s−1 ) on a background of 22 µmol photons m−2 s−1 , which
was presented to the plants for 2 h prior to the first light fleck and which interrupted the individual
light flecks. Arrows indicate increase (↑) and decrease (↓) of PPFD respectively. (Pearcy et al.
1985)
300
250
LUE ( % )
200
150
100
50
0
1 10 100
Light fleck length ( s )
Fig. 4.12 Light use efficiency (LUE) during light flecks as a function of light fleck duration (log
scale on the x-axis) in induced (closed symbols) and uninduced (open symbols) leaves of three
tropical rain forest understory shrub species. (After data of Valladares et al. 1997)
Fig. 4.13 Efficiency of the utilization of 5-s light flecks (PPFD = 500 µ mol photons m−2 s−1 )
on a background irradiance of PPFD varied between 0 and 10 µ mol photons m−2 s−1 given for
60 s after each light fleck. In each case, leaves had been conditioned to reach a steady-state CO2
assimilation at 500 µ mol photons m−2 s−1 before applying the series of light flecks (Kirschbaum
and Pearcy 1988b)
irradiance light flecks, namely post irradiance CO2 fixation (Kirschbaum and Pearcy
1988a; Leakey et al. 2005).
What is the nature of these conditioning processes, which include induction,
the use of short-time high irradiance and the apparent after-effects during low back-
ground irradiation? The nature and mechanism of the important after effects are
understood looking at a 20-s light fleck experiment with the tropical shade-plant
Alocasia macrorrhiza where CO2 and O2 gas exchange were recorded simultane-
ously before, during and after the light fleck (Fig. 4.14). At the beginning of the light
fleck, O2 evolution increased very rapidly, and during the first second it attained
about twice the rate of steady state CO2 -uptake. Subsequently it dropped again and
matched the rate of CO2 -uptake after 2.5 s. This suggests that light-dependent elec-
Fig. 4.14 Time course of photosynthetic O2 evolution and CO2 uptake of a fully induced leaf
of Alocasia macrorrhiza during a 20-s light fleck (PPFD = 500 µ mol photons m−2 s−1 ). Arrows
indicate duration of light fleck (Kirschbaum and Pearcy 1988a)
tron transport, indicated by photosynthetic O2 evolution, may proceed rapidly to fill

up pools of reduced compounds in a very short initial period after stepped increase
in irradiance, before it becomes limited by reactions of CO2 -reduction. Biophysi-
cal light-reactions of photosynthesis (Box 4.2B) are extremely fast. Therefore, in
light flecks the slower processes of induction are the biochemical reactions of CO2 -
fixation and assimilation as well as stomatal responses. Among them, the regen-
eration of ribulose-bis-phosphate the CO2 -acceptor in photosynthesis, is compara-
tively fast. It was found to be 60 – 120 s under transient light conditions as typical
for light flecks, while light-activation of the activity of the carboxylase (RuBISCO)
and stomatal reactions occurred in the range of 10 – 30 min (Sassenrath-Cole and
Pearcy 1992). On the other hand, the mechanisms which show slower induction
are also subject to slower decay and may remain active during intermittent light
flecks. At the end of a light fleck, when there is a stepped decrease of irradiance,
O2 -evolution drops immediately as photosynthetic electron transport stops. How-
ever, CO2 -uptake declines only gradually showing an after effect of the high irra-
diance during the light-fleck due to a surplus of reduced compounds formed during
high PPFD. In this way the consequence of after effects is that high light of light
flecks can be used more effectively when the high irradiance is not continuous be-
cause this allows a more efficient use of the intrinsic potential for reduction built
up during absorption of light at high intensity. Thus, this phenomenon explains the
particular efficiency of short duration light fleck utilisation as a co-ordination of
more rapid responses (photosynthetic electron transport) and more sluggish
processes (photosynthetic CO2 assimilation), with both adjusted to some extent
in series so that each process may not be entirely simultaneous. Of course, the light
flecks must be short, if this is to be relatively important quantitatively. The different
time constants of the processes involved also explain the conditioning to intermit-
tent light (Fig. 4.11). Thus, the specific dynamics of transients after light intensity
is stepped up and down, make light flecks a quantitatively more important energy
source for forest-floor photosynthetic carbon assimilation than one might expect
from their intensity and duration alone. The dynamics of fluctuating light, stomatal
conductance and biochemical activation and pools of key photosynthetic interme-
diates are also convincingly simulated by mathematical models (Kirschbaum et al.
1997). Another factor which is also involved is respiration the activity of which is
regulated by ATP-demand of the plants during light fleck dynamics (Noguchi et al.
2001a).
One may also ask whether light fleck responses are different in shade and sun
plants, since one might expect that the latter are less dependent on intermittent light.
Indeed, such differences have been observed. They are largely based on stomatal
dynamics. Stomatal relations are very important as the induction of photosynthesis
during dynamic light flecks depends on stomatal opening and CO2 availability to the
mesophyll. When stomata are already open at low background irradiance induction
may be faster than when first a stomatal opening movement of guard cells is required
(Valladares et al. 1997).
Responses of plants to light flecks are also species specific (Leakey et al. 2005).
Species dependence of stomatal responses are also seen in Fig. 4.10. A compari-
son of two dipterocarp rain forest tree species showed that one species could use
light flecks quickly and the other more slowly but instead performed photosynthesis
continuously at low light (Zipperlen and Press 1997). In species of Piper, accli-
mation of stomatal responses to different light intensities was observed, which was
important for the performance of the plants in varying light environments (Tinoco-
Ojanguren and Pearcy 1992). A comparison of Piper auritum, a pioneer tree, and
Piper aequale, a shade tolerant shrub of Mexican tropical forests, showed that dif-
ferences in induction of photosynthesis could be accounted for by differences in
stomatal behaviour. The shade tolerant shrub, P. aequale, had the larger and more
rapid response of stomatal conductance (gH2 O ) to light flecks, which was shown to
improve carbon gain during subsequent light flecks for shade adapted plants. Con-
versely, low-light acclimated plants of the pioneer tree, P. auritum, showed even
slower and smaller conductance responses than sun-acclimated plants, and there
was no significant improvement in use of subsequent light flecks (Tinoco-Ojanguren
and Pearcy 1993a,b). Another comparison was provided by Poorter and Oberbauer
(1993), who studied saplings of a climax tree species, Dipteryx panamensis, and
a pioneer tree, Cecropia obtusifolia, in a rainforest of Costa Rica. The results of
their comparative investigation are compiled in Table 4.4. Remembering the differ-
ences in general photosynthesis characteristics of these groups of plants (Sect. 4.1.1)
it appears that the climax-tree saplings exploit temporal variation in light availabil-
ity by refining the speed of the induction response. In contrast, the pioneer species
adjust by realising higher rates of light-saturated photosynthesis under high irradia-
tion.
Different light fleck responses have also been reported in relation to leaf-lon-
gevity (Kursar and Coley 1993). In shade-tolerant species with short lived leaves
(1 year) induction to attain 90% of maximum photosynthetic rates took 3 – 6 min,
while 11 – 36 min were needed in long-lived leaves (> 4 years). In this case, how-
ever, RuBISCO activation seemed to be the time-limiting factor.
Table 4.4 Comparison of the responses of saplings of two Costa Rican rainforest tree species to
light flecks in situ (Poorter and Oberbauer 1993)
Dipteryx panamensis Cecropia obtusifolia
Character Climax species Pioneer species
in bright microsites
Induction time needed to reach 90% 16 min 10 min
of light-saturated rate of photosyn-
thesis in the morning
Daily average induction time needed Shorter than in C.o. Longer than in D. p.
Duration of maintenance of high Longer than in C.o. Shorter than in D. p.
levels of induction
Behaviour when grown in shaded Faster rates of induction No difference in rates of
sites as compared to bright sites induction
No difference in light- Lower light-saturated
saturated rates of photo- rates of photosynthesis
synthesis
On the forest floor plants heat up in light flecks and this much adds to sud-
den light stress (Leakey et al. 2003, 2005). Thus, light flecks may not only have
beneficial effects and it can not be generalized that light fleck activity is directly
associated with greater carbon gain (Leakey et al. 2005). It is an intriguing ques-
tion if the plants adapted to life in the deep shade of forest floors do not get un-
der severe problems of photoinhibition and photodamage when subjected to high
irradiance in the light flecks. Shade leaves of some plants do not appear to be pho-
toinhibited during light flecks, they have a limited xanthophyll-cycle strategy and
use increased synthesis of D1 -protein (Schiefthaler et al. 1999). However, mostly it
is observed that there is photoinhibition due to NPQ but no photodamage. This is
managed because of the short time constants of zeaxanthin functions (Sect. 4.1.4).
With the rise and fall of the transthylakoid pH in relation to incident irradi-
ance xanthophyll-cycle dependent energy dissipation is engaged rapidly during light
flecks preventing photooxidative damage and disengaged rapidly after light flecks
pass (Demmig-Adams et al. 1996; Logan et al. 1997; Watling et al. 1997; Adams et
al. 1999).
4.2.2 Light Quality: Signalling Functions of Light
As already mentioned above (Sect. 3.4.1), light quality changes in relation to hori-
zontal and vertical structure of forests.
Ultraviolet light (UV) is interesting because blue light and UV may exert sig-
nalling functions. Shade adapted tropical tree seedlings can gradually adapt to UV
(Krause et al. 2003b) and UV radiation may be attenuated by UV absorbing sub-
stances and plant cuticles (Krauss et al. 1997). Often, however, natural UV-A and
UV-B radiation may cause photoinhibition and photodamage particularly in shade

leaves exposed to full sunlight (Krause et al. 1999, 2003a).
Most interesting in relation to signalling, however, is the other end of the spec-
trum because filtration by canopies eliminates the red light (R) from the solar spec-
trum much more effectively than the far-red light (FR). Sunlight has a mean R/FR
ratio of 1.2 but under green canopies the ratio may be reduced to levels below 0.5
(Vázquez-Yanes and Orozco-Segovia 1993). This affects all processes regulated by
the phytochrome system. Irradiation with red light generates the active PFR form
of phytochrome, which elicits various photomorphogenetic responses. Far-red light
inactivates the phytochrome, shifting the phytochrome equilibrium towards the in-
active PR -form (Box 4.7). The light intensities required in phytochrome effects are
often extremely low. It is the signalling function of light, which is sensed by the
phytochrome system and not its function as an energy source.
Box 4.7 The reversible phytochrome system

4.3 Seedlings: Germination, Establishment and Growth
4.3.1 Regulation of Seed Dormancy and Germination
Among the many processes governed by phytochrome (Sect. 4.2.2) one of the most
well known is the germination of the seeds of “light-germinators”, i.e. positive pho-
toblastic seeds. In this way phytochrome also plays an important role in the regu-
lation of succession and regeneration in tropical forests, because light dependence
of seed germination is one of the most fundamental differences between pioneer
species and late successional or climax species, where competition is not only ex-
plained by the substrate and energy aspects of irradiance but also its signalling func-
tions (Aphalo et al. 1999).
Only seeds of late successional and climax species can germinate and establish
seedlings under deep canopy shade. These seeds germinate very soon after dispersal
and also remain alive in the soil only for a short time. The mean ecological longevity
of seeds in the tropical rainforest may be one of the shortest of any plant community
(Vázquez-Yanes and Orozco-Segovia 1993). The advantage of this behaviour lies in
the fact that seeds are more threatened by predators and parasites in the soil environ-
ment of the tropical rainforest with continuous moisture and high temperature, than
are seedlings. Thus, the seed banks in rainforest soils are depleted of seeds of late
succession and climax species. On the other hand, seedlings may grow extremely
slowly and a persistent nursery of small plants is built up, i.e. a seedling bank
instead of a seed bank. Flores (1992) has studied two species of the cloud forest of
the northern coastal range of Venezuela and his observations give a good idea about
the actual longevity of tree seedlings after germination:
• Aspidosperma fendleri (Apocynaceae), an emergent species, which grows its
crowns above the canopy (see Fig. 3.25),
– germination time 5 days,
– longevity of cotyledons 2 months,
– longevity of 1st leaf pair 2 years.
• Richeria grandis (Euphorbiaceae), a canopy species,
– germination time 20 days,
– longevity of cotyledons 2.5 years,
– longevity of leaves 3 years.
The small plants remain in a state of slow growth until a canopy gap provides an
opportunity for stimulation of growth (see Sect. 4.3.2). The survival of the seedlings
is independent of photosynthetic parameters and largely determined by morpholog-
ical characteristics which are likely to provide protection from and enhance defence
against herbivores and pathogens, i.e. dense and tough leaves, a well established root
system and a high wood density. Furthermore, reserves of non-structural carbohy-
drates in stems, roots and storage cotyledons support long-term survival of seedlings
of shade tolerant species enabling them to cope with periods of biotic and abiotic
4.3 Seedlings: Germination, Establishment and Growth 139
stress (Myers and Kitajima 2007). Seedlings in gaps proved to be more resistant
to herbivory than seedlings in the undergrowth of tropical forests (Blumwald and
Peart 2001). Seedling survival of 13 tropical tree species was found to be negatively
correlated to relative growth rate (RGR), i.e. both low RGR of plants raised in the
shade and high RGR of plants in the sun, and to leaf area ratio and positively corre-
lated to root/shoot ratios and wood density (Kitajima 1994; Fig. 4.15). Fast growing
pioneer plants need lower stem support and afford a lower wood density, but this
reduction of support costs is related to higher mortality rates (Fig. 4.15E; King et al.
2006).
In contrast to late successional species seeds of woody pioneer species are ca-
pable of dormancy. They are often the most abundant components of the soil seed
bank in tropical forests (Vázquez-Yanes and Orozco-Segovia 1993). Dormancy may
be enforced by hard seed coats which are impermeable to water and oxygen and
need many weeks for breakdown by weathering and microbial action. However,
germination is mainly determined by light.
Light may act via temperature effects, especially via temperature alternations,
which are required by some seeds for germination. Canopy gaps and clearings lead
to greater fluctuations of soil surface temperatures due to direct insolation (see
Fig. 3.29). Often, however, germination is regulated by light quality and the involve-
ment of the phytochrome system rather than by light intensity. A higher proportion
of red light activates and a higher proportion of far-red light inactivates phytochrome
and the reversibility of phytochrome effects may be important in excluding reactions
to short light flecks (Sect. 4.2.1) and sensing true light gaps. The photoreversibility
of phytochrome mediated germination within certain time limits may be essen-
tial to prevent germination resulting from light flecks (Vázquez-Yanes and Orozco-
Segovia 1993).
The sophisticated regulation of dormancy and germination, respectively, is most
frequent among pioneer species and gap colonizers, with germination inhibited un-
der closed canopies and stimulated in clearings.
4.3.2 Growth of Seedlings
Seedling growth depends on availability of water (Sect. 5.1) and nutrients (Sect.
3.4.4) but most drastically on light (Medina 1998; Poorter 1998). After germina-
tion and establishment of seedlings often forming nurseries of slow growing plants
(Sect. 4.3.1), further growth will depend very much on photosynthesis, the light
powered fixation of CO2 and reduction to organic material, as well as allocation
of photosynthetic products (Zipperlen and Press 1996; Scholes et al. 1997). This,
of course, applies to both pioneer and climax species. Both may differ though, in
responses to light intensity, which are representative of sun and shade plant char-
acteristics (see Sect. 4.1.1).
However, in some cases the differences in photosynthetic capacity related to light
intensity between seedlings of pioneer and climax species, or between mature early
and late successional species, has been found to be surprisingly small (Riddoch
Fig. 4.15 Seedling mortality

rate during the first year in
the shade of 13 tropical tree
species related to relative
growth rate (RGR) in shade
and sun (A,B), leaf area ratio
(LAR) (C), root : shoot ratio
(D) and wood density (E).
(After Kitajima 1994)
4.3 Seedlings: Germination, Establishment and Growth 141
Fig. 4.16 Leaf-anatomy of

seedlings of a pioneer or
early succession tree, Nau-
clea diderrichii (A,B), and
a late succession tree Entan-
drophragma angolense (C,D),
grown at high light (A,C) and
at low light (B,D). The white
bars denote 50 µ m. (Riddoch
et al. 1991)
et al. 1991). Huber (1978) examined photosynthetic characteristics, e.g. the light
compensation point (Sect. 4.1.1), of 54 vascular plant species in Rancho Grande
(Venezuela). He found that by this criterion the majority of the species growing in
the lower forest strata did not belong to extreme shade-adapted plant types, but pos-
sessed a wide capacity for response to the highly variable irradiance in this montane
cloud forest. It may be noted generally that a schematic distinction of pioneer sun
plants/shade plants in this context is too simple. Changes can occur during develop-
ment (Turnball 1991; Agyeman et al. 1999; Kyereh et al. 1999; Poorter et al. 2005).
Mature shade leaves of seedlings can substantially acclimate to full sunlight em-
ploying mechanisms of energy dissipation, UV absorbing substances etc. (Krause et
al. 2004). Availability of water and nutrients, especially nitrogen (Sect. 4.1.2) play
a role in this (Castro et al. 1995; Bungard et al. 2000).
Complex regulation is involved. For a fluctuating tropical environment with fre-
quent disturbance by typhoons and canopy opening a trade off between acclima-
tion ability and plasticity has been considered (Yamashita et al. 2002). Strauss-
Debenedetti and Bazzaz (1991) have suggested that plasticity and acclimation
should be distinguished as follows:
• late successional species often cannot acclimate to high light intensities when
transferred from low-light to high light (low acclimation) but may grow well if
kept continuously under low and high light respectively (high plasticity),
• pioneer species may grow at low and high light and show a considerable stimu-
lation after transfer from low light to high light (high plasticity and high accli-
mation).
The expression of low-light and high-light forms of a species may also be deter-
mined by the phytochrome system (Smith et al. 1993), but in particular blue-light
photoreceptors are also involved in this regulation (Lichtenthaler et al. 1981; Hum-
beck and Senger 1984; see Lüttge et al. 1986). Leaf-anatomical features often show
pronounced differences; sun leaves are thicker than shade leaves and have addi-
tional layers of palisade parenchyma. In a comparison of young seedlings of the
tropical trees Nauclea diderrichii (De Wilde.) Merrill, a pioneer species, and En-
tandrophragma angolense (Welw.) C.DC., a climax species, both from West Africa,
differences in acclimation and photosynthetic capacity at high light intensity were
only small. However, there were marked morphogenetic effects on leaf anatomy in
plants grown in the sun and in the shade respectively, in the pioneer species N. dider-
richii but not so much in the climax species E. angolense (Fig. 4.16).
References
Adams WW, Demmig-Adams B, Logan BA, Barker DH, Osmond CB (1999) Rapid changes in
xanthophyll cycle-dependent energy dissipation and photosystem II efficiency in two vines,
Stephania japonica and Smilax australis, growing in the understory of an open Eucalyptus
forest. Plant Cell Environ 22:125–136
References 143
Agyeman VK, Swaine MD, Thompson J (1999) Responses of tropical forest tree seedlings to
irradiance and the derivation of a light response index. J Ecol 87:815–827
Anderson JM, Thomson WW (1989) Dynamic molecular organization of the plant thylakoid mem-
brane. Photosynthesis. Alan R Liss, New York, pp 161–182
Aphalo PJ, Ballaré CL, Scopel AL (1999) Plant-plant signalling, the shade-avoidance response and
competition. J Exp Bot 50:1629–1634
Augustynowicz J, Gabrỳs H (1999) Chloroplast movements in fern leaves: correlation of move-
ment dynamics and environmental flexibility of the species. Plant Cell Environ 22:1239–1248
Bailey S, Walters RG, Jansson S, Horton P (2001) Acclimation of Arabidopsis thaliana to the light
environment: the existence of separate low light and high light responses. Planta 213:794–801
Baroli I, Do AD, Yamane T, Niyogi KK (2003) Zeaxanthin accumulation in the absence of a func-
tional xanthophyll cycle protects Chlamydomonas reinhardtii from photooxidative stress.
Plant Cell 15:992–1008
Barth C, Krause GH, Winter K (2001) Responses of photosystem I compared with photosystem II
to high-light stress in tropical shade and sun leaves. Plant Cell Environ 24:163–176
Bilger W, Björkman O (1994) Relationships among violaxanthin deepoxidation, thylakoid mem-
brane conformation, and non-photochemical chlorophyll fluorescence quenching in leaves of
cotton (Gossypium hirsutum L.). Planta 193:238–246
Bilger W, Schreiber U, Bock M (1995) Determination of the quantum efficiency of photosystem
II and of non-photochemical quenching of chlorophyll fluorescence in the field. Oecologia
102:425–432
Björkman O, Demmig B (1987) Photon yield of O2 evolution and chlorophyll fluorescence char-
acteristics at 77 K among vascular plants of diverse origins. Planta 170:489–504
Blumwald AG, Peart DR (2001) Growth strategies of a shade-tolerant tropical tree: the interactive
effects of canopy gaps and simulated herbivory. J Ecol 89:608–615
Bonal D, Barigah TS, Granier A, Guehl JM (2000) Late-stage canopy tree species with extremely
low δ 13 C and high stomatal sensitivity to seasonal soil drought in the tropical rainforest of
French Guiana. Plant Cell Envoiron 23:445–459
Büch K, Stransky H, Bigus H-J, Hager A (1994) Enhancement by artificial electron acceptors of
thylakoid lumen acidification and zeaxanthin formation. J Plant Physiol 144:641–648
Bungard RA, Press C, Scholes JD (2000) The influence of nitrogen on rain forest dipterocarp
seedlings exposed to a large increase in irradiance. Plant Cell Environ 23:1183–1194
Castro Y, Fetcher N, Fernández DS (1995) Chronic photoinhibition in seedlings of tropical trees.
Physiol Plant 94:560–565
Cheng L (2003) Xanthophyll cycle pool size composition in relation to the nitrogen content of
apple leaves. J Exp Bot 54:385–393
Chow WS, Qian L, Goodchild DJ, Anderson JM (1988) Photosynthetic acclimation of Alocasia
macrorrhiza (L.) G. Don. to growth irradiance: structure, function and composition of chloro-
plasts. Aust J Plant Physiol 15:107–122
Critchley C, Russell AW (1994) Photoinhibition of photosynthesis in vivo: the role of protein
turnover in photosystem II. Physiol Plant 92:188–196
Demmig-Adams B (1990) Carotenoids and photoprotection: a role for the xanthophyll zeaxanthin
cycle. Biochim Biophys Acta 1020:1–24
Demmig-Adams B, Adams WW (1992) Photoprotection and other responses of plants to high light
stress. Annu Rev Plant Physiol Plant Mol Biol 43:599–626
Demmig-Adams B, Adams WW (1993) The xanthophyll cycle, protein turnover and the high light
tolerance of sun-acclimated leaves. Plant Physiol 103:1413–1420
Demmig-Adams B, Adams WW (1994) Capacity for energy dissipation in the pigment bed in
leaves with different xanthophyll cycle pools. Aust J Plant Physiol 21:575–588
Demmig-Adams B, Gilmore AM, Adams WW (1996) In vivo functions of carotenoids in higher
plants. FASEB J 10:403–412
Doley D, Unwin GL, Yates DJ (1988) Spatial and temporal distribution of photosynthesis and tran-
spiration by single leaves in a rainforest tree, Argyrodendron peralatum. Aust J Plant Physiol
15:317–326
Eschenbach C, Glauner R, Kleine M, Kappen L (1998) Photosynthesis of selected tree species in

lowland dipterocarp forest of Sabah, Malaysia. Trees 12:356–365
Evans JR (1988) Acclimation by the thylakoid membranes to growth irradiance and the partitioning
of nitrogen between soluble and thylakoid proteins. Aust J Plant Physiol 15:93–106
Fetene M, Lee HSJ, Lüttge U (1990) Photosynthetic acclimation in a terrestrial CAM bromeliad,
Bromelia humilis Jacq. New Phytol 114:399–406
Field CB (1988) On the role of photosynthetic responses in constraining the habitat distribution of
rainforest plants. Aust J Plant Physiol 15:343–358
Flores S (1992) Growth and seasonality of seedlings and juveniles of primary species of a cloud
forest in northern Venezuela. J Trop Ecol 8:299–305
Genty B, Briantais J-M, Baker NR (1989) The relationship between the quantum yield of photo-
synthetic electron transport and quenching of chlorophyll fluorescence. Biochin Biophy Acta
990:87–92
Gilmore AM (1997) Mechanistic aspects of xanthophyll cycle-dependent photoprotection in higher
plant chloroplasts and leaves. Physiol Plant 99:197–209
Gilmore AM, Govindjee (1999) How higher plants respond to excess light: energy dissipation in
photosystem II. In: Singhal GS, Renger G, Sopory SK, Irrgang K-D, Govindjee (eds) Concepts
in photobiology: photosynthesis and photomorphogenesis. Narosa Publ House, New Delhi, pp
513–548
Gilmore AM, Yamasaki H (1998) 9-Aminoacridine and dibucaine exhibit competitive interactions
and complicated inhibitory effects that interfere with measurements of pH and xanthophyll
cycle-dependent photosystem II energy dissipation. Photosynthesis Res 57:159–174
Gilmore AM, Hazlett TL, Debrunner PG, Govindjee (1996) Comparative time-resolved photo-
system II chlorophyll a fluorescence analyses reveal distinctive differences between photoin-
hibitory reaction center damage and xanthophyll cycle-dependent energy dissipation. Pho-
tochem Photobiol 64:552–563
Gilmore AM, Shinkarev VP, Hazlett TL, Govindjee (1998) Quantitative analysis of the effects
of intrathylakoid pH and xanthophyll cycle pigments on chlorophyll a fluorescence lifetime
distributions and intensity in thylakoids. Biochemistry 73:13582–13593
Gorton H, Williams WE, Vogelmann TC (1999) Chloroplast movement in Alocasia macrorrhiza.
Hager A (1980) The reversible, light-induced conversions of xanthophylls in the chloroplast. In:
Czygan FC (ed) Pigments in plants. G Fischer, Stuttgart, pp 57–79
Hager A, Holocher K (1994) Localization of the xanthophyll-cycle enzyme violaxanthin de-
epoxidase within the thylakoid lumen and abolition of its mobility by a (light-dependent) pH
decrease. Planta 192:581–589
He J, Chow WS (2003) The rate coefficient of repair of photosystem II after photoinactivation.
He J, Chee CW, Goh CJ (1996) ‘Photoinhibition’ of Heliconia under tropical conditions: the im-
portance of leaf orientation for light interception and leaf temperature. Plant Cell Envoiron
19:1238–1248
Holt NE, Zigmantas D, Valkunas L, Li X-P, Niyogi KK, Fleming GR (2005) Carotenoid cation
formation and the regulation of photosynthetic light harvesting. Science 307:433–436
Horton P, Ruban A (2005) Molecular design of the photosystem II light-harvesting antenna: pho-
tosynthesis and photoprotection. J Exp Bot 56:365–373
Horton P, Ruban AV, Walters RG (1994) Regulation of light harvesting in green plants. Indication
by nonphotochemical quenching of chlorophyll fluorescence. Plant Physiol 106:415–420
Huber O (1978) Light compensation point of vascular plants of a tropical cloud forest and an
ecological interpretation. Photosynthetica 12:382–390
Huc R, Ferhi A, Guehl JM (1994) Pioneer and late stage tropical rainforest tree species (French
Guiana) growing under common conditions differ in leaf gas exchange regulation, carbon
isotope discrimination and leaf water potential. Oecologia 99:297–305
References 145
Humbeck K, Senger H (1984) The blue light factor in sun and shade plant adaptation. In: Senger
H (ed) Blue light effects in biological systems. Springer, Berlin Heidelberg New York, pp
344–351
King DA, Davies SJ, Tan S, Noor NSMD (2006) The role of wood density and stem support costs
in the growth and mortality of tropical trees. J Ecol 94:670–680
Kirschbaum MUF, Pearcy RW (1988a) Concurrent measurements of oxygen- and carbon-dioxide
exchange during light flecks in Alocasia macrorrhiza (L.) G. Don. Planta 174:527–533
Kirschbaum MUF, Pearcy RW (1988b) Gas exchange analysis of the fast phase of photosynthetic
induction in Alocasia macrorrhiza. Plant Physiol 87:818–821
Kirschbaum MUF, Küppers M, Schneider H, Giersch C, Noe S (1997) Modelling photosynthesis
in fluctuating light with inclusion of stomatal conductance, biochemical activation and pools
of key photosynthetic intermediates. Planta 204:16–26
Kitajima K (1994) Relative importance of photosynthetic traits and allocation patterns as correlates
of seedling shade tolerance of 13 tropical trees. Oecologia 98:419–428
Kitajima K, Hogan KP (2003) Increases of chlorophyll a/b ratios during acclimation of tropical
woody seedlings to nitrogen limitation and high light. Plant Cell Environ 26:857–865
Königer M, Harris GC, Pearcy RW (1998) Interaction between flux density and elevated tempera-
tures on photoinhibition in Alocasia macrorrhiza. Planta 205:214–222
Koten O van, Snel JFH (1990) The use of chlorophyll fluorescence nomenclature in plant stress
physiology. Photosynthesis Res 25:147–150
Krause GH, Virgo A, Winter K (1995) High susceptibility to photoinhibition of young leaves of
tropical forest trees. Planta 197:583–591
Krause GH, Schmude C, Garden H, Koroleva O, Winter K (1999) Effects of solar ultraviolet ra-
diation on the potential efficiency of photosystem II in leaves of tropical plants. Plant Physiol
121:1349–1358
Krause GH, Koroleva OY, Dalling JW, Winter K (2001) Acclimation of tropical tree seedlings to
excessive light in simulated tree-fall gaps. Plant Cell Environ 24:1345–1352
Krause GH, Galle A, Gademann R, Winter K (2003a) Capacity of protection against ultraviolet
radiation in sun and shade leaves of tropical forest plants. Funct Plant Biol 30:533–542
Krause GH, Grube E, Virgo A, Winter K (2003b) Sudden exposure to solar UV-B radiation reduces
net CO2 uptake and photosystem I efficiency in shade-acclimated tropical tree seedlings. Plant
Physiol 131:745–758
Krause GH, Grube E, Koroleva OY, Barth C, Winter K (2004) Do mature shade leaves of tropical
tree seedlings acclimate to high sunlight and UV radiation? Funct Plant Biol 31:743–756
Krause GH, Gallé A, Virgo A, García M, Bucic P, Jahns P, Winter K (2006) High-light stress
does not impair biomass accumulation of sun-acclimated tropical tree seedlings (Calophyllum
longifolium Willd. and Tectona grandis L.f.). Plant Biol 8:31–41
Krauss P, Markstätter C, Riederer M (1997) Attenuation of UV radiation by plant cuticles from
woody species. Plant Cell Environ 20:1079–1085
Krieger-Liszkay A, Trebst A (2006) Tocopherol is the scavenger of singlet oxygen produced by
the triplet state of chlorophyll in the PSII reaction centre. J Exp Bot 57:1677–1684
Kursar TA, Coley PD (1993) Photosynthetic induction times in shade-tolerant species with long
and short-lived leaves. Oecologia 93:165–170
Kyereh B, Swaine MD, Thompson J (1999) Effect of light on the germination of forest trees in
Ghana. J Ecol 87:772–783
Leakey ADB, Press MC, Scholes JD (2003) High-temperature inhibition of photosynthesis is
greater under sunflecks than uniform irradiance in a tropical rain forest tree seedling. Plant
Cell Environ 26:1681–1690
Leakey ADB, Scholes JD, Press MC (2005) Physiological and ecological significance of sunflecks
for dipterocarp seedlings. J Exp Bot 56:469–482
Lichtenthaler HK, Buschmann C, Döll M, Fietz H-J, Bach T, Kozel U, Meier D, Rahmsdorf U
(1981) Photosynthetic activity, chloroplast ultrastructure and leaf characteristics of high-light
and low-light plants and of sun and shade leaves. Photosynth Res 2:115–141
Logan BA, Barker DH, Demmig-Adams B, Adams WW (1996) Acclimation of leaf carotenoid
composition and ascorbate levels to gradients in the light environment within an Australian
rainforest. Plant Cell Environ 19:1083–1090
Logan BA, Barker DH, Adams WW, Demmig-Adams B (1997) The response of xanthophyll cycle-
dependent energy dissipation in Alocasia brisbanensis to sunflecks in a subtropical rainforest.
Aust J Plant Physiol 24:27–53
Lüttge U (1985) Epiphyten:Evolution und Ökophysiologie. Naturwissenschaften 72:557–566
Lüttge U, Ball E, Kluge M, Ong BL (1986) Photosynthetic light requirements of various tropical
vascular epiphytes. Physiol Vég 24:315–331
Lüttge U, Berg A, Fetene M, Nauke P, Peter D, Beck E (2003) Comparative characterization of
photosynthetic performance and water relations of native trees and exotic plantation trees in
an Ethiopian forest. Trees 17:40–50
Lüttge U, Kluge M, Bauer G (2005) Botanik, 5th edn. VCH, Weinheim
Matsubara S, Gilmore AM, Osmond CB (2001) Diurnal and acclimatory responses of violaxanthin
and lutein epoxide in the Australian mistletoe Amyema miquelii. Aus J Plant Physiol 28:793–
800
Matsubara S, Morosinotto T, Bassi R, Christian A-L, Fischer-Schliebs E, Lüttge U, Orthen B,
Franco AC, Scarano FR, Förster B, Pogson BJ, Osmond CB (2003) Occurrence of the lutein-
epoxide cycle in mistletoes of the Loranthaceae and Viscaceae. Planta 217:868–879
Matsubara S, Naumann M, Martin R, Nichol C, Rascher U, Morosinotto T, Bassi R, Osmond B
(2005) Slowly reversible de-epoxidation of lutein-epoxide in deep shade of a tropical tree
legume may ‘lock-in’ lutein-based photoprotection during acclimation to strong light. J Exp
Bot 56:461–468
Maxwell K, Johnson GN (2000) Chlorophyll fluorescence – a practical guide. J Exp Bot 51:659–
668
Maxwell C, Griffiths H, Young AJ (1994) Photosynthetic acclimation to light regime and water
stress by the C3 -CAM epiphyte Guzmania monostachia: gas exchange characteristics, photo-
chemical efficiency and the xanthophyll cycle. Funct Ecol 8:746–754
Maxwell C, Griffiths H, Borland AM, Young AJ, Broadmeadow MSJ, Fordham MC (1995) Short-
term photosynthetic responses of the C3 -CAM epiphyte Guzmania monostachia var. monos-
tachia to tropical seasonal transitions under field conditions. Aust J Plant Physiol 22:771–781
Medina E (1998) Seedling establishment and endurance in tropical forests: ecophysiology of stress
during early stages of growth. In: Scarano FR, Franco AC (eds) Ecophysiological strategies
of xerophytic and amphibious plants in the neotropics. Oecologia Brasiliensis, vol IV. PPGE-
UFRJ, Rio de Janeiro, pp 23–43
Myers JA, Kitajima K (2007) Carbohydrate storage enhances seedling shade and stress tolerance
in a neotropical forest. J Ecol 95:383–395
Newell EA, McDonald EP, Strain BR, Denslow JS (1993) Photosynthetic responses of Miconia
species to canopy openings in a lowland tropical rainforest. Oecologia 94:49–56
Nielsen SL, Enríquez S, Duarte CM (1997) Control of PAR-saturated CO2 exchange rate in some
C3 and CAM plants. Biol Plant 40:91–101
Noguchi K, Nakajima N, Terashima I (2001a) Acclimation of leaf respiratory properties in Aloca-
sia odora following reciprocal transfers of plants between high- and low-light environments.
Plant Cell Environ 24:831–839
Noguchi K, Go C-S, Terashima I, Ueda S, Yoshinari T (2001b) Activities of the cyanide-resistant
respiratory pathway in leaves of sun and shade species. Aust J Plant Physiol 28:27–35
Osmond CB, Grace SC (1995) Perspectives on photoinhibition and photorespiration in the field:
quintessential inefficiencies of the light and dark reactions of photosynthesis? J Exp Bot
46:1351–1362
Pearcy RW (1983) The light environment and growth of C3 and C4 tree species in the understory
of a Hawiian forest. Oecologia 58:19–25
Pearcy RW, Osteryoung K, Calkin HW (1985) Photosynthetic responses to dynamic light environ-
ments by Hawaiian trees. Plant Physiol 79:896–902
References 147
Pfündel E, Bilger W (1994) Regulation and possible function of the violaxanthin cycle. Photosynth
Res 42:89–109
Poorter L (1998) Seedling growth of Bolivian rain forest species in relation to light and water
availability. PhD Thesis, Utrecht (ISBN 90-393-1759-3)
Poorter L, Oberbauer SF (1993) Photosynthetic induction responses of two rainforest tree species
in relation to light environment. Oecologia 96:193–199
Poorter L, Bongers F, Sterck FJ, Wöll H (2005) Beyond the regeneration phase: differentiation of
height-light trajectories among tropical tree species. J Ecol 93:256–267
Ramalho JC, Compos PS, Quartin VL, Silva MJ, Nunes MA (1999) High irradiance impairments
on photosynthetic electron transport, ribulose-1,5-bisphosphate carboxylase/oxygenase and
N assimilation as a function of N availability in Coffea arabica L. plants. J Plant Physiol
154:319–326
Reich PB, Walters MB, Ellsworth DS, Uhl C (1994) Photosynthesis-nitrogen relations in Amazo-
nian tree species. I. Patterns among species and communities. Oecologia 97:62–72
Riddoch I, Lehto T, Grace J (1991) Photosynthesis of tropical tree seedlings in relation to light and
nutrient supply. New Phytol 119:137–147
Sack L, Tyree MT, Holbrook NM (2005) Leaf hydraulic architecture correlates with regeneration
irradiance in tropical rainforest trees. New Phytol 167:403–413
Sassenrath-Cole GF, Pearcy RW (1992) The role of ribulose-1,5-bisphosphate regeneration in the
induction requirement of photosynthetic CO2 exchange under transient light conditions. Plant
Physiol 99:227–234
Schäfer C, Schmid V (1993) Pflanzen im Lichtstreß. Biol Unserer Zeit 23:55–62
Schiefthaler U, Russell AW, Bolhàr-Nordenkamf HR, Critchley C (1999) Photoregulation and pho-
todamage in Schefflera arboricola leaves adapted to different light environments. Aust J Plant
Physiol 26:485–494
Schindler C, Lichtenthaler HK (1996) Photosynthetic CO2 -assimilation, chlorophyll fluorescence
and zeaxanthin accumulation in field grown maple trees in the course of a sunny and a cloudy
day. J Plant Physiol 148:399–412
Scholes JD, Press MC, Zipperlen SW (1997) Differences in light energy utilization and dissipation
between dipterocarp forest tree seedlings. Oecologia 109:41–48
Schreiber U, Bilger W (1987) Rapid assessment of stress effects on plant leaves by chlorophyll
fluorescence measurements. In: Tenhunen JD, Catarino FM, Lange OL, Oechel WC (eds)
Plant responses to stress. Functional analysis in Mediterranean ecosystems. NATO-ASI-Series
G, Ecological sciences, vol 15. Springer, Berlin Heidelberg New York, pp 27–53
Schreiber U, Bilger W (1993) Progress in chlorophyll fluorescence research: major developments
during the past years in retrospect. Prog Bot 54:151–173
Sims DA, Pearcy RW (1994) Scaling sun and shade photosynthetic acclimation of Alocasia macr-
orrhiza to whole-plant performance. – I. Carbon balance and allocation at different daily pho-
ton flux densities. Plant Cell Environ 17:881–887
Sims DA, Gebauer RLE, Pearcy RW (1994) Scaling sun and shade photosynthetic acclimation
of Alocasia macrorrhiza to whole-plant performance. – II. Simulation of carbon balance and
growth at different photon flux densities. Plant Cell Environ 17:889–900
Smith H, Samson G, Fork DC (1993) Photosynthetic acclimation to shade:probing the role of
phytochromes using photomorphogenic mutants of tomato. Plant Cell Environ 16:929–937
Stegemann J, Timm H-C, Küppers M (1999) Stimulation of photosynthetic plasticity in response
to high fluctuating light: an empirical model integrating dynamic photosynthetic induction and
capacity. Trees 14:145–160
Stephens GR, Waggoner PE (1970) Carbon dioxide exchange of a tropical rainforest. Part I. Bio-
Science 20:1050–1053
Strauss-Debenedetti S, Bazzaz FA (1991) Plasticity and acclimation to light in tropical Moraceae
of different successional positions. Oecologia 87:377–387
Thiele A, Winter K, Krause GH (1997) Low inactivation of D1 protein of photosystem II in young
canopy leaves of Anacardium excelsum under high-light stress. J Plant Physiol 151:286–292
Thiele A, Krause GH, Winter K (1998) In situ study of photoinhibition of photosynthesis and
xanthophyll cycle activity in plants growing in natural gaps of the tropical forest. Aust J Plant
Physiol 25:189–195
Timm H-C, Stegemann J, Küppers M (2002) Photosynthetic induction strongly affects the light
compensation point of net photosynthesis and coincidentally the apparent quantum yield.
Trees 16:47–62
Timm H-C, Küppers M, Stegemann J (2004) Non-destructive analysis of architectural expansion
and assimilate allocation in different tropical tree saplings: consequences of using steady-state
and dynamic photosynthesis models. Ecotropica 10:101–121
Tinoco-Ojanguren C, Pearcy RW (1992) Dynamic stomatal behaviour and its role in carbon gain
during lightflecks of a gap phase and an understory Piper species acclimated to high and low
light. Oecologia 92:222–228
Tinoco-Ojanguren C, Pearcy RW (1993a) Stomatal dynamics and its importance to carbon gain in
two rainforest Piper species. I. VPD effects on the transient stomatal response to light flecks.
Tinoco-Ojanguren C, Pearcy RW (1993b) Stomatal dynamics and its importance to carbon gain in
two rainforest Piper species. II. Stomatal versus biochemical limitations during photosynthetic
induction. Oecologia 94:395–402
Tóth VR, Mészáros I, Veres S, Nagy J (2002) Effects of the available nitrogen on the photosynthetic
activity and xanthophyll cycle pool of maize in the field. J Plant Physiol 159:627–634
Turnball MH (1991) The effect of light quantity and quality during development on the photosyn-
thetic characteristics of six Australian rainforest tree species. Oecologia 87:110–117
Tyystjärvi E, Aro E-M (1996) The rate constant of photoinhibition measured in lincomycin-treated
leaves is directly proportional to light intensity. Proc Nat Acad Sci USA 93:2213–2218
Valladares F, Allen MT, Pearcy RW (1997) Photosynthetic responses to dynamic light under
field conditions in six tropical rainforest shrubs occurring along a light gradient. Oecologia
111:505–514
Vázquez-Yanes C, Orozco-Segovia A (1993) Patterns of seed longevity and germination in the
tropical rainforest. Annu Rev Ecol Syst 24:69–87
Vogelmann TC, Bornman JF, Yates DJ (1996) Focussing of light by leaf epidermal cells. Physiol
Plant 98:43–56
Warren CR, Adams MA, Chen ZL (2000) Is photosynthesis related to concentration of nitrogen
and Rubisco in leaves of Australian native plants? Aust J Plant Physiol 27:407–416
Watling JR, Robinson SA, Woodrow IE, Osmond CB (1997) Responses of rainforest understory
plants to excess light during sunflecks. Aust J Plant Physiol 24:17–25
Yamashita N, Koike N, Ishida A (2002) Leaf ontogenetic dependence of light acclimation in inva-
sive and native subtropical trees of different successional status. Plant Cell Environ 25:1341–
1356
Zipperlen SW, Press MC (1996) Photosynthesis in relation to growth and seedling ecology of two
dipterocarp rain forest tree species. J Ecol 84:863–876
Zipperlen SW, Press MC (1997) Photosynthetic induction and stomatal oscillations in relation to
the light environment of two dipterocarp rain forest tree species. J Ecol 85:491–503
Chapter 5
Tropical Forests.
III. Ecophysiological Responses to Drought
5.1 Drought in Moist Tropical Forests
Seasonality of rainfall (Sect. 3.1) can lead to the formation of tree rings in wet
tropical forests (Worbes 1999; Dünisch et al. 2003). Seasonal drought may occur
regularly in these moist forests and not only in dry tropical forests. In Central and
South America it can be enforced by southern oscillation or El Niño events (En-
gelbrecht et al. 2002). After extreme dry periods tropical rain forests may even be
threatened by fire (van Nieuwstadt and Sheil 2005, Sect. 10.3). Along the 65 km
across the isthmus of Panamá, where B. E NGELBRECHT and colleagues have stud-
ied tree seedling performance in relation to drought stress in the moist forests there
is a gradient of forest formation due to a moisture gradient from the drier Pacific
to the wetter Atlantic side (Fig. 5.1A). The number of days when no precipitation
reaches the forest floor (days with lower than 3 mm rain) range from 20 to more
than 90 in the moist forests across the isthmus (Fig. 5.1B).
Responses of tree seedlings were studied in detail because they were expected
to be the most drought sensitive life stage of trees in the moist forests due to their
shallow roots and limited access to soil water. However, it was found that once
established seedlings were remarkably tolerant to severe drought events and desic-
cation and wilting (Engelbrecht et al. 2002; Tyree et al. 2002). Three cases could be
distinguished where effects of drought on survival of seedlings and growth were not
correlated:
• drought had little effect on both growth and survival in well adapted species,
• drought had little effect on survival but a strong effect on growth in species which
shed their leaves under dry conditions and are facultatively deciduous,
• drought strongly affected survival in species where the surviving individuals
show an intermediate growth reduction
(Engelbrecht and Kursar 2003). Among the factors of environmental pressure of
light, herbivory and pathogens soil moisture proved to be the main one determin-
ing seedling survival or mortality (Engelbrecht and Kursar 2003; Engelbrecht et
150 5 Tropical Forests. III. Ecophysiological Responses to Drought
Fig. 5.1A, B Rainfall gradient 4000

across the isthmus of Panamá. A
A Annual rainfall from the
drier Pacific (0 km) to the
Rainfall (mm )
wetter Atlantic (60 km) side 3000
of the isthmus. B Number
of times (y-axis) in 27 years
when years with 20 to about
90 days occurred where the 2000
daily precipitation was below
3 mm (x-axis) so that due to
interception by the canopy
and evaporation no rain water 1000
0 10 20 30 40 50 60
reached the ground. (A: by
Distance from Pacific ( km )
courtesy of B. Engelbrecht,
B: Tyree et al. 2002, by per-
Number of times event has occurred
mission of Oxford University 100

B
Press)
10
1
20 40 60 80 100
Number of days without rain reaching understory
Fig. 5.2 Relation of drought

resistance to lethal leaf water
potential in wilted leaves.
Low and high lethal leaf wa-
ter potential are related to low
and high mortality, respec-
tively. (Figure by courtesy
of B. Engelbrecht and T.A.
Kursar; for details see Tyree
et al. 2003)
al. 2005). Differences of species in their responses to drought could be basically

explained by tolerance of wilting and desiccation and there was a linear relation
between lethal leaf water potential and drought resistance (Fig. 5.2; Tyree et al.
2003). Thus, via the effects on the seedlings drought has an important impact on
5.2 Drought in Dry Tropical Forests 151
plant population dynamics (Engelbrecht and Kursar 2003) and species distribution
(Engelbrecht et al. 2005) in the forest types along the isthmus of Panamá.
Species may react to drought stress by adaptive reduction of leaf area (Phillips
et al. 2001; Sobrado 2003). Stomatal regulation plays a role. Among 110 species of
mature trees in a 12-year-old tropical rainforest mesocosm model, Rascher et al.
(2004) found functional diversity with different reactions of different species, with
leaf fall, reduced maximum rates of photosynthetic electron flow and stomatal clo-
sure. Tropical evergreen rainforest species show isohydric regulation where stom-
ata are sensitive to vapour pressure deficit and respond to short term water stress
(Cunningham 2004) and stomatal responses to humidity are based on sensing the
transpiration rate itself (Meinzer et al. 1993).
5.2 Drought in Dry Tropical Forests
“From wet to dry” species in the tropics face increasing problems of water availabil-
ity (Holbrook and Franco 2005). The dominating ecophysiological stress parameters
in dry tropical forests are strong seasonal drought (H2 O) and high irradiance (hν,
see Chapter 4 for a detailed discussion) with strong interactions between them and
with other environmental parameters, i.e.
1. hν ↔ T : Absorption of radiation by leaves leads to heating.
2. H2 O ↔ hν: Heating and drying of the atmosphere, increases the leaf-air water-
vapour pressure gradient and thus leads to increased transpirational water loss.
3. H2 O ↔ T : Water loss can be controlled by closure of stomata, but this then
reduces transpirational cooling by evaporation, and leaves heat up further.
4. H2 O ↔N: Soil water deficit reduces the availability of N; the transpiration stream
serves distribution of nutrients in the plant.
In their adaptations plants combine phenological, structural, physiological and bio-
chemical responses.
5.2.1 Leaf Shedding and Hydraulic Architecture
In dry tropical forests some trees may have access to deep water sources and can
maintain their water-use during drought. These trees do not have a great seasonal
variation in their leaf fall (Meinzer et al. 1999). In other trees leaf shedding is an
avoidance strategy in terms of the biological stress concept and like for savannas
(Sect. 10.1.2.1) phenological cycles play an important role. Many dry tropical forest
trees shed leaves at the onset of the dry season. A reduction of hydraulic conduc-
tance of the leaves precedes senescence and possibly causes senescence (Sobrado
1993; Brodribb and Holbrook 2003). Trees may flush new leaves before the onset
of the rainy season protecting the young growth from herbivory (see Sect. 3.4.4.3)
by insects which emerge with the rains (Murali and Sukumar 1993). However, this
is an overall somewhat simplified picture. In fact, there is a large range of phe-
nological behaviours. A co-occurrence of tree species with phenological patterns
ranging from deciduous to completely evergreen is often observed, and one can find
species that are leafless for as much as 6 months standing near to ones that retain
full crown foliage despite the near absence of rainfall for many months (Brodribb
and Holbrook 2005; Diaz and Grandillo 2005). Leaves may be shed throughout the
year and trees shed and flush leaves almost every month in a pattern associated with
sporadic rainfall events (Diaz and Grandillo 2005). The phenological variability of
leafless and fully leaved trees occurring side by side is not necessarily due to differ-
ent rooting depths but can be explained by differences in the capability to use small
episodic rainfall events especially with water wetting the leaf surfaces rather than
coming from the soil (Diaz and Grandillo 2005).
Hydraulic architecture needs to be adapted. By shedding leaves drought-decidu-
ous species avoid significant plant water loss during the driest and hottest months
but they must cope with larger seasonal water potential fluctuations in their leaves
and require a higher water transport efficiency which leads to seasonal occurrence of
xylem embolisms (Sobrado 1993). The principle differences seen when deciduous
dry rainforest trees are compared with evergreen ones are the following:
• deciduous species have larger xylem vessel diameters,
• therefore, deciduous species have lower wood density,
• deciduous trees are more vulnerable to xylem vessel cavitation or embolism
blocking water transport,
• deciduous species have a higher water storage capacity.
These correlations can be found to be borne out in nature (Sobrado 1993; Choat et al.
2005). However, on the other hand there is also an overwhelming functional diver-
sity and variation in hydraulic strategies, such as stem and leaf specific conductance
and vulnerability to embolism, among dry forest species co-occurring side by side
(Brodribb et al. 2002, 2003). In some deciduous species drought-induced embolism
is avoided prior to leaf shedding, whereas in others leaf shedding and xylem em-
bolism are closely linked (Brodribb et al. 2002), where leaf hydraulic conductance
and water potential are not correlated with leaf life spans (Brodribb and Holbrook
2005). Xylem cavitation of leaf veins can elicit a feed forward signal to stomata
causing stomatal closing responses that may lead to depression of gas exchange
(Brodribb and Holbrook 2004) and are one of the reasons for a midday-depression
(see Sects. 5.2.2.1 and 10.1.2.3).
Among various other physiological reactions a high emission of isoprene gas
has been noted, which may be linearly related to irradiance up to 2,500 µ mol m−2 s−1
and constitute considerable parts of the entire carbon budget of the plants (Lerdau
and Keller 1997). It is under the control of the endogenous circadian biological clock
(Wilkinson et al. 2006). As a hydrophobic gas this has been discussed in relation to
reduction of evapotranspiratory water loss. However, isoprene my also function as
an antioxidant (Peñuelas et al. 2005; Affek and Yakir 2002) protecting against ex-
cessive light and damage by heat and reducing oxidative stress (Sect. 4.1.2) (Sharkey
and Yeh 2001).
In nutrient poor habitats of dry forests low soil water may amplify nutritional
problems because more than actual nutrient availability water may control uptake
of nutrients (Rentería et al. 2005). Before leaves are shed nutrients such as nitro-
gen are remobilized in the senescing leaves and stored in twigs, the N-resources of
which may provide some N for later reconstruction of the canopy (Sobrado 1995).
RuBISCO activity may be reduced by water stress (Parry et al. 2002).
5.2.2 Ecophysiological Responses of Plants with

C3 -Photosynthesis and Crassulacean Acid Metabolism
(CAM)
In their reactions to the interacting stress effects of drought and high irradiance the
plants must optimise responses to the particular limitations given. This may lead to
disadvantages; for example closing stomata at low water availability and high irradi-
ance reduces water loss but also causes increased heating; CO2 -uptake is prevented
and this leads to the dangers of photoinhibition (Sect. 4.1.7). C3 - and CAM-plants
may respond by stomatal closure in the middle of the days when challenged by high
irradiance, heating and transpiratory loss of water peaks, but this has very different
implications in both modes of photosynthesis.
5.2.2.1 The Midday Depression of C3 -Plants
The C3 -bromeliad Pitcairnia integrifolia grows in the thornbush-forest of Trinidad

and smaller adjacent islands. Its performance on a clear and very hot day demon-
strates the implications of the strategy of midday stomatal closure in C3 -photosyn-
thesis (Lüttge et al. 1986; Fig. 5.3). Photosynthetic CO2 -uptake rose after dawn as
light-intensity increased and reached the highest rate at about 09.00 h. During this
time temperature increased from about 23 ◦ C to about 36 ◦ C, but leaf temperature
remained very close to air temperature. Beyond that point stomata began to shut
and had fully closed by noon, when leaf conductivity to water vapour, gH2 O , was
zero (not shown in Fig. 5.3). At this time, and until about 15.00 h irradiance had
attained its highest level around 2,000 µ mol photons m−2 s−1 and leaf temperature
now increased much above air temperature with the highest value close to 52 ◦ C
and almost 8 ◦ C higher than air temperature. If inhibition of CO2 -uptake was only
due to stomatal closure, one would have expected intercellular CO2 -concentration
i
( pCO 2
) to have remained at low levels during this period. However, pCO
i
2
rose and
this shows that there were likely to be photoinhibitory responses occurring as well
as the well documented change in carboxylation efficiency at this time. Later in the
afternoon, when irradiance and temperatures declined again, stomata re-opened and
i
pCO dropped, but CO2 -uptake only reached less than a third of the rate attained
2
in the morning. Hence, strain during the hottest time of the day was only partially
elastic and had a strong plastic component. Only during the subsequent night water
Fig. 5.3 Daily course of CO2 uptake (JCO2 ), intercellular CO2 concentration ( pCOi ), leaf temper-
2
ature (TL ), air temperature (TA ) and solar irradiance (L) for a plant of Pitcairnia integrifolia in
Trinidad and photograph of a plant with the head of the porometer attached, which was used for
measurements. (Lüttge et al. 1986)
uptake and rehydration as well as possible repair mechanisms may restore photo-
synthetic capacity.
The phenomenon of reduced gas exchange during the hottest time of the day is
called midday-depression. It is very frequent among trees, shrubs and herbs in hot
and arid regions (Schulze et al. 1974, 1975a,b; Tenhunen et al. 1980, 1981, 1984;
Pathre et al. 1998) and also frequently observed among trees in savannas and cer-
rados (Sect. 10.1.2.3, Figs. 10.10 – 10.13). The midday-depression may be smaller
or larger. Gas exchange may be totally absent during this time by full stomatal clo-
sure. Moreover, recovery in the afternoon, as shown for example in Fig. 5.3, may be
expressed to different extents and with increasing drought it may not occur at all.
Usually nocturnal rehydration may provide more effective recovery but this as well
will be reduced as drought becomes increasingly severe.
In addition to mechanisms of stress tolerance, there are also means of stress
avoidance. P. integrifolia for example may roll its leaves, exposing only the lower
abaxial surface to the sun. This surface is densely covered by silvery trichomes.
Bromeliad trichomes have evolved for absorption of water and nutrients (Sect. 6.4).
The trichomes of P. integrifolia are non-absorbent and composed of dead cells which
effectively reflect the light. Compared to white paper (100% reflectance) the re-
flectance of the abaxial leaf surface with scales was found to be 46.5% but only
19.8% when the scales were removed. As an alternative role of the non-absorbent
bromeliad trichomes functioning as a water repellent has been discussed rather than
reflection of excessive light and reduction of photoinhibition (Pierce et al. 2001).
5.2.2.2 CAM: Escape from the Dilemma Desiccation or Starvation
Choosing between limiting the effects of any one stress represents a daily “damage
limitation exercise” such that plants with C3 -photosynthesis face the dilemma of
desiccation or starvation, when under water stressed conditions. With the midday-
depression, the strategy is to try to avoid desiccation by stomatal closure at the
expense of CO2 -supply for photosynthesis. Desiccation is always more rapid and
is the more immediate danger than starvation. One escape from this dilemma is
provided by the evolution of crassulacean acid metabolism (CAM) (Box 5.1),
where CO2 is fixed during the night, when water-vapour pressure saturation deficit
of the atmosphere is much lower than during the day, and hence stomatal opening
has a smaller effect on the water budget of the plants. The CO2 fixed is stored in
chemical form of organic acids mainly as malic acid, remobilized again during the
day and made available for photosynthesis, so that the plants can utilize the light
energy of solar irradiance for CO2 -assimilation behind closed stomata.
Fig. 5.4 Phylogenetic tree of plant families with Crassulacean acid metabolism
This mode of photosynthesis was first discovered in plants of the genus Kalan-
choë (see Sect. 2.5), which belong to the family of the Crassulaceae, and hence the
name. However, it has evolved independently several times, i.e. polyphyletically,
since there are CAM-performing taxa on almost all branches of the phylogenetic tree
of vascular plants (Fig. 5.4). Among the plants of the thornbush-succulent forests
many are CAM-plants, i.e. the cacti in the new world (Figs. 3.11B and 3.13), the
succulent Euphorbiaceae in the old world, the Didieraceae (Fig. 3.12) and many of
the rosette plants in the Bromeliaceae (which may cover the whole floor of neotrop-
ical dry forests like the rosettes of Bromelia humilis in Fig. 3.10A), Agavaceae and
Liliaceae, to name the major ones.
However, CAM may not only operate in the simple day-night fashion described
above. In fact it provides an enormous range of plasticity in form and function, al-
lowing responses to environmental conditions to be optimised (see also Sect. 2.5).
The best way of describing these options is by reference to the four phases of
CAM according to the nomenclature introduced by Osmond (1978; see Box 5.1).
Phase I represents nocturnal stomatal opening with CO2 -uptake, fixation and stor-
age as malic acid, whereas during phase III daytime stomatal closure with CO2 -
remobilization and assimilation occurs. Phases II and IV are transitional phases in
the early morning and in the afternoon. Phase IV often plays an important role, be-
cause when CAM plants are well watered it may be quite extensive. Then CAM
plants take up CO2 directly from the atmosphere and assimilate it directly by the
C3 -mode of photosynthesis via RuBISCO. This can make a major contribution to
their productivity.
Conversely, water stress may become so severe that even CAM plants face the
dilemma of desiccation or starvation. Then, stomata may be closed even during
the night, and CAM represents an option for survival by recycling CO2 internally.
The CO2 evolved nocturnally during respiratory metabolism is refixed and stored
as malic acid; the day-time remobilization and reassimilation, using solar radiation,
recycles carbohydrate reserves for the subsequent night (Box 5.1). Under severe
drought stress cacti, for instance, keep stomata closed continuously for many months
(see also Sect. 8.2.3.2.1). By CO2 -recycling they do not gain carbon, but very little is
lost and solar energy can be used to maintain metabolism and remain competent un-
til water is available again. At the same time, with totally closed stomata, the plants
lose only a little water via cuticular transpiration. Water storage tissues in cacti and
other succulents also provide reserves and help to overcome drought periods.
In a drought deciduous forest in western Mexico, Lerdau et al. (1992) studied the
performance of the arborescent cactus Opuntia excelsea. In the dry season, when
trees had shed their leaves, the cactus had a competitive advantage, as there was
no light limitation. However, a factor associated with plant size, possibly water sta-
tus, limited carbon gain during the dry season. Larger individuals were able to uti-
lize water stored in their trunks and main branches (see also Sect. 8.2.3.2.1). Light
availability in the forest understorey constrained CO2 -assimilation of the cactus in
the wet season.
Daytime CO2 -remobilization from nocturnally stored organic acids behind closed
stomata also participates in controlling photoinhibition which would be amplified
Box 5.1 Crassulacean acid metabolism (CAM)

In CAM plants there are two ways of primary CO2 fixation, namely via the
enzymes phosphoenolpruvate-carboxylase (PEPC) and ribulosebisphosphate-
carboxylase oxygenase (RuBISCO). In its typical performance CAM has four
phases (Osmond 1978):
• Phase I:
Nocturnal dark fixation of CO2 via PEPC generating malic acid, which
is translocated into the vacuole by proton pumps (H+ -ATPase and H+ -
pyrophosphatase – PPi ase – transporting protons) and an inward rectifying
malate anion channel (transporting malate2− ) at the tonoplast.
• Phase II:
A transition phase in the early morning, after light energy becomes available,
with primary CO2 fixation partially via PEPC and RuBISCO, respectively.
• Phase III:
Efflux of non-dissociated malic acid from the vacuole, malate decarboxyla-
tion and refixation of the CO2 via RuBISCO behind closed stomata.
• Phase IV:
Opening of stomata in the afternoon, when nocturnally accumulated malic
acid is consumed, and primary CO2 fixation via RuBISCO.
CAM may play a role as a water-conserving mechanism at different levels of
drought stress.
• In the typical performance dominating nocturnal CO2 uptake reduces tran-
spirational loss of water related to CO2 acquired and thus increases water-use
efficiency, because the evaporative demand on leaves with open stomata is
smaller in the dark than in the light.
• At increased drought stress first phase IV and then also phase II are elimi-
nated, and stomata remain closed for the whole light period, further restrict-
ing transpirational loss of water.
• At still more severe drought, stomata may also be partially or totally closed
during the dark period. In this situation the CO2 fixed nocturnally for the ac-
cumulation of malic acid partially or totally may come from internal sources,
i.e. mainly respiration (CO2 recycling). This further reduces transpirational
loss of water but also limits carbon acquisition.
The scheme of CAM () shows the key reactions in metabolism. With PYR
pyruvate; PEP phosphoenolpyruvate; OAA oxaloacetate; MAL malate; Pi inor-
ganic phosphate; [CH 2 O] carbohydrate and transport across the tonoplast: with
MC malate transporter; the H+ -ATPase and H+ -PPi ase, and passive malic acid
efflux.
Net CO2 exchange by the CAM-plant Kalanchoë daigremontiana () with
increasing drought stress: ◦—◦ well-watered; +—+ low and •—• high drought
stress. Phases I to IV are indicated. Phase II and IV CO2 exchange is expressed
only in the well-watered plant; onset of phase I CO2 exchange is delayed in the
severely stressed plant (Smith and Lüttge 1985).
Box 5.1 (Continued)

when internal CO2 -levels at high irradiance were low as in the midday depression of
C3 -plants (Sect. 5.2.2.1) (Osmond 1982; Adams and Osmond 1988; Griffiths 1989).
i
In fact, internal CO2 -levels ( pCO ) behind closed stomata during the light period of
2
CAM may be very high and reach up to a few percent (Cockburn et al. 1979; Kluge
et al. 1981; see Lüttge 1987, 2002). However, in correlation with the internal CO2 -
concentrating mechanism of organic acid remobilization from the vacuoles and the
Fig. 5.5 Chlorophyll-

fluorescence variables (see
Box 4.6 for explanation)
in Clusia minor in the C3
state (–·–·–) and in the CAM
state (– – – phase II, · · · · ·
phase IV, —— phase III).
(Haag-Kerwer 1994)
related high rates of CO2 -reduction, high internal oxygen concentrations also build
up, i.e. close to 40% or twice the atmospheric O2 -concentration (Lüttge 2002). Thus,
other protective mechanisms of energy dissipation (Sects. 4.1.4 and 4.1.6) must also
be active.
Experiments measuring chlorophyll fluorescence in the neotropical facultative
CAM-tree Clusia minor, which can perform both CAM and C3 -photosynthesis
(Sect. 6.6.2.3), have shown that photoinhibition, if it occurs, is most likely to be
observed during phase IV of CAM, when stomata are open and plants fix CO2 via
RuBISCO rather than in phase III. Light response characteristics of chlorophyll-
fluorescence variables (see Box 4.6) in phases II and IV were similar to those ob-
served with C. minor in the C3 -state and very different to those of phase III of CAM
(Haag-Kerwer 1994; Fig. 5.5).
When nocturnal accumulation of malic acid occurs from recycled CO2 alone
(= 100% recycling), this is an extreme case. However, stomata may only be partially
closed during the night and malic acid accumulation may be due to both recycled
CO2 and CO2 -uptake from the atmosphere. Since the stoichiometry of CO2 -fixed to
malic acid formed is unity, recycling can be calculated in absolute terms as
malic acid accumulated minus CO2 taken up
or in relative terms (% recycling) as

malic acid accumulated minus CO2 taken up
× 100 .
malic acid accumulated
The degree of recycling may then depend on the severity of drought stress. This
is illustrated in Fig. 5.6 by a study of Aechmea aquilega and its higher altitude coun-
Fig. 5.6 Net nocturnal CO2 uptake from the atmosphere and internal CO2 recycling of Aechmea
(A. aquilega at the three lower altitudes and A. fendleri at the highest altitude) in relation to altitude
and precipitation in Trinidad. (After data of Griffiths et al. 1986)
terpart Aechmea fendleri along a gradient of altitude and precipitation in Trinidad.

A. aquilega grows both terrestrially and epiphytically from very dry deciduous
thornbush-forests to quite wet forests, and A. fendleri is epiphytic in wet forests.
Figure 5.6 shows that with increasing altitude and precipitation total CO2 -uptake by
the Aechmeas increased and relative CO2 -recycling decreased considerably.
References
Adams WW, Osmond CB (1988) Internal CO2 supply during photosynthesis of sun and shade
grown CAM plants in relation to photoinhibition. Plant Physiol 86:117–123
Affek HP, Yakir D (2002) Protection by isoprene against singlet oxygen in leaves. Plant Physiol
129:269–277
Brodribb TJ, Holbrook NM (2003) Changes in leaf hydraulic conductance during leaf shedding in
seasonally dry tropical forest. New Phytol 158:295–303
Brodribb TJ, Holbrook NM (2004) Diurnal depression of leaf hydraulic conductance in a tropical
tree species. Plant Cell Environ 27:820–827
Brodribb TJ, Holbrook NM (2005) Leaf physiology does not predict leaf habit; examples from
tropical dry forest. Trees 19:290–295
Brodribb TJ, Holbrook NM, Gutiérrez MV (2002) Hydraulic and photosynthetic co-ordination in
seasonally dry tropical forest trees. Plant Cell Environ 25:1435–1444
Brodribb TJ, Holbrook NM, Edwards EJ, Gutiérrez MV (2003) Relations between stomatal clo-
sure, leaf turgor and xylem vulnerability in eight tropical dry forest trees. Plant Cell Environ
26:443–450
Choat B, Ball MC, Luly JG, Holtum JAM (2005) Hydraulic architecture of deciduous and ever-
green dry forest tree species from north-eastern Australia. Trees 19:305–311
Cockburn W, Ting IP, Sternberg LO (1979) Relationships between stomatal behaviour and the
internal carbon dioxide concentrations in crassulacean acid metabolism plants. Plant Physiol
63:1029–1032
Cunningham SC (2004) Stomatal sensitivity to vapour pressure deficit of temperate and tropical
evergreen rainforest trees of Australia. Trees 18:399–407
Diaz M, Granadillo E (2005) The significance of episodic rains for reproductive phenology and
productivity of trees in semiarid regions of north-western Venezuela. Trees 19:336–348
Dünisch O, Montóia VR, Bauch J (2003) Dendrochronological investigations on Swietenia macro-
phylla King and Cedrela odorata L. (Melicaceae) in the central Amazon. Trees 17:244–250
Engelbrecht BMJ, Kursar TA (2003) Comparative drought-resistance of seedlings of 28 species of
co-occurring tropical woody plants. Oecologia 136:383–393
Engelbrecht BMJ, Wright SJ, Steven D de (2002) Survival and ecophysiology of tree seedlings
during El Niño drought in a tropical moist forest in Panama. J Trop Ecol 18:569–579
Engelbrecht BMJ, Kursar TA, Tyree MT (2005) Drought effects on seedling survival in a tropical
moist forest. Trees 19:312–321
Griffiths H (1989) Carbon dioxide concentrating mechanisms and the evolution of CAM in vas-
cular epiphytes. In: Lüttge U (ed) Vascular plants as epiphytes: evolution and ecophysiology.
Ecological studies, vol 76. Springer, Berlin Heidelberg New York, pp 42–86
Griffiths H, Lüttge U, Stimmel K-H, Crook CE, Griffiths NM, Smith JAC (1986) Comparative
ecophysiology of CAM and C3 bromeliads. III. Environmental influences on CO2 assimilation
and transpiration. Plant Cell Environ 9:385–393
Haag-Kerwer A (1994) Photosynthetische Plastizität bei Clusia und Oedematopus. Dr. rer.-nat.-
Thesis, Darmstadt)
Holbrook NM, Franco AC (2005) From wet to dry: tropical trees in relation to water availability.
Trees 19:280–281
References 163
Kluge M, Böhlke C, Queiroz O (1981) Crassulacean acid metabolism (CAM) in Kalanchoë.

Changes in intracellular CO2 concentration during continuous light or darkness. Planta
152:87–92
Lerdau MT, Keller M (1997) Controls on isoprene emission from trees in a subtropical dry forest.
Lerdau MT, Holbrook NM, Mooney HA, Rich PM, Whitbeck JL (1992) Seasonal patterns of acid
fluctuations and resource storage in the arborescent cactus Opuntia excelsea in relation to light
availability and size. Oecologia 92:166–171
Lüttge U (1987) Carbon dioxide and water demand: crassulacean acid metabolism (CAM) a ver-
satile ecological adaptation exemplifying the need for integration in ecophysiological work.
New Phytol 106:593–629
Lüttge U (2002) CO2 -concentrating: consequences in crassulacean acid metabolism. J Exp Bot
53:2131–2142
Lüttge U, Klauke B, Griffiths H, Smith JAC, Stimmel K-H (1986) Comparative ecophysiology of
CAM and C3 bromeliads. V. Gas exchange and leaf structure of the C3 bromeliad Pitcairnia
integrifolia. Plant Cell Environ 9:411–419
Meinzer FC, Goldstein G, Holbrook NM, Jackson P, Cavellier J (1993) Stomatal and environmental
control of transpiration in a lowland tropical forest tree. Plant Cell Environ 16:429–436
Meinzer FC, Andrade JL, Goldstein G, Holbrook NM, Cavelier J, Wright SJ (1999) Partitioning
of soil water among canopy trees in a seasonally dry tropical forest. Oecologia 121:293–301
Murali KS, Sukumar R (1993) Leaf flushing phenology and herbivory in a tropical dry deciduous
forest, southern India. Oecologia 94:114–119
Nieuwstadt MGL van, Sheil D (2005) Drought, fire and the survival in a Borneo rain forest. J Ecol
93:191–201
Osmond CB (1978) Crassulacean acid metabolism: a curiosity in context. Annu Rev Plant Physiol
29:379–414
Osmond CB (1982) Carbon cycling and stability of the photosynthetic apparatus in CAM. In: Ting
IP, Gibbs M (eds) Crassulacean acid metabolism. American Society of Plant Physiologists,
Rockville, pp 112–127
Parry MAJ, Andralojc PJ, Khan S, Lea PJ, Keys AJ (2002) Rubisco activity: effects of drought
stress. Ann Bot 89:833–839
Pathre U, Sinha AK, Shirke PA, Sane PV (1998) Factors determining the midday depression of
photosynthesis in trees under monsoon climate.Trees 12:472–481
Peñuelas J, Llusià J, Asensio D, Munné-Bosch S (2005) Linking isoprene with plant thermotoler-
ance, antioxidants and monoterpene emissions. Plant Cell Environ 28:278–286
Phillips N, Bond BJ, Ryan MG (2001) Gas exchange and hydraulic properties in the crowns of two
tree species in a Panamanian moist forest. Trees 15:123–130
Pierce S, Maxwell K, Griffiths H, Winter K (2001) Hydrophobic trichome layers and epicuticular
wax powders in Bromeliaceae. Am J Bot 88:1371–1389
Rascher U, Bobich EG, Lin GH, Walter A, Morris T, Naumann M, Nichol CJ, Pierce D, Bil
K, Kudeyarov V, Berry JA (2004) Functional diversity of photosynthesis during drought in
a model tropical rainforest – the contributions of leaf area, photosynthetic electron transport
and stomatal conductance to reduction in net ecosystem carbon exchange. Plant Cell Environ
27:1239–1256
Rentería LY, Jaramillo VJ, Martínez-Yrézar A, Pérez-Jiménez A (2005) Nitrogen and phosphorus
resorption in trees of a Mexican tropical dry forest. Trees 19:431–441
Schulze E-D, Lange OL, Evenari M, Kappen L, Buschbom U (1974) The role of air humidity
and leaf temperature in controlling stomatal resistance of Prunus armeniaca L. under desert
conditions. I. A simulation of the daily course of stomatal resistance. Oecologia 17:159–170
Schulze E-D, Lange OL, Evenari M, Kappen L, Buschbom U (1975a) The role of air humidity
conditions. III. The effect on water use efficiency. Oecologia 19:303–314
Schulze E-D, Lange OL, Kappen L, Evenari M, Buschbom U (1975b) The role of air humidity
conditions. II. The significance of leaf water status and internal carbon dioxide concentration.
Sharkey TD, Yeh SS (2001) Isoprene emission from plants. Annu Rev Plant Physiol Plant Mol
Biol 52:407–436
Smith JAC, Lüttge U (1985) Day-night changes in leaf water relations associated with the rhythm
of crassulacean acid metabolism in Kalanchoë daigremontiana. Planta 163:272–282
Sobrado MA (1993) Trade-off between water transport efficiency and leaf life-span in a tropical
forest. Oecologia 96:19–23
Sobrado MA (1995) Seasonal differences in nitrogen storage in deciduous and evergreen species
of a tropical dry forest. Biol Plant 37:291–295
Sobrado MA (2003) Hydraulic characteristics and leaf water use efficiency in trees from tropical
montane habitats. Trees 17:400–406
Tenhunen JD, Lange OL, Braun M, Meyer A, Lösch R, Pereira JS (1980) Midday stomatal clo-
sure in Arbutus unedo leaves in a natural macchia under simulated habitat conditions in an
environmental chamber. Oecologia 47:365–367
Tenhunen JD, Lange OL, Braun M (1981) Midday stomatal closure in mediterranean type scle-
rophylls under simulated habitat conditions in an environmental chamber. II. Effect of the
complex of leaf temperature and air humidity on gas exchange of Arbutus unedo and Quercus
ilex. Oecologia 50:5–11
Tenhunen JD, Lange OL, Gebel J, Beyschlag W, Weber JA (1984) Changes in photosynthetic ca-
pacity, carboxylation efficiency, and CO2 -compensation point associated with midday stom-
atal closure and midday depression of net CO2 exchange of leaves of Quercus suber. Planta
162:193–203
Tyree MT, Vargas G, Engelbrecht BMJ, Kursar TA (2002) Drought until death do us part: a case
study of the desiccation-tolerance of a tropical moist forest seedling-tree, Licania platypus
(Hemsl.) Fritsch. J Exp Bot 53:2239–2247
Tyree MT, Engelbrecht BMJ, Vargas G, Kursar TA (2003) Desiccation tolerance of five tropical
seedlings in Panama. Relationship to a field assessment of drought performance. Plant Physiol
132:1439–1447
Wilkinson MJ, Owen SM, Possell M, Hartwell J, Gould P, Hall A, Vickers C, Hewitt CN (2006)
Circadian control of isoprene emissions from oil palm (Elaeis guineensis). Plant J 47:960–968
Worbes M (1999) Annual growth rings, rainfall-dependent growth and long-term growth patterns
of tropical trees from the Caparo Forst Reserve in Venezuela. J Ecol 87:391–403
Chapter 6
Tropical Forests. IV. Lianas, Hemi-Epiphytes,
Epiphytes and Mistletoes
6.1 The Conquest of Space:

Cryptogams and a Diversity of Life Forms of Vascular Plants
Perhaps epiphytism could be thought to be primarily the utilization of any possi-

ble surface for holdfast and establishment, i.e. a conquest of space with epiphytes
found in aquatic and terrestrial habitats made up of various combinations of lower
and higher plants. In aquatic habitats, i.e. lakes, rivers and the sea, there are always
algae growing on each other. This not only applies to unicellular and filamentous
forms and their colonies, but also to macroalgae like kelp and red algae. In the mesic
terrestrial climate many lower plants are epiphytic, like mosses and lichens and
also some forms of small pleurococcoid aerial green algae as well as cyanobacteria
(blue green algae). In the tropics lower plants may constitute massive formations
of epiphytic biomass, e.g. the bryophytes the biomass of which increases with alti-
tude (Freiberg and Freiberg 2000) in upper montane cloud forests (“moss-forests”,
Fig. 6.1). Even the surfaces of leaves of plants in such forests may harbour a diverse
flora or phyllosphere with bacteria, cyanobacteria, fungi, green algae, bryophytes
and lichens and occasional seedlings of vascular plants (Ruinen 1961, 1974; Coley
et al. 1993; Freiberg 1998) (Fig. 6.2).
Life forms of vascular plants that evolved in the conquest of space are lianas,
hemi-epiphytes and the mistletoes, where the latter not only compete for space but
also combine epiphytism with parasitism. In the temperate zone we have climbers
and vines often especially in moist gallery forests, but if we exclude the parasites of
the mistletoes among the vascular plants the fern Polypodium vulgare is the only
known epiphyte, and moreover, it is only facultatively epiphytic. In contrast, the
popular view of tropical rainforests is determined by the image of an abundant flora
of epiphytes, vines and lianas, climbers with hanging and host-strangling shoots and
curtains of aerial roots (Fig. 6.3).
This image, although intuitively correct, needs to be carefully differentiated.
Growth of lianas, climbers and vines is particularly rich at the perimeter of forests,
along rivers, roads and around clearings. They are often light-demanding plants.
Epiphytes are often although not generally found to be more abundant in montane
166 6 Tropical Forests. IV. Lianas, Hemi-Epiphytes, Epiphytes and Mistletoes
Fig. 6.1 Upper montane cloud forest at 2,000 m a.s.l. (Sierra Maigualida 05◦ 30 N, 65◦ 15 W) with
epiphytic mosses, “moss forest”
rainforests and in cooler upper montane fog and cloud forests, where air moisture is
always high, than in the hot lowland rainforests (Freiberg and Freiberg 2000).
Remembering that vascular plants evolved from aquatic ancestors during the con-
quest of land and that then they were subject to many new kinds of stress with re-
spect to water and nutrient relations, it may not be surprising that there is very little
6.1 The Conquest of Space: Cryptogams and a Diversity of Life Forms of Vascular Plants 167
Fig. 6.2A, B Epiphylls. A On a leaf of Clusia sp., with cyanobacteria, algae, mosses and lichens
(Sierra Maigualida 05◦ 30 N, 65◦ 15 W). B On a leaf in a cloud forest above Lake Coté (Costa
Rica) with mosses and a higher-plant seedling
fossil record of epiphytism. Epiphytism must be a fairly recent development among

vascular plants. Most epiphyte diversity dates from the Pliocene-Pleistocene (Benz-
ing 1989a, 1990; Lüttge 1989). Against this background, when plants may even live
in the air, this may be considered a rather extreme case of the conquest of space. The
Fig. 6.3A–C Rainforests with epiphytes and lianas. A Plate from Martius’ Flora Brasiliensis (Mar-
tius 1840–1906). B A cloud forest above Lake Coté, Costa Rica. C Curtain of aerial roots of
a strangler fig (Ficus) (Queensland, Australia)
6.1 The Conquest of Space: Cryptogams and a Diversity of Life Forms of Vascular Plants 169
Fig. 6.4A, B Atmospheric bromeliads of Tillandsia flexuosa (A) and Tillandsia recurvata (B) on
telephone wires (Falcon State, Venezuela)
so-called atmospheric bromeliads constitute such life forms. They have given up
any contacts with substrates supplying water and nutrients other than from the at-
mosphere. They may hang down from the branches of phorophytes (see Fig. 6.15D
below) or may even get holdfast on wires of fences or telephone lines (Fig. 6.4).
Table 6.1 Taxonomic diversity of epiphytes. (Data from Kress 1989)

Number of taxa % epiphytes of total taxa
with vascular epiphytes of vascular plants
Species 23,466 10
Genera 879 7
Families 84 19
Orders 44 45
Evolution of epiphytism has clearly occurred many times and has been poly-
phyletic, since taxonomic diversity of epiphytes is quite substantial (Table 6.1).
The best-known families with epiphytes are headed by the Orchidaceae, although
the Araceae, Piperaceae and Bromeliaceae are also important, and epiphytic taxa
are abundant among ferns. In some tropical forests up to 50% of all leaf biomass
may be due to epiphytes, and of the known species of lianas 90% are native to
the tropics. The meagre fossil record of epiphytes (Mägdefrau 1956) does not offer
support for the discussion of evolution. The upper Devonian arborescent horesetail,
Pseudoborina ursina about 360 million years ago had a stem diameter of 0.12 m
and carried an epiphyte Codonophytum epiphytum. The phylogenetic relationships,
however, are uncertain. In Permian formations in Germany from about 260 million
years ago leafy organs resembling basket forming mantle leaves of extant epiphytic
ferns of the genera Platycerium and Polypodium are preserved. However, these ex-
tant fern genera are leptosporangiate, and such forms did not exist at that time. The
most original extant pteridophytes are two species each of the genera Psilotum and
Tmesipteris and all of them are epiphytes. They belong to the Psilotatae and are
often considered as relicts of evolution. Although they are very similar to the earli-
est land plants, i.e. the Psilophytatae, which first conquered land and died out again
in the middle Devonian 360 million years ago, the exact phylogenetic relationships
again are not clear. From the very poor palaeontological record alone one may read-
ily conclude that epiphytism of vascular plants is a very recent event in geological
history.
6.2 Cryptogams
6.2.1 Bacteria and Cyanobacteria
Microclimate and, foremost, the moisture and availability of liquid water is the
most important parameter determining the frequency and biodiversity of epiphylls
(Freiberg 1999). Bacteria may make the cuticle of leaves more permeable for water
as they secrete surfactants and extracellular enzymes (Schreiber et al. 2005). Cover-
age of the host leaves by mats of cyanobacteria can have adverse effects on light uti-
lization by the leaves. However, it has been suggested that nitrogen fixation by the
epiphyllic cyanobacteria may also supplement nutrients available to the host (Ru-
6.2 Cryptogams 171
inen 1965; Benzing 1990) or to the whole ecosystem (Freiberg 1998). Dinitrogen
(N2 ) reduction in the phyllosphere mainly depends on light and water. Maximum
rates measured are 5 nM N m−2 leaf surface s−1 (Freiberg 1998). In a premontane
tropical rainforest in Costa Rica N-supply by N2 -fixation in the phyllosphere was
calculated as 2 to 7 kg N ha−1 year−1 (Freiberg 1994) but the higher values (30 to
60 kg N ha−1 year−1 ) which have been presented in the literature for other areas are
discussed critically (see also Sect. 10.2.3.2.1).
6.2.2 Bryophytes and Lichens
As shown above (Sect. 6.1, Fig. 6.1) bryophytes and lichens may constitute a con-
siderable floristic diversity and biomass among the epiphytes in tropical rainforests,
especially in the cloud forests at higher elevations with their cooler nights (Seifriz
1924; Sipman 1989). Epiphyllous liverworts are involved in a reciprocal transfer
of nitrogen with their host leaves, where the amount of nitrogen obtained by the
epiphyllous liverworts from the host leaves varies between 1% and 57% of their en-
tire demand, and vice versa host leaves obtain mostly up to 2.5% of their nitrogen
demand from leachates of the epiphylls (Wanek and Pörtl 2005).
Green and Lange (1994) have provided a comparison of photosynthesis in
mosses and lichens. A major difference between the two groups is the effect of wa-
ter relations on photosynthesis. In mosses, the CO2 -exchange surface is external,
and the mosses have special water storage volumes, i.e. special cells – often dead
cells – and capillary structures, which are separated from the gas exchange areas.
Therefore bryophytes may constitute most important water stores in the epiphytic
habitat (Freiberg 1997). Lichens have an internal CO2 -exchange surface with the
phycobionts embedded in a relatively compact fungal tissue, and any water storage
will tend to hinder gas exchange either within the compact tissue or at the outer
lichen surface. Therefore, as compared to mosses, lichens tend to have lower max-
imal water content on a dry weight basis and there is the risk that high thallus wa-
ter content impairs CO2 -uptake and assimilation. This difference between the two
groups possibly explains the particular dominance of bryophytes in very wet habi-
tats, while lichens can also be very successful in dry habitats. Bryophytes are also
well adapted to the large variation of temperatures in their montane, cloud and elfin
forest habitats. They show constitutive temperature resistance to both rather cool
temperatures often encountered in these habitats and quite hot temperatures that
may occur during light fleck events (Lösch and Mülders 2000).
Floristics, taxonomy and habitat occupation by tropical lichens has been well
studied (Galloway 1991). Less work is available on their ecophysiology. However,
ecophysiological studies have investigated lichens in the temperate rainforests, e.g.
in New Zealand (Green and Lange 1991; Green et al. 1991; Lange et al. 1993) and
also in forests of the wet tropics (Lange et al. 1994; Zotz and Winter 1994a). Lange
et al. (1994) have studied the gas exchange, water relations and potential productiv-
ity of the cyanobacterial basidiolichen Dictyonema glabratum living epiphytically,
saxicolously and terrestrially in a lower montane tropical rainforest in Panamá with

an annual precipitation of 4,000 – 4,500 mm and a mean annual temperature of 21 –
22 ◦ C (Fig. 6.5). This lichen occurs both in shaded and exposed sites, and it is quite
frequent in this forest. It has a rather unusual ecophyiological behaviour with respect
to water saturation of its thallus and a number of additional traits, which explain its
high productivity in the habitat.
Normally photosynthetic net CO2 -uptake is impaired in lichen thalli including
lichens in tropical forests (Lange et al. 2000, 2004) when they are oversaturated
with water due to diffusion limitation from surface water or blocked air channels in
the mycelium (e.g. see Sect. 11.2.2 and Fig. 11.22). However, not in all lichens wa-
ter saturation inhibits CO2 -diffusion. The reasons are not known. It is not related to
a formation of secondary chemical lichen compounds and must be due to morpho-
logical features (Lange and Green 1997; Lange et al. 1997). D. glabratum maintains
maximum rates of net-CO2 uptake when it is fully hydrated up to a water content of
1,000% of its dry weight (Fig. 6.6). This allows maximum benefit from the heavy
rain storms occurring in the habitat. The lichen also possesses a mechanism for
concentrating internal inorganic carbon by energy dependent transport, which oc-
curs in many algae and also higher water plants (see Griffiths 1989; Badger et al.
1993). This mechanism allows photosynthesis at elevated intracellular CO2 -levels.
In corticulous crustose green algal lichens in the understorey of a lowland tropical
rainforest periods of thallus suprasaturation with water are reduced by the presence
Fig. 6.5 The lichen Dictyonema glabratum (syn. Cora pavonia) growing epiphytically on a 6-m-
tall bush in a lower montane tropical rainforest in Panama. (Photograph courtesy B. Büdel; see
Lange et al. 1994)
6.2 Cryptogams 173
Fig. 6.6 Net photosynthesis

of the lichen Dictyonema
glabratum in relation to water
content in percent of dry
weight. (Lange et al. 1994)
Fig. 6.7 A daily course of

net CO2 exchange and water
content of the lichen Dicty-
onema glabratum with air
temperature and light inten-
sity (PPFD) in its natural
habitat in a lower montane
rainforest in Panamá; mea-
sured on 23 September 1993
by Lange et al. (1994)
of water-repelling surface structures of the hyphae of the mycobiont as well as the

production of a hydrophobic fungal protein, hydrophobin (Lakatos et al. 2006).
Net CO2 -exchange on many days is typically bimodal with a peak in the morning,
when thalli are wetted from dew and early fog, and another peak after midday when
heavy showers may occur (Fig. 6.7). In between, the thallus may dry out and CO2 -
uptake ceases. In fact, drying out occurs for a few hours almost every day, and sim-
ilar to most lichens, D. glabratum is also desiccation tolerant (Sect. 11.4.2). How-
ever, unlike many other lichens it does not reactivate photosynthetic CO2 -fixation
immediately after rewetting following desiccation, but it needs a recovery period of
about 60 min. Thus it appears that lichens of the very moist lower montane rainfor-
est show subtle changes in rehydration and reactivation characteristics as compared
to lichens from temperate or arid habitats (Sect. 11.4.2).
Another important trait is the thermophily of D. glabratum. It has been noted
above that epiphytic mosses and lichens are particularly abundant in forests at higher
elevations. The better supply of water in the epiphytic habitat by dew and fog may
only be one reason for this distribution. An even more critical factor may be the
reduction of respiration at lower night temperatures because respiration is a critical
factor in the net productivity of lichens in the tropics where respiratory losses may
be substantial. Indeed, the cooler nights much reduce respiration and thus nocturnal
loss of carbon, which is a considerable factor in overall productivity, may be de-
cisive for the general preponderance of epiphytic mosses and lichens in cloud and
fog forests in the tropics. In D. glabratum respiration increases only slightly with
temperature up to 22 ◦ C, but then increases sharply with higher temperatures. Net
photosynthesis increases up to 22 ◦ C and then declines in parallel with increasing
respiration, so that gross photosynthesis calculated from net photosynthesis and res-
piration remains at a constant high level up to 40 ◦ C. The balance of net photosyn-
thesis is positive up to 35 ◦ C (Fig. 6.8). In fact the maximum rates of net CO2 -uptake
in D. glabratum are quite high even in comparison to sun plants among vascular epi-
phytes (see below: Table 6.6). On a thallus area basis the highest rate observed in the
field was 8 µ mol m−2 s−1 . Calculations have suggested that the annual relative pro-
duction under the habitat conditions of D. glabratum in Panamá is 2.28, i.e. a gain of
2.28 g of carbon per 1 g of initial thallus carbon. Thus, even with leaching of carbon
under the influence of regular heavy rain which is frequently observed in lichens
(Bruns-Strenge and Lange 1992), D. glabratum must retain sufficient surplus to al-
low rapid growth.
Fig. 6.8 Net photosynthesis,

dark respiration and calcu-
lated gross photosynthesis
of the lichen Dictyonema
glabratum in relation to tem-
perature at a light intensity
(PPFD) of 150 µ mol m−2 s−1
and at high thallus water
content. (Lange et al. 1994)
6.3 Lianas, Climbers, Vines and Hemi-Epiphytes 175
This behaviour of a tropical rainforest lichen is quite remarkable in comparison

to the slow growth otherwise observed among lichens. The observations of Lange
et al. (1994) put the biomass production of lichens and its ecological importance
in tropical fog and cloud forests in a new perspective. Other lichens may have sim-
ilar capacities for high productivity (Zotz et al. 1998; Lange et al. 2000, 2004).
Lichens with cyanobacteria as phycobionts are particularly frequent in the moist
tropics (Büdel et al. 2000). In the cyanobacterial lichens dinitrogen fixation is
an additional advantage. If we average maximum rates of net photosynthesis given
by Lange et al. (2000, 2004), a superior performance of cyanobaterial lichens com-
pared to green alga lichens is observed with rates in µmol m−2 s−1 of 5.3 ± 1.9 (11)
and 3.0 ± 1.1 (3), respectively (SD with number of measurements including differ-
ent species). In the balance of overall productivity, however, it must be noted that
maximum rates are only possible for short periods each day (Fig. 6.7) and respi-
ratory losses are significant (Fig. 6.8; Lange et al. 2000, 2004). In the deep shade
of the understorey of tropical rain forests lichens may also make effective use of
light-flecks (Lakatos et al. 2006; Sect. 4.2.1).
6.3 Lianas, Climbers, Vines and Hemi-Epiphytes
Lianas, climbers and vines are rooted in the soil and use other plants, especially
trees, as support for growth away from the ground (Holbrook and Putz 1996a). It is
mostly assumed that the particular advantage of this habit is to allow these plants
to escape from deep shade and to reach the upper canopy of forests. This implies
that their seeds would germinate in the shade and seedlings initially would grow
upwards and develop in the shade. In the tropics, lush growth of lianas and climbers,
however, is mostly found adjacent to clearings and in sites disturbed by man, and
it appears that these forms need high irradiance for establishment and development.
Lianas in fact can slow down successions in the reinvasion of gaps (Sect. 3.3.3), in
that they suppress climax species, and thus, support the growth of pioneer species
(Schnitzer et al. 2000). Growth of saplings of trees is not only inhibited by lianas
via competition for light but also below ground via competition in the root medium
(Schnitzer et al. 2005).
The plants climb using tendrils formed from modified leaves or parts of leaves,
shoots or adventitious roots. Shoots wind around branches or form coils, which are
then modified by the secondary thickening, so that the wood develops in the form
of bands (Fig. 6.9), or it is fragmented to individual strands forming rope- or cable-
like structures which are resistant to torsion.
Some species in the aroid genera Philodendron and Monstera and the Cyclan-
thaceae Asplundia begin their life with rooting in the soil and climbing up a phoro-
phyte, but later their old roots degenerate. By growing at the tip and continuously
degenerating the base of their shoots, they literally crawl up their hosts. Hence,
they begin as lianas and later become epiphytes. They have been termed secondary
hemi-epiphytes. However, this term is not all that convincing; strictly they are sec-
Fig. 6.9 Spirally twisted flat and band-like shoots of lianas
ondarily epiphytes. Moreover, in some cases aerial adventitious roots can be formed
again, which may hang down from these plants like curtains (Fig. 6.3C) and es-
tablish contact with the ground for a second time. Thus, secondary hemi-epiphytes
would become primary hemi-epiphytes, a term used for plants which start their
life epiphytically but subsequently establish soil-contact.
The latter include the stranglers, a true group of “murderers”. Among them is
a genus with an extreme plasticity, namely Clusia (Clusiaceae, Order Theales) (see
also below: Sect. 6.6.2.3). Clusia-seedlings may get established terrestrially and
grow directly as independent trees. However, these plants, like other stranglers, may
begin their seedling stage as humus epiphytes, using accumulations of humus in
knotholes or between branches of phorophytes for establishment (Fig. 6.10A). Al-
ternatively, they germinate in epiphyte gardens together with several other epiphytic
species (Fig. 6.10B–D), where tanks of bromeliads or nest and basket-forming ferns
provide the required substrate. Then, adventitious roots develop, some of which
grow positively gravitropically to the ground whilst others are attached to the host
tree (Fig. 6.11). First they may only compete with their host for light. Subsequently,
after rooting in the ground, they also compete for nutrients in the soil. Eventually
they strangle and kill their host. Their adventitious roots surrounding the trunk of the
host tree hinder the secondary thickening and clamp the phloem in the bark with
the sieve tubes, the tender pathways for long distance transport of assimilates. Pre-
vented from adequate partitioning of supplies, the phorophyte dies. It seldom falls
Fig. 6.10A–D Epiphytic seedlings of Clusia rosea in humus accumulation of tree forks (A) in an
epiphytic garden (B), and inside tanks of the bromeliad Aechmea lingulata (C, D). In D the tank
of A. lingulata has been cut open showing the accumulated humus and the root system of a Clusia
seedling
Fig. 6.11 Adventitious root-system of the strangler Clusia rosea
down. The roots of its ungrateful visitor often form a veritable net with anastomoses
via parenchyma bridges, and inside this hollow cylinder of adventitious roots the
stem of the former host rots away. Thus, the originally epiphytic strangler becomes
an independent tree with a pseudostem of adventitious roots. Such behaviour is not
only observed by Clusia, of course, but equally by other genera with stranglers, e.g.
Ficus species (Fig. 6.12). Other species of Ficus, i.e. F. pertusa and F. trigonata
live in palm trees. They generate negatively gravitropic roots which suspend them
in the crowns of the palms, where they find humus between the leaf bases (Putz
and Holbrook 1986). A most successful form is represented by Ficus bengalensis.
The original aerial roots of a single plant, after gaining ground-contact, may form
an entire forest of pseudostems. Walter and Breckle (1984) described an individual,
which was only 26 m tall but had an average crown diameter of 170 m, a crown cir-
cumference of 530 m and a crown area of 22,000 m2 . (See also Ficus microcarpa in
Fig. 6.12C.)
By clasping other plants lianas are saving investment in thick stems which would
provide independent support for their heavy plant biomass. Only the first shoots
produced after germination are self supporting and their Young’s modulus is high
indicating higher stiffness or lower elasticity, then the shoots start to climb which is
associated with anatomical changes and a decreasing Young’s modulus (Rowe and
Speck 1996; Speck 1997). The reduction of supporting tissues in lianas is also true
for their leaves which are found to have a lower leaf mass per unit leaf area and
this is also beneficial for optimising nitrogen use in relation to photosynthesis of
Fig. 6.12A–C Network of anastomosing strangler roots of Ficus sp. (A) and interior of the hollow
cylinder marking the stem of the original host (B) in a cloud forest above Lake Coté, Costa Rica.
“Forest” of pseudostems of one individual tree of Ficus microcarpa (Foster Botanical Garden,
Honolulu, Hawaii) (C)
fast growing lianas (Kazda and Salzer 2000). Using the support by their host plants
lianas can afford to have very wide xylem vessels reducing friction for the tran-
spiration stream and facilitating transport over long distances (Ewers et al. 1990).
Vessels of up to 0.7 mm in diameter have been observed. The xylem sap readily
flows out of these wide vessels when they are cut open. Wandering around one may
serve oneself from such lianas for a cool drink. Vareschi (1980) reports on a 1 m
long piece from which 205 ml of water were collected within 3 min.
Thus, lianas and hemi-epiphytes evidently develop a particular hydraulic archi-
tecture to support a huge leaf biomass via relatively thin shoots. The theoretical
expectation is that compared to free standing trees (Zotz et al. 1997):
• lianas and hemi-epiphytes show significantly higher specific stem conductiv-
ity, K s ,
• lianas and hemi-epiphytes show less wood cross-section per unit leaf area, i.e.
lower Huber-values, Hv,
• lianas and hemi-epiphytes tend to have less conductive stems per unit leaf
area, K l .
A quantitative comparison between free standing trees of three species and hemi-
epiphytes of four species of Ficus, respectively, and the hemi-epiphyte Clusia uvi-
tana has been presented by Patiño et al. (1995), where Ficus species are C3 -plants
and C. uvitana is a C3 /CAM-intermediate species (Table 6.2). The ranges of K s val-
ues for the hemi-epiphytic and the terrestrial Ficus species overlap, but as expected
much higher maximum values are reached in the hemi-epiphytes. However, for the
hemi-epiphytic C. uvitana the K s value is much lower. As also expected K l values
are lower in hemi-epiphytic Ficus species than in free standing trees, but the values
for C. uvitana are still much lower. The Huber-values, Hv, are similar in both the C3
and C3 /CAM-hemi-epiphytes and as expected lower than in the free standing trees.
The differences in K s and K l values between the hemi-epiphytes with the different
modes of photosynthesis show that in addition to hemi-epiphytism hydraulic archi-
tecture is also related to the performance of CAM and that due to the water saving
mechanism of CAM (Sect. 5.2.2.2, Box 5.1) C. uvitana can afford lower specific
stem conductivity, K s , and lower leaf specific conductivity of stems, K l .
The wide vessels of lianas are potentially prone to cavitation and embolism
which would lead to loss of conductivity. However, Andrade et al. (2005) observed
that maximum sap flow densities in co-occurring lianas and free standing trees were
comparable at a similar stem diameter. In the tropical vine-like bamboo Rhipido-
Table 6.2 Hydraulic architecture parameters of free standing trees and hemi-epiphytes of the genus
Ficus and the hemi-epiphyte Clusia uvitana (rounded up values from Patiño et al. 1995)
Ficus C. uvitana
Free standing Hemi-Epiphytes Hemi-Epiphyte
trees
Specific stem conductivity 11 to 14 7 to 34 1.1
K s (kg s−1 m−1 MPa−1 )
Cross section per unit leaf area 2.0 to 6.1 1.0 to 2.2 1.4
Huber-value, Hv × 104
Conductive stem per unit leaf area 23 to 52 7 to 23 1.5
K l (kg s−1 m−1 MPa−1 )
cladum racemiflorum (Cochard et al. 1994) the rather wide xylem vessels were
found to be surprisingly resistant to cavitation. Experimentally, xylem water po-
tentials (see Box 6.1 for explanation of terminology) of −45 bar were required
to induce 50% loss of hydraulic conductivity, but at the end of the 1993 dry sea-
son potentials of only − 37.5 bar were reached with a loss of conductivity of 10%
due to cavitation and embolism. In the wet season high root pressures possibly re-
pair cavitation. When transpiration was low in the rainy season, i.e. at night and
during rain events positive hydrostatic potentials up to 1.2 bar were built up. The
hemi-epiphyte Monstera acuminata built up pressures of 2.25 bar in the aerial roots
reaching the soil, which are among the highest root pressures known (López-Portillo
et al. 2000).
In contrast to the vine R. racemiflorum, C. uvitana is found experimentally to be
very vulnerable to cavitation loosing 50% of hydraulic conductivity at a stem water
potential of only −13 bar (as compared to −45 bar in R. racemiflorum). It is possible
that CAM is important in helping to prevent additional damage by insuring low
transpiration rates and avoiding dehydration. Indeed, in comparison with two other
epiphytes with different adaptive strategies, C. uvitana performed equally well in
terms of long-term carbon gain. The comparison (Zotz and Winter 1994b) included:
• the evergreen C3 /CAM intermediate C. uvitana,
• the drought-deciduous C3 -orchid Catasetum viridiflavum,
• the evergreen C3 -fern Polypodium crassifolium.
In C. uvitana, during the four months dry season, mean daily carbon gain was re-
duced by ca. 40% following the shift from C3 -photosynthesis to CAM, with strongly
decreased daytime CO2 -uptake. C. viridiflavum grew new leaves in the second half
of the dry season with greatly reduced carbon gain. In the wet season rates of CO2 -
uptake by these leaves doubled. Growth occurred until the end of the wet season,
when leaves were shed again. In P. crassifolium the daily carbon balance was nega-
tive during the dry season, and the epiphyte showed characteristics of chronic pho-
toinhibition. Nevertheless, in all of the three species there were similar rates of an-
nual carbon gain (1,000 g CO2 m−2 year−1) and long-term nitrogen-use efficiency
(i.e. annual carbon gain/mean leaf N content was around 1.1 g CO2 kg N−1 year−1 ).
The long-term water use efficiency (WUE) of net CO2 uptake in C. uvitana was
more than twice that in the other two species.
Box 6.1 Some basic principles of water relations of plants
The water potential ψ is defined as

μH2 O − μ0H2 O
• ψ= ,
V H2 O
where μ0H2 O is the chemical potential of pure water, μH2 O the actual chemical
potential of water in a solution and V H2 O the partial molar volume of water.
Box 6.1 (Continued)
ψ is best explained using an osmotic system of two chambers separated by

a semipermeable membrane each having a vertical tube. The solvent particles
(• = H2 O molecules) can pass the semipermeable membrane, whereas the solute
particles (◦ = solute molecules) cannot permeate.
The difference of osmotic pressure between the solutions of the two chambers,
π, drives water across the semipermeable membrane from the chamber of
the lower osmotic pressure, π, to the chamber of the higher π. The flow of
water is a volume flow Jv . The associated volume change leads to the ascent of
solution in the vertical tube of the chamber with the higher π. The column of
solution exerts a hydrostatic pressure on the solution, P , which counteracts
the osmotic water flow driven by π. At thermodynamic equilibrium
• P = π and ψ = 0 ,
i.e. there is no water potential difference between the two chambers since at any
given time
• ψ = P − π .
All parameters have the physical unit of a pressure. π is related to the solute
concentration, c, as follows
• π = c· R·T ,
where T is the temperature in Kelvin and R the universal gas constant.
In plant cells P is built up at the elastic cell walls and is called turgor pres-
sure.
6.4 Epiphytes 183
Box 6.1 (Continued)
Measurements of ψ can be made by psychrometric techniques or using a pres-

sure chamber. In the latter case a plant shoot or leaf is tightly sealed in the
chamber, with the cut stem or petiole protruding to the outside. When pres-
sure is exerted on the air in the chamber, xylem sap is expressed from the cut
end. The equilibrium pressure, where the sap just reaches the cut end, can, with
certain precautions, be related to the water potential of the stem or leaf.
P can be measured by inserting small glass capillaries into cells which are
adjoined to a pressure transducing read-out system (pressure probe). Often it
is also calculated from ψ and π, but this needs to be interpreted with care.
π is obtained from psychrometry or freezing point determinations of cell
sap. Plasmolysis studies are also applied.
(See textbooks, e.g. Nobel 1983; Lüttge et al. 2005.)
6.4 Epiphytes
As noted above, the only vascular epiphyte in the mesic climate, the fern Polypodium
vulgare, is a facultative epiphyte. This means that there are gradations between
the terrestrial and epiphytic habit (Gessner 1956; Richards 1996). In the tropics
many species, e.g. among bromeliads and aroids, grow equally well terrestrially and
epiphytically (Fig. 6.13; see also Table 6.4).
Benzing (1989a) has conceived five different schemes alternatively classifying
epiphytes in categories based on:
I. relationships to the host (or “phorophyte”),
II. growth habit,
III. humidity,
IV. light
V. phorophyte-provided media
(Table 6.3). The epiphytic life form is effectively encompassed in categories I and II.
The other three categories refer to the three major stress factors of epiphytic plant
life (see Sect. 6.6). The entire system of five schemes is very useful as it offers
a good summary of the great morphological and ecophysiological diversity among
epiphytes and their associates. On the other hand, it suffers from the general problem
of attempts of this kind of casting the diversity of life into schematic systems. Thus,
the study of case stories may prove more appealing.
One of the most exciting case stories is offered by the Bromeliaceae (for mono-
graphs see Martin 1994; Benzing 2000). They operate with tanks and epidermal
scales or trichomes. The tanks are made up by densely overlapping leaf-bases of
the rosette-forming bromeliads and depending on life form there is a gradation in
effectiveness of water storing capacity. The scales are epidermal structures which
developed increasing complexity during the evolution of bromeliads. They consist
of living basal or foot cells, stalk cells, which may be living or dead in the mature
stage of the trichomes, and the actual scales comprised of dead cells (Fig. 6.14).
Fig. 6.13A,B The bromeliad Aechmea lingulata growing both terrestrially (A) and epiphytical-
ly (B). (St. John-Island, US Virgin Islands, Lesser Antilles)
Fig. 6.14A,B Schemes of scales of bromeliads. A Top view. B Cross-section the living cells of
the scale dotted and with a nucleus. The black line along cells in B indicates cutinization of the
epidermal cells and the outer walls of the trichome stalk cells which allows entry of solutes into
the leaves only via a specific pathway enforcing membrane passage and cytoplasmic control over
the solutes taken up. (After Sitte 1991 with permission of G. Fischer-Verlag)
6.4 Epiphytes 185
Table 6.3 Five different schemes alternatively classifying epiphytes. (After Benzing 1989a,b)
I. Relationships to the phorophyte
1. Autotrophs, using the phorophyte only for support
1.1 Accidental
1.2 Facultative
1.3 Hemi-epiphytic
1.3.1 Primary
1.3.1.1 Strangling
1.3.1.2 Non-strangling
1.3.2 Secondary
1.4 Genuinely epiphytic
2. Parasites
II. Growth habit
1. Trees
2. Shrubs
3. Suffrutescent to herbaceous forms
3.1 Tuberous
3.1.1 Storage, woody and herbaceous
3.1.2 Myrmecophytic, mostly herbaceous
3.2 Broadly creeping: woody or herbaceous
3.3 Narrowly creeping: mostly herbaceous
3.4 Rosulate, herbaceous
3.5 Root/leaf tangle, herbaceous
3.6 Trash-basket, herbaceous
III. Humidity
1. Poikilohydrous (mostly lower plants)
2. Homoiohydrous
2.1 Hygrophytes
2.2 Mesophytes
2.3 Xerophytes
2.3.1 Drought-endurers
2.3.2 Drought-avoiders
2.4 Impounders
IV. Light
1. Exposure types
2. Sun types
3. Shade-tolerant types
V. Phorophyte-provided media
1. Relatively independent of rooting medium
1.1 Atmospheric forms
1.2 Twig and bark inhabitants
1.3 Forms creating substitute soils or attracting ant colonies
2. Utilizing preexisting specific rooting media
2.1 Humus-dependent
2.2.1 Shallow humus forms
2.2.2 Deep humus forms
2.2.3 Ant-nest garden and plant catchment inhabitants
2.2 Parasites
By the structure of tanks and scales we may distinguish four different life forms
of bromeliads (Table 6.4):
• Type I: Soil Root.

Some bromeliads which are obligately terrestrial do not form tanks; often these
forms are highly xeromorphic; they may be densely covered with scales; however
the scales do not function in water and nutrient absorption but may rather serve
reflection of light (see Sect. 5.2.2.1, Figs. 5.3 and 6.15A).
• Type II: Tank Root.
Other obligately terrestrial bromeliads have rudimentary tanks, which have lim-
ited water and litter collecting capacity; the scales make only minor contributions
to water and solute uptake; however, in addition to the soil-roots, plants of this
type develop stem-borne “tank-roots” growing up between the overlapping leaf
bases into the tanks (Figs. 4.3 and 6.15B).
• Type III: Tank-Absorbing Trichome.
The roots are conditionally absorbent but mostly have only mechanical func-
tions for holdfast in these epiphytic forms and may even secrete a cement-like
lipopolysaccharide (Brighigna et al. 1990), the tanks effectively collect rain wa-
ter and decomposing debris; scales are found most densely on the leaf bases in
the tank, where they serve water and nutrient uptake (Figs. 6.13 and 6.15C).
• Type IV: Atmospheric-Absorbing Trichome.
Tanks in these forms are mostly absent and only occasionally poorly developed;
Table 6.4 Life-forms of Bromeliaceae, their characterization and distribution among the three
subfamilies Pitcairnioideae, Bromelioideae and Tillandsioideae. (After Pittendrigh 1948; Smith et
al. 1986a; Smith 1989; Benzing 2000; note that the latter author has recently separated two different
groups out of type III and distinguishes five types)
Designation Root Tank Epidermal Growth habit Taxa
of life-form system trichomes
Type I Absorbent Lacking Unspecialized Obligately Great majority
soil roots and terrestrial of Pictarnioideae/
non-absorbent many Bromel-
ioideae
Type II Absorbent Rudi- Relatively Obligately All terrestrial
soil roots mentary unspecialized terrestrial Bromelioideae
and tank
roots
Type III Usually Well Specialized and Most obligately All the epiphytic
only devel- absorbent; (some facul- Bromelioideae,
mechanical oped concentrated tatively) epiphytic majority
on leaf base of Tillandsioideae
Type IV Exclusively Often Specialized and Obligately Tillandsioideae:
mechanical entirely absorbent; often epiphytic Several species of
lacking cover entire shoot (or saxicolous) Vriesea, otherwise
species exclu-
sively Tillandsia
6.4 Epiphytes 187
Fig. 6.15A–D Life forms of bromeliads. A Type I, soil root, Pitcairnia integrifolia, Trinidad. B
Type II, tank root, Bromelia humilis, Falcon, Venezuela. The basal leaves were removed and the
rosette was turned upside down for photography, so that the tank roots growing upwards between
the leaves can be seen. C Type III, tank-absorbing trichome, Tillandsia fasciculata, Cerro Santa
Ana, Paraguana Peninsula, Falcon, Venzuela. D Type IV, atmospheric-absorbing trichome, Tilland-
sia usneoides, Merida, Venezuela
the entire leaf surface is covered by highly specialized scales, which provide the
only route for uptake of water and minerals from rain and dust in the atmosphere;
in some forms roots are lacking entirely (Figs. 6.4 and 6.15D).
These life forms of bromeliads provide an interesting example of how the vegeta-
tive plant form has been shaped by evolution towards epiphytism; and in this case,
particularly driven by the need for water and nutrient acquisition in the epiphytic
habitat.
6.5 Mistletoes
Mistletoes growing on bushes and trees are not literally epiphytes, which originally
use the phorophytes only as a holdfast. Mistletoes are true parasites. They largely
belong to two families of the Order Santalales, namely the Loranthaceae (∼900
species and 65 genera) and the Viscaceae (∼400 species). Mistletoes occur ubiqui-
tously in the temperate zone, in arid regions as well as in the wet tropics (Sallé et
al. 1993). The majority of mistletoe taxa occur in the tropics. Although ecophysiol-
ogy of mistletoes is increasingly well studied (Popp and Richter 1997), apparently
it is not known why they have such a particularly high diversity and biomass in the
tropics (Benzing 1990).
When germinating on the host trees, haustoria of mistletoes penetrate through
the bark and join the host cambium, where they form a cambium themselves, which
keeps pace with that generating the secondary thickening of the host so that the
haustoria gradually become incorporated in the host’s wood (Sallé et al. 1993).
Via the haustoria the mistletoes establish vascular contacts with the host. Very few
mistletoes have phloem connections, since the contacts are predominantly apoplas-
tic between the xylem elements of host and parasite. Thus, the standard view is that
mistletoes are hemi-parasites on the xylem and transpiration-stream taking only wa-
ter and nutrients from the host, while they are photosynthetically competent and ca-
pable of their own assimilation. The idea that mistletoes might have evolved from
terrestrial root hemi-parasites sucking the xylem of host roots has been discussed
(Benzing 1990).
In order to direct part of the transpiration stream from the host to their own shoot
system for water and nutrient supply, mistletoes need to establish the required driv-
ing force. Indeed, it has been shown that they have a more negative leaf-water poten-
tial (see Box 6.1) and a larger leaf-conductance for water vapour and hence a higher
transpiration rate, than the host leaves (Schulze et al. 1984; Ziegler 1986; Richter
et al. 1995; Popp and Richter 1997). The difference between the leaf conductances
in mistletoes and in their hosts respectively, can also be demonstrated by carbon-
isotope analysis, because in C3 -plants the variable rate of CO2 -diffusion via stomata
primarily determines overall changes in 13 C-discrimination during photosynthesis,
i.e. more negative δ 13 C-values indicate higher life time stomatal conductance, gH2 O ,
higher average internal CO2 -partial pressures pCO i , and lower water use efficiency,
2
WUE (Sect. 2.5). Lüttge et al. (1998) measured 21 host mistletoe pairs in Brazil and
6.5 Mistletoes 189
found that consistently δ 13 C-values were more negative in the mistletoes than in the
host leaves documenting higher transpiration rates and gH2 O of the mistletoes and
their operation at higher pCOi (Table 6.5). The higher transpiration rates also con-
2
tributed to higher transpirational cooling of the mistletoe leaves as compared to the
host leaves (Table 6.5), which certainly is an additional advantage of the parasites
in hot and dry sites.
If mistletoes have similar or lower CO2 -assimilation rates as compared to host
leaves, this also implies that the mistletoes may have considerably lower water-use-
efficiencies (WUE = CO2 assimilated : H2 O transpired) at the expense of the host.
Net CO2 uptake is generally considered to be lower in mistletoes as compared to
their hosts (Popp and Richter 1997). However, in the study of Lüttge et al. (1998)
photosynthetic capacity assessed from measurements of chlorophyll fluorescence
parameters (Box 4.6) in mistletoe leaves proved to be similar to that of host leaves.
Only in very exposed open sites photosynthetic capacity of mistletoe leaves was
inferior, but that was due to a pronounced sun-type expression of host plant leaves.
Mistletoes may even grow on mangrove associates like Conocarpus erectus (Orozco
et al. 1990; see Chap. 7) and true mangroves, where they must establish a water
potential gradient large enough to allow movement of water downhill from the salt-
loaded halophilic host to their own leaves, but it is also observed that increasing
drought and salinity stress may make hosts less suitable for invasion by mistletoes
(Miller et al. 2003).
The view that mistletoes exclusively are parasites for water and nutrients, needs
to be modified since carbon gain of mistletoes from the host can be significant
(Richter et al. 1995; Escher et al. 2004). Studies of partitioning of dry matter and
mineral nutrients (Pate et al. 1991a,b), which included analyses of carbon-isotope
ratios (see Sect. 2.5; Marshall and Ehleringer 1990), showed that 24% (Pate et al.
1991a) to 62% (Marshall and Ehleringer 1990) of the mistletoe carbon may be de-
rived from the host, and a tabulation of Popp and Richter (1997) even lists values
up to 87%. This is not yet the ceiling though, because a fully holoparasitic het-
erotrophic Loranthaceae mistletoe, Tristerix aphyllus, has been discovered, which
grows on the tissue of cactus stems (Kraus et al. 1995). First the transfer of carbon-
compounds from the hosts to the mistletoes is partially due to the fact, that under
various circumstances the xylem sap itself may also carry organic compounds. Sec-
ond the mistletoe tissue may take up organic material from the host phloem by
phloem unloading via apoplastic pathways and active membrane-transport, where
the mistletoe becomes a sink for source substrates from the host. The involvement
Table 6.5 Comparison of 21 mistletoe/host-pairs in the cerrado belt of Brazil. Data are averages of
differences between mistletoes and hosts leaves, i.e. mistletoe minus host values are shown. δ 13 C
i
and pCO are derived from carbon isotope analyses of dried leaf material, leaf temperatures, Tleaf ,
2
are from instantaneous measurements in the field. (Averages ±SD(n)). (After data of Lüttge et al.
1998)
δ 13 C (‰) i
pCO 2
(Pa/MPa) Tleaf (◦ C)
−3.25 ± 1.59 (21) 52 ± 25 (21) −2.0 ± 1.4 (17)
of active, membrane-controlled transport can make acquisition of both mineral ions

and organic compounds by the mistletoe from the host a highly selective process
(Pate et al. 1991b; Rey et al. 1991; Escher et al. 2003). Phloem mobile mineral nu-
trients can also arrive in the mistletoes via the internal phloem-xylem-circulation of
the host (Bannister et al. 2002).
An interesting morphological feature of mistletoes, related to parasite-host nu-
trient relations, is the often observed strong resemblance between parasite and host
leaves (Ehleringer et al. 1986; Fig. 6.16). The mimicry of host leaves is common
Fig. 6.16 The Loranthaceae Phthirusa ovata (recognisable by its inflorescence) in a host tree,
Brazil
6.6 Stressors Driving Ecophysiological Adaptation of Epiphytes and Hemi-Epiphytes 191
since mistletoes often have higher nitrogen contents than their hosts, and hence re-
duces the likelihood of mistletoe herbivory. Mimicry is absent when mistletoes are
poorer in N than the host.
6.6 Stressors Driving Ecophysiological Adaptation of Epiphytes

and Hemi-Epiphytes
The major factors, which limit epiphytic life and thus may become stressors
(Box 3.1), are:
• light,
• water,
• mineral nutrients.
Light (Sect. 6.6.1) is highly variable, similar to the situation of other species op-
erating in different strata of the forest canopy (see Sect. 3.4.1). Water (Sect. 6.6.2)
and mineral nutrients (Sect. 6.6.3) are particularly difficult resources to obtain by
epiphytes having no roots in the soil and therefore availability of water may be con-
sidered the major constraint in the epiphytic habitat.
6.6.1 Light and the Evolution of Plants to Epiphytism
A generally encountered view assumes that climbing and epiphytism in plants is

a struggle for light in an escape from the darkness of forest floors. This goes back
to A.F.W. Schimper, who concluded his observations in forests of the American
tropics with the hypothesis that epiphytic bromeliads evolved from shade adapted
terrestrial forms (Schimper 1888). However, as we have already mentioned above
(Sect. 6.1), lianas and vines are most frequently light demanding plants of pioneer
successions. Epiphytes occupy sites of variable light exposure. Studies of the distri-
bution of epiphytic orchids on phorophytes in a West African rainforest have shown
that only a small percentage of the total number of species are found in the upper
canopy, and most species dwell within the crowns of trees (Fig. 6.17).
Pittendrigh (1948) grouped the epiphytic bromeliads of Trinidad in three cate-
gories according to their light demand:
• an exposure group,
• a sun group,
• a shade tolerant group.
Using stable carbon isotope analysis (Sect. 2.5), Griffiths and Smith (1983) have de-
termined the distribution of C3 -photosynthesis and CAM among these 40 species.
They related the mode of photosynthesis to Pittendrigh’s light-demanding categories
and the annual precipitation at the sites where they occur in Trinidad. The result of
Fig. 6.17 Distribution of epiphytic orchids on trees in a West African rainforest. Numbers of
species found in the different zones of the phorophyte related to total orchid species counted.
The zones of the phorophyte are: A the basal part of the stem up to 3 m above ground level; B the
stem up to the first ramifications; C, D and E the canopy divided into three equal parts along the
length of the branches from inside to outside. (Goh and Kluge 1989, after Johansson 1975)
the survey is depicted in Fig. 6.18 (see also Fig. 6.23). At the wettest site (> 6.4 m
precipitation per year) the shade group is not represented at all, with one species of
the exposure group, six species of the sun group and only one CAM species being
present. At somewhat lower annual precipitation, the sun group prevails with a total
of eight species and three CAM species among them. Under intermediate precip-
itation, only C3 species comprise the exposure and shade groups. However, at the
driest sites one finds only CAM plants of the exposure and sun groups. Thus, CAM
among epiphytic bromeliads is clearly correlated with reduced water availability
and sun exposure which exacerbates drought stress.
Together with the development and specialization of epidermal trichomes (see
Sect. 6.4, Table 6.4, Fig. 6.14), which can be considered as an evolutionary trait
(Mez 1904; Tietze 1906), Pittendrigh (1948) used the abundance and distribution
of species in the three categories for consideration of the evolution of epiphytism
among bromeliads. He suggested that epiphytic bromeliads did not evolve from
shade demanding ancestors of the forest floor but rather were derived from terrestrial
ancestors of open habitats originally adapted to sun exposure and at least temporary
drought.
The observation that CAM occurs only among bromeliads of the sun and expo-
sure groups supports this interpretation because phylogenetically CAM is consid-
ered to be rather an advanced physiological trait. The family of the Bromeliaceae
Fig. 6.18 Distribution of epiphytic Bromeliaceae of the exposure group (Ex), the sun group (Su)
and the shade-tolerant group (Sh) with C3 photosynthesis (open parts of the bars) and CAM (closed
parts of the bars) in Trinidad related to annual rainfall. (After Griffiths and Smith 1983)
Fig. 6.19 Scheme of putative phylogenetic relationships within the Bromeliaceae based on the tax-
onomic distribution of CAM and C3 photosynthesis at the level of individual genera and molecular
systematics. The scheme shows that both the epiphytic habit (E) and CAM must have arisen more
than once during evolution of the present-day forms. Within the Bromelioideae there are indica-
tions of a progressive loss of CAM in some genera. T = terrestrial, E = epiphytic forms, C3 =
plants with C3 -photosynthesis. (Smith 1989; Crayn et al. 2000, 2004)
is monophyletic. Among the three subfamilies most likely the Bromelioideae and
possibly the Tillandsioideae are monophyletic while the Pitcarnioideae are poly-
phyletic (Crayn et al. 2000, 2004; Horres et al. 2000). The family comprises about
2800 species (Luther and Sieff 1998). Martin (1994) has identified the photosyn-
thetic pathway of 249 species of which 69% show CAM capacity with various de-
grees of CAM expression. CAM and epiphytism have evolved independently and
polyphyletically several times within each subfamily (Fig. 6.19; Smith 1989; Crayn
et al. 2000). From a terrestrial C3 ancestor epiphytism evolved first. In the branch
giving rise to the Tillandsioideae from the epiphytic C3 plants CAM evolved, so that
this subfamily is exclusively epiphytic with both C3 and CAM species. In the branch
starting from the terrestrial C3 plants CAM evolved giving rise to the exclusively
terrestrial Pitcairnioideae with both C3 and CAM plants. From the branch starting
Table 6.6 Cardinal points of light-response curves of various epiphytes compared to genuine sun
and shade plants. Epiphytes were related to sun and shade plants respectively, by evaluating all
of the three given criteria (light-compensation point, light-saturation of photosynthesis, rate of
photosynthesis at saturation) together, because coordination is not simple when using single criteria
individually. (After Lüttge 1985)
Plant type or species Light compensa- Light saturation Rate of photosyn-
tion point (µ mol of photosynthesis thesis at saturation
photons m−2 s−1 ) (µ mol photons (µ mol CO2 or O2
m−2 s−1 ) m−2 s−1 )
Sun plantsa 20–30 400–600 10–20

Epiphytes:
Platycerium grande (C3 fern)b 20
520 >3
Anthurium hookeri (C3 aroid)b 5–40 180–375 1–6
Kalanchoë uniflora 25–75 275–500 0.5–4
(C3 /CAM Crassulaceae)b
Phalaenopsis violacea 16–20 180–200 4–8
(CAM orchid)c
Phalaenopsis grandifolia 14–20 240–260 6–8
(CAM orchid)c
Shade plantsa 0.5–10 60–200 1–3
Epiphytes:
Aglaomorpha heracleum 5 160 3
(C3 fern)b
Pyrrossia longifolia (CAM fern)b 8 100–150 1–2
Pyrrosia lanceolata (C3 fern)b 15 200–300 –
Drymoglossum piloselloides 8 300–500 –
(CAM fern)b
Nepenthes × hookeriana 5–10 150–225 3–5
(C3 carnivorous plant)b
a Data as given also in Table 4.1
b Data for glass-house grown plants (Lüttge et al. 1986)
c Data for field-plants in Singapore (Goh and Kluge 1989).
from terrestrial CAM plants epiphytism evolved giving rise to the exclusively CAM
Bromelioideae with both terrestrial and epiphytic species.
In conclusion, at least for the bromeliads, Schimper’s idea of epiphyte evolution
towards light from shade demanding understorey plants is most probably not valid. It
was more likely to have represented a conquest of space by plants already adapted to
the ecophysiological problems of exposed nutrient-poor habitats. The shade-adapted
epiphytic bromeliads are mostly shade-tolerant and not shade-demanding, and they
probably constitute a later development.
However, Schimper’s hypothesis may still apply to the evolution of epiphytism
in other groups of plants. Where this is the case, one would expect to find not only
shade-tolerant but also clearly shade-demanding species. This can be assessed by
comparing the cardinal points of light-dependence curves, which distinguish shade
and sun plants (see Sect. 4.1.1, Table 6.6). By the criteria of light-compensation
point, light-saturation and maximal rates of photosynthesis at least some epiphytic
ferns and orchids are found to be typical shade plants, while other ferns and orchids,
Kalanchoë uniflora and Nepenthes appear to be sun types. Mechanisms of protec-
tion from photodamage involving zeaxanthin and antioxidants, such as tocopherol
(Sects. 4.1.3 and 4.1.4) have been studied in various epiphytic ferns (Tausz et al.
2001).
6.6.2 Water
6.6.2.1 Acquisition of Water Shaping Life Forms of Epiphytes

and Hemi-epiphytes
The availability of water is the most pronounced problem for epiphytes and hemi-
epiphytes which have no root-soil contact (Zotz and Hietz 2001). Rada and Jaimez
(1992) compared terrestrial and epiphytic plants of the facultatively epiphytic Ara-
ceae Anthurium bredmeyeri growing close to each other in a tropical Andean cloud
forest. The epiphytic plants were affected to a greater degree by the decrease in
water availability during the dry season. They showed a larger decrease in leaf con-
ductance and lower leaf water potentials during the dry season than the terrestri-
ally growing plants as well as a reduction in stomatal densities in new leaf growth.
Clearly, the water factor can have a large influence on life form of epiphytes.
Most of the lower plant epiphytes, i.e. aerial algae, lichens, bryophytes and
even some ferns, are poikilohydrous and desiccation tolerant (Table 6.3: III 1; see
also Sect. 11.4.2). They are truly resistant to drought stress, because they can dry
out without suffering damage, overcoming drought periods in a non-hydrated state
and becoming viable again when water can be absorbed from precipitation. Of the
lichens only those, which have green algae as the photoautothrophic symbionts, are
able to acquire their water and reactivate photosynthesis from the water vapour in
the gas phase (Lange et al. 1986, 1988). This also holds for pleurococcoid aerial
green algae (Bertsch 1966). However, lichens having cyanobacteria as symbionts
require water in liquid form to reactivate photosynthesis.
Tillandsoid trichomes have no effect on leaf boundary layers and any associated
reduction in transpirational water loss (Benz and Martin 2006). Some atmospheric
bromeliads may take up water from the gas phase of the atmosphere by equilibration
of the hygroscopic cell walls of the dead scale cells in the trichomes which densely
cover their surface. Thus, one may observe a peak of water-vapour uptake when
the relative air humidity (RH) increases at the beginning of the night (Fig. 6.20).
However, this is matched again by a loss of water vapour at the beginning of the day
when RH decreases and therefore the bromeliad leaf cells do not have a net gain of
water from this mechanism (Schmitt et al. 1989).
In consequence, angiosperm epiphytes have developed a range of other adapta-
tions which often are equally related to the nutrient “stress” factor (Sect. 6.6.3), e.g.
formation of tanks or humus collecting baskets, in which they effectively create
their own soil with a limited water storage capacity. Water demanding animals like
small frogs may even live in tanks of bromeliads (Fig. 6.21), which in some species
Fig. 6.20A–C Night-day cycle of water-vapour exchange by plants of Tillandsia recurvata L.

A Water-vapour exchange of normal living plants shows a peak of net uptake (negative values
of JH2 O ) as the dew-point temperature (T ) decreases and relative air humidity (R H ) increases at
the onset of the dark period, and a peak of net release (positive values of JH2 O ) with the oppo-
site changes of T and R H at the beginning of the light period. B These peaks are also observed
with plants killed in boiling water. They are restricted to passive hygroscopic equilibration of dead
structures. C Subtraction of JH2 O by the dead plants from that of the living plants shows true tran-
spirational water-vapour loss, which is much higher throughout most of the dark period in this
CAM bromeliad than during the light period. (Schmitt et al. 1989, from Lüttge 1989)
Fig. 6.21 Small frog in a tank of a flowering plant of Bromelia humilis
can impound 5–10 l of water. Water storage tissues in leaves and stems may also
be prominent, so that leaf and stem succulence occurs in most bromeliads (Horres
and Zizka 1995), orchids and the epiphytic cacti (Fig. 6.22). In this relation it has
been underlined that independent of age plant size of epiphytes matters a lot be-
cause availability and especially storage capacity of water is highly size dependent
and this has effects on many other functions including allocation and partitioning
of nutrients and area based photosynthetic capacity. Identical environmental condi-
tions impose different degrees of stress on co-occurring smaller and larger plants
(Zotz and Andrade 1998; Schmidt and Zotz 2001; Schmidt et al. 2001; Zotz et al.
2001, 2004; Zotz and Hietz 2001).
6.6.2.2 CAM and Water Relations Parameters
The pre-eminent role of water in limiting the life of epiphytes has resulted in the
frequent occurrence of CAM, the mode of photosynthesis which conserves water
(see Sect. 5.2.2.2). At Barro Colorado Island, Panamá, 25% of the epiphyte flora are
CAM plants (Zotz and Ziegler 1997). Some authorities have counted about 13,500
species of epiphytes with CAM. This corresponds to 57% of all epiphyte species,
while only 10% of all vascular plants are CAM species. The advantage of CAM for
epiphytic life is:
Fig. 6.22A–C Stem succulent epiphytic cacti (A Epiphyllum, B Selenicereus inermis growing
through a termite nest), and adaxial water storage tissue of a leaf succulent bromeliad (C)
• water saving, i.e. a high water use efficiency,

• provision of an osmotic driving force for water uptake by nocturnal acid accu-
mulation,
• flexibility in the mode of carbon acquisition.
See Sect. 5.2.2.2.
A census of epiphytic bromeliad species in Trinidad has related the frequency
of bromeliad epiphytes and the relative number of CAM species to annual rainfall
and the prevailing type of forest (Fig. 6.23). Very dry deciduous seasonal forest sus-
tains low epiphytic bromeliad biomass and the small number of species are CAM
plants. The abundance of the epiphytic bromeliad species is highest in the ever-
green seasonal forest and the lower montane rainforest (Fig. 6.23). This is consis-
tent with the general observation that epiphyte richness is highest at mid elevation,
e.g. at 1,000 m a.s.l. in a Costa Rican study covering the altitudinal range from 0 to
2,500 m a.s.l., and strongly correlated to rainfall and not to temperature and light in
the canopy (Cardelús et al. 2006). In the Trinidadian study the relative contribution
of CAM species to the total number of species declines rapidly as forests get wet-
ter and the water saving function of CAM becomes less important. A decrease of
CAM epiphytes with increasing altitudes as seen in Fig. 6.23 is frequently described
(Earnshaw et al. 1987; Hietz et al. 1999).
Table 6.7 summarizes some water-relation parameters (Box 6.1) of epiphytes.
Since most of those studied to date are also CAM-species, it is difficult to decide
Fig. 6.23 Relations between total number of epiphytic bromeliad species, the relative number of
CAM species among them, annual rainfall and prevailing forest types in Trinidad. (Smith 1989)
whether these are typical properties of epiphytes or general characteristics of CAM

plants. The high relative water content of epiphytes as compared to various C3 -
crop plants and trees is noteworthy. A high relative water content is a typical fea-
ture of CAM-plants and is associated with high water-storage capacity and succu-
lence (Fig. 6.24). For the epiphytic ferns and orchids of Australia, Winter et al.
(1983) could demonstrate correlations between succulence and CAM expression
(Fig. 6.24). The highest osmotic pressures (π) of epiphytes in Table 6.7 are some-
what above 20 bar; in C3 -desert plants they may reach 100 bar. The lowest, i.e. most
negative, water potentials (ψ) of epiphytes are at −10 bar, in C3 -desert plants val-
ues below −150 bar may be found. Hence, the cell sap of epiphytes is diluted, i.e.
osmotic pressures are relatively low (see also Martin et al. 2004), the water potential
is high and the turgor pressure (P) is low. In this respect epiphytic C3 - and CAM-
bromeliads are little different, and also terrestrial CAM-plants show values in this
range. The epiphytic C3 /CAM intermediate Clusia uvitana in a rainforest in Panama
has leaf water potentials in the same range, i.e. −7 to −9 bar (Zotz et al. 1994).
Clusia (see Sect. 6.6.2.3) and Ficus are genera of hemiepiphytes and stranglers,
with very similar habits, but the latter has been studied much less in terms of physi-
ological ecology of photosynthesis and water relations. This is astonishing because
Ficus appears to be as successful in tropical forests as Clusia. Both have very dif-
ferent strategies though. Most species of Clusia have CAM-capacity but as far as
it is known to date all species of Ficus are obligate C3 -plants. Holbrook and Putz
(1996b) have made interesting intraspecific comparisons of water relations in the
life forms of epiphytes and terrestrial trees in the genera Ficus and Clusia. They
found that in five species of Ficus the epiphytic life forms as compared to terrestrial
trees had:
• several-fold higher specific leaf area (m2 g−1 ), which may also be taken as higher
degree of “succulence”,
• two- to fourfold lower stomatal densities, which may be discussed in relation to
the need of reduced transpiration at lower availability of water in the epiphytic
habitat,
Table 6.7 Water relation parameters of epiphytes as compared to terrestrial plants. (Simplified
from Table 3 of Lüttge 1985). Values were obtained by various authors with different methods as
the Scholander pressure chamber technique, plasmolysis measurements and cryoscopy
Plants Relative water ψ P π
contenta (bar)
Epiphytic CAM plants 0.80 to 0.94 –0.8 to –9.9 2.0 to 10.0 1.7 to 23.4
(ferns, orchids, bromeliads)
Terrestrial CAM plants 0.93 to 0.96 –2.4 to –7.9 1.4 to 5.4 4.8 to 12.0
Epiphytic C3 bromeliad – –2.0 to –3.8 2.3 to 3.9 5.2 to 6.1
Terrestrial crop plants 0.58 to 0.75 – – –
North American trees 0.60 to 0.85 – – –
a Ratio of the volume of intracellular water (i.e. inside the plasmalemma) at incipient plasmolysis
to the volume of intracellular water at maximum turgor pressure.
Fig. 6.24 Correlation be-

tween δ 13 C values as a yard-
stick for CAM expression
(see Sect. 2.5) and leaf thick-
ness indicative of the degree
of succulence for Australian
epiphytic ferns and orchids.
(Winter et al. 1983)
• osmotic pressures (π) about 6 bars lower,

• a bulk modulus of cell wall elasticity (ε) about 50% lower.
We must note that cell wall elasticity is inversely related to ε, i.e. the higher ε the
more elastic and the lower ε the stiffer is the cell wall. With the relationship of
V
P = ε , (6.1)
V
where P and V are turgor pressure and volume changes, respectively, and V is
cell volume, it then follows that a given change in volume (V ) leads to a lower
change in pressure when ε is larger or cell walls are stiffer, i.e. in the epiphytic Ficus
plants a larger volume of water can be lost before turgor is lost than in the terrestrial
trees. As a result leaves of the more succulent epiphytes and the conspecific less
succulent trees of Ficus species lost turgor at approximately the same relative water
content.
With the water potential
ψ = P −π (6.2)
(see Box 6.1) at P = 0, or zero turgor, ψ = −π, and this is substantially higher
(less negative) in the epiphytes due to the lower osmotic pressure π. These observa-
tions agree with the general trends for higher succulence, higher ψ and lower π in
epiphytes (see above and Table 6.7).
In contrast to Ficus species, in Clusia differences in all of these water relation
parameters between epiphytes and trees were very small. Thus, while Clusia has
instantaneous plasticity of responding to changing water supply and evaporative
demand by photosynthetic options (C3 -CAM transitions) Ficus shows intrinsic de-
velopmental changes during the transformation from epiphyte to tree which is asso-
ciated with improved acquisition of water.
In CAM plants water relation parameters ψ, P and π also oscillate together with
the day-night malic acid rhythm. Figure 6.25 describes an experiment with the atmo-
Fig. 6.25 Experiment showing the capacity of cells of the atmospheric CAM bromeliad Tillandsia
usneoides for water uptake during a day-night cycle. Plants were weighed (initial FW) dipped for
10 min into water, dried superficially and then weighed at intervals to determine the point where
rapid evaporation of surface water is completed and water is only lost from the living cells by
transpiration, which allows to estimate water uptake by extrapolation (A). It is seen that osmotic
pressure π (B) and malate levels (C) increase during the night, and increased water uptake (D) and
turgor pressure (E) measured directly with an intracellular pressure probe are associated with this.
(see Lüttge 1987)
spheric CAM-bromeliad Tillandsia usneoides, showing that nocturnal accumulation

of malic acid provides an osmoticum, which may drive cellular water uptake. It can
be seen that cell-sap osmotic pressure (π) increases together with malic-acid lev-
els. Water uptake, measured after dipping the plants for a short period into water as
shown in Fig. 6.25A, clearly increased during the night together with π, and this
also led to an increase in turgor pressure. It should be noted, that while atmospheric
bromeliads could occur in rather dry habitats, they are often found at sites where
fog forms during the later part of the night and in the early morning. Water from
condensed fog and dew is then available at times when malic-acid concentration in
the cells is high and can lead to osmotic uptake of water.
6.6.2.3 CAM and Flexibility: The Case Study of Clusia
A major advantage of CAM in habitats where there are large short term and seasonal
variations in water availability is the inherent flexibility in this mode of carbon ac-
quisition. The different expression of the four CAM phases (see Box 5.1) in con-
stitutive CAM plants already allows highly variable responses. If water supply were
to range from very severe to moderate and low drought stress, there may be, re-
spectively total stomatal closure and CO2 -recycling (also called CAM-idling), pre-
dominant nocturnal opening of stomata (Phase I) or increasing use of phase IV and
phase II CO2 -uptake during the daytime hours. There may even be continuous CO2
uptake day and night under well watered conditions. In addition there are species
which are true C3 -CAM intermediates. They can switch from C3 photosynthesis
to CAM as drought stress increases, and back again when the stress is released.
Among the epiphytic bromeliads Guzmania monostachia is such a C3 -CAM inter-
mediate and it is the only one in the Bromeliaceae family (Maxwell 2002; Maxwell
et al. 1994, 1995, 1999). However, there are other taxa with C3 -CAM-intermediate
epiphytes. The epiphytic fern Pyrrosia confluens and the Crassulaceae Kalanchoë
uniflora belong to this group (Griffiths 1989) as well as species of Peperomia (Sipes
and Ting 1985; Ting et al. 1985; Holthe et al. 1987). The plants showing the most
flexible response, however, are in the hemi-epiphyte and strangler genus Clusia.
A separate book is monographically devoted to Clusia (Lüttge 2007a) and it is only
briefly used here as case study. Each of the photosynthetic modes mentioned above
are expressed in Clusias:
• pure C3 -photosynthesis,
• pure CAM,
• C3 /CAM-intermediate behaviour with environmentally controlled reversible
changes between the two modes of photosynthesis,
• night time stomatal closure with fixation of respiratory CO2 and vacuolar malate
accumulation plus day-time stomatal opening and reduction of CO2 both from
the atmosphere and the nocturnally stored malate (CAM-cycling),
• stomatal closure around the clock and only recycling of respiratory CO2 (CAM-
idling),
(see Fig. 6.26).
Fig. 6.26A–I Modes of photosynthetic CO2 exchange in Clusia.

Left panel. Comparison of four species under identical conditions in a phytotron, Clusia venosa
with C3 photosynthesis (A), Clusia minor with CO2 uptake day and night (B), Clusia major and
Clusia alata both with CAM but differing in the development of phase IV (C,D).
Center panel. Clusia minor in a growth chamber with C3 photosynthesis under well-watered condi-
tions at high irradiance (1,700 µ mol photons m−2 s−1 ) and medium leaf/air water vapour pressure
difference (W = 6.6 mbar bar−1 ) (E); CAM with the well-expressed four phases (I – IV) under
drought stress at low irradiance (400 µ mol photons m−2 s−1 ) and high W (13.5 mbar bar−1 ) (F);
and CO2 uptake day and night under well-watered conditions, low irradiance (400 µ mol photons
m−2 s−1 ) and low W (3.4 mbar bar−1 ) (G).
Right panel. Clusia rosea in the field with C3 photosynthesis (H) and CAM with an extended
phase II in the first half of the day (I).
Black bars on the abscissa indicate the dark periods. (Lüttge 1991)
Variability of ecophyiological response is observed:

• between different species under given environmental conditions (Fig. 6.26, left
panel),
• for a given species under different environmental conditions (Fig. 6.26, center
and right panels),
• even for the two different leaves of a given node in the same plant when they are
kept under different conditions (Fig. 6.27).
The rapid changes between C3 -photosynthesis and CAM, that may be performed by
Clusia are determined by the external control parameters:
i) water relations,
ii) day/night temperature regime,
iii) light.
Fig. 6.27A, B CO2 gas exchange of two opposite leaves at the same node of Clusia minor in
a growth chamber. After four days of drought stress the plant was rewatered on the fifth day (arrow
head). The leaf kept in a cuvette at high VPD (W = 13.1 mbar bar−1 ) continued to perform
CAM (A) with all phases I – IV expressed. The other leaf (B) at low VPD (W = 6.2 mbar bar−1 )
rapidly switched to daytime CO2 uptake and suppressed nocturnal CO2 uptake. The black bars on
the abscissa indicate the dark period. (Lüttge 1991)
i) In an experiment with a plant of C. minor rewatered after a period of several

days of drought, it was possible to get two opposite leaves at a given node to
perform C3 -photosynthesis and CAM respectively, at the same time. One leaf,
when maintained in an atmosphere with a low leaf-air water vapour pres-
sure difference (W or VPD), i.e. kept under a low transpiratory demand,
switched to C3 -photosynthesis a few hours after watering with CO2 uptake
Fig. 6.28 Change of Clusia minor from C3 photosynthesis to CAM as night-time temperatures
are lowered to give an increasing day/night temperature difference (T ). JCO2 net CO2 exchange,
gH2 O leaf conductance for water vapour. Day-time temperature was always 30 ◦ C. (Haag-Kerwer
et al. 1992)
markedly reduced in the subsequent night. The other leaf, kept at high VPD,
continued to perform CAM with the four phases clearly noticeable, as both
leaves had done during the drought period before watering (Fig. 6.27).
ii) By varying the temperature regime, it was found that a certain day-night
temperature difference was important for expression of CAM in C. minor
(Fig. 6.28). The shift between CAM and C3 -photosynthesis was fully re-
versible when the temperature regimes were changed between equal day/night
temperature and day/night temperature differences (Fig. 6.29).
iii) In well-watered plants a drastic increase in light intensity led to an elimina-
tion of nocturnal dark-CO2-fixation and an increase in daytime C3 -photosyn-
thesis. Obviously, this represents an optimal use of high light energy provided
that water is not limiting (Fig. 6.30).
Clusia spp. are also remarkable in several other ways:
• showing the highest nocturnal acid accumulation ever observed for CAM
plants (Table 6.8),
• accumulating large amounts of citric acid during the dark period additionally or
alternatively to malic acid (Table 6.8).
The latter observation also requires a comparative evaluation of the relative
ecophysiological advantages of malic and citric acid accumulation during CAM
(Table 6.9). Consideration of intermediary metabolism suggests that different com-
partmentation and different contributions of mitochondrial and cytosolic reactions
may both be involved. Citric acid accumulation, in contrast to malic acid accumu-
lation, does not lead to a net gain of carbon, although it contributes to carbon recy-
cling. However, carbon recycling via citric acid may be favourable because daytime
breakdown of citric acid may possibly result in the liberation of more CO2 than
Fig. 6.29A–C Change of Clusia minor from C3 photosynthesis to CAM and back to C3 photo-
synthesis as a day/night temperature difference of 10 ◦ C is introduced and removed again. Gas
exchange after seven days at 25/25 ◦ C day/night (A) followed by five days at 30/20 ◦ C (B) and
by four days at 25/25 ◦ C (C). JCO2 net CO2 -exchange, gH2 O leaf conductance for water vapour.
(Haag-Kerwer et al. 1992)
Fig. 6.30 Elimination of dark CO2 fixation (D) and stimulation of light CO2 fixation (L) in a well-
watered plant of Clusia minor by transfer from lower to high irradiation (hν). (Data from Schmitt
et al. 1988)
Table 6.8 The highest nocturnal acid accumulation (mmol titratable H+ /l) and the highest noc-
turnal citrate accumulation (mmol citrate/l) observed in CAM. These records were measured in
epiphytes (Aechmea nudicaulis) and hemi-epiphytes (Clusia species)
Titratable acidity
Aechmea nudicaulis Field Trinidad, March 1983 625a

Clusia rosea, phytotron 1120c
Clusia minor Field, Trinidad, March 1990 1410b
Citrate
Clusia minor Field Trinidad, March 1990 125b

Clusia rosea, phytotron 200c
Data from a Smith et al. 1986b, b Borland et al. 1992, c Franco et al. 1992
that of malic acid, and the availability of this internal CO2 could prevent photoin-
hibition (see Sect. 4.1.7) more effectively when light intensity is high. In fact it has
been observed for four different Clusia species that the ratio of malic acid : citric
acid accumulated during the dark period decreased in response to drought stress,
relatively favouring carbon recycling via citrate. Since citrate accumulation does
not contribute to C-acquisition, naturally it also does not help to improve water-
use-efficiency (WUE). It also adds less to changes in cell-sap osmotic pressure
than malate accumulation, because only one mole of citrate is formed per mole
of hexose consumed. However, citrate is known to be an effective buffering sub-
Table 6.9 Comparative evaluation of the ecophysiological advantages of nocturnal accumulation

of malate ( malate) and citrate ( citrate) respectively, in CAM. (Franco et al. 1992; Haag-
Kerwer et al. 1992)
malate citrate
Carbon acquisition Yes No
H2 O-saving during C acquisition Yes No
π with possible H2 O acquisition Yes Limited
Nocturnal recycling CO2 Carbon skeletons
Daytime recycling CO2 CO2
Buffering capacity Small Large
Ecophysiological functions H2 O-saving; preventing Effectively preventing
photoinhibition to some photoinhibition
extent
stance. This may sustain the very high nocturnal vacuolar acid accumulation ob-
served in Clusia with day-night changes of titratable proton levels of more than 1 M
(Table 6.8).
The diversity of the hemi-epiphyte and strangler Clusia in the tropics has pre-
sented us with many unexpected observations and stimulating new reflections on
the nature of ecophysiological adaptations. Even individual species can occur as the
life forms of free standing terrestrial trees, stranglers, hemi-epiphytes and epiphytes.
There are ca. 350 – 400 species of Clusia occupying a wide range of habitats, e.g.
coastal rocks and sand dunes, savannas, gallery forests, open shrub land, dry low
land forest, secondary shrub forest, dry montane karstic limestone forest, montane
rain forest, upper montane rain forest, cloud and elfin forest, rock outcrops (insel-
bergs; see Sect. 11.3.2; Fig. 6.31, see also Fig. 3.6D), and again individual species
may be found in several of these types of sites (Table 6.10). Perhaps Clusia is so
successful because of the high degree of physiological plasticity. This also makes it
suitable for reclamation of tropical land by afforestation and as an ornamental tree
even in the center of cities.
Fig. 6.31A–E Habitat diversity of Clusia. A Clusia fluminensis on sand dunes in the restinga
formation on the Atlantic coast near Rio de Janeiro, Brazil. B,C Clusia rosea on granite rocks on
the British Virgin Island Virgin Gorda (Lesser Antilles), with aerial adventitious root systems in
the rock furrows in C. D Clusia sp. Gran Sabana, Venezuela, with an epiphytic bromeliad Catop-
sis berteroniana. E Clusia rosea in montane rainforest on the US Virgin Island St. John (Lesser
Antilles)
Table 6.10 Ecological amplitude of individual species of Clusia, C. multiflora (C. mu.), C. parv-
iflora (C.p.), C. rosea (C.r.), C. fluminensis (C.f.), C. minor (C.mi.), C. criuva (C.c.) with their
modes of photosynthesis indicated (From Lüttge 2007b)
C.mu. C.p. C.r. C.f. C.mi. C.c.
C3 C3 CAM CAM C3 /CAM weak
CAM
inducible
Restinga • •
Coastal rocks •
Savanna/cerrado • •
Gallery forest – •
cerrado ecotone
Open shrub land •
Dry low land forest • • •
Secondary shrub forest • •
Dry montane karstic • • •
limestone forest
Montane rain forest • • •
Upper montane •
rain forest
Cloud forest/fog forest/ •
elfin forest
Inselberg • •
6.6.3 Mineral Nutrients
Some special adaptations to the poor nutrient supply in the epiphytic habitat have al-
ready been mentioned above in the discussion of life forms of epiphytes (Sects. 6.3
and 6.4) and in relation to water stress (Sect. 6.6.2). They include the use of hu-
mus accumulation in trees (Fig. 6.10A) and morphological features of the plants
for collecting humus such as the formation of baskets and nests (Fig. 6.10B) as
well as tanks (Figs. 6.15 and 6.10C,D). Scales (epidermal trichomes) of bromeliads
(Fig. 6.14) and the velamen radicum of aerial roots of aroids and orchids serving
atmospheric nutrition also belong to the specialised plant structures formed for nu-
trient and water uptake (Goh and Kluge 1989; Reinert 1998). The velamen is a mul-
tilayered peripheral structure, which is readily infiltrated by water from throughfall
or stemflow (Fig. 6.32).
It has even been argued that the successful trapping of rain and throughfall,
enriched by leachates from leaves and stems, is nutritional piracy, depriving
Fig. 6.32A–F Velamen radicum in epiphytic orchids. A Velamen of a Dendrobium species. V ve-
lamen; Ex exodermis; C cortex. B Detail from A showing the dead velamen cells with the typically
perforated walls. C Detail from A showing the exodermis with an aeration cell (AC). D Surface
view of an aerial root of Vanda tricolor with dry velamen. The air-filled velamen cells appear
homogeneously whitish. E The same detail as in D; however, after wetting the velamen. With
the exception of the pneumatothodes (PN), the velamen cells are filled with water and thus appear
dark. F Cross-Sect. through the water-imbibed velamen of Vanda tricolor in the region of the pneu-
matothode (PN) and aeration cells (AC). The air-filled cells of the pneumatothode appear white.
(Goh and Kluge 1989)
host trees of resources which otherwise would reach their rooting medium (Ben-
zing 1989a,b, 1990). A comparison of the nitrogen content in leaves of faculta-
tive epiphytes at adjacent sites, showed significantly lower N-levels in two aroid
species growing epiphytically as compared to their terrestrial counterparts. N-
content was similar in two tank-forming bromeliads, whereas in seedlings of Clu-
sia low N-content was also related to growth directly on the phorophyte and not to
growth inside bromeliad tanks (Fig. 6.33). Epiphytes, especially in the Orchidaceae,
may form mycorrhizas (Reinert 1998). The fungal hyphae penetrate the epiphytes as
well as the decaying bark of the host phorophyte. This is evidently piracy of a more
overt kind and the tree effectively becomes the pedosphere of the epiphyte (Ruinen
1953; Johannson 1977; Benzing 1982; Benzing and Atwood 1984). An interesting
example of counter-piracy is found when phorophytes produce adventitious canopy
roots which exploit the nutrient debris collected within the epiphyte cover (Nadkarni
1981).
Analyses of the stable isotope 15 N have been used to trace several possible
sources of nitrogen and processes of nitrogen acquisition in tropical epiphytes as
compared to associated soil rooted trees (Stewart et al. 1995; Reinert 1998; Hietz
et al. 1999; Wania et al. 2002). The relations are complex. N-isotope signatures of
epiphytes vary with canopy position and the related water supply. The variations
Fig. 6.33 Comparison of nitrogen levels in cohabitant epiphytic and terrestrial-life-forms of two
aroid and two bromeliad species and of Clusia rosea. Data show N content in leaves of epiphytic
minus leaves of terrestrial plants of the same species (numbers are p values of a t test for statisti-
cally significant differences or n.s. = non-significant). (Ball et al. 1991)
are attributable in part to altered 15 N-discrimination during N-acquisition, i.e. atmo-

spheric deposition, leachates and biological dinitrogen fixation, and to changes in
partitioning of N isotopes within the plant. Epiphytes may make considerable use
of biological N2 -fixation by cyanobacteria and free living N2 -fixing bacteria of the
phyllosphere (see Sect. 6.1) including their own leaves. Brighigna et al. (1992) de-
scribed the trichome layers of bromeliad leaves as a favourable habitat for microbes
including N2 fixing bacteria.
Two additional strategies involve interactions with animals, one of which is
predatory and the other one symbiotic, namely:
• carnivory,
• mutualism with ants.
Carnivory by plants is quite frequent in the tropics and subtropics (Fig. 6.34). In the
plant kingdom, carnivory is generally assumed to be a mechanism for the acquisition
of mineral nutrients, especially N, P and S, by photosynthetically autotrophic plants
living in nutrient-poor habitats such as peat bogs (Schmucker and Linnemann 1959;
Lüttge 1983). Hence, one would assume that carnivorous plants would be rather
frequent in the canopy habitat. However, this is not the case. Carnivorous plants have
developed special organs for the capture of prey, glands for digestion and absorp-
Fig. 6.34 Global distribution of carnivorous plants. The genus Drosera is almost ubiquitous and
occurs between the two lines drawn in the north and in the south respectively, except in the areas left
white. The exclusively tropical and subtropical distribution of the genera Nepenthes and Byblis is
clearly seen and the islands of endemic Heliamphora in Northern South America should be noted.
Distribution of carnivorous genera of the family Lentibulariaceae is not shown on this map, namely
Pinguicula: Northern temperate zone; Genlisea: Tropical South America, South Africa; Utricu-
laria: cosmopolitan; Biovularia: Cuba, Tropical South America; Polypompholyx: South America,
Australia. (After Schmucker and Linnemann 1959)
tion of low molecular compounds obtained from the prey, and mechanisms for the
attraction of small animals such as showy and colourful appendages and production
of scent and nectar (see also Sect. 10.2.3.3). Thus, in cost/benefit analyses the rarity
of carnivorous plants in epiphytic habitats can be explained by the high costs for
investment and maintenance of these complex attraction and capture mechanisms
(Givnish et al. 1984; Ellison 2006). Since other factors, particularly water and often
light, are equally limiting, a cost-benefit analysis suggested carnivory would not be
effective under these circumstances.
On the other hand, there are a few examples of carnivorous plants among
climbers and epiphytes. The pitcher plant genus of Nepenthes is native in the Mala-
ian archipelago and an exclusively tropical genus (Fig. 6.34). There are 71 species of
Nepenthes altogether, among which a few are purely terrestrial, but 6 are epiphytic
and many are climbers (Fig. 6.35). The pitchers attract prey by their shiny and often
Fig. 6.35 Nepenthes gracilis (Malaysia)
Fig. 6.36 Nepenthes. A Pitcher cut open showing the velvety covering with loose wax particles
in the upper part and the gland zone below. B Part of the upper pitcher region with wax particles
and protrusions. C Scale-covered gland
colourful rim, which also bears nectaries towards the inside of the pitcher opening.
Small animals, predominantly insects, having fallen over the slippery collar into the
pitcher lumen, rapidly drown in the digestive fluid produced by glands on the bot-
tom, which contains a protease secreted by the plant and other enzymes provided
by microorganisms participating in prey digestion. Escape via the pitcher walls
is prevented by downward pointing scale-like tissue over the glands (Fig. 6.36C),
modified stomata with protrusions towards the pitcher lumen and a lubrication with
small and loose wax particles in the upper part of the pitcher (Gorb and Gorb 2006;
Fig. 6.36A,B). Substances obtained from the digested prey are absorbed via the
gland cells.
Some of these traits are also shared by tanks of bromeliads. They often contain
dead and putrefying insects and may absorb substances like amino acids from such
prey via their scales. In some cases, such as the epiphytic bromeliad Catopsis bert-
eroniana, there is also wax at the adaxial leaf surfaces lubricating the tank interior
(Benzing 1989b). However, these plants have no glands and do not secrete digestive
enzymes, so that at most there is only the initial development towards the carnivo-
rous syndrome (see also Sect. 10.2.3.3).
A more sophisticated example is Utricularia. Many species in this genus are
aquatic, forming small bladders from modified leaves along the stems. The bladders
are tightly closed by a trap-door, and actively transport ions across the trap wall into
the outer medium to drive the osmotic loss of water from the trap lumen. This cre-
ates tension in the bladder-walls, which sets the bladder trap. Small animals trigger
the opening of the trap door by touching the antennae-like protuberances and are
swept into the trap as the tension in the trap wall is released (Fig. 6.37). The animals
are then digested inside the bladders. In the tropics, Utricularia species often live
Fig. 6.37 Schematic drawing of a longitudinal section of a trap of Utricularia. i Bladder lumen;
o outer medium; D trap door; M trap mouth; SH sensitive hairs of the trap door = trigger hairs;
A, antenna. Dotted arrow opening and closing of the trap door; the tissue beneath the trap door
prevents opening from the inside. H1 , H2 , H3 various types of gland hairs serving prey digestion
and transport functions. (After Schmucker and Linnemann 1959)
Fig. 6.38A–C Utricularia humboldtii in the tanks of Brocchinia tatei. A Inflorescence of U. hum-
boldtii emerging from a tank of B. tatei. B Outer tank leaves removed to show the basal system
of U. humboldtii bladders on stems and leaves with petioles and lamina emerging from the tank.
C Larger and smaller U. humboldtii-bladders
Fig. 6.39A–D Ant-house epiphytes. A, B The orchid Schomburgkia humboldtiana with pseudob-
ulbs cut open in B to show the ants nest. C The bromeliad Tillandsia flexuosa epiphytic on the
cactus Pilosocereus ottonis, where ants are nesting in the inflated basal part of the tank. D The
Asclepiadeaceae Dischidia rafflesiana with adventitious roots in modified leaves
References 219
epiphytically between mosses on stems of trees. Most cunning is Utricularia hum-

boldtii, which lives inside the tanks of the bromeliad Brocchinia tatei (Fig. 6.38).
Another tropical habitat which is often nutrient limited are the savannas. We
will refer to carnivorous plants again below, when we discuss their role and their
contribution to nutrient turnover in this habitat (Sect. 10.2.3.3).
In symbiotic mutualisms of epiphytes with ants (Huxley 1980) we may distin-
guish two forms, which among other benefits, provide mineral nutrition to plants,
namely (Davidson and Epstein 1989; Benzing 1989b, 1990):
• ant garden epiphytes,
• ant house epiphytes.
Ants frequently construct nests in trees using various materials which are rich in
nutrients (forming an ant-nest “carton”). Seeds of plants may germinate directly
from such a nutritive carton. Since the plants offer various goods in return, such as
nectar, fruits and seeds, the ants often disperse and plant the seeds of their epiphytes
in ant gardens. Conversely, plants themselves may also provide nesting facilities for
ants, e.g. cavities in various parts of the plant body or hollow stems (Figs. 3.23 and
6.39; Sect. 3.3.4.4). Among the epiphytes there are many myrmecophytic species
with such ant houses, e.g. orchids with ant nests in pseudobulbs and bromeliads
with inflated tank leaves (Fig. 6.39A–C). The ants carry soil and other decaying
material into the nests and add their faeces, which gives a debris from which the host
plants may absorb nutrients. The pitchers of some species of the Asclepiadaceae
Dischidia are most sophisticated, since adventitious roots grow into the soil and
debris accumulated by ants inside these containers. In effect, epiphytic Dischidias
literally construct their own flower pots (Fig. 6.39D). A study with the Malaysian
Dischichia major using stable isotopes to trace sources of N and C (Sect. 2.5) has
shown that 29% of the host nitrogen is derived from debris deposited into the leaf
pitchers by ants and that 39% of the carbon assimilated by the host is derived from
ant-related respiration (Treseder et al. 1995). Ant respiration may increase the CO2
concentration in the pitcher lumen above atmospheric levels. Since the host has
stomata on the inner pitcher-wall surface this can be directly used for CO2 -fixation
even in the dark in this obligate CAM-plant.
In many cases these plant-ant interactions are true symbioses with obligate mutu-
alism, since the partners are no longer successful individually. The epiphytic plants
benefit nutritionally and may be protected from herbivores, while the ants obtain
nest-sites and various items of food.
References
Andrade JL, Meinzer FC, Goldstein G, Schnitzer SA (2005) Water uptake and transport in lianas
and co-occurring trees of a seasonally dry tropical forest. Trees 19:282–289
Badger MR, Pfanz H, Büdel B, Heber U, Lange O (1993) Evidence for the functioning of photo-
synthetic CO2 -concentrating mechanisms in lichens containing green algal and cyanobacterial
photobionts. Planta 191:57–70
Ball E, Hann J, Kluge M, Lee HSJ, Lüttge U, Orthen B, Popp M, Schmitt A, Ting IP (1991) Eco-
physiological comportment of the tropical CAM-tree Clusia in the field. I. Growth of Clusia
rosea Jacq. on St.John, US Virgin Islands, Lesser Antilles. New Phytol 117:473–481
Bannister P, Strong GL, Andrew I (2002) Differential accumulation of nutrient elements in some
New Zealand mistletoes and their hosts. Funct Plant Biol 29:1309–1318
Benz BW, Martin CE (2006) Foliar trichomes, boundary layers, and gas exchange in 12 species of
epiphytic Tillandsia (Bromeliaceae) J Plant Physiol 163:648–656
Benzing DH (1982) Mycorrhizal infections of epiphytic orchids in southern Florida. Am Orchid
Soc Bull 51:618–622
Benzing DH (1989a) The evolution of epiphytism. In: Lüttge U (ed) Vascular plants as epiphytes.
Evolution and ecophysiology. Ecological Studies, vol. 76. Springer, Berlin Heidelberg New
York, pp 15–41
Benzing DH (1989b) The mineral nutrition of epiphytes. In: Lüttge U (ed) Vascular plants as epi-
phytes. Evolution and ecophysiology. Ecological studies, vol 76. Springer, Berlin Heidelberg
Benzing DH (1990) Vascular epiphytes. Cambridge University Press, Cambridge
Benzing DH (2000) Bromeliaceae: profile of an adaptive radiation. Cambridge University Press,
Cambridge
Benzing DH, Atwood JT (1984) Orchidaceae: ancestral habitats and current status in forest
canopies. Syst Bot 9:155–165
Bertsch A (1966) CO2 -Gaswechsel und Wasserhaushalt der aerophilen Grünalge Apatococcus lo-
batus. Planta 70:46–72
Borland AM, Griffiths H, Maxwell C, Broadmeadow MSJ, Griffiths NM, Barnes JD (1992) On the
ecophysiology of the Clusiaceae in Trinidad: expression of CAM in Clusia minor L. during the
transition from wet to dry season and characterization of three endemic species. New Phytol
122:349–357
Brighigna L, Fiordi AC, Palandri MR (1990) Structural comparison between free and anchored
roots in Tillandsia (Bromeliaceae) species. Caryologia 43: 27–42
Brighigna L, Montaini P, Favilli F, Trejo AC (1992) Role of the nitrogen-fixing bacterial microflora
in the epiphytism of Tillandsia (Bromeliaceae) Am J Bot 79:723–727
Bruns-Strenge S, Lange O (1992) Photosynthetische Primärproduktion der Flechte Cladonia por-
tentosa an einem Dünenstandort auf der Nordseeinsel Baltrum. III. Anwendung des Photosyn-
thesemodells zur Simulation von Tagesläufen des CO2 -Gaswechsels und zur Abschätzung der
Jahresproduktion. Flora 186:127–140
Büdel B, Meyer A, Salazar N, Zellner H, Zotz G, Lange OL (2000) Macrolichens of montane rain
forests in Panama, Province Chiriquí. Lichenologist 32:539–551
Cardelús CL, Colwell RC, Watkins JE (2006) Vascular epiphyte distribution patterns: explaining
the mid-elevation richness peak. J Ecol 94:144–156
Cochard H, Ewers FW, Tyree MT (1994) Water relations of a tropical vine-like bamboo (Rhipi-
docladum racemiflorum) root pressures, vulnerability to cavitation and seasonal changes in
embolism. J Exp Bot 45:1085–1089
Coley PD, Kursar TA, Machado J-L (1993) Colonisation of tropical rainforest leaves by epiphylls:
effect of site and host plant leaf lifetime. Ecology 74:619–623
Crayn DM, Terry RG, Smith JAC, Winter K (2000) Molecular systematic investigations in Pit-
carnioideae (Bromeliaceae) as a basis for understanding the evolution of crassulacean acid
metabolism (CAM). In: Wilson KL, Morrison DA (eds) Monocots: systematics and evolution.
CSIRO, Melbourne, pp 569–579
Crayn DM, Winter K, Smith JAC (2004) Multiple origins of crassulacean acid metabolism and the
epiphytic habit in the Neotropical family Bromeliaceae. Proc Natl Acad Sci US 101:3703–
3708
References 221
Davidson DW, Epstein WW (1989) Epiphytic associations with ants. In: Lüttge U (ed) Vascular
plants as epiphytes. Evolution and ecophysiology. Ecological studies, vol. 76. Springer, Berlin
Earnshaw MJ, Winter K, Ziegler H, Stichler W, Cruttwell NEG, Kerenga K, Cribb PJ, Wood J,
Croft JR, Carver KA, Gunn TC (1987) Altitudinal changes in the incidence of crassulacean
acid metabolism in vascular epiphytes and related life forms in Papua New Guinea. Oecologia
73:566–572
Ehleringer JR, Ullmann I, Lange OL, Farquhar GD, Cowan IR, Schulze ED, Ziegler H (1986)
Mistletoes: a hypothesis concerning morphological and chemical avoidance of herbivory. Oe-
cologia 70:234–237
Ellison AM (2006) Nutrient limitation and stoichiometry of carnivorous plants. Plant Biol 8:740–
747
Escher P, Eiblmeier M, Hetzger I, Rennenberg H (2003) Seasonal and spatial variation of reduced
sulphur compounds in mistletoes (Viscum album) and the xylem sap of its hosts (Populus x
euamericana and Abies alba).Physiol Plant 117:72–78
Escher P, Eiblmeier M, Hetzger I, Rennenberg H (2004) Seasonal and spatial variation of carbo-
hydrates in mistletoes (Viscum album) and the xylem sap of its hosts (Populus x euamericana
and Abies alba). Physiol Plant 120:212–219
Ewers FW, Fisher JB, Chiu S-T (1990) A survey of vessel dimensions in stems of tropical lianas
and other growth forms. Oecologia 84:544–552
Franco AC, Ball E, Lüttge U (1992) Differential effects of drought and light levels on accumulation
of citric and malic acids during CAM in Clusia. Plant Cell Environ 15:821–829
Freiberg ER (1994) Stickstofffixierung in der Phyllosphäre tropischer Regenwaldpflanzen in Costa
Rica. Dissertation, Ulm
Freiberg E (1998) Microclimatic parameters influencing nitrogen fixation in the phyllosphere in
a Costa Rica premontane rain forest. Oecologia 117:9–18
Freiberg E (1999) Influence of microclimate on the occurrence of cyanobacteria in the phyllosphere
in a premontane rain forest of Costa Rica. Plant Biol 1:244–252
Freiberg M (1997) Spatial and temporal pattern of temperature and humidity of a tropical premon-
tane rain forest tree on Costa Rica. Selbyana 18:77–84
Freiberg M, Freiberg E (2000) Epiphyte diversity and biomass in the canopy of lowland and mon-
tane forests in Ecuador. J Tropical Ecol 16:673–688
Galloway DJ (ed) (1991) Tropical lichens: their systematics, conservation, and ecology. The Sys-
tematics Association Spec, vol 43. Clarendon Press, Oxford, pp 275–277
Gessner F (1956) Der Wasserhaushalt der Epiphyten und Lianen. In: Ruhland W (ed) Handbuch
der Pflanzenphysiologie, Bd III. Pflanze und Wasser. Springer, Berlin Göttingen Heidelberg,
pp 915–950
Givnish TJ, Burkhardt EL, Happel RE, Weintraub JD (1984) Carnivory in the bromeliad Broc-
chinia reducta, with a cost/benefit model for the general restriction of carnivorous plants to
sunny, moist, nutrient-poor habitats. Am Nat 124:479–497
Goh CJ, Kluge M (1989) Gas exchange and water relations in epiphytic orchids. In: Lüttge U
(ed) Vascular plants as epiphytes. Evolution and ecophysiology. Ecological studies, vol 76.
Springer, Berlin Heidelberg New York, pp 137–166
Gorb EV, Gorb SN (2006) Physicochemical properties of functional surfaces in pitchers of the
carnivorous plant Nepenthes alata Blanco (Nepenthaceae). Plant Biol 8:841–848
Green TGA, Lange OL (1991) Ecophysiological adaptations of the lichen genera Pseudocyphel-
laria and Sticta to south temperate rainforests. Lichenologist 23:267–282
Green TGA, Lange OL (1994) Photosynthesis in poikilohydric plants: a comparison of lichens and
bryophytes. In: Schulze E-D, Caldwell MC (eds) Ecophysiology of photosynthesis. Ecological
studies, vol 100. Springer, Berlin Heidelberg New York, pp 319–341
Green TGA, Kilian E, Lange O (1991) Pseudocyphellaria dissimilis: a desiccation-sensitive,
highly shade-adapted lichen from New Zealand. Oecologia 85:498–503
Griffiths H (1989) Carbon dioxide concentrating mechanisms and the evolution of CAM in vas-
cular epiphytes. In: Lüttge U (ed) Vascular plants as epiphytes: evolution and ecophysiology.
Ecological studies, vol 76. Springer-Verlag, Berlin Heidelberg New York, pp 42–86
Griffiths H, Smith JAC (1983) Photosynthetic pathways in the Bromeliaceae of Trinidad: relations
between life forms, habitat preference and the occurrence of CAM. Oecologia 60:176–184
Haag-Kerwer A, Franco AC, Lüttge U (1992) The effect of temperature and light on gas exchange
and acid accumulation in the C3 -CAM plant Clusia minor L. J Exp Bot 43:345–352
Hietz P, Wanek W, Popp M (1999) Stable isotopic composition of carbon and nitrogen content in
vascular epiphytes along an altitudinal transect. Plant Cell Environ 22:1435–1443
Holbrook NM, Putz FE (1996a) Physiology of tropical vines and hemiepiphytes: plants that climb
up and plants that climb down. In: Mulkey SS, Chazdon RL, Smith AP (eds) Tropical forest
plant ecophysiology. Chapman and Hall, New York, pp 363–394
Holbrook NM, Putz FE (1996b) From epiphyte to tree: differences in leaf structure and leaf water
relations associated with the transition in growth form in eight species of hemiepiphytes. Plant
Holthe PA, Sternberg L da SL, Ting IP (1987) Developmental control of CAM in Peperomia scan-
dens. Plant Physiol 84:743–747
Horres R, Zizka G (1995) Untersuchungen zur Blattsukkulenz bei Bromeliaceae. Beih Biol
Pflanzen 69:43–76
Horres R, Zizka G, Kahl G, Weising K (2000) Molecular phylogenetics of Bromeliaceae: evidence
from trnL(UAA) intron sequence of the chloroplast genome. Plant Biol 2:306–315
Huxley C (1980) Symbiosis between ants and epiphytes. Biol Rev 55:321–340
Johansson DR (1975) Ecology of epiphytic orchids in West African rain forests. Am Orchid Soc
Bull 44:125–136
Johansson DR (1977) Epiphytic orchids as parasites of their host trees. Am Orchid Soc Bull
46:703–707
Kazda M, Salzer J (2000) Leaves of lianas and self-supporting plants differ in mass per unit area
and in nitrogen content. Plant Biol 2:268–271
Kraus R, Trimborn P, Ziegler H (1995) Tristerix aphyllus, a holoparasitic Loranthacea. Naturw
82:150–151
Kress WJ (1989) The systematic distribution of vascular epiphytes. In: Lüttge U (ed) Vascular
plants as epiphytes. Evolution and ecophysiology. Ecological studies, vol 76. Springer, Berlin
Lakatos M, Rascher U, Büdel B (2006) Functional characteristics of corticolous lichens in the
understory of a tropical lowland rainforest. New Phytol 172:679–695
Lange OL, Green TGA (1997) High thallus water contents can limit productivity of crustose
lichens in the field. In: Türk R, Zorer R (eds) Progress and problems in lichenology in the
nineties – IAL3. Bibliotheca Lichenologica 68:81–99, J. Cramer in der Gebr. Borntraeger Ver-
lagshandlg., Berlin-Stuttgart
Lange OL, Kilian E, Ziegler H (1986) Water vapor uptake and photosynthesis of lichens: per-
formance differences in species with green and blue-green algae as phycobionts. Oecologia
71:104–110
Lange OL, Green TGA, Ziegler H (1988) Water status related photosynthesis and carbon, iso-
tope discrimination in species of the lichen genus Pseudocyphellaria with green or blue-green
photobionts and in photosymbiodemes. Oecologia 75:494–501
Lange OL, Büdel B, Heber U, Meyer A, Zellner H, Green TGA (1993) Temperate rainforest lichens
in New Zealand: high thallus water content can severely limit photosynthetic CO2 exchange.
Lange OL, Büdel B, Zellner H, Zotz G, Meyer A (1994) Field measurements of water relations and
CO2 exchange of the tropical cyanobacterial basidiolichen Dictyonema glabratum in a Pana-
manian rainforest. Bot Acta 107:279–290
Lange OL, Green TGA, Reichenberger H, Hesbacher S, Proksch P (1997) Do secondary substances
in the thallus of a lichen promote CO2 diffusion and prevent depression of net photosynthesis
at high water content? Oecologia 112:1–3
References 223
Lange OL, Büdel B, Meyer A, Zellner H, Zotz G (2000) Lichen carbon gain under tropical con-
ditions: water relations and CO2 exchange of three Leptogium species of a lower montane
rainforest in Panama. Flora 195:172–190
Lange OL, Büdel B, Meyer A, Zellner H, Zotz G (2004) Lichen carbon gain under tropical condi-
tions: water relations and CO2 exchange of Lobariaceae species of a lower montane rainforest
in Panama. Lichenologist 36:329–342
López-Portillo J, Ewers FW, Angeles G, Fisher JB (2000) Hydraulic architecture of Monstera
acuminata: evolutionary consequences of the hemi-epiphytic growth form. New Phytol 145:
289–299
Lösch R, Mülders P (2000) New aspects in cryptogamic research. Contributions in honour of
Ludger Kappen. In: Schroeter B, Schlensong M, Green TGA (eds) Bibliotheca Lichenologica
75:253–263. J. Cramer in der Gebr. Borntraeger Verlagshandlg., Berlin-Stuttgart
Luther HE, Sieff E (1998) An alphabethical list of bromeliad binomials. The Bromeliac Society,
Newberg
Lüttge U (1983) Ecophysiology of carnivorous plants. In: Lange OL, Nobel PS, Osmond CB,
Ziegler H (eds) Physiological plant ecology III. Responses to chemical and biological envi-
ronment. Encyclopedia of plant physiology NS. Springer, Berlin Heidelberg New York, pp
489–517
Lüttge U (1985) Epiphyten: Evolution und Ökophysiologie. Naturwissenschaften 72:557–566
Lüttge U (1987) Carbon dioxide and water demand: crassulacean acid metabolism (CAM), a ver-
satile ecological adaptation exemplifying the need for integration in ecophysiological work.
New Phytol 106:593–629
Lüttge U (1989) Vascular epiphytes: setting the scene. In: Lüttge U (ed) Vascular plants as epi-
phytes. Evolution and ecophysiology. Ecological studies, vol 76. Springer, Berlin Heidelberg
New York, pp 1–14
Lüttge U (1991) Clusia: Morphogenetische, physiologische und biochemische Strategien von
Baumwürgern im tropischen Wald. Naturwissenschaften 78:49–58
Lüttge U (ed) (2007a) Clusia: a woody neotropical genus of remarkable plasticity and diversity.
Ecol Studies, vol 194. Springer, Berlin Heidelberg New York
Lüttge U (2007b) Physiological ecology. In: Lüttge U (ed) Clusia: a woody neotropical genus of
remarkable plasticity and diversity. Ecol Studies, vol 194. Springer, Berlin Heidelberg New
York, pp 187–234
Lüttge U, Ball E, Kluge M, Ong BL (1986) Photosynthetic light requirements of various tropical
vascular epiphytes. Physiol Vég 24:315–331
Lüttge U, Haridasan M, Fernandes GW, Mattos EA de, Trimborn P, Franco AC, Caldas LS, Ziegler
H (1998) Photosynthesis in mistletoes in relation to their hosts at various sites in tropical
Brazil. Trees 12:167–174
Lüttge U, Kluge M, Bauer G (2005) Botanik, 5. Aufl. VCH, Weinheim
Mägdefrau K (1956) Paläobiologie der Pflanzen. G Fischer, Jena
Marshall JD, Ehleringer JR (1990) Are xylem-trapping mistletoes partially heterotrophic? Oecolo-
gia 84:244–248
Martin CE (1994) Physiological ecology of the Bromeliaceae. Bot Rev 60:1–82
Martin CE, Lin T-C, Lin K-C, Hsu C-C, Chiou W-L (2004) Causes and consequences of high
osmotic potentials in epiphytic higher plants. J Plant Physiol 161:1119–1124
Martius CFP von (1840–1906) Flora brasiliensis, vol 1–15. München and Leipzig
Maxwell K (2002) Resistance is useful: diurnal patterns of photosynthesis in C3 and crassulacean
acid metabolism epiphytic bromeliads. Funct Plant Biol 29:679–687
Maxwell C, Griffiths H, Young AJ (1994) Photosynthetic acclimation to light regime and water
stress by the C3 -CAM epiphyte Guzmania monostachia: gas exchange characteristics, photo-
chemical efficiency and the xanthophylls cycle. Funct Ecol 8:746–754
Maxwell C, Griffiths H, Borland AM, Young AJ, Boradmeadow MSJ, Fordham MC (1995) Short-
term photosynthetic responses to tropical seasonal transitions under field conditions. Aust J
Plant Physiol 22:771–778
Maxwell C, Marrison JL, Leech RM, Griffiths H, Horton P (1999) Chloroplast acclimation in
leaves of Guzmania monostachia in response to high light. Plant Physiol 121:89–95
Mez C (1904) Physiologische Bromeliaceen-Studien. I. Die Wasser-Ökonomie der extrem atmo-
sphärischen Tillandsien. Jahrb Wiss Bot 40:157–229
Miller AC, Watling JR, Overton IC, Sinclair R (2003) Does water status of Eucalyptus largiflorens
(Myrtaceae) affect infection by mistletoe Amyema miquelii (Loranthaceae)? Funct Plant Biol
30:1239–1247
Nadkarni NM (1981) Canopy roots: convergent evolution in rainforest nutrient cycles. Science
124:1023–1024
Nobel PS (1983) Biophysical plant physiology and ecology. Freeman, San Francisco
Orozco A, Rada F, Azocar A, Goldstein G (1990) How does a mistletoe affect the water, nitrogen
and carbon balance of two mangrove ecosystem species? Plant Cell Environ 13:941–947
Pate JS, True KC, Kuo J (1991a) Partitioning of dry matter and mineral nutrients during a repro-
duction cycle of the mistletoe Amyema linophyllum (Fenzl.) Tieghem parasitizing Casuarina
obesa Miq. J Exp Bot 42:427–439
Pate JS, True KC, Rasins E (1991b) Xylem transport and storage of amino acids by S.W. Australian
mistletoes and their hosts. J Exp Bot 42:441–451
Patiño S, Tyree MT, Herre EA (1995) Comparison of hydraulic architecture of woody plants of dif-
fering phylogeny and growth form with special reference to free-standing and hemi-epiphytic
Ficus species from Panama. New Phytol 129:125–134
Pittendrigh CS (1948) The bromeliad-Anopheles-malaria complex in Trinidad. I. The bromeliad
flora. Evolution 2:58–89
Popp M, Richter A (1997) Ecophysiology of xylem-tapping mistletoes. Progr Bot 59:657–674
Putz FE, Holbrook NM (1986) Notes on the natural history of hemiepiphytes. Selbyana 9:61–69
Rada F, Jaimez R (1992) Comparative ecophysiology and anatomy of terrestrial and epiphytic
Anthurium bredmeyeri Schott in a tropical andean cloud forest. J Exp Bot 43:723–727
Reinert F (1998) Epiphytes: photosyntheiss, water balance and nutrients. In: Scarano FR, Franco
AC (eds) Ecophysiological strategies of xerophytic and amphibious plants in the neotropics.
Oecologia Brasiliensis, vol IV. PPGE-UFRJ, Rio de Janeiro, pp 87–108
Rey L, Sadik A, Fer A, Renandiu S (1991) Trophic relations of the dwarf mistletoe Arcenthobium
oxycedri with its host Juniperus oxycedrus. J Plant Physiol 138:411–416
Richards PW (1996) The tropical rainforest. An ecological study, 2nd edn. Cambridge University
Press, London
Richter A, Popp M, Mensen R, Stewart RG, von Willert DJ (1995) Heterotrophic carbon gain of
the parasitic angiosperm Tapinanthus oleifolius. Aust J Plant Physiol 22:537–544
Rowe NP, Speck T (1996) Biomechanical characteristics of the ontogeny and growth habit of the
tropical liana Condylocarpon guianense (Apocynaceae). Int J Plant Sci 157:406–417
Ruinen J (1953) Epiphytosis. A second view on epiphytism. Ann Bogor 1:101–157
Ruinen J (1961) The phyllosphere. I. An ecologically neglected milieu. Plant Soil 15:81–109
Ruinen J (1965) The phyllosphere. III. Nitrogen fixation in the phyllosphere. Plant Soil 22:375–395
Ruinen J (1974) Nitrogen fixation in the phyllosphere. In: Quispel A (ed) The biology of nitrogen
fixation. North Holland Publishing, Amsterdam, pp 121–167
Sallé G, Frochot H, Audary C (1993) Le gui. Recherche 24:1334–1342
Schimper AFW (1888) Botanische Mitteilungen aus den Tropen. II. Epiphytische Vegetation
Amerikas. G Fischer, Jena
Schmidt G, Zotz G (2001) Ecophysiological consequences of differences in plant size – in situ
carbon gain and water relations of the epiphytic bromeliad, Vriesea sanguinolenta. Plant Cell
Environ 24:101–112
Schmidt G, Stuntz S, Zotz G (2001) Plant size – an ignored parameter in epiphyte ecophysiology.
Plant Ecol 153:65–72
Schmitt AK, Lee HSJ, Lüttge U (1988) The response of the C3 -CAM tree Clusia rosea, to light
and water stress. I. Gas exchange characteristics. J Exp Bot 39:1581–1590
References 225
Schmitt AK, Martin CE, Lüttge U (1989) Gas exchange and water vapour uptake in the atmo-
spheric CAM bromeliad Tillandsia recurvata L.: the influence of trichomes. Bot Acta 102:80–
84
Schmucker T, Linnemann G (1959) Carnivorie. In: Handbuch der Pflanzenphysiologie, vol XI.
Springer, Berlin Göttingen Heidelberg, pp 198–283
Schnitzer SA, Dalling JW, Carson WP (2000) The impact of lianas on the regeneration in tropical
forest canopy gaps: evidence for an alternative pathway of gap-phase regeneration. J Ecol
88:655–666
Schnitzer SA, Kuzee ME, Bongers F (2005) Disentangling above- and below-ground competition
between lianas and trees in a tropical forest. J Ecol 93:1115–1125
Schreiber L, Krimm U, Knoll D, Sayed M, Auling G, Kroppenstedt RM (2005) Plant-microbe
interactions: identification of epiphytic bacteria and their ability to alter leaf surface perme-
ability. New Phytol 166:589–594
Schulze E-D, Turner NC, Glatzel G (1984) Carbon, water and nutrient relations of two mistletoes
and their hosts: a hypothesis. Plant Cell Environ 7:293–299
Seifriz W (1924) The altitudinal distribution of lichens and mosses on Mt. Gedeh, Java. J Ecol
12:307–313
Sipes DL, Ting IP (1985) Crassulacean acid metabolism and crassulacean acid metabolism modi-
fication in Peperomia camptotricha. Plant Physiol 77:59–63
Sipman HJM (1989) Lichen zonation in the Parque los Nevados transect. Stud Trop Andean
Ecosyst 3:461–483
Sitte P (1991) Morphologie. In: Sitte P, Ziegler H, Ehrendorfer F, Bresinsky A (eds) Strasburger
Lehrbuch der Botanik. G Fischer, Stuttgart, pp 13–238
Smith JAC (1989) Epiphytic bromeliads. In: Lüttge U (ed) Vascular plants as epiphytes. Evolution
and ecophysiology. Ecological studies, vol. 76: Springer, Berlin Heidelberg New York, pp
109–138
Smith JAC, Griffiths H, Lüttge U (1986a) Comparative ecophysiology of CAM and C3 bromeliads.
I. The ecology of the Bromeliaceae in Trinidad. Plant Cell Environ 9:359–376
Smith JAC, Griffiths H, Lüttge U, Crook CE, Griffiths NM, Stimmel K-H (1986b) Compara-
tive ecophysiology of CAM and C3 bromeliads. IV. Plant water relations. Plant Cell Environ
9:395–410
Speck T (1997) Ecobiomechanics: biomechanical analyses help to understand aut- and synecology
of plants. In: Jeronimidis G, Vincent JFV (eds) Plant biomechanics. Centre for Biomimetics,
Univ of Reading, Reading, pp 9–15
Stewart GR, Schmidt S, Handley LL, Turnbull MH, Erskine PD, Joly CA (1995) 15 N natural
abundance of vascular rainforest epiphytes:implications for nitrogen source and acquisition.
Tausz M, Hietz P, Briones O (2001) The significance of carotenoids and tocopherols in photo-
protection of seven epiphytic fern species of a Mexican cloud forest. Aust J Plant Physiol
28:775–783
Tietze M (1906) Physiologische Bromeliaceen-Studien. II. Die Entwicklung der wasseraufneh-
menden Bromeliaceen-Trichome. Zeitschrift für Naturwissenschaften, Halle 78:1–50
Ting IP, Bates L, O’Reilly Sternberg L, DeNiro MJ (1985) Physiological and isotopic aspects of
photosynthesis in Peperomia. Plant Physiol 78:246–249
Treseder KK, Davidson DW, Ehleringer JR (1995) Absorption of ant-provided carbon dioxide and
nitrogen by a tropical epiphyte. Nature 375:137–139
Walter H, Breckle S-W (1984) Ökologie der Erde, vol. 2. Spezielle Ökologie der tropischen und
Wanek W, Pörtl K (2005) Phyllosphere nitrogen relations: reciprocal transfer of nitrogen between
epiphyllous liverworts and host plants in the under-story of a lowland tropical wet forest in
Costa Rica. New Phytol 166:577–588
Wania R, Hietz P, Wanek W (2002) Natural 15 N abundance of epiphytes depends on the position
within the forest canopy: source signals and isotope fractionation. Plant Cell Environ 25:581–
589
Winter K, Wallace BJ, Stocker GC, Roksandic Z (1983) Crassulacean acid metabolism in Aus-
tralian vascular epiphytes and some related species. Oecologia 57:129–141
Ziegler H (1986) Control of photosynthesis by variation of diffusion resistance in mistletoes and
their hosts. In: Marcello R, Clijsters H, Poucke M van (eds) Biological control of photosyn-
thesis. Advances in agricultural biotechnology. Martinus Nijnhoff, Dordrecht, pp 171–185
Zotz G, Andrade J-L (1998) Water relations of two co-occurring epiphytic bromeliads. J Plant
Physiol 152:545–554
Zotz G, Hietz P (2001) The physiological ecology of vascular epiphytes: current knowledge, open
questions. J Exp Bot 52:2067–2087
Zotz G, Winter K (1994a) Photosynthesis and carbon gain of the lichen, Leptogium azureum, in
a lowland tropical forest. Flora 189:179–186
Zotz G, Winter K (1994b) Annual carbon balance and nitrogen-use efficiency in tropical C3 and
CAM epiphytes. New Phytol 126:481–492
Zotz G, Ziegler H (1997) The occurrence of crassulacean acid metabolism among vascular epi-
phytes from Central Panama. New Phytol 137:223–229
Zotz G, Tyree MT, Cochard H (1994) Hydraulic architecture, water relations and vulnerability to
cavitation of Clusia uvitana Pittier: a C3 -CAM tropical hemiepiphyte. New Phytol 127:287–
295
Zotz G, Patiño S, Tyree MT (1997) Water relations and hydraulic architecture of woody hemi-
epiphytes. J Exp Bot 48:1825–1833
Zotz G, Büdel B, Meyer A, Zellner H, Lange OL (1998) In situ studies of water relations and CO2
exchange of the tropical macrolichen, Sticta tomentosa. New Phytol 139:525–535
Zotz G, Hietz P, Schmidt G (2001) Small plants, large plants: the importance of plant size for the
physiological ecology of vascular epiphytes. J Exp Bot 52:2051–2056
Zotz G, Enslin A, Hartung W, Ziegler H (2004) Physiological and anatomical changes during the
early ontogeny of the heteroblastic bromeliad, Vriesea sanguinolenta, do not concur with the
morphological change from atmospheric to tank form. Plant Cell Environ 27:1341–1350
Chapter 7
Tropical Forests. V. Mangroves
7.1 Phytogeography
Mangroves are a characteristic and important type of tropical and subtropical

forests, with a unique capacity to tolerate large short-term changes of salinity. The
name comes from the Spanish “mangle” for Rhizophora, a mangrove genus, and the
English “grove”. Mangroves may also be considered as “tide-forests”, since their
ecology is determined primarily by the tides at the three typical sites where they
occur (Fig. 7.1):
• coastal mangroves,
• estuarine mangroves and
• coral mangroves,
i.e. mangroves along the coastlines, in river estuaries and around coral reefs and
coral islands. However, salinity in mangroves is not only influenced by the tides,
but also by the climate. At high tide salinity in the rooting medium of mangroves,
of course, will be determined by sea water. At low tide, however, it will be higher
or lower depending on the climatic conditions, i.e.
• humid climate with rainfall frequently diluting and leaching salt,
• arid climate with salt normally concentrated,
so that in any case mangrove sites are characterized by conditions of very variable
salinity, which may even change rhythmically. Climate impacts on mangrove trees
are also reflected in an annual cyclicity. Although as tropical trees mangrove trees
lack distinct growth rings, high resolution profiles of stable carbon isotope ratios
(δ 18O, δ 13 C) in the wood of the stems reveal a seasonal cyclicity related to phys-
iological processes under the environmental driving forces of salinity and water
Fig. 7.1A–F Coastal mangroves of the Morocoy National Park near Tucacas at the Caribbean
coast of Venezuela (A, B) and in Queensland, Australia (C). Estuary mangroves in Trinidad (D)
and Costa Rica (E Las Baulas, F Rio Parrita)
228 7 Tropical Forests. V. Mangroves
7.1 Phytogeography 229
Fig. 7.1 Continued

potential of soils and other factors such as relative air humidity (Verheyden et al.
2004, 2005).
Mangroves delimit most tropical coast-lines and also extend into the subtropics
(Fig. 7.2), such that 60 – 75% of all tropical coast-lines are occupied by mangroves
(Popp 1991). The area covered by mangroves is 140,000 km2 , which is about 0.1%
of the total land surface of the earth. The total global biomass of mangroves is
estimated to be 8.7 gigatons dry weight (Twilley et al. 1992). There are about 50 – 75
Table 7.1 Na+ and Cl− levels in different mangrove species collected from all over the world.
(Popp et al. 1984)
7.1 Phytogeography 231
Fig. 7.2 Global distribution of mangroves with numbers indicating the approximate number of
tree and shrub species in the mangrove vegetation. (After Vareschi 1980, with kind permission of
R Ulmer)
Fig. 7.3 Co-occurring mangrove species in the Morocoy Park of Venezuela (see Fig. 7.1A,B):
Avicennia germinans, Laguncularia racemosa, Rhizophora mangle
different species of mangroves in 20 – 26 genera in 16 – 20 families (Ellison 2002),

some of which are depicted in Fig. 7.3 and listed in Table 7.1. Floristic diversity is
poor in the Americas (1 – 5 tree species) and also in Africa (four species in West
Africa, eight species in East Africa and Madagascar) but quite respectable in Asia
(about 25 species in India and 30 species in SE-Asia), although in general terms
mangroves are floristically much poorer than other tropical forests.
7.2 Site Characteristics and Contrasts in Salinity
Mangroves are characterized by their trees. Trees in mangrove forests may be-
come quite tall although often mangroves have a scrub-like physiognomy (e.g. com-
pare Fig. 7.1B and F). The woody mangrove species are frequently distributed in
a banded zonation pattern oriented in parallel to the shore line. This pattern is cor-
related with the frequency and duration of tidal immersion modulating the degrees
of salinity stress, and it is also influenced by dispersal of propagules, competition
among mangrove species and herbivory (Ball 2002). Taller trees are formed in the
Fig. 7.4 The mangrove fern

Acrostichum aureum (Costa
Rica)
7.3 Morphological Characteristics of the Mangrove Tree Life Form 233
fringe forest near the water’s edge and dwarf forms further inland at higher eleva-
tion in the intertidal zone (Cheeseman and Lovelock 2004; Lovelock et al. 2006a;
see also Sect. 7.6). Lin and Sternberg (1992a,b, 1993) have compared scrub and
tree life-forms of the mangrove species Rhizophora mangle with trees in the fringe
forest at lower levels (24 cm above sea level) and the scrub formation at higher lev-
els (60 cm a.s.l.). The scrub form is associated with high salinities occurring at the
higher levels during the dry season. In the rainy season, the scrub mangroves can
also take up fresh water from rain, and R. mangle is therefore a facultative halo-
phyte. However, frequent stress is caused by the changes in salinity following shifts
between flooding by ocean water and fresh water introduced by rain. Such varia-
tions can lead to a significant decrease in photosynthesis and plant growth in the
scrub mangroves, in contrast to constant salinity which maintains the salt load in
the substratum.
On the silty substrate of mangroves the undergrowth of vascular plants is usually
poor (Ball 1996) although the vigorous growth of large terrestrial ferns of the genus
Acrostichum often is a striking feature (Fig. 7.4). These “mangrove ferns” are shade-
tolerant plants, which, however, have their maximum development and productivity
under full exposure. Acrostichum aureum is quite salt tolerant, although perhaps
somewhat less than the mangrove trees. However, the gametophytes are only resis-
tant to mild salinity stress and can survive the full salinity of sea water only for
short periods, so that establishment is a problem and the fern remains restricted to
the landward side of mangrove swamps (Medina et al. 1990; Li and Ong 1998; Sun
et al. 1999).
7.3 Morphological Characteristics

of the Mangrove Tree Life Form
At the littoral habitats the wood of the trunks of mangrove trees needs to resist
particularly strong winds as well as the pressure of tides. The major stress factors
related to morphological characteristics of the mangrove tree life form, however, are
salinity and the additional stress of low O2 -partial pressure resulting from the inun-
dation of the silty hypoxic substratum in which they root. This causes particular
demands on morphology of roots for aeration and hydraulic architecture for lifting
water to the shoots against the low water potential of the saline substratum.
7.3.1 Hypoxia in Inundated Swampy Soils, Root Morphology

and Aeration
The root systems of mangrove trees are most remarkable. In comparison to other
tropical forest communities root biomass is greater in mangroves (Ball 1996).
Highly conspicuous is the diverse range of strangely shaped bizarre root systems
above the soil surface. They must have evolved to provide anchorage as well as
aeration in the silty muddy soils. With respect to the latter function they have been
named pneumatophores. Diffusion of gases is highly limited in the inundated soil.
Therefore, only contact of the root system with the atmosphere or with the sea water,
depending on tidal level allows gaseous exchange. Figures 7.5 and 7.6 show a va-
riety of some of the most frequently observed aerial root systems with stilt roots,
planks and buttresses and finger- or knee-like protrusions above ground.
The root aeration provided by these pneumatophores is reinforced by a physio-
logical mechanism. The exposed parts of these roots usually have lenticels, which
Fig. 7.5A–C Root types of mangroves. A Stilt-, B Buttress-type roots. C Knee- or finger-like
pneumatophores. (Vareschi 1980, with kind permission of R. Ulmer)
Fig. 7.6A–D Stilt roots of Rhizophora mangle (A–C) and finger-like pneumatophores of Avicennia
germinans (D)
7.3 Morphological Characteristics of the Mangrove Tree Life Form 235
are openings in the bark where gas but not water can penetrate. The influx of water
is prevented by surface tension in the intercellular spaces of the lenticels. A pneu-
matophore aerenchyma may occupy as much as 70% of total root volume (Curran
1985). During high tide, root respiration reduces the O2 -concentration in the inter-
cellular spaces of the root aerenchyma to a hypoxia of as little as 4 – 8% (Kitaya
et al. 2002). Photosynthetically active cells at the surface of the pneumatophores
can feed some O2 into the aerenchyma for root respiration (Aiga et al. 1995; Ki-
taya et al. 2002), but pneumatophore photosynthesis is decreased with flooding due
to reduced irradiance (Kitaya et al. 2002). Thus, the low O2 -concentration in the
submerged pneumatophores of only a quarter to less than half of the atmospheric
concentration causes considerable O2 gradients along the roots and gas-pressure
deficits in the aerenchyma to minus 1.7 kPa (Chiu and Chou 1993; Skelton and
Allaway 1996; Youssef and Saenger 1996; Lösch and Busch 1999), because the O2
consumed in respiration can not be reabsorbed readily from the sea water, while
the CO2 liberated can be dissolved as bicarbonate and released. At low tide, when
the roots establish contact with the air again, the pressure deficit effectively leads
to air being sucked into the root air spaces via the lenticels. Hypoxia (see also
Sect. 3.2.3) is an additional stress to salinity in mangrove trees and, in view of
the energy costs of salinity tolerance (e.g. salt exclusion, K+ /Na+ -selectivity, see
Sects. 7.4 and 7.6), such mechanisms for the control of hypoxia at the root level are
quite important in addition to ventilation from the photosynthesizing shoot via the
aerenchyma.
7.3.2 Hydraulic Architecture and Xylem Sap Flow
For assessment of water potential gradients in plants as driving forces for xylem sap
flow in trees S CHOLANDER and coworkers introduced the pressure chamber tech-
nique. In view of the particular situation of mangroves in their saline substratum
they devoted much of their pioneering work to mangrove trees, where xylem ten-
sions of 3.8 – 5.2 MPa were recorded and accepted as sufficiently exceeding the os-
motic pressure of sea water (2.5 MPa) to produce the driving force for sap ascent and
a transpiration stream according to the cohesion-tension theory (Scholander 1968;
Scholander et al. 1965, 1966). Conversely, Zimmermann et al. (1994a, 2002) argue
that the balancing pressure used in the pressure chamber technique overestimates
xylem tension. Using staining techniques they detected high-molecular-weight poly-
meric polysaccharide-mucilage in the xylem vessels of Rhizophora mangle. They
argue that such mucilage would tend to strongly support gas bubble formation,
which would prevent stable xylem tensions larger than 0.1 MPa. Moreover, the mu-
cilage would hinder a mass flow of water. Xylem conductivity would decrease with
saps of extreme salinities controlled by swelling and shrinking of pectin-based hy-
drogels in the pit membranes (López-Portillo et al. 2005), but also reduced leaf
conductivity for water vapour with increasing salinity is involved due to stomatal
regulation (Sobrado 2001). Zimmermann et al. (1994a, 2002) observed sap flow
rates in mangrove trees of 0.05 – 0.14 mm s−1 which they consider to be quite low.
7.4 Exclusion, Inclusion and Excretion of Salt 237
They suggest then, that sap ascent is driven by gravity-independent streaming at

gas/water interfaces (given by the gas bubbles) as well as a gradient of chemical ac-
tivity of water established by the potentially highly hygroscopic mucilage attracting
and holding the water. This raised much dispute in relation to the standard cohesion-
tension theory of sap flow in the xylem (Zimmermann et al. 1994b; Tyree 1997;
Lösch 1998; Wei et al. 1999a,b), and although this is not the place to get involved
in detail in this controversy it is noteworthy that mangroves are important players
in this game. Thus, Becker et al. (1997) note that xylem clogging by mucilage in
mangroves cannot be generalized and they report sap flow rates in mangrove trees
of 0.09 – 0.16 mm s−1 , which they think are not that low and compare well with
those of other tropical trees. We may conclude with Becker et al. (1997) that “like
plants of other vegetation types, mangrove species will probably exhibit a range of
transpirational behaviours in response to their saline habitat once they have been
more fully investigated”.
Hydraulic architecture plays a large role in such comparisons (Ball 1996). So-
brado (2000) found that the hydraulic systems of the three mangrove species Avicen-
nia germinans, Laguncularia racemosa and Rhizophora mangle were comparable to
the lowest end of the range reported for tropical trees. Wood and bark anatomy are
adapted to water availability, salinity and oxygen supply in relation to the frequency
and duration of flooding periods (Yáñez-Espinoza et al. 2001). Specific hydraulic
conductivity of leaves declines with increasing salinity (Lovelock et al. 2006b). Un-
der salinity stress controlled by the phytohormone auxin trees tend to form xylem
vessels with smaller diameters which also applies to mangroves (Junghans et al.
2006). Mangrove species of the Rhizophoraceae have smaller vessel diameters than
non-mangrove species of the same family (Janssonius 1950).
7.3.3 Vivipary
Some mangrove species are viviparious (Fig. 7.7). After fertilization they develop
from the zygotes as embryos and then seedlings, which grow out of the flowers and
fruits and remain for some while on the mother plant. Once liberated the viviparious
seedlings can establish directly in the sediment at low tide or float in the sea water
and are dispersed. Establishment appears to be particularly important since traits
related to it appear to be stronger predictors of distribution than those associated
with dispersal (Clarke et al. 2001). In general, however, advantages of vivipary are
not clear since it is observed in only some mangrove tree taxa (e.g. Rhizophora
mangle).
7.4 Exclusion, Inclusion and Excretion of Salt
Mangroves, like all other halophytes (which are plants growing in saline habitats),
utilize strategies where they function as
Fig. 7.7 Vivipary in Rhi-

zophora mangle
• salt excluders or
• salt includers with intracellular salt dilution (succulence) and compartmenta-
tion,
and in addition operate with
• salt excretion
(Popp et al. 1993).
Salt exclusion normally only affords resistance against mild or intermediate salin-
ity stress, mainly for osmotic reasons (see Box 6.1). In order to maintain osmotic
balance and keep a water potential gradient from the substratum to the plants, salt
excluding plants would have to synthesise alternative organic solutes, which would
consume energy and tie-up a large amount of important resources in terms of carbon
skeletons, nitrogen and sulphur (see Box 7.1). Thus, the alternative is salt inclusion
whereby the salt itself is used as a readily available and “cheap” osmoticum. In
species which have special salt glands on their leaves, surplus salt may be elimi-
nated by salt excretion.
The relative effectiveness of these mechanisms is illustrated by comparing salt
levels in the leaves and in the xylem sap of mangrove species with and without salt
glands (Fig. 7.8). Species with salt glands appear to have higher salt concentrations
in the xylem sap than species without salt glands. Analyses in the field suggest that
Rhizophora mucronata, is a salt excluder as shown by the rather low Cl− -levels in
the xylem sap while Aegialitis annulata is a salt includer as indicated by the larger
xylem sap Cl− -concentration. In contrast to the salt excluder, A. annulata has salt
glands and is capable of salt excretion. Irrespective of the large differences in xylem
sap salt concentrations adult leaves had similar salt levels in both species. Hence,
the different strategies for dealing with salt, whether by exclusion at the root level or
excretion at the leaf level and dilution via succulence (see below) lead to the same
salt level in leaves.
However, the distinction between salt excluders and salt includers is only rela-
tive. There is always some control of salt uptake at the root level. This is the case in
all mangroves, and the salt concentration in the xylem sap is always much smaller
than in seawater, where the Cl− concentration is over 500 mM (Scholander 1968;
Fitzgerald and Allaway 1991). The levels of Na+ and Cl− in the xylem sap of the
examples shown in Fig. 7.8 are five to more than ten times less than those of sea wa-
ter, in contrast to the levels in leaves. Table 7.1 gives a compilation of NaCl-levels
in 23 different mangrove species collected from all over the world. Generally, the
salt concentrations in the leaves were similar to that of seawater. Deviations of
tissue contents of Na+ and Cl− from the average contents of these ions in sea water
are small and mostly not larger than ca. ±100 mM although in some cases devi-
ations of ca. +200 mM and ca. −400 mM have been reported (Table 7.1). This is
Fig. 7.8 Na+ and Cl− concentrations in leaves (white columns) and xylem sap (X, black columns)
of mangrove tree species analysed in the field (Rhizophora mucronata, Aegialitis annulata: Atkin-
son et al. 1967) and grown in sea water in a glass house (Laguncularia racemosa, Aegiceras cor-
niculatum: Polanía 1990), respectively, with and without salt glands, respectively, in their leaves.
(From Lüttge 2002)
due to accumulation of salt from the lower concentrations in the xylem sap into the
leaf cells which is fast during leaf expansion but continues gradually in mature and
senescing leaves (Cram et al. 2002). However, a comparison of salt concentrations
in the xylem sap and in the sea water rooting medium shows that strictly speaking
at the root level all mangroves are salt excluders. This can add to salinization of
the substratum, which, may have ecophysiological implications for photosynthetic
CO2 -uptake and transpiration (Passioura et al. 1992; see Sect. 7.5.1). Salt accu-
mulation in the leaves almost equally affects both ions Na+ and Cl− with a small
tendency to a larger Cl− accumulation. While the average Cl− /Na+ -ratio in sea wa-
Fig. 7.9A, B Cross-sections of a young (A) and a mature (B) leaf of the mangrove Sonneratia
sp. The mature leaf is much thicker, having a much larger water content to area ratio (= leaf
succulence) due to enlargement and high vacuolization of the inner mesophyll cells. (Lear and
Turner 1977, with kind permission of University of Queensland Press)
ter is ca. 1.2, the Cl− /Na+ -ratios of the mangrove trees shown in Table 7.1 average
at 1.4 ± 0.2, with the exception of the three species where the salt content was ca.
400 mM less than that of sea water.
Many mangrove tree species can also grow in fresh water and behave as facul-
tative halophytes. As in other halophytes, up to a certain level salinity stimulates
growth, but high salinities inhibit growth to different extents in different mangrove
species (Ball 1996, 2002). The optimum salt concentration for growth may be well
below the NaCl-concentration of sea water, e.g. in the mangrove tree Avicennia ger-
minans it was found to be at 170 mM and higher concentrations (680 and 940 mM)
were inhibitory (Suárez and Medina 2005). Thus, there is a range of comportments
from moderate to high salt tolerance and obligate halophily (Ball 1996).
Salt accumulation as a consequence of salt inclusion and the concentrating effect
of transpiration has important correlates at the cellular level, namely salt compart-
mentation and dilution. Salt is sequestered (compartmented) in the cell sap vac-
uoles where it can be diluted by osmotic uptake of water. However, this requires an
increased volume if the overall salt concentration were to be maintained at a con-
stant level. Therefore, such salt dilution is associated with succulence (“salt succu-
lence”), with the formation of large central vacuoles. This supports maintenance of
water relations and turgor pressure according to the relationship
ψ = P − π , (7.1)
where ψ is water potential, P turgor potential and π osmotic potential (see Box 6.1).
Succulence of mangrove leaves may increase as leaves age (Cram et al. 2002), and
Fig. 7.10 Cl− content

(mol m−2 leaf surface) and
Cl− concentration (mol l−1
tissue water at saturation) in
leaves of the mangrove La-
guncularia racemosa related
to the degree of succulence.
The latter is given by the ra-
tio of the leaf-water content
at water saturation and the
surface of both sides of the
leaves (kg m−2 ). (After Biebl
and Kinzel 1965, from Kinzel
the author and R. Ulmer)
this is mainly due to an enlargement of leaf cells, providing larger vacuoles in which
salt can be accumulated and diluted to some extent (Fig. 7.9). Thus, the total chlo-
ride content of leaves, when expressed on a leaf area basis, increases considerably,
whereas chloride concentration remains rather constant as succulence increases.
This clearly demonstrates the dilution effect enabled by succulence (Fig. 7.10).
An anatomical disadvantage of salt succulence is a reduction of CO2 -diffusion
within the leaves to the chloroplasts decreasing photosynthetic capacity (Parida et
al. 2004a).
Box 7.1 Compatible Solutes
As for all halophytes, the cytoplasm, the enzymes and membranes of mangrove
cells are as equally sensitive to higher Na+ -concentrations as those of glycophytes
(Ball and Anderson 1986; Sommer et al. 1990). Due to their large hydration shells,
Na+ ions disturb the molecular water structures, i.e. the specific arrangement of
the dipole molecules of H2 O at the surfaces of proteins and membranes. This leads
to the requirement for compartmentation. The NaCl taken up is sequestered in
the vacuoles, where it is accumulated and may be effectively diluted as shown
above. However, an osmotic balance is required in the cytoplasm because turgor
pressure (Eq. 7.1) can only build up at the plasmalemma/cell wall boundary. The
tonoplast membrane itself does not offer enough elastic resistance, and there can-
not be a gradient of π across the tonoplast, i.e. πcytoplasm must equal πvacuole. To
this end, halophytes normally synthesise small organic molecules which serve as
osmolytes, and are also called compatible solutes, as they both serve as cytoplas-
mic osmotica and are compatible with water structures. Their function in stabiliz-
ing cytoplasmic water structures is based on their molecular electron and charge
distribution being similar enough to water-dipoles to be compatible with mainte-
nance of cytoplasmic structures. Box 7.1 presents a variety of compounds, which
are known to function as compatible solutes. Sorbitol, mannitol and pinitol are par-
ticularly frequent among mangroves (Popp 1984; Popp and Polanía 1989; Richter et
al. 1990). Accumulation of compatible solutes in the cytoplasm alone is much more
efficient in terms of resources and energy needed than if organic molecules were
used throughout the whole cell as osmotica to withstand salt stress of the medium.
In succulent tissues the relative volume of the cytoplasm is only 1 – 2% of the to-
tal cell volume, so that vacuolar salt accumulation accompanied by cytoplasmic
accumulation of compatible solutes is a very cost effective mechanism of osmotic
adjustment.
Fig. 7.11A–F Development of the salt gland hairs of the mangrove Avicennia marina. A–E various
stages of development. F Mature salt gland (Fahn and Shimony 1977, with kind permission of
the author and Linnean Society). A Terminal cell; Ba basal cell; C cuticle; E epidermal cell; S
secretory cell; St stalk cell; W cell wall
Fig. 7.12 Leaves of Avicennia germinans with salt crystals (above) and dissolving salt at high air
humidity (below)
The mechanism of salt excretion by glands has been studied extensively in non-
tropical halophytes (Lüttge 1975). It is an energy dependent, active transport pro-
cess, moving ions against large gradients of their electro-chemical potential. Fig-
ure 7.11 shows the development of the glandular hairs of the mangrove Avicennia
marina (Fahn and Shimony 1977). The mature salt gland is covered and encircled
by a cuticle, so that an apoplastic, cell-wall route of salt excretion is not available
(Fitzgerald and Allaway 1991). The salt is moved via basal cells, often called “col-
lecting cells”, and stalk cells to the secretory cells, which excrete it into a subcutic-
ular space at the head of the gland. Water follows osmotically. The pressure of the
excreted fluid increases in the subcuticular space, and the salt is eventually released
through pores in the cuticle opening under the hydrostatic pressure. During hot and
dry days numerous salt crystals form on the leaves as the excreted salt solution dries
(Fig. 7.12). Conversely, in the early morning, when air humidity is high, the excreted
salt on the leaf surface hygroscopically absorbs water and a salty “rain” may drip
down from the mangrove trees.
Excretion is also under adaptive regulation. With increases of xylem osmolality
due to drought in Avicennia germinans (Sobrado 2002) or salinity in Laguncularia
racemosa (Sobrado 2004) excretion tends to rise exponentially (Fig. 7.13), and at
Xylem osmolality ( mol m -3 )
1.0 50
Excretion ( mmol-2 d-1 )
0.5 25
0 0
0 10 20 30
Salinity ( ‰ )
Fig. 7.13 Osmolality of xylem sap (triangles) and rates of salt excretion (circles) by Laguncularia
racemosa in relation to salinity (in ‰ sea water strength). Open and closed symbols are observa-
tions in the field and in the laboratory, respectively. (After data of Sobrado 2004)
the same time leaf conductance for water vapour decreases hyperbolically providing
a trade-off between enhancement of salt excretion and control of water loss (Sobrado
2002, 2004).
7.5 Photosynthesis
7.5.1 CO2 -Exchange and Stomatal Conductance
Measurements of gas exchange of mangrove trees in relation to the degree of sub-

stratum salinity are summarized in Fig. 7.14, where for comparison the units of
reference of different authors were unified so that salinity is roughly indicated as
that of 1/10, 1/2 and 1/1 of sea water. (This also applies to Sect. 7.5.2 below.)
An effect of salinity on net CO2 -uptake ( JCO2 ) is not very pronounced up to 1/2-
strength of sea water; only Avicennia corniculatum seems to be more sensitive than
Avicennia marina and the 19 species averaged. Full strength (1/1) sea water then re-
duces JCO2 as stomatal conductance of the leaves (gH2 O ) is also declining, but these
i
effects are not dramatic. Internal CO2 -partial pressure ( pCO ) remains between 150
2
i
and 250 Pa/MPa. The increase of pCO2 in A. corniculatum, while stomata partially
closed (reduced gH2 O ) and JCO2 strongly decreased at 1/1-strength sea water can be
explained by photoinhibition (see Sect. 7.5.3) preventing fixation of internal CO2 .
In any of the cases shown in Fig. 7.14 comparatively high rates of photosynthesis
are still maintained at full strength sea water. These rates compare well with rates
of glycophytic C3 -plants and even C4 -plants in the absence of salinity (Table 7.2),
which underlines the strong capacity of mangroves to perform effectively under high
salinity.
Comparative studies of Conocarpus erectus and Avicennia germinans during the
rainy season and the dry season offer additional insights into the success of man-
Table 7.2 Maximum rates of net-CO2 -uptake (JCO2 ) and water-use-efficiency ratios (WUEratio )
for mangrove trees at various salinities (as summarized in Figs. 7.14 and 7.19) in comparison to
C3 -, C4 - and CAM-plants (Black 1973)
Maximum JCO2 WUEratio × 103

(µmol m−2 s−1 )
Mangroves
1/10 sea water 11 – 14 5 – 16
1/2 sea water 8 – 14 4 – 14
1/1 sea water 4 – 10 2 – 12
C3 -plants 10 – 25 0.6 – 1.3
C4 -plants 25 – 50 1.7 – 2.4
CAM-plants
Darkness 0.5 – 2.5 6 – 30
Light 7–8 1–4
7.5 Photosynthesis 247
Fig. 7.14 Net CO2 -uptake

(JCO2 ), leaf conductance for A.marina A.corni- 19 spec.
15
J CO ( μmol m-2 s-1 )

water vapour (gH2 O ) and leaf culatum
internal CO2 partial pressure
i
( pCO ) of the mangrove tree
2 10
species Avicennia marina
and Avicennia corniculatum
(after Ball and Farquhar 5
2
1984a,b) and of 19 different
1/ 1 1 1/ 1 1 1/ 1/
species studied in the field 10 / 2 / 1 10 / 2 / 1 2 1
in Australia and Papua New 0
Guinea (after Clough and
( mmol m-2 s-1 )

300
Sim 1989) at approximately
1/10-, 1/2- and 1/1-strength
sea water as indicated in the 200
columns. (From Lüttge 2002)
100
1/ 1 1 1/ 1 1 1/ 1/
2O
10 / 2 / 1 10 / 2 / 1 2 1
piCO ( Pa / MPa ) gH
0
300
200
100
2
1/ 1/
10 1 / 2 1 / 1 10 1 / 2 1 / 1 1/
2
1/
1
0
groves under strongly varying conditions of salinity. C. erectus is not a true man-
grove, but rather a mangrove associate or ally. It does not grow as close to salt-water
lagoons and estuaries or tidal plains as the mangroves sensu strictu. However, at
places its distribution overlaps with that of mangroves, for instance in alluvial sand
plains at the Caribbean coast of Venezuela. There are various sizes of vegetation
islands on these sand plains (see Sect. 8.2.1), where C. erectus and A. marina are
found in close proximity (Fig. 7.15). In the rainy season the sand plains may be
flooded by fresh water to a depth of 0.5 m, but in the dry season they dry out be-
coming hypersaline and covered with a crust of salt (Sects. 8.2.1 and 8.2.3.4). These
two species were found here growing on the same vegetation island, and Fig. 7.16
shows that in the rainy season, both species had similar CO2 -uptake rates, JCO2 , but
that A. germinans operated at considerably lower conductance, gH2 O , and internal
CO2 -concentration, pCO i , than C. erectus. In the dry season, JCO2 in the morning
2
was similar to that measured in the wet season for A. germinans. There was a mid-
day depression (see Sects. 5.2.2.1 and 10.1.2.3), which was followed, however, by
considerable recovery in the afternoon; pCO i was similar to that in the wet season
2
although gH2 O was somewhat reduced. Conversely, CO2 -uptake in C. erectus in the
dry season was greatly reduced, with only a small peak in the morning, and gH2 O
i
and pCO were low throughout the day. It is evident that with similar JCO2 for both
2
species in the rainy season, the smaller reduction of JCO2 in the dry season allowed
A. germinans to maintain productivity under salinity-stress and drought better than
Fig. 7.15A,B Avicennia germinans (A) and Conocarpus erectus (B) on an alluvial sand plain at the
Caribbean coast of Venezuela
the mangrove-associate C. erectus as the true mangrove maintained similar rates of

i
JCO2 to C. erectus at lower pCO in the rainy season and then JCO2 in C. erectus was
2
i
greatly reduced as pCO2 declined in the dry season (Fig. 7.16).
Although C. erectus does not grow close to the shoreline or reach the tidally in-
undated mud plains, it may grow around sand dunes. It is then subject to salinity
from salt spray and develops very thick succulent leaves at the windward side of the
Fig. 7.16A,B Leaf conductance for water vapour, gH2 O , net CO2 -uptake, JCO2 , and internal CO2
i
partial pressure, pCO , in leaves of the same plants of Avicennia germinans (A) and Conocarpus
2
erectus (B) studied during the rainy season (◦) and the dry season (•). (After Smith et al. 1989)
bushes, while leaves on the sheltered side are non-succulent (Fig. 7.17). The non-
succulent leaves protected from the salt spray have higher JCO2 and transpiration,
JH2 O , during the second part of the day than the salt-exposed leaves (Fig. 7.18).
Similar observations were described by Naidoo et al. (2002) who compared the
mangrove associate Hibiscus tiliaceus with the true mangroves Avicennia marina
Fig. 7.17A,B Wind shaped bushes of Conocarpus erectus on sand dunes of the Paraguana Penin-
sula (A) and Chichiriviche (B) on the Caribbean coast of Venezuela
and Bruguiera gymnorhiza at sites with low and high salinities, respectively, and
found that H. tiliaceus showed better photosynthetic performance than the man-
groves at the low salinity site and vice versa at the high salinity site.
Fig. 7.18 Net CO2 -uptake,

JCO2 , and transpiration, JH2 O ,
of Conocarpus erectus on
coastal sand dunes. • Succu-
lent leaves on the windward
side of the bushes, ◦ sheltered,
on the leeward side. (Smith et
al. 1989)
7.5.2 Water Use Efficiency
In relation to salinity and osmotic stress of mangrove trees water-use-efficiency

(WUE) is of great interest. The WUEratio , as defined by JCO2 /JH2 O , decreases with
salinity only slightly in A. marina and more pronouncedly at 1/1-strength sea wa-
ter in A. corniculatum. WUEratio is not only affected by substrate salinity but also
by leaf-to-air water vapour pressure difference (VPD) with considerable decreases
as VPD increases (Fig. 7.19). However, Ball (1986) argues that notwithstanding
these reductions of WUEratio the values observed remain exceptionally high. In-
deed, the WUEratio values summarized for all conditions of salinity and VPD given
in Fig. 7.19 are higher than in glycophytic C3 -plants as well as C4 -plants, and most
remarkably they are in the same range as obtained for the highly water saving noc-
turnal CO2 -uptake by CAM plants (Table 7.2).
Since stomatal control affects JCO2 and JH2 O and hence WUEratio a calculation of
intrinsic WUE, WUEintrinsic, is performed to include a consideration of the driving
a , and internal, p i
forces for JCO2 , i.e. the difference between external, pCO2 CO2 , CO2 -
partial pressure, and for JH2 O , i.e. VPD, and also the CO2 -compensation point of
photosynthesis, ,
a
pCO − pCO
i
WUEintrinsic = 2 2
(7.2)
a
pCO2
−
Fig. 7.19 Water-use- A.marina A. corniculatum

efficiency ratios (WUEratio )
of Avicennia marina and 15 6
Avicennia corniculatum at ap-
WUE - Ratio x103

6
proximately 1/10-; 1/2- and
1/1-strength sea water salin-
ity and different leaf-to-air 10
vapour pressure differences 12
(VPD) as indicated by the
12
numbers (in Pa/kPa) in the
5 24 24
graphs. (After Ball and Far-
quhar 1984a; from Lüttge
2002)
0
1/ 1/ 1/ 1/ 1/ 1/
10 2 1 10 2 1
strength sea water
and since
1.6 JCO2
i
pCO2
= pCO
a
2
− (7.3)
gH2 O
JH2 O
gH2 O = (7.4)
VPD
a
pCO2
− pCO
i
2
= 1.6 WUEratio VPD , (7.5)
intrinsic water use efficiency is obtained as

1.6 WUEratio VPD
WUEintrinsic = . (7.6)
a
pCO2
−
(The factor 1.6 accounts for the ratio of diffusivities of water vapour to CO2
in air.) The results show that WUEintrinsic increases with increasing salinity and
VPD (Fig. 7.20) and suggest improved use of water as stomata partially close and
VPD increases. This supports the conclusions reached from considering WUEratio
(Fig. 7.19), because assuming constant pCO a
2
and (denominator in (7.6))
WUEintrinsic for the values in Fig. 7.19 in A. marina at all salinities would increase
from 6 Pa/kPa to 24 Pa/kPa VPD by a factor of 1.3 to 1.5 and in A. corniculatum at
1/10- and 1/2-strength sea water by a factor of 1.25. At 1/1-strength sea water and
VPD from 6 Pa/kPa to 12 Pa/kPa the increase in WUEintrinsic would be by a factor
of 2.
Hence, in terms of both WUEratio and WUEintrinsic mangrove trees prove well
equipped for economic water use in their habitats which are characterized by high
salinity and solar radiation leading to high VPD. With respect to the role of VPD it is
also necessary to mention leaf and air temperatures which in addition to atmospheric
water vapour partial pressure are essential determinants of VPD. Reduced transpi-
ration with increased WUE would reduce transpirational cooling (Sect. 5.2.2.1).
Leaf angle position towards solar radiation and morphological characteristics of
WUE - intrinsic 0.8
0.6
0.4
0.2
0
1/ 1/2 1/ 0 10 20 30 40
10 1
strength sea water VPD ( Pa / kPa )
Fig. 7.20 Intrinsic water-use-efficiencies (WUEintrinsic ) of averaged 19 different mangrove tree
species studied in the field in Australia and Papua New Guinea in relation to approximate sea
water strength salinity and leaf-to-air water vapour pressure differences (VPD). (After Clough and
Sim 1989, from Lüttge 2002)
leaves are additional attributes in optimization of these relations by mangrove trees

(Ball 1996).
The relationships discussed above may be interpreted as illustrating the compro-
mise of the desiccation-starvation dilemma (see Sect. 5.2.2). Since the flow of salt
into the leaves is proportional to salinity and transpiration, control of transpiration
also reduces the salt load and the danger of serious water deficits and salt toxicity
in the leaves. The reduction of CO2 -gain, which also follows partial stomatal clo-
i
sure, is partially offset by effective CO2 -fixation at low pCO , whilst maintaining
2
high WUE (as noted by Ball 1986). Interestingly, it was also observed that of all the
mangroves studied, the salt-excreting species (Avicennia marina) afford the highest
rates of CO2 -uptake and water-vapour conductances, although the salt load of leaves
may be similar in excreting and non-excreting species (Fig. 7.14).
7.5.3 High Irradiance, Photoinhibition and Oxidative Stress
Kitao et al. (2003) observed a correlation of light saturated electron transport

rates (ETRmax ) of mangrove trees in a gradation from pioneer species, to interme-
diate species and shade tolerant climax species (Table 7.3). [This is also a good
example for the potential of ETRmax -analyses for the assessment of intrinsic photo-
synthetic capacities of plants adapted to various sites (Sect. 4.1.7).]
Photoinhibition has manifold protective functions as well as potentially being
irreversibly destructive (Sect. 4.1.7). As shown by measurements of potential quan-
tum yield of photosystem II (Fv /Fm of PSII, see Sect. 4.1.7) mangroves are of-
ten highly resistant against photoinhibition. Cheeseman (Cheeseman 1994; Cheese-
man et al. 1991) did not observe photoinhibition in Rhizophora mangle under water
stress in the greenhouse and in Bruguiera parviflora in the field. Under extremely
challenging conditions Fv /Fm -values of ∼ 0.8 were detected in Rhizophora stylosa
(Cheeseman et al. 1997). Sobrado (1999) studied Avicennia germinans during the
rainy and the dry season at a site with high salinity (30 – 50‰ in the wet season,
60‰ in the dry season) and at a site with low salinity (5 – 15‰ and 40‰ in the wet
and dry season, respectively; where 30 – 35‰ correspond to 0.52 – 0.55 M NaCl,
i.e. the salinity of sea water). Predawn values of Fv /Fm were ∼ 0.75 under all con-
ditions indicating only very mild chronic photoinhibition not reversible over night.
Similar predawn values of Fv /Fm were measured with Avicennia marina under 1/1-
strength sea water salinity and hyper-salinity of 2/1-strength sea water (Sobrado and
Ball 1999), and there was no evidence for pronounced chronic photoinhibition un-
der severe salinity. The shade tolerant plants of Table 7.3 showed very slight chronic
photoinhibition after darkening of 5 h or longer (Fv /Fm = 0.78). Thus, mangrove
tree chloroplasts must be well protected against chronic photoinhibition and pho-
todestruction.
On the other hand, during high insolation mangrove trees certainly can become
subject to acute photoinhibition, which is not reversible after short periods of dark-
ening. Björkman et al. (1988) reported a large decrease of Fv /Fm for various man-
grove species at high solar radiation in the field. In the different seasons and sites
where Sobrado (1999) did not observe chronic photoinhibition as mentioned above
she detected Fv /Fm values as low as 0.45 – 0.55 at midday, i.e. severe acute pho-
toinhibition, which then was largely reversible over night. This may be related to
the protective functions of acute photoinhibition where excess photosynthetic ex-
citation energy is dissipated in a harmless way, mainly in the form of heat (see
Sect. 4.1.4). The involvement of xanthophylls in this protective process has been
shown in mangroves (Christian 2005). The depression of Fv /Fm in mangrove-tree
leaves at midday was found to be correlated with the concentration of zeaxanthin
per unit leaf area (Lovelock and Clough 1992). This was not seen, however, in
a study, where the performance of Avicennia marina was compared at 1/1- and 2/1-
strength sea water salinity. The hyper saline condition reduced net photosynthetic
CO2 -uptake ( JCO2 ) from 7.6 to 4.3 µ mol m−2 s−1 and stomatal conductance for wa-
ter vapour from 123 to 53 mol m−2 s−1 . Despite the much reduced CO2 -assimilation
under the hyper saline regime, xanthophyll pool sizes and epoxidation states as well
as non-photochemical energy dissipation (i.e. not connected to CO2 -assimilation)
Table 7.3 Photosynthetic electron transport rates (ETRmax ) for pioneer, intermediate and shade-
tolerant climax species of mangrove trees at an irradiance of 1,000 µ mol m−2 s−1 (i.e. at saturation
for the intermediate and shade tolerant species and near saturation for the pioneer species). (After
data of Kitao et al. 2003)
Mangrove trees ETRmax
(µmol m−2 s−1 )
Pioneer species
Sonneratia alba 95
Rhizophora stylosa 90
Intermediate species
Bruguiera gymnorrhiza 60
Rhizophora apiculata 55
Shade-tolerant climax species
Xylocarpus granatum 40
were similar in both sets of plants, i.e. at 1/1- and 2/1-strength sea water salinity.
This suggests that in the hyper saline regime more excitation energy may have been
dissipated via photorespiration. Other alternative electron sinks may have been also
involved. In any case, under the conditions of this study there was no increased
xanthophyll-cycle dependent photoprotection or non-photochemical dissipation of
excess excitation energy despite the 43% decrease in CO2 -assimilation rate with the
doubling of the salinity level.
Over energization of the photosynthetic apparatus also causes oxidative stress
due to the formation of reactive oxygen species (ROS, see Sect. 4.1.3). Protec-
tion mechanisms comprise the ascorbate/glutathione cycle and antioxidative en-
zymes, such as superoxide dismutases (SODs) and others, which are found to be
active in mangroves (Cheeseman et al. 1997; Takemura et al. 2002; Parida et al.
2004b). Comparing the mangrove species Rhizophora stylosa and Rhizophora man-
gle with pea (Pisum sativum), Cheeseman et al. (1997) found that total SOD ac-
tivities were 38 and 6 times higher, respectively, in the two mangrove species than
Tissue cyclitol concentration (mM)
20 A
15
10
Fig. 7.21 Salinity depen-

5
dent accumulation of poly-
ols and correlated increase
of potential quantum yield 0
of photosystem II (Fv /Fm )
after 30 min of dark adapta-
tion in gametophytes of the 150 200 250 300 350
mangrove fern Acrostichum Tissue Na+-concentration (mM)
aureum. (Closed circles to- 0.75
tal polyols, open circles D - B
Fv / F m
pinitol; errors are SE) (A),

0.70
correlation of polyol and
Na+ concentration in the
tissue (B, C), gametophytes 0.65
grown under the external 0 50 100 150
salinities indicated by the External NaCl concentration (mM)
absicissa in (B) and then
transferred to 340 mM NaCl 90 C
( % of initial )
for two days. Hardening by

Fv / F m
growth at increased salin-

ity (A) is demonstrated by 80
increased Fv /Fm (B) and this
is correlated to tissue polyol
concentration (C). (After Sun 70
et al. 1999, from Lüttge 2002) 0 5 10 15 20
Tissue cyclitol concentration (mM)
in pea. Another interesting observation is that polyols, which are the dominating
compatible solutes in mangroves (Box 7.1) may have an additional function as ef-
fective radical scavengers (Orthen et al. 1994). Li and Ong (1998) and Sun et al.
(1999) studied this in the gametophytes of the mangrove fern Acrostichum aureum.
Polyol concentration was strongly correlated with Na+ accumulation in the tissue
as determined by substrate salinity (Fig. 7.21A). This involves hardening to salin-
ity stress as it is seen when the gametophytes grown at up to 150 mM NaCl are
transferred to 340 mM NaCl for two days and then, Fv /Fm is measured at an irra-
diance of 400 µ mol m−2 s−1 following 30 min dark adaptation (Fig. 7.21B) where
high Fv /Fm is correlated with cyclitol content (Fig. 7.21C).
Different strategies in niche occupation are also involved. Lovelock and Clough
(1992) give an example for mangroves of the Daintree River in Australia (17◦ S,
147◦ E). In Rhizophora there is stress avoidance as leaves are oriented nearly ver-
tically and thus reduce light absorption. Bruguiera parviflora has small horizontal
leaves, which are rich in xanthophylls functioning in dissipation of excitation en-
ergy. Larger horizontally arranged leaves of Bruguiera gymnorhiza tend to heat up
more strongly and are therefore more subject to photodamage. Thus, B. parviflora
dominates the canopy, whereas B. gymnorhiza is less abundant at the top of the
canopy.
7.5.4 Interacting Factors: Salinity, Irradiance, Elevated CO2
Environmental factors can interact. Salinity and irradiance stress may be additive,
e.g. between two sympatric mangrove species at saline sites Aegiceras corniculatum
was found to be favoured where excess radiation was less frequent and Avicennia
marina under conditions of persistent excess irradiance (Christian 2005). Very in-
teresting findings on the interactions between salinity and irradiance were obtained
in studies of seedling establishment and growth (Ball 2002). Mangrove seedlings
need full sunlight and the formation of gaps in the mangrove forests is essential
for regeneration. At full sunlight for most species there were no substantial differ-
ences in seedling survival after 12 months at low and high salinity except for 2
species Bruguiera parviflora and Ceriops australis. At 30% sunlight both species
showed similarly high survival rates of seedlings at high and low salinity; however,
at high irradiance they required high salinity in addition to the full sunlight. Ob-
viously low salinity conditions induced sensitivity to high irradiance in these two
mangrove species (Fig. 7.22).
Elevated atmospheric CO2 -concentration, pCO a , modulates water use and car-
2
bon gain and one might expect that it affects salt tolerance of mangrove trees. How-
ever, a comparative study of two mangrove species differing in salt tolerance, i.e.
Rhizophora apiculata and Rhizophora stylosa, showed that when relative growth
rates were limited by salinity, i.e. 350 mM as compared to 125 mM NaCl at the
a
root level, pCO elevated from 340 to 700 ppm had little effect on the growth rates.
2
However, elevated pCO a stimulated growth when it was limited by air humidity,
2
7.6 Nutrition 257
Fig. 7.22 Survival of

seedlings of Bruguiera parvi-
flora (triangles) and Ceriops
australis (circles) at low and
high salinity, respectively,
under full sunlight (A) and in
the shade at 30% full sunlight
(B), respectively. (After data
of Ball 2002)
a
i.e. at 43% as compared to 86% relative air humidity. Thus, elevated pCO 2
in the
future could modify competitive potentials of different mangrove trees along salin-
ity × aridity gradients, but it is unlikely that it will allow mangroves to expand
into areas with salinities much more extreme than currently tolerated (Ball et al.
1997).
7.6 Nutrition
Mineral nutrition of mangroves is much determined by decomposition of litter

(Mfilinge et al. 2002; Ochieng and Erftemeijer 2002) and the activity of micro-
bial mats (Sect. 7.7.2). Mangroves may occasionally suffer phosphorus and nitro-
gen limitation especially in dwarf mangrove formations further inland of zonations
(Cheeseman and Lovelock 2004; Parida and Das 2004; Lovelock et al. 2006a, b).
Phosphorus deficiency inhibits water transport and hampers water relations of man-
groves (Lovelock et al. 2006b, c). High salinity inhibits uptake and reduction of ni-
trate (Pariada and Das 2004), and nitrogen fertilization may stimulate growth (Kao
et al. 2001; Yates et al. 2002). However, other stress factors are dominating so that
mineral supply is not very likely to become the limiting factor (Yates et al. 2002;
Alongi et al. 2003). An interesting ion is K+ because under salinity stress when
Na+ is accumulated K+ levels are normally reduced and this can lead to ion imbal-
ances with adverse effects also in mangroves (Naidoo et al. 2002). However, since
seawater has a K+ -concentration of 10 mM this may not be a particular problem of
stress in mangroves (Cram et al. 2002).
7.7 Aquatic Communities
7.7.1 Macroalgae in Mangroves
Macroalgae in mangroves grow between the roots but mainly epiphytically on

the pneumatophores and trunks of trees (Post 1963). Species diversity is mainly
given by red algae of the genera Bostrychia, Caloglossa and Stictosiphonia, al-
though brown algae may also occur, e.g. mats of Hormosira banksii in SE-Australia
(Karsten 1995). About 15 – 20% of the total biomass of mangrove communities is
represented by these macroalgae (Karsten 1995).
The macroalgae are subject to the same stress conditions and even more so than
the woody mangrove plants, e.g. changing salinity and desiccation at low tide. They
show a broad salinity tolerance between 1/5- and 2/1-strength sea water (Karsten
and West 1993). Some mangrove algae are also desiccation tolerant (Biebl 1962;
see Sect. 11.4), e.g. Stictosiphonia arbuscula can lose up to 95% of its tissue wa-
ter and recover within several hours when rewetted (Karsten 1995). Accumulation
of compatible solutes (Sect. 7.4) plays a large role in the red algae in response
to salinity and shows a rather high chemical diversity including floridoside, dige-
neaside, D-sorbitol, D-dulcitol, D-mannitol and isethionic acid (Box 7.1) (Karsten
1995, 1996; Karsten et al. 1995a,b, 1997a,b). Different compatible solute spectra
have been found in Bostrychia tenuisissima from different geographic provenance
in Australia, i.e. sorbitol plus dulcitol and sorbitol plus digeneaside, respectively,
and this was controlled genetically (Karsten et al. 1995b). In the mangrove fern
Acrostichum aureum the gametophyte uses D-pinitol and the sporophyte D-1-O-
methyl-muco-inositol (Sun et al. 1999).
Low irradiance is an important factor limiting the growth of macroalgae in man-
grove forests. While the canopy of the trees may receive photosynthetic photon
flux densities (PPFD) up to 2,500 µ mol m−2 s−1 , the algae may not obtain more
than 60–100 µ mol m−2 s−1 . This is not only due to shading by the trees but also
to turbid water with organic materials and debris (Karsten 1995). Bostrychia sim-
pliciuscula and species of Caloglossa may still show positive relative growth rates
at the very low PPFD of 2.5 µ mol m−2 s−1 (Karsten and West 1993; Karsten et al.
1994).
7.8 Mangroves as Endangered Ecosystems 259
7.7.2 Microbial Mats
Microbial mats are an essential aspect of mangrove ecosystems. The surface muds
are zones of net heterotrophy. Light limitation beneath mangrove forests might mean
that photosynthesis by benthic microalgae only makes a minor contribution to pri-
mary productivity (Alongi 1994), but Karsten (1995) arrives at the estimate that
microbial mats account for 5 – 20% of the total mangrove productivity. They are
largely composed of diatoms, cyanobacteria, sulphur bacteria, purple sulphur bac-
teria and sulphate reducing bacteria (Table 7.4). Their thickness as given in Table
7.4 is 10 – 12 mm but can also be as much as 80 – 120 mm (Hussain and Khoja
1993). Root associations of mangroves with halotolerant N2 -fixing bacteria have
been shown to improve N-supply and to contribute to the high productivity of
mangrove ecosystems (Zuberer and Silver 1978, 1979; Sengupta and Chaudhuri
1991).
The main stress factors, as for the other mangrove communities are large varying
amplitudes of salinity, irradiance and desiccation. Cyanobacteria use glycosylglyc-
erol as compatible solute (Karsten 1996) and form scytonemin, pterins and myco-
sporine-like amino acid compounds as ultraviolet light protectives (Karsten et al.
1998; see Sect. 11.2.1.2).
Table 7.4 Layers in microbial mats of mangroves (after Karsten 1995)
Thickness from top (mm) Organisms

0–2 Fine sand plus diatoms
2–4 Cyanobacteria
4–6 Sulphur bacteria
6–8 Purple sulphur bacteria
8–11 Sulphate reducing bacteria
> 11 Sand
7.8 Mangroves as Endangered Ecosystems with Numerous

Benefits for Man and the Need for their Conservation
Mangroves are among the most endangered ecosystems on earth (Springer 2002).
They have been frequently considered to be useless and are disappearing rapidly.
However, they are very unique, and with their characteristic physiognomic beauty
they are among the outstanding natural heritages we have. Moreover, they have nu-
merous direct benefits for us (Oo 2002). They are pioneer communities at the inter-
face between sea and land and stabilize coastlines. Some mangrove trees provide
useful wood for fuel and the production of charcoal and particularly resistant tim-
ber for construction purposes. They provide fodder and medicine. They serve as
nursery grounds for breeding of marine life, for fish and crabs of coral reefs, and
sustain the economical basis of coastal fisheries (Ellison 2002). They are also used
for establishing ponds for the culture of fish and prawns.
Mangroves are among the most productive ecosystems of the world. If we take
the productivity of macroalgae as 15 – 20% and that of microbial mats as 5 – 20%,
the productivity of trees would be 80 – 60%. This underlines the diversity of the
mangrove communities.
References
Aiga I, Nakano Y, Ohki ST, Kitaya Y, Yabuki K (1995) Photosynthetic CO2 fixation in pneu-
matophores of grey mangrove, Avicennia marina. Environ Control Biol 33:97–101
Alongi DM (1994) Zonation and seasonality of benthic primary production and community respi-
ration in tropical mangrove forests. Oecologia 98:320–327
Alongi DM, Clough BF, Dixon P, Tirendi F (2003) Nutrient partitioning and storage in arid-zone
forests of the mangroves Rhizophora stylosa and Avicennia marina. Trees 17:51–60
Atkinson MR, Findlay GP, Hope AB, Pitman MG, Saddler HDW, West KR (1967) Salt regulation
in the mangroves Rhizophora mucronata Lam. and Aegialitis annulata R.Br. Aust J Biol Sci
20:589–599
Ball MC (1986) Photosynthesis in mangroves. Wetlands (Aust) 6:12–22
Ball MC (1996) Comparative ecophysiology of mangrove forest and tropical lowland moist rainfor-
est. In: Mulkey SS, Chazdon RL, Smith AP (eds) Tropical forest plant ecophysiology. Chap-
man and Hall, New York, pp 461–496
Ball MC (2002) Interactive effects of salinity and irradiance on growth: implications for mangrove
forest structure along salinity gradients. Trees 16:126–139
Ball MC, Anderson JM (1986) Sensitivity of photosytem II to NaCl in relation to salinity tolerance.
Comparative studies with thylakoids of the salt-tolerant mangrove, Avicennia marina, and the
salt-sensitive pea, Pisum sativum. Aus J Plant Physiol 13:689–698
Ball MC, Farquhar GD (1984a) Photosynthetic and stomatal responses of two mangrove species,
Aegiceras corniculatum and Avicennia marina, to long term salinity and humidity conditions.
Plant Physiol 74:1–6
Ball MC, Farquhar GD (1984b) Photosynthetic and stomatal responses of the grey mangrove,
Avicennia marina, to transient salinity conditions. Plant Physiol 74:7–11
Ball MC, Cochrane MJ, Rawson HM (1997) Growth and water use of the mangroves, Rhizophora
apiculata and R. stylosa, in response to salinity and humidity under ambient and elevated
concentration of atmospheric CO2 . Plant Cell Environ 20:1158–1166
Becker P, Asmat A, Mohamed J, Moksin M, Tyree MT (1997) Sap flow rates of mangrove trees
are not unusually low. Trees 11:432–435
Biebl R (1962) Protoplasmatisch-ökologische Untersuchungen an Mangrovenalgen von Puerto
Rico. Protoplasma 55:572–606
Biebl R, Kinzel H (1965) Blattbau und Salzgehalt von Laguncularia racemosa (L) Gaertn. f. und
anderer Mangrovebäume auf Puerto Rico. Österr Bot Z 112:56–93
Björkman O, Demmig B, Andrews TJ (1988) Mangrove photosynthesis: response to high-irradiance
stress: Aust J Plant Physiol 15:43–61
Black CC (1973) Photosynthetic carbon fixation in relation to net CO2 uptake. Annu Rev Plant
Physiol 24:253–286
Cheeseman JM (1994) Depressions of photosynthesis in mangrove species. In: Baker NR, Bowyer
JR (eds) Photoinhibition of photosynthesis. From molecular mechanisms to the field. Bios
Scientific Publishers, Oxford, pp 379–391
Cheeseman JM, Lovelock CE (2004) Photosynthetic characteristics of dwarf and fringe Rhi-
zophora mangle L. in a Belizean mangrove. Plant Cell Environ 27:769–780
References 261
Cheeseman JM, Clough BF, Carter DR, Lovelock CE, Eong OJ, Sim RG (1991) The analysis of
photosynthetic performance in leaves under field conditions – a case study using Bruguiera
mangroves. Photosyn Res 29:11–22
Cheeseman JM, Herendeen LB, Cheeseman AT, Clough BF (1997) Photosynthesis and photopro-
tection in mangroves under field conditions. Plant Cell Environ 20:579–588
Chiu C-Y, Chou C-H (1993) Oxidation in the rhizosphere of mangrove Kandelia candel seedlings.
Soil Sci Plant Nutr 39:725–731
Christian R (2005) Interactive effects of salinity and irradiance on photoprotection in acclimated
seedlings of two sympatric mangroves. Trees 19:596–606
Clarke PJ, Kerrigan RA, Westphal CJ (2001) Dispersal potential and early growth in 14 tropical
mangroves: do early life history traits correlate with patterns of adult distribution? J Ecol
89:648–659
Clough BF, Sim RG (1989) Changes in gas exchange characteristics and water use efficiency of
mangroves in response to salinity and vapour pressure deficit. Oecologia 79:38–44
Cram WJ, Torr PG, Rose DA (2002) Salt allocation during leaf development and leaf fall in man-
groves. Trees 16:112–119
Curran M (1985) Gas movements in the roots of Avicennia marina (Forsk.) Vierh. Aust J Plant
Physiol 12:97–108
Ellison AM (2002) Macroecology of mangroves: large-scale patterns and processes in tropical
coastal forests. Trees 16:181–194
Fahn A, Shimony C (1977) Development of the glandular and non-glandular leaf hairs of Avicennia
marina (Forsskål) Vierh. Bot J Linn Soc 74:34–46
Fitzgerald MA, Allaway WG (1991) Apoplastic and symplastic pathways in the leaf of the grey
mangrove Avicennia marina (Forssk.) Vierh. New Phytol 119:217–226
Hussain MI, Khoja TM (1993) Intertidal and subtidal blue-green-algal mats of open and mangrove
areas in the Farasan Archipelago (Saudi Arabia), Red Sea. Bot Mar 36:377–388
Janssonius HH (1950) The vessels in the wood of Javan mangrove trees. Blumea 6:465–469
Junghans U, Polle A, Düchting P, Weiler E, Kuhlman B, Gruber F, Teichmann T (2006) Adaptation
to high salinity in poplar involves changes in xylem anatomy and auxin physiology. Plant Cell
Environ 29:1519–1531
Kao W-Y, Tsai H-C, Tsai T-T (2001) Effect of NaCl and nitrogen availability on growth and pho-
tosynthesis of seedlings of a mangrove species, Kadelia candel (L.) Druce. J Plant Physiol
158:841–846
Karsten U (1995) Mangrovenalgen. Biol in unserer Zeit 25:51–58
Karsten U (1996) Growth and organic osmolytes of geographically different isolates of Mi-
crocoleus chthonoplastes (cyanobacteria) from benthic microbial mats: response to salinity
change. J Phycol 32:501–506
Karsten U, West JA (1993) Ecophysiological studies on six species of the mangrove red algal genus
Caloglossa. Aust J Plant Physiol 20:729–739
Karsten U, Koch S, West JA, Kirst GO (1994) The intertidal red alga Bostrychia simpliciuscula
Harvey ex J. Agardh from a mangrove swamp in Singapore: acclimation to light and salinity.
Aquatic Bot 48:313–323
Karsten U, Barrow KD, Mostaert AS, King RJ (1995a) The osmotic significance of the hetero-
side floridoside in the mangrove alga Catenella nipae (Rhodophyta: Gigartinales) in Eastern
Australia. Estuarine Coastal Shelf Sci 40:239–247
Karsten U, Bock C, West JA (1995b) Low molecular weight carbohydrate patterns in geograph-
ically different isolates of the eulittoral red alga Bostrychia tenuisissima from Australia. Bot
Acta 108:321–326
Karsten U, Barrow KD, Nixdorf O, West JA, King RJ (1997a) Characterization of mannitol
metabolism in the mangrove red alga Caloglossa leprieurii (Montagne) J. Agardh. Planta
201:173–178
Karsten U, Barrow KD, West JA, King RJ (1997b) Mannitol metabolism in the intertidal mangrove
red alga Caloglossa leprieurii: salinity effects on enzymatic activity. Phycologia 36:150–
156
Karsten U, Maier J, Garcia-Pichel F (1998) Seasonality in UV-absorbing compounds of cyanobac-

terial mat communities from an intertidal mangrove flat. Aquatic Microbial Ecol 16:37–44
Kinzel H (1982) Pflanzenökologie und Mineralstoffwechsel. Ulmer, Stuttgart, pp 376–382
Kitao M, Utsugi H, Kuramoto S, Tabuchi R, Fujimoto K, Lihpai S (2003) Light-dependent photo-
synthetic characteristics indicated by fluorescence in five mangrove species native to Pohnpei
Island, Micronesia. Physiol Plant 117:376–382
Kitaya Y, Sumiyoshi M, Kawabata Y, Monji N (2002) Effect of submergence and shading of
hypocotyls on leaf conductance in young seedlings of the mangrove Rhizophora stylosa. Trees
16:147–149
Lear R, Turner T (1977): Mangroves of Australia. University of Queensland Press, St Lucia
Li X-P, Ong B-L (1998) Responses of photosynthesis to NaCl in gametophytes of Acrostichum
aureum. Physiol Plant 102:119–127
Lin G, Sternberg L da SL (1992a) Effect of growth form, salinity, nutrient and sulfide on photo-
synthesis, carbon isotope discrimination and growth of red mangrove (Rhizophora mangle L.).
Aust J Plant Physiol 19:509–517
Lin G, Sternberg L da SL (1992b) Comparative study of water uptake and photosynthetic gas
exchange between scrub and fringe red mangroves Rhizophora mangle L. Oecologia 90:399–
403
Lin G, Sternberg L da SL (1993) Effects of salinity fluctuation on photosynthetic gas exchange and
plant growth of the red mangrove (Rhizophora mangle L.). J Exp Bot 44:9–16
López-Portillo J, Ewers FW, Angeles G (2005) Sap salinity effects on xylem conductivity in two
mangrove species. Plant Cell Environ 28:1285–1292
Lösch R (1998) Plant water relations. Progr Bot 60:193–233
Lösch R, Busch J (1999) Plant functioning under waterlogged conditions. Progr Bot 61:255–268
Lovelock CF, Clough BF (1992) Influence of solar radiation and leaf angle on leaf xanthophyll
concentrations in mangroves. Oecologia 91:518–525
Lovelock CE, Ball MC, Choat B, Engelbrecht BMJ, Holbrook NM, Feller IC (2006a) Linking
physiological processes with mangrove forest structure: phosphorus deficiency limits canopy
development, hydraulic conductivity and photosynthetic carbon gain in dwarf Rhizophora
mangle. Plant Cell Environ 29:793–802
Lovelock CE, Ball MC, Feller IC, Engelbrecht BMJ, Ewe ML (2006b) Variation in hydraulic con-
ductivity of mangroves: influence of species, salinity, and nitrogen and phosphorus availability.
Lovelock CE, Feller IC, Ball MC, Engelbrecht BMJ, Ewe ML (2006c) Differences in plant function
in phosphorus- and nitrogen-limited mangrove ecosystems. New Phytol 172:514–522
Lüttge U (1975) Salt glands. In: Baker DA, Hall JL (eds) Ion transport in plant cells and tissues.
North-Holland Publishing, Amsterdam, pp 335–376
Lüttge U (2002) Mangroves. In: Läuchli A, Lüttge U (eds) Salinity: environment – plants –
molecules. Kluwer, Dordrecht, pp 113–135
Medina E, Cuevas E, Popp M, Lugo AE (1990) Soil salinity, sun exposure, and growth of Acros-
tichum aureum, the mangrove fern. Bot Gaz 151:41–49
Mfilinge PL, Atta N, Tsuchiya M (2002) Nutrient dynamics and leaf litter decomposition in a sub-
tropical mangrove forest at Oura Bay, Okinawa, Japan. Trees 16:172–180
Naidoo G, Tuffers AV, Willert DJ von (2002) Changes in gas exchange and chlorophyll fluores-
cence characteristics of two mangroves and a mangrove associate in response to salinity in the
natural environment. Trees 16:140–146
Ochieng CA, Erftemeijer PLA (2002) Phenology, litterfall and nutrient resorption in Avicennia
marina (Forssk.) Vierh. in Gazi Bay, Kenya. Trees 16:167–171
Oo NW (2002) Present state and problems of mangrove management in Myanmar. Trees 16:218–
223
Orthen B, Popp M, Smirnoff N (1994) Hydroxyl radical scavenging properties of cyclitols. Proc R
Soc Edinburgh Biol Sci 102:267–272
Parida AK, Das AB (2004) Effects of NaCl stress on nitrogen and phosphorus metabolism in a true
mangrove Bruguiera parviflora grown under hydroponic culture. J Plant Physiol 161:921–928
References 263
Parida AK, Das AB, Mittra B (2004a) Effects of salt on growth, ion accumulation, photosynthesis
and leaf anatomy of the mangrove, Bruguiera parviflora. Trees 18:167–174
Parida AK, Das AB, Mohanty P (2004b) Defense potentials to NaCl in a mangrove, Bruguiera
parviflora: differential changes of isoforms of some antioxidative enzymes. J Plant Physiol
161:531–542
Passioura JB, Ball MC, Knight JH (1992) Mangroves may salinize the soil and in so doing limit
their transpiration rate. Funct Ecol 6:476–481
Polanía J (1990) Anatomische und physiologische Anpassungen von Mangroven. Dissertation,
Vienna
Popp M (1984) Chemical composition of Australian mangroves. II. Low molecular weight carbo-
hydrates. Z Pflanzenphysiol 113:411–421
Popp M (1991) Mangroven – Wälder der Gezeitenzone in den Tropen und Subtropen. Praxis
Naturw 40:16–22
Popp M, Polanía J (1989) Compatible solutes in different organs of mangrove trees. Ann Sci For
46:842s–844s
Popp M, Larher F, Weigel P (1984) Chemical comparison of Australian mangroves. III. Free amino
acids, total methylated onium compounds and total nitrogen. Z Pflanzenphysiol 114:15–25
Popp M, Polanía J, Weiper M (1993) Physiological adaptations to different salinity levels in man-
grove. In: Lieth H, Al Masoom A (eds) Towards the rational use of high salinity tolerant plants,
vol 1. Kluwer, Dordrecht, pp 217–224
Post E (1963) Zur Verbreitung und Ökologie der Bostrychia-Caloglossa-Assoziation. Inter Rev
Ges Hydrobiol 48:47–152
Richter A, Thonke B, Popp M (1990) 1S-1O-Methylmucoinositol in Viscum album and members
of the Rhizophoraceae. Phytochemistry 29:1785–1786
Scholander PF (1968) How mangroves desalinate sea water. Physiol Plant 21:251–261
Scholander PF, Hammel HT, Bradstreet ED, Hemmingsen EA (1965) Sap pressure in vascular
plants. Science 148:339–346
Scholander PF, Bradstreet ED, Hammel HT, Hemmingsen EA (1966) Sap concentrations in halo-
phytes and some other plants. Plant Physiol 41:529–532
Sengupta A, Chaudhuri A (1991) Ecology of heterotrophic dinitrogen fixation in the rhizosphere
of mangrove plant community at the Ganges River estuary in India. Oecologia 87:560–564
Skelton NJ, Allaway WG (1996) Oxygen and pressure changes measured in situ during flooding
in roots of the grey mangrove Avicennia marina (Forsk.) Vierh. Aquat Bot 54:165–175
Smith JAC, Popp M, Lüttge U, Cram WJ, Diaz M, Griffiths H, Lee HSJ, Medina E, Schäfer C,
Stimmel K-H, Thonke B (1989) Ecophysiology of xerophytic and halophytic vegetation of
a coastal alluvial plain in northern Venezuela. VI. Water relations and gas exchange of man-
groves. New Phytol 111:293–307
Sobrado MA (1999) Drought effects on photosynthesis of the mangrove, Avicennia germinans,
under contrasting salinities. Trees 13:125–130
Sobrado MA (2000) Relation of water transport to leaf gas exchange properties in three mangrove
species. Trees 14:258–262
Sobrado MA (2001) Hydraulic properties of a mangrove Avicennia germinans as affected by NaCl.
Biol Plant 44:435–438
Sobrado MA (2002) Effect of drought on leaf gland secretion of the mangrove Avicennia germinans
L. Trees 16:1–4
Sobrado MA (2004) Influence of external salinity on the osmolality of xylem sap, leaf tis-
sue and leaf gland secretion of the mangrove Laguncularia racemosa (L.) Gaertn. Trees
18:422–427
Sobrado MA, Ball MC (1999) Light use in relation to carbon gain in the mangrove, Avicennia
marina, under hypersaline conditions. Aust J Plant Physiol 26:245–251
Sommer C, Thonke B, Popp M (1990) The compatibility of D-pinitol and 1D-1-O-methylmuco-
inositol with malate dehydrogenase activity. Bot Acta 103:270–273
Springer (2002) Trees: structure and function, vol 16, no 2/3. Special issue. Mangroves. Springer,
Suárez N, Medina E (2005) Salinity effect on plant growth and leaf demography of the mangrove
Avicennia germinans L. Trees 19:721–727
Sun WQ, Li X-P, Ong B-L (1999) Preferential accumulation of D-pinitol in Acrostichum aureum
gametophytes in response to salt stress. Physiol Plant 105:51–57
Takemura T, Hanagata N, Dubinsky Z, Karube I (2002) Molecular characterization and response
to salt stress of mRNAs encoding cytosolic Cu/Zn superoxide dismutase and catalase from
Bruguiera gymnorrhiza. Trees 16:94–99
Twilley RR, Chen R, Hargis T (1992) Carbon sinks in mangroves and their implication to carbon
budget of tropical ecosystems. Water Air Soil Pollut 64:265–288
Tyree MT (1997) The cohesion-tension theory of sap ascent: current controversies. J Exp Bot
48:1753–1765
Verheyden A, Helle G, Schleser GH, Dehairs F, Beeckman H, Koedam N (2004) Annual cyclicity
in high-resolution stable carbon and oxygen isotope ratios in the wood of the mangrove tree
Rhizophora mucronata. Plant Cell Environ 27:1525–1536
Verheyden A, de Ridder F, Schmitz N, Beeckman H, Koedam N (2005) High-resolution time series
of vessel density in Kenyan mangrove trees reveal a link with climate. New Phytol 167:425–
435
Wei C, Steudle E, Tyree MT (1999a) Water ascent in plants: do ongoing controversies have a sound
basis. Trends Plant Sci 4:372–375
Wei C, Tyree MT, Steudle E (1999b) Direct measurement of xylem pressure in leaves of intact
maize plants. A test of the cohesion-tension theory taking hydraulic architecture into consid-
eration. Plant Physiol 121:1191–1205
Yáñez-Espinoza L, Terrazas T, López-Mata L (2001) Effects of flooding on wood and bark
anatomy of four species in a mangrove forest community. Trees 15:91–97
Yates EJ, Ashwath N, Midmore DJ (2002) Responses to nitrogen, phosphorus, potassium and
sodium chloride by three mangrove species in pot culture. Trees 16:120–125
Youssef T, Saenger P (1996) Anatomical adaptive strategies to flooding and rhizosphere oxidation
in mangrove seedlings. Aust J Bot 44:297–313
Zimmermann U, Zhu JJ, Meinzer FC, Goldstein G, Scheider H, Zimmermann G, Benkert R,
Thürmer F, Melcher P, Webb D, Haase A (1994a) High molecular weight organic com-
pounds in the xylem sap of mangroves: implications for long-distance water transport. Bot
Acta 107:218–229
Zimmermann U, Meinzer FC, Benkert R, Zhu JJ, Schneider H, Goldstein G, Kuchenbrod E, Haase
A (1994b) Xylem water transport: is the available evidence consistent with the cohesion the-
ory? Plant Cell Environ 17:1169–1181
Zimmermann U, Wagner H-J, Heidecker M, Mimietz S, Schneider H, Szimtenings M, Haase A,
Mitlöhner R, Kruck W, Hoffmann R, König W (2002) Implications of mucilage on pressure
bomb measurements and water lifting in trees rooting in high-salinity water. Trees 16:100–111
Zuberer DA, Silver WS (1978) Biological dinitrogen fixation (acetylene reduction) associated with
Florida mangroves. Appl Environ Microbiol 35:567–575
Zuberer DA, Silver WS (1979) Nitrogen fixation (acetylene reduction) and the microbial coloniza-
tion of mangrove roots. New Phytol 82:467–472
Chapter 8
Ecosystems of Coastal Sand Plains
8.1 Restingas
8.1.1 Geological History and Vegetation Physiognomy
The Atlantic rain forest of Brazil is one of the 25 biodiversity hotspots in the world.
In the south eastern state of Rio de Janeiro it is surrounded by various marginal plant
communities, one of which is the restinga (Scarano 2002; Scarano et al. 2005a). It
stands on quaternary terrains, i.e. sandy coastal plains between the rain forest and the
sea (Martin et al. 1993; Scarano et al. 1997). The vegetation has its origins mostly
associated to the rain forest and there are hardly any endemic plant species (Rizzini
1979; Araujo 2000). Migration from the Atlantic forest to the restingas was success-
ful for ecologically plastic species, which were able to adjust to the more extreme
and seasonal conditions imposed to the restinga habitats (Scarano 2002). Approx-
imately 80% of the plant species occurring in the restingas of the state of Rio de
Janeiro are also found in montane rainforests (Araujo 2000). Along the coastline
of Rio de Janeiro the quaternary sandy deposits and dunes date mostly from the
Holocene, having been established and re-established from 5000 to 3000 years BP,
but further north under the influence of the Paraiba do Sul river in some areas marine
sandy deposits date from the Pleistocene (120,000 years BP) and remained acquir-
ing their final shape after a series of invasions and regressions of the sea during the
Holocene (Martin et al. 1993).
The restingas comprise a mosaic of plant communities ranging from open, patchy
formations to forests (Fig. 8.1A,B). Across a transect from the sea inland, successive
sand dune ridges, vegetation islands on the sand plain and dune forests, and where the
ground water table is high, open fresh water lagoons and swamp forests contribute to
the physiognomy of the landscape (Reinert et al. 1997; Duarte et al. 2005). The very
diverse plant communities found in the restingas are subjected to conditions as extreme
as seasonal drought and oligotrophy on the one hand, and permanent flooding on the
other (Henriques et al. 1986). Thus, the flora of a given restinga plant community often
is very different from the flora of closely adjacent communities (Araujo et al. 1998).
266 8 Ecosystems of Coastal Sand Plains
Fig. 8.1A, B Restinga vegetation physiognomies. A Patchy restinga vegetation with vegetation
islands viewed from a sand dune ridge with a dune forest in the foreground (dry restinga of Mas-
sambaba, 22◦ 56 S, 42◦ 13 W). B Fresh water lagoon with a swamp forest in the background (at
the intermediate restinga of the Jurubatiba National Park, 22◦ 00 – 22◦ 23 S, 41◦ 15 – 41◦ 45 W)
8.1.2 The Nurse Plant Syndrome

and Dynamics of Vegetation Islands
The nurse plant syndrome is effective when plant species shelter seedlings, young
and/or adult individuals of other species through their ontogeny (Dias and Scarano
2007). In the restingas nurse plants are pioneers on the bare sand, and they com-
prise some small palms (e.g. Allagoptera arenaria; Fig. 8.2A) and bromeliads (e.g.
Aechmea nudicaulis; Fig. 8.2B) but most importantly the shrubs of Clusia species,
especially C. hilariana which is the dominant plant in the restingas of Rio de Janeiro
(Sampaio et al. 2005). Therefore, restingas have also been called Clusia scrubs (Ule
1901). Characteristically nurse plants of the restingas are often terrestrial forms
8.1 Restingas 267
Fig. 8.2A, B Establishment of new vegetation islands by nurse plants? A Allagoptera arenaria, the
wind drawing circles on the sand by moving the leaves around (dry restinga of Massambaba, see
Fig. 8.1). B Aechmea nudicaulis (foreground centre), Allagoptera arenaria (foreground left), dune
forest (background) (intermediate restinga of Jurubatiba Park, see Fig. 8.1)
of typical epiphytes of the neighbouring Atlantic rain forest possibly pre-adapted

to stress related to missing or limited supplies of water and nutrients from a pe-
dosphere (Scarano 2002). Many of them perform crassulacean acid metabolism
(CAM), which is a biochemical adaptation of photosynthesis to stress due to high
insolation and partially problematic supply of water (Sect. 5.2.2.2), i.e. bromeli-
ads and Clusias including C. hilariana (Reinert et al. 1997; Scarano 2002; Lüttge
2007a,b).
Underneath the canopy of C. hilariana a higher species richness was found than
under any other woody species (Zaluar 1997; Sampaio et al. 2005; Dias et al. 2005,
2007). The higher density and richness under scrubs of C. hilariana is mainly due
to effects on seed dispersers and activation of dispersal (Dias et al. 2007). Aechmea
nudicaulis does not germinate and grow seedlings on the open sand, among other
reasons probably due to the high temperatures reached on the bare sand. It only ger-
minates within vegetation islands. Via directional growth of rhizomes and ramets,
however, it then acquires space and selects its own habitats (Fig. 8.2B; Sampaio
et al. 2004, 2005; Dias et al. 2005). Conversely, however, it is not only C. hilar-
iana scrub that nurses A. nudicaulis. The Clusia may germinate inside the tanks
of the bromeliad and is nursed itself (Dias and Scarano 2007). These reciprocal
interactions between different nurse plants generating vegetation islands underline
the non-linear dynamics of the spatiotemporal patchiness (see also Sects. 3.3.1,
3.3.3 and 8.2.2) of the restinga ecosystem.
8.1.3 Ecophysiology of Photosynthesis of Restinga Plants
The performance of the special mode of photosynthesis CAM is quite frequent

among the plants of the restingas with many species of bromeliads, orchids, cacti
and Clusia. Comparative ecophysiological studies have been performed of various
shrubs which are much determining the physiognomy of the restingas (Figs. 8.1A,
8.2B) such as the Clusiaceae Rheedia brasiliensis, Calophyllum brasiliense, Clu-
sia hilariana and Clusia fluminensis, the Myrsinaceae Myrsine parviflora and the
Fabaceae Andira legalis (Duarte et al. 2005; Geßler et al. 2005; Scarano et al.
2005b). Restingas where these plants were investigated show a moisture gradient
dependent on annual rainfall, the ground water table and the degree of ground-
water salinity. Table 8.1 summarizes some relations of maximum apparent rates
of photosynthesis (ETRmax ) deduced from measurements of instant light response
curves (Sect. 4.1.7) to such moisture gradients, where intrinsic photosynthetic ca-
pacity given by ETRmax increases in three of the species shown at the drier sites and
shows a decrease or no response in a fourth species (A. legalis).
Table 8.1 Maximum apparent rates of photosynthetic electron transport ETRmax (µ mol m−2 s−1 )
obtained from light curve measurements of four restinga shrubs along a moisture gradient. Within
vertical columns different letters at the numbers indicate statistically significant differences. (Data
from Duarte et al. 2005; Geßler et al. 2005; Scarano et al. 2005b)
Vegetation type Myrsine Rheedia Clusia Andira
parviflora brasiliensis fluminensis legalis
Swamp forest 63a 136a

at an intermediate restinga
Wet restinga 198a 288a
Intermediate restinga 90b 113b
Dry restinga 176c 198b 172a 159b
8.2 Salinas 269
8.2 Salinas
8.2.1 Formation of Coastal Salt Marshes

and Vegetation Physiognomy
Coastal salt marshes at the northern coast of South-America are formed in bays,
where the sand-laden waves returning from the beach to the sea are driven back to-
wards the shore by the North-Eastern trade winds. First a sandbar is formed leading
Fig. 8.3 Formation of salt marshes from coastal bays and lagoons. (After Walter and Breckle 1984,
with kind permission of S.-W. Breckle and G. Fischer-Verlag)
to a sandbank and to separation of a lagoon. Subsequently, this lagoon fills in with

sand, drying out and becoming a salt marsh. Both fixed and mobile sand dunes may
also form at the coast behind the salt marsh (Fig. 8.3). A typical example are the
inland salt-marshes near the northern Caribbean coast of Venezuela first described
briefly by Walter (Walter and Breckle 1984) and later studied ecophysiologically in
some detail (Lüttge et al. 1989a,b; Medina et al. 1989; Smith et al. 1989). These salt
marshes are much more extreme habitats than the restingas (Sect. 8.1) with a very
strong seasonality. The most salt-resistant plants found first in such sites are man-
groves, which begin to surround the lagoon (Fig. 8.3), an example being Avicennia
germinans at the lagoon and salt marshes near Chichiriviche on the Caribbean coast
of Venezuela.
The flat alluvial sand plain covering areas previously occupied by the lagoon is
subject to marked seasonality because there is a pronounced rainy season in October
to December and a strong dry season during the rest of the year interrupted only by
a small and short wet period in April (Fig. 8.4). During the rainy season the sand
plain may be covered by several decimeters of fresh water, whereas during the dry
season the surface is dry and a considerable salt crust may form (Fig. 8.5). The
very salty groundwater, with an NaCl-concentration several times that of seawater
(Fig. 8.6), percolates upwards to the surface where the water evaporates and leaves
behind the dissolved salt. The vegetation of the sand plain can be described by
distinguishing five units (Fig. 8.7):
a) the vegetation-free sand and salt flats (Fig. 8.7A),
b) a halophyte zone with Batis maritima and Sesuvium portulacastrum as the
dominating species (Fig. 8.7B),
c) a grass-land zone with Sporobolus virginicus and Oxycarpha suaedifolia as the
characteristic plants (Fig. 8.7D),
d) vegetation islands with the mangrove associate Conocarpus erectus and the
cactus Subpilosocereus ottonis as the physiognomically determinant species
(Fig. 8.7A, C), [enumeration continuing on page 274]
Fig. 8.4 Average monthly

values of rainfall, evap-
oration and temperature
near the northern Caribbean
coast of Venezuela close to
Chichiriviche for a 22-year
period. (Medina et al. 1989)
8.2 Salinas 271
Fig. 8.5A,B Alluvial sand plain with vegetation islands near the northern Carribbean coast of
Venezuela at Chichiriviche. A In the rainy season (November 1985) covered with fresh water. B In
the dry season (February 1983) covered with a thick salt crust (background). (See Medina et al.
1989)
Fig. 8.6 A Ionic composition

of the ground water from pits
excavated in the dry season
on the sand plain at some
distance from (I and II) and
close to the open lagoon (III)
respectively, of Chichiriviche,
Venezuela, as compared to
seawater. B Example of a soil
pit dug to examine the ground
water (Medina et al. 1989).
(For the position of the three
soil pits see also the transect
of Fig. 8.8)
8.2 Salinas 273
Fig. 8.7A–D Vegetation units of the alluvial plain at Chichiriviche, Venezuela. A Vegetation island
on the sand plain. B Halophyte zone with Sesuvium portulacastrum (front) and Batis maritima
(middle ground) around a vegetation island with Conocarpus erectus in the background. C Veg-
etation island with bushes of Conocarpus erectus and Subpilosocereus ottonis. D Grassland with
deciduous forest in the background
e) a deciduous forest characterized by species of Capparis, Caesalpinia coriaria,

Prosopis juliflora, Jacquinia revoluta, Maytenus karstenii, Erythroxylon cuma-
nense, Croton sp. and Pereskia guamacho (Fig. 8.7D).
A transect presenting finer details is shown in Fig. 8.8. In addition to the seasonal
differences in the water table the figure also gives an indication of the seasonal
Fig. 8.8 Transect of a part of the alluvial plain of Chichiriviche, Venezuela, showing topographical
variations with the major vegetation units (centre), the water tables (centre) and top soil chloride
concentrations in the rainy season and the dry season respectively (bottom) and the distribution of
the most frequent plant species (top). (After Medina et al. 1989)
8.2 Salinas 275
changes in salt content of the upper 10 cm of top-soil. Overall salt concentrations in

the soil as well as seasonal fluctuations tend to be highest in the bare areas and de-
crease with the sequence of vegetation units as follows: halophyte zone – grassland
zone – vegetation islands – deciduous forest.
8.2.2 Dynamics of Vegetation Islands
The most conspicuous feature of these salinas are the small vegetation islands with
a diameter of 3 – 10 m and a soil surface 10 – 40 cm higher than the sand plain. Ob-
servers have been tempted to consider these islands as a particular stage in a progres-
sive succession, which starts from the bare sand plain, then leads to the halophyte
vegetation, followed by island vegetation and finally on to grassland and deciduous
forest (Walter 1973).
However, a closer examination extending over several years has shown that there
is no such one-way progressive succession towards a stable climax community at the
end. There are oscillations between the various vegetation units in time which may
be determined by medium-term climatic fluctuations; e.g. the years between 1966
and 1975 appeared to be wetter, and the years between 1976 and 1986 were drier
than the long term average (Fig. 8.9). The whole ecosystem is highly dynamic and
provides an excellent example of an oscillating mosaic, as opposed to a stable climax
equilibrium (see Sect. 3.3.3). By following the development of a given island, we
see that islands not only grow into savannas and forest but also die, being eroded
and eventually disappearing into the bare sand plain (Fig. 8.10). Thus, these islands
appear to be metastable states between the more stable states of the forest and the
sand plain, respectively.
Fig. 8.9 Variations of to-

tal annual rainfall near
the northern Caribbean
coast of Venezuela close
to Chichiriviche for the years
1964 – 1986. (Medina et al.
1989)
8.2 Salinas 277
Fig. 8.10 A Primordial or decaying vegetation island? B–E Continuous observation of a decaying
and eroding vegetation island, i.e. the same island in November 1985 (B), March 1986 (C), April
1987 (D) and January 1989 (E). North Coast of Venezuela near Chichiriviche
8.2.3 Strategies of Adaptation of Plants

in the Different Vegetation Units
8.2.3.1 Small Perennial Halophytes: Salt Inclusion and Stress Tolerance
The zone of small perennial halophytes surrounding the edges of vegetation islands
and bordering the vegetation free salt flats is dominated by three species, namely
Portulaca rubricaulis, Sesuvium portulacastrum and Batis maritima (Fig. 8.11). The
creeping fruticose stems of the first two species contrast with B. maritima, which
has a more upright growth habit. The three species are also primary colonizers of
8.2 Salinas 279
Fig. 8.11A–E Portulaca rubricaulis (A rainy season Nov. 1985); Sesuvium portulacastrum (B rainy
season, Nov. 1985; C dry season March 1986); Batis maritima (D rainy season Nov. 1985; E dry
season March 1986) in the sand plain of Chichiriviche, Venezuela
the sand plain. They occupy the most extreme habitats in the sand plain, which range
from being flooded with fresh water in the rainy season to dry, salt-encrusted soil
in the dry season when the water table may drop to 1 m below ground (Lüttge et al.
1989b; Figs. 8.5 and 8.8).
All of the three halophytes are salt includers and accumulate NaCl in their highly
succulent leaves. However, P. rubricaulis combines this strategy of stress tolerance
with that of stress avoidance in that it is deciduous and sheds its leaves during the
dry season. Among the three species, it has the lowest salt concentrations in its leaf
sap, viz. 230 mM Cl− and 60 mM Na+ , while the other two species have much
higher salt levels in their leaves, i.e. 260 – 1,080 mM Cl− and 370 – 720 mM Na+ in
the wet season, and 540 – 1,410 mM Cl− and 920 – 1,590 mM Na+ in the dry sea-
son, respectively. However, P. rubricaulis is a C4 -plant while the other two species
are C3 -plants. Thus P. rubricaulis may use the higher instantaneous productivity
of C4 -photosynthesis (see Box 10.2) for production of enough perennial shoots to
compete in the habitat effectively.
B. maritima also accumulated sulphate, with a twofold increase of leaf-sap con-
centrations in the dry season. In S. portulacastrum, Na+ accumulation exceeded
Cl− accumulation by far and synthesis of the organic acid anion oxalate is found to
serve in maintaining charge balance. Increased salt accumulation in the leaves of B.
maritima and S. portulacastrum in the dry season is accompanied by a 1.5- to 2-fold
increase in leaf succulence (see also Sect. 7.4). S. portulacastrum was also shown
to use compatible solutes (see Sect. 7.4 and Box 7.1) such as proline and pinitol
which augments the tolerance of salt inclusion (Fig. 8.12).
In leaves of the erect stems of B. maritima photosynthetic gas exchange, mea-
sured as net CO2 uptake and transpirational loss of water vapour, shows little re-
sponse to the transitions between the rainy and the dry season (Fig. 8.13). In con-
trast, photosynthesis in the leaves of the prostrate stems of S. portulacastrum is
severely impaired in the dry season, showing a pronounced midday depression of
gas-exchange (see Sects. 5.2.2.1 and 10.1.2.3) and about 40% inhibition of light
saturated rates of photosynthesis (Fig. 8.13). Thus, S. portulacastrum clearly suf-
fers more under the stress of the dry season but it is also more of a pioneer coloniser
of the sand plain as it occupies the outermost edges of the vegetation islands and
larger areas of the flat plain (Figs. 8.7B and 8.11C).
In summary, the three species of halophytes, which are subject to very similar
challenges by extreme environmental conditions, have different strategies of adap-
tation to stress. Notwithstanding the similarities in life-forms, with small fruticose
stems and succulent leaves, the differences in ecophysiological comportment reflect
ecological diversity, and once again prove to be a basis for species diversity (see
Sect. 3.3.2).
Fig. 8.12 Concentrations of

the compatible solutes proline
and pinitol in the leaf sap of
Sesuvium portulacastrum in
the wet season (W ) and the
dry season (D). (Lüttge et al.
1989b)
8.2 Salinas 281
Fig. 8.13 Net CO2 exchange

(JCO2 ) and water-vapour
conductance of leaves (gH2 O )
of Sesuvium portulacastrum
(•, ◦) and Batis maritima
(, ) in the sand plain of
Chichiriviche, Venezuela, in
the wet season (◦, ) and the
dry season (•, ). (Lüttge et
al. 1989b)
8.2.3.2 Terrestrial CAM Plants: Salt Exclusion and Stress Avoidance
There are two different life forms of terrestrial CAM plants on the sand plain, which
are stem succulent cacti and tank forming bromeliads. They are salt excluders. In
this way they avoid the physiological stress of salinity, and their use of the flexibility
of the CAM cycle (see Sect. 5.2.2.2 and Box 5.1) is important in this strategy in the
salinas (Lüttge 1993).
8.2.3.2.1 Columnar Ceroid Cacti
The dominating cactus of the salinas at the North coast of Venezuela is the colum-
nar cactus Subpilosocereus ottonis with strongly branching individuals more than
6 m tall (Fig. 8.14A). Small seedlings are frequently found at the rim of vegeta-
tion islands (Fig. 8.14B) as well as among other vegetation and within the tanks
Fig. 8.14A–F Large plants (A), small seedling (B) and seedling rerooted after injury in the field (C)
of Subpilosocereus ottonis in the alluvial plain of Chichiriviche, Venezuela. Experimental seedlings
of Cereus validus (D), one rotting during an extended salt treatment (E), and another one dried out
at the bottom and totally insulated from the substratum (F)
8.2 Salinas 283
of bromeliads. A similar type of cactus occurs in the subtropical salinas, Sali-

nas Grandes in Argentina, namely Cereus validus. Experiments with 10 cm tall
seedlings of C. validus (Fig. 8.14D) and analyses of S. ottonis in the field show
that these cacti are strong salt excluders (Fig. 8.15). When subject to salinity the
roots rapidly accumulate large amounts of NaCl. However, there is no salt export
from the roots to the shoots, so that the peripheral green stem chlorenchyma and
the central water storage parenchyma of the stems receive very little additional salt
during a salinity treatment of up to 14 days and 600 mM NaCl in the root medium
(Fig. 8.15).
It is quite obvious that the fine absorptive roots of the cacti die under the stress of
salinity and the rest of the cactus becomes quite isolated from the substratum. When
the salinity treatment is extended for several months, in some cases salt solution may
diffuse upwards killing the stem tissue so that eventually the whole cactus seedling
rots away and dies (Fig. 8.14E). In other cases, however, the base of the shoot dries
out and then finally becomes insulated very effectively from the ground (Fig. 8.14F).
The major part of the green stem survives. But by what means and for how long?
Certainly, the cacti cannot survive indefinitely without water and nutrient supply
from the substratum. In the strongly seasonal salinas, of course, they only need
Fig. 8.15A,B Effects of NaCl solutions supplied to the roots of small seedlings of Cereus validus
(see Fig. 6.14D–F) in pot culture with sand on Na+ and Cl− levels in the roots, the water storage
stem parenchyma and the peripheral green chlorenchyma. A NaCl concentrations in the watering
solution were increased by daily increments of 50 mM up to 600 mM (upper abscissa), and the
plants were analyzed as soon as the respective concentrations were reached at the times indicated
(lower abscissa). B NaCl concentrations in the watering solution were increased by daily incre-
ments of 50 mM. At any given concentration indicated on the abscissa, plants were kept for 14 days
after this concentration was reached and then analyzed. (Nobel et al. 1984)
Fig. 8.16A,B Root system of Subpilosocereus ottonis as shown on a seedling (A) and a tall fallen
cactus (B) with extended horizontal root system and vertical tap roots
8.2 Salinas 285
to overcome the dry season when salinity-stress is present, particularly if they are
able to form functional roots again in the wet season. Indeed, cacti are known to
be capable of rapid adventitious root regeneration (Fig. 8.14C). S. ottonis also
develops a large horizontal root system from which strong vertical tap roots protrude
into the soil, and from which fine absorptive roots have a seasonal turnover related
to substratum salinity (Fig. 8.16).
To overcome an extended dry season, the insulated stems of the cacti use the
possibility of nocturnal recycling of respiratory CO2 provided by the CAM-
mechanism (see Sect. 5.2.2.2 and Box 5.1). The experimental seedlings of C. validus
under salt stress reduced photosynthetic gas exchange (Fig. 8.17), and the contribu-
tion of nocturnal CO2 -recycling to total night-time malate accumulation increased
from 20% in the controls to 50% in the NaCl-treated plants. In the extreme case of
total insulation, stomata may close permanently during both day and night, reducing
gas exchange to an absolute minimum. In this way the plants do not gain carbon,
but they minimize loss. Carbon from respiratory CO2 is recycled into malate during
the night and, after decarboxylation of malate, in photosynthetic CO2 -fixation and
carbohydrate synthesis through the Calvin cycle during the day. Thus, metabolism
Fig. 8.17 Effect of NaCl on photosynthetic gas exchange of small seedlings of Cereus validus.
The control plants were irrigated with water; the NaCl-treated plants received NaCl solutions of
daily increments of 50 mM until 400 mM NaCl was reached and were then kept for 16 days at this
NaCl level. The horizontal black bar indicates the dark period. (Nobel et al. 1984)
Fig. 8.18 Loss of fresh

weight, i.e. loss of water,
by small plants of Subpi-
losocereus ottonis (about
0.3 m tall) derooted and
placed in full sun exposure
in the dry season starting on
day 0. Errors are SD, n was 6
to 18. (Lüttge et al. 1989a)
is maintained by respiratory and photosynthetic energy turnover, and the only input
under these conditions is light energy, which keeps them alive.
Naturally, some loss of water vapour occurs via cuticular transpiration and
leads to a gradual reduction of the water reserves in the water-storage parenchyma
of the cactus stems (Fig. 8.18) and a decline of vitality. Its has been shown that cacti
survive when up to 54% of tissue water content is lost, although any subsequent loss
is lethal (Holthe and Szarek 1985). Thus, the chance of a small cactus to survive
the salinity stress of the dry season is much smaller than that of a large cactus.
A certain minimal biomass, with a sufficiently large water reserve in the water-
storage parenchyma, is required. Of course, survival also depends on the length of
the dry and wet season, respectively. If the wet season is longer and the dry season
relatively short, newly established seedlings have a better chance of survival. Thus,
as shown for cacti and agaves in the deserts of North America (Jordan and Nobel
1979, 1982) seedlings do not survive every year, and one observes age classes of
larger plants, which indicates the wetter periods when seedlings were able to become
established.
For the larger cacti growing on the vegetation islands, salinity stress in the top
soil is reduced even in the dry season. The major tap roots of 6 – 7 m tall plants of S.
Fig. 8.19 Profile of Cl− levels

in the soil of a large plant of
Subpilosocereus ottonis on
a vegetation island in the
sand plain of Chichiriviche,
Venezuela, during the dry
season. (Medina et al. 1989)
8.2 Salinas 287
ottonis are no longer than 50 cm, and even during the dry season they do not extend
below the point where salinity becomes 150 meq Cl− kg−1 air-dried soil (Fig. 8.19).
In conclusion, the major problem for the cacti in the salinas really is to survive the
vulnerable seedling stage.
8.2.3.2.2 Tank-Forming Bromeliads
Some tank-forming terrestrial bromeliads occur on the salinas. At the northern coast
of Venezuela by far the most frequent is Bromelia humilis, with the type II or tank-
root life form (see Sect. 6.4 and Fig. 6.15B). Since it does not necessarily need to
form soil roots, the leaf rosettes may simply lie on the ground, and thus, B. humilis
is effectively also a salt excluder and stress avoider.
Within tanks and through tank roots B. humilis can collect and utilize water.
Therefore, in contrast to the columnar cacti (Sect. 8.2.3.2.1), B. humilis can re-
plenish water reserves even from very small spells of rain during the dry season.
There is a marked peripheral water-storage parenchyma of thin-walled, non-green
and highly vacuolated cells, with little cytoplasm at the adaxial surface of the leaves
(see Fig. 6.22C). The leaf tissue loses water during the dry season, and the leaves be-
come less succulent and have increased dry weight : fresh weight ratios (Fig. 8.20).
Overall, the CAM plant B. humilis demonstrates water storage at three different
time scales:
• short term storage based on the osmotic effects of nocturnal malate accumulation
in the leaf cells (see Sect. 6.6.2.2),
• medium term storage in the tanks,
• long term storage in the water parenchyma.
Fig. 8.20 The degree of succulence (fresh weight : area) and dry weight : fresh weight (DW/FW)
ratios in leaves of shaded and exposed plants of Bromelia humilis in the wet season (W ) and in the
dry season (D) in the sand plain of Chichiriviche, Venezuela. (Lee et al. 1989)
B. humilis occurs in different vegetation units on the sand plain and expresses three
different phenotypes of growth form and pigmentation (see Sect. 4.1.2), namely the
dark green phenotype shaded under shrubs and trees of the deciduous forest, the yel-
low phenotype exposed on bare soil islands or in the grassland of the sand plain, and
a light-green intermediate phenotype, which also grows in relatively exposed condi-
tions (Fig. 4.3). The differential characteristics of shade and sun plants (Sect. 4.1.2)
are fully expressed in these phenotypes. In the dry season net CO2 exchange was
reduced with 7.2 mmol m−2 day−1 in the yellow and 22.2 mmol m−2 day−1 in the
green and shaded plants (Fig. 8.21), while an average rate obtained for all pheno-
types in the rainy season was 33.0 mmol m−2 day−1 .
In the dry season the plants operated with increasing internal CO2 -recycling
which corresponded to 56% in the green and 87% in the yellow phenotype (Fig.
Fig. 8.21 Net CO2 exchange (JCO2 ) during the night and in the early morning of a green shaded (•)
and a yellow exposed plant (◦) of Bromelia humilis in the sand plain of Chichiriviche, Venezuela.
The horizontal black bar indicates the night-period in the dry season. (Lee et al. 1989)
Table 8.2 Data on productivity of exposed and shaded plants of Bromelia humilis in the alluvial
plain of Chichiriviche, Venezuela. (Lee et al. 1989)
Exposed Shaded
Potential max. productivity (t DW ha−1 year−1 ) 6 6

Actual productivity (not including roots and ramets) (t DW ha−1 year−1 ) 0 3
FW/plant (kg) 0.18 0.54
Leaves/plant 42 70
New leaves/plant × year 36 48
Lost leaves/plant × year 36 23
New ramets/plant × year 0.24 0.67
Informationon productivity was obtained from DW/FW determinations and repeated measure-
ments and leaf counts of tagged plants over a period of 16 months from the end of one rainy
season to the end of a dry season and again in the following year. DW = dry weight, FW = fresh
weight
8.2 Salinas 289
8.21), while recycling for all phenotypes in the wet season was only 21%. Clearly,
the yellow exposed plants are under the most severe stress of
• high irradiance,
• drought,
• low nutrient supply.
The latter applies because the exposed plants also lack-supplies from decomposing
litter falling between the plants and into the tanks of the shaded plants in the decid-
uous forest. This is reflected in their productivity. It is seen that losses and gains are
more or less balanced and net productivity is close to zero in the yellow exposed
plants, as compared to the green shaded plants which have a distinct primary net
productivity (Table 8.2).
8.2.3.3 Epiphytic CAM Plants: Avoidance of Salinity Stress
On the large cacti and the shrubs of the vegetation islands one frequently finds
epiphytic CAM plants, especially the bromeliad Tillandsia flexuosa and the orchid
Schomburgkia humboldtiana (see Fig. 6.39). Although there may be some salt spray
driven inland by stronger winds, in their epiphytic habitat these plants largely avoid
salinity stress. CAM serves adaptation to drought stress (Sect. 5.2.2.2). Addition-
ally both species are myrmecophilous (Sect. 6.6.3). CO2 -acquisition in S. humbold-
tiana is greatly reduced in the dry season. In contrast, rates of CO2 -uptake are con-
stantly low in T. flexuosa over both rainy and dry seasons (Fig. 8.22). Internal CO2 -
Fig. 8.22 Net CO2 exchange (JCO2 ) of Schomburgkia humboldtiana (•, ◦) and Tillandsia flexuosa
(, ) in the wet season (◦, ) and in the dry season (•, ). The horizontal black bar indicates the
night period. (Griffiths et al. 1989)
recycling is similar in both plants at 65 – 76% and independent of season. T. flexuosa

is more frequent and so, despite lower potential maximum productivity, the physio-
logical characteristics maintain carbon acquisition continuously over the seasons.
8.2.3.4 Mangroves and Associates
The shrubby vegetation of the islands is dominated by the mangrove associate Cono-
carpus erectus. The true mangrove Avicennia germinans also plays a limited role,
particularly on vegetation islands closer to the lagoon with more permanent salt
stress. A germinans appears to be much more salt tolerant than C. erectus. The
more detailed ecophysiological comportment of the two species is discussed in
Sect. 7.5.1.
References
Araujo DSD (2000) Análise Florística e Fitogeográfica das Restingas do Estado do Rio de Janeiro.
D.Sc. Thesis, Universidade Federal do Rio de Janeiro, Rio de Janeiro
Araujo DSD, Scarano FR, Sá CFC, Kurtz BC, Zaluar HLT, Montezuma RCM, Oliveira RC (1998)
Communidades vegetais do Parque Nacional da Restinga de Jurubatiba. In: Esteves FA (ed)
Ecologia das Lagoas Costeiras do Parque Nacional da Restinga de Jurubatiba e do Município
de Macaé (RJ). Nupem-UFRJ, Rio de Janeiro, pp 39–62
Dias ATC, Scarano FR (2007) Clusia as nurse plant. In: Lüttge U (ed) Clusia. Ecological studies,
vol. 194. Springer, Berlin Heidelberg New York, pp 55–71
Dias ATC, Zaluar HLT, Ganade G, Scarano FR (2005) Canopy composition influencing plant patch
dynamics in a Brazilian sandy coastal plain. J Trop Ecol 21:343–347
Duarte HM, Geßler A, Scarano FR, Franco AC, Mattos EA de, Nahm M, Rennenberg H, Rodrigues
PJFP, Zaluar HLT, Lüttge U (2005) Ecophysiology of six selected shrub species in different
plant communities at the periphery of the Atlantic Forest of SE-Brazil. Flora 200:456–476
Geßler A, Duarte HM, Franco AC, Lüttge U, Mattos EA de, Nahm M, Scarano FR, Zaluar HLT,
Rennenberg H (2005) Ecophysiology of selected tree species in different plant communities
at the periphery of the Atlantic Forest of SE-Brazil II. Spatial and ontogenetic dynamics in
Andira legalis, a deciduous legume tree. Trees 19:510–522
Griffiths H, Smith JAC, Lüttge U, Popp M, Cram WJ, Diaz M, Lee HSJ, Medina E, Schäfer C,
Stimmel K-H (1989) Ecophysiology of xerophytic and halophytic vegetation of a coastal allu-
vial plain in northern Venezuela. IV. Tillandsia flexuosa Sw. and Schomburgkia humboldtiana
Reichb., epiphytic CAM plants. New Phytol 111:273–282
Henriques RPB, Araújo DSD De, Hay JD (1986) Descrição e classificação dos tipos de vegetação
da restinga de Carapebus, Rio de Janeiro. Rev Brasil Bot 9:173–189
Holthe PA, Szarek SR (1985) Physiological potential for survival of propagules of Crassulacean
acid metabolism species. Plant Physiol 79:219–224
Jordan PW, Nobel PS (1979) Infrequent establishment of seedlings of Agave deserti (Agavaceae)
in the northwestern Sonoran desert. Am J Bot 66:1079–1084
Jordan PW, Nobel PS (1982) Height distributions of two species of cacti in relation to rainfall,
seedling establishment and growth. Bot Gaz 143:511–517
Lee HSJ, Lüttge U, Medina E, Smith JAC, Cram WJ, Diaz M, Griffiths H, Popp M, Schäfer C,
a coastal alluvial plain in northern Venezuela. III. Bromelia humilis Jacq., a terrestrial CAM
bromeliad. New Phytol 111:253–271
References 291
Lüttge U (1993) The role of Crassulacean acid metabolism (CAM) in the adaptation of plants to
salinity. New Phytol 125:59–71
Lüttge U (2007a) Photosynthesis. In: Lüttge U (ed) Clusia. Ecological studies, vol. 194. Springer,
Lüttge U (2007b) Physiological ecology. In: Lüttge U (ed) Clusia. Ecological studies, vol. 194.
Springer, Berlin Heidelberg New York, pp. 187–234
Lüttge U, Medina E, Cram WJ, Lee HSJ, Popp M, Smith JAC (1989a) Ecophysiology of xerophytic
and halophytic vegetation of a coastal alluvial plain in northern Venezuela. II. Cactaceae. New
Phytol 111:245–251
Lüttge U, Popp M, Medina E, Cram WJ, Diaz M, Griffiths H, Lee HSJ, Schäfer C, Smith JAC,
Stimmel K-H (1989b) Ecophysiology of xerophytic and halophytic vegetation of a coastal
alluvial plain in northern Venezuela. V. The Batis maritima – Sesuvium portulacastrum vege-
tation unit. New Phytol 111:283–291
Martin L, Suguio K, Flexor JM (1993) As flutuações de nível do mar durante o quaternário su-
perior e a evolução geológica de “deltas” brasileiros. Boletim do Instituto de Geografia da
Universidade de São Paulo, Publicação Especial 15:1–186
Medina E, Cram WJ, Lee HSJ, Lüttge U, Popp M, Smith JAC, Diaz M (1989) Ecophysiology of
xerophytic and halophytic vegetation of a coastal alluvial plain in northern Venezuela. I. Site
description and plant communities. New Phytol 111:233–243
Nobel PS, Lüttge U, Heuer S, Ball E (1984) Influence of applied NaCl on crassulacean acid
metabolism and ionic levels in a cactus, Cereus validus. Plant Physiol 75:799–803
Reinert F, Roberts A, Wilson JM, Ribas L de, Cardinot G, Griffiths H (1997) Gradation in nutri-
ent composition and photosynthetic pathways across the restinga vegetation Brazil. Bot Acta
110:135–142
Rizzini CT (1979) Tratado de Fitogeografia do Brasil, vol 2. Edusp, São Paulo
Sampaio MC, Araújo TF, Scarano FR, Stuefer JF (2004) Directional growth of a clonal bromeliad
species in response to spatial habitat heterogeneity. Evol Ecol 18:429–442
Sampaio MC, Picó FX, Scarano FR (2005) Ramet demography of a nurse bromeliad in Brazilian
restingas. Am J Bot 92:674–681
Scarano FR (2002) Structure, function and floristic relationships of plant communities in stressful
habitats marginal to the Brazilian Atlantic rain forest. Ann Bot 90:517–524
Scarano FR, Duarte HM, Franco AC, Geßler A, Mattos EA de, Rennenberg H, Lüttge U (2005a)
Physiological synecology of tree species in relation to geographic distribution and ecophysio-
logical parameters at the Atlantic forest periphery in Brazil: an overview. Trees 19:493–496
Scarano FR, Duarte HM, Franco AC, Geßler A, Mattos EA de, Nahm M, Rennenberg H, Zaluar
HLT, Lüttge U (2005b) Ecophysiology of selected tree species in different plant communities
at the periphery of the Atlantic Forest of SE-Brazil I. Performance of three different species of
Clusia in an array of plant communities. Trees 19:497–509
Scarano FR, Ribeiro KT, Moraes LFD, Lima HC (1997) Plant establishment on flooded and un-
flooded patches of a freshwater swamp forest in souteastern Brazil. J Trop Ecol 14:793–803
Smith JAC, Popp M, Lüttge U, Cram WJ, Diaz M, Griffiths H, Lee HSJ, Medina E, Schäfer C,
a coastal alluvial plain in northern Venezuela. VI. Water relations and gas exchange of man-
groves. New Phytol 111:293–307
Ule E (1901) Die Vegetation von Cabo Frio an der Küste von Brasilien. Bot Jahrb Syst 28:511–528
Walter H (1973) Die Vegetation der Erde in ökophysiologischer Betrachtung, vol 1. Die tropischen
und subtropischen Zonen. G Fischer, Jena
Walter H, Breckle S-W (1984) Ökologie der Erde, vol. 2. Spezielle Ökologie der tropischen und
Zaluar HLT (1997) Espécies focais e a formação de moitas na restinga abertga de Clusia, Carpebus,
RJ. M.Sc. Dissertation, Universidade Federal do Rio de Janeiro, Rio de Janeiro
Chapter 9
Savannas. I. Physiognomy, Terminology
and Ecotones: Why Do Savannas Exist?
9.1 Physiognomy and Terminology
Savannas are open habitats typically dominated by grasses and often strongly af-
fected by seasonal changes of rainfall. The term savanna, sabana in Spanish and sa-
vana (or campo) in Portuguese, is a West Indian expression of uncertain Caribbean
origin (Huber 1987). A classical example of a savanna is the Llanos north of the
Orinoco in Venezuela (Fig. 9.1). Alexander von Humboldt vividly described the
seasonal contrasts in this large Venezuelan savanna-area:
“When under the vertical rays of the sun, never covered by clouds the combusted layer of
grass has fallen into dust, the hardened soil cracks as if it were shaken by mighty earth-
quakes.”1
“The uniform vision of these steppes has some greatness but also some tristesse and depres-
sion in it. It is as if the whole nature would be frozen; scarcely every now and then the shade
of a small cloud, which hurries across the zenith and announces the near rainy season, falls
over the savanna. One hardly can get used to the vision of the Llanos, which offer a picture
like the surface of the sea. . . ”2
“. . . as a saying goes here: ‘The large ocean of greenery’ (‘los Llanos son como un mar de
yerbas’)”2
“When after a long drought the beneficial rainy season sets in, the scene in the steppe sud-
denly changes. When the surface of the earth is just wetted the fragrant steppe covers itself
with Kyllingias with highly panicled Paspalum and with a diversity of grasses. Stimulated
1 “Wenn unter dem senkrechten Strahl der niebewölkten Sonne die verkohlte Grasdecke in Staub
zerfallen ist, klafft der erhärtete Boden auf, als wäre er von mächtigen Erdstößen erschüttert. . . ”
2 “Der einförmige Anblick dieser Steppen hat etwas Großartiges, aber auch etwas Trauriges und
Niederschlagendes. Es ist, als ob die ganze Natur erstarrt wäre, kaum daß hin und wieder der Schat-
ten einer kleinen Wolke, die durch den Zenith eilend die nahe Regenzeit verkündet, auf die Savanne
fällt. Nur schwer gewöhnt man sich an den Anblick der Llanos, die. . . ein Bild der Meeresflächen
bieten.”
“. . . wie man hier oft sagen hört, ‘dem großen Meer von Grün’. . . ‘los Llanos son como un mar de
yerbas’. . . ” (Südamerikanische Reise, A. v. Humboldt 1808/1982 – Humboldt von 1982)
294 9 Savannas. I. Physiognomy, Terminology and Ecotones: Why Do Savannas Exist?
Fig. 9.1A–C The Llanos, Venezuela (A, B) and a savanna in Costa Rica, protected within the
borders of the Santa Rosa Park (C)
9.1 Physiognomy and Terminology 295
Fig. 9.2 Physiognomy of savannas with examples from all over the world. Transects of Vareschi
(Vareschi 1980, with kind permission of R. Ulmer)
by the light, herbaceous mimosas open their folded dormant leaves and greet the rising sun
like the early song of the birds.”3
In order to give an impression of the vegetation, Fig. 9.1 shows a picture of the
Llanos in Venezuela and a savanna in Central America, and the drawings of transects
in Fig. 9.2 present various types of savannas from all over the world. H. Walter
(Walter and Breckle 1984) distinguished between savanna and park-land as follows:
• Savanna: homogeneous plant communities with scattered woody plants (trees,
shrubs and bushes) in a grass layer closed in a greater or less degree over the soil
and with a few herbs in between.
• Park-land: a mosaic of forest islands in an open grassland with few woody plants
where the forest is associated with biotopes (e.g. river banks, valley bottoms,
hills), which are ecologically different from the grassland.
In contrast, the geographer C. Troll (see Walter and Breckle 1984) considered var-
ious types of landscapes as savannas which constitute a macro-mosaic of grassland
with different tree-formations (Fig. 9.3). Some of them are illustrated in the accom-
panying photographs:
• the gallery forest along rivers (Fig. 9.4),
• the forest of gullies,
• the termite savanna (Fig. 9.5),
• the “morichales” (Fig. 9.6).
Fig. 9.3 Macro-mosaics of plant formations with grassland in landscapes called savannas by C.
Troll. (Walter and Breckle 1984, with kind permission of S.-W. Breckle and G. Fischer-Verlag)
3 “Tritt endlich nach langer Dürre die wohltätige Regenzeit ein, so verändert sich plötzlich die
Szene in der Steppe. Kaum ist die Oberfläche der Erde benetzt, so überzieht sich die duftende
Steppe in Kyllingien, mit vielrispigem Paspalum und mit mannigfaltigen Gräsern. Vom Lichte
gereizt, entfalten krautige Mimosen ihre gesenkt schlummernden Blätter und begrüßen die aufge-
hende Sonne, wie der Frühgesang der Vögel.” (Ansichten der Natur, A. v. Humboldt 1849/1986 –
Humboldt von 1986)
Fig. 9.4A, B Gallery forests. A Rio Parupa, Gran Sabana, Venezuela. B In the cerrados near Brasília
(Fazenda Agua Limpa), Brazil
Fig. 9.5A, B Termite savannas. A After a fire, Queensland, Australia. The termite nests scattered
over the field are well visible after the vegetation has been burnt. B Acacia wooded savanna, Great
Rift Valley, Ethiopia
Fig. 9.6 Savanna with dense stands of the moriche palm, Mauritia flexuosa. Such morichales are
restricted to wet marshy parts of savannas. (Gran Sabana, Venezuela.) Alexander von Humboldt
has described the morichales as follows: The palm Mauritia flexuosa “at moist places forms mag-
nificent groups of fresh and shiny greenery, which recalls the green of our alder bushes. With their
shade these trees maintain the moisture in the soil . . . ”
“Sie bildet an feuchten Orten herrliche Gruppen von frischem glänzendem Grün, das an das Grün
unserer Ellergebüsche erinnert. Durch ihren Schatten erhalten die Bäume die Nässe des Bodens
...“
(A. v. Humboldt, Ansichten der Natur 1849/1986 – Humboldt von 1986)
This already illustrates the problems inherent in the use of the term savanna,
which has been applied to a wide range of habitats.
Attempts to delineate the term more precisely have been made repeatedly (Huber
1982, 1987, 1990). Clearly, in the open habitats of the savannas, the herbaceous
ground-layer is the ecologically decisive stratum. There may be shrubs and trees,
scattered or in small groups, but they never form a closed canopy. Thus, the major
contribution to the input of energy into the whole ecosystem (by capture of solar
radiation and primary biomass production) comes from the herbaceous ground-
stratum. The herbaceous layer itself may be dominated by grasses and sedges or by
broad leaved herbs so that one can distinguish between:
• grass savannas (Fig. 9.1), i.e. the savannas sensu strictu, and
• herb savannas (Fig. 9.7), which as “yerbazal” in Spanish, may not be considered
as savannas in a strict sense but then would require use of a rather awkward term
in English, e.g. “broad-leaved meadows”.
Typical tropical grass savannas are restricted to low and medium elevations not ex-
ceeding 1,000 – 1,200 m, while herb savannas may occur at higher elevations. In
Fig. 9.7 Yerbazal (herb savanna), Sierrania Parú (04◦ 25 N, 65◦ 32 W, 1,250 m a.s.l.), domi-
nated by Stegolepis hitchcockii (broad flat leaves), Brocchinia hechtioides (slender tank-forming
bromeliad) and Bonnetia crassa (small shrub)
either case it is essential that we are dealing with tropical or subtropical ecosys-
tems (Fig. 1.3A), where in contrast to the tropical forests with a closed canopy of
trees (Chap. 3) the ground stratum of grasses and/or herbs dominates energy
turnover (Huber 1982, 1987, 1990). This may also occur in ecosystems outside the
tropics, which then are distinguished as prairies, pampas or steppes. This strict dis-
tinction was not then familiar to Alexander von Humboldt who used “savanna” and
“steppe” synonymously. On the other hand, the term savanna has also been used for
physiognomic characterization of vegetation outside the tropics (Eiten 1986).
Even in the neighbouring countries of South America, Venezuela and Brazil,
classification of savanna-like vegetation has led to different terminologies (Sar-
miento 1984; Eiten 1972, respectively). For the vegetation in Brazil the cerrado-
concept was developed, which is somewhat narrower than the more general sa-
vanna-concept. In Brazil, 20% of the area of the whole country and 40% of the non-
Amazonian part are covered by cerrados (Eiten 1972; Gottsberger and Silberbauer-
Gottsberger 2006a,b). They are geographically and ecologically intermediate be-
tween tropical rainforest and tropical/subtropical desert. The annual precipitation
averages between 1200 and 1,600 mm ranging from 800 to 2,000 mm in the driest
and wettest parts, respectively. There is strong seasonality with more than 90% of
the rain falling in 7 months (October – April). The cerrado soil is very infertile and
can vary from less than 5% to over 95% sand, the rest is clay and a little silt (Eiten
1972, 1986).
Table 9.1 attempts a systematic comparison of the terms used to describe Vene-
zuelan and Brazilian savannas and cerrados (see also Figs. 9.1 and 9.8). The density
Table 9.1 Physiognomic description of various types of Venezuelan savannas and Brazilian cerra-
dos. (After Sarmiento 1984; Eiten 1972)
Terminology
Description Venezuela Brazil
1. Savannas without woody species taller Grass savanna Campo limpo
than the herbaceous stratum
2. Savannas with low woody species (< 8 m)
forming a more or less open stratum
a) Shrubs and trees isolated or in small Tree and shrub Campo sujo
groups, < 2% of total surface savanna
b) Shrubs and trees Woodland Campo cerrado
2 – 15% of total surface or bush savanna
c) Trees > 15% of total surface Woodland Cerrado (sensu strictu)
3. Savannas with tall trees (> 8 m)
a) Isolated trees, < 2% of total surface Tall tree savanna
b) Trees 2 – 15% of total surface Tall tree savanna
woodland
c) Trees 15 – 30% of total surface Tall tree wooded
grassland
d) Trees > 30% of total surface Tall woodland Cerradão
4. Savannas with large trees in small groups Park savanna Campo coperto
5. Mosaic of units of savannas and forests Park
Fig. 9.8A–D Aspects of the cerrado (A,B) and the cerradão (C,D) near Brasília (Fazenda Agua
Limpa), Brazil
and size of woody plants, i.e. shrubs and trees is an important feature of this system.
In Africa one also encounters the distinction between “wooded savanna”, where the
trees stand more or less isolated, and “woodland savanna,” where the canopies of
individual trees touch each other (Fig. 9.9).
Fig. 9.9A,B Wooded savanna (A) and woodland savanna (B), Great Rift Valley, Ethiopia
9.2 Seasonality
Water and nutrient availability distinguish different types of savannas (Baruch

2005). In his climatic-hydrological classification Sarmiento (1984) separates mainly
four types of savannas based on water status and seasonality (Fig. 9.10):
• a semi-seasonal savanna with a long rainy period but without excess of water
(i.e. flooding) and a short period with a water deficit,
• a seasonal savanna with changes between periods with sufficient water and pe-
riods of drought,
• a hyperseasonal savanna, where periods of excess of water and of drought pro-
vide strong seasonal contrast (see quotations of A. von Humboldt in Sect. 9.1),
• a marsh savanna, where long periods of water excess are interrupted by drier
periods, when water, however, still is in sufficient supply.
The latter merges into wetland ecosystems of which there is a large variety of types
which are difficult to separate and define (Esteves 1998). Many of them are river
floodplains in wet tropical forests (Sect. 3.2.3) and also associated with ponds and
lakes.
Fig. 9.10 Scheme of the water budgets of savannas in the climatic-hydrological classification of
Sarmiento (1984). Following the annual cycle described by the circumference of each circle shows
the extensions of annual cycles dominated by water excess (hatched area), normal water availabil-
ity (white area) and water deficit (dotted area) (Reprinted by permission of Harvard University
Press)
9.3 The Savanna Problem: Why Do Savannas Exist? 305
9.3 The Savanna Problem: Why Do Savannas Exist?
A more fundamental problem is why there are savannas at all. Why is closed forest
not growing all over these sites in the tropics? Are savannas natural plant communi-
ties or only products of human activities? There is no generally accepted hypothesis,
and a number of possibilities are listed by Huber (1982) as follows:
• The climatic hypothesis. This must be rejected for several reasons, but most
simply because of the co-occurrence of forest with closed canopies of trees and
open savanna under the same climatic conditions.
• The edaphic hypothesis, especially including the importance of the nutrient lim-
itation.
• The fluvial hypothesis, i.e. colonization of ancient riverbeds by savanna.
• The hydrological hypothesis, i.e. the important influence of the water regime
including limitations due to insufficient or excessive drainage.
• The relict or refuge hypothesis, where savannas are considered as relicts of
a formerly more widespread dry vegetation type.
• The anthropogenic hypothesis, implying the role of man in establishing, main-
taining and extending savannas especially by forest clearing and burning.
A steady state model of grass/tree coexistence based on separated rooting niches,
where trees have sole access to water in deeper soil horizons and grasses have prefer-
ential access to and are superior competitors for water in the surface soil (Sect. 10.1),
is criticized by Higgins et al. (2000). Instead they propose a non-equilibrium non-
linear model which is a very detailed model essentially based on a “storage” func-
tion considering tree seedling establishment and recruitment. Variations in rainfall
(Sect. 10.1.2), where establishment of tree seedlings requires several humid years
in sequence (Baruch et al. 1996) and fire (Sect. 10.3) on a background of low levels
of adult tree mortality allow the storage effect. The grass/tree coexistence is then
supported by the limited opportunities for tree seedlings to escape both drought and
the flame zone during fires into the adult stage. This prevents forest formation but
stores those individuals that have escaped.
In any event, savannas are not only man made. The cerrado of central Brazil also
is a natural, original vegetation and not derived from tropical mesophytic forest by
man’s destruction. Cerrados and natural savannas in the tropics are highly valuable
biotopes both floristically and ecologically. It is mostly overlooked, that they are
under enormous economic pressure and just as much threatened by the current de-
struction as tropical forests (Skole et al. 1994; Ribeiro et al. 2005). The cerrados of
Brazil belong to the 25 biodiversity hotspots of the world’s vegetation (Myers et al.
2000; Oliveira and Marquis 2002; Gottsberger and Silberbauer-Gottsberger 2006a).
They originally covered 1,783,200 km2 and are now already reduced to 20% of their
original area (356,630 km2 , 22,000 km2 or 6.2% of which are protected) and they
have 10,000 plant species, 4400 of which are endemic (Myers et al. 2000). In the
20 years from 1975 to 1996 the cerrado area covered by the crops soybean, maize,
rice and beans increased from 4.20 × 106 (15%) to 9.17 × 106 (28%) ha, where
the numbers in brackets give the cerrado area in per cent of the total area which
is covered in Brazil by these crops, and it is seen that the whole increment be-
tween 1975 and 1996 was due to cerrado cultivation (Resck et al. 2000; Gottsberger
and Silberbauer-Gottsberger 2006a). There may even arise some kind of unsavoury
contest in that destruction will increasingly turn towards savannas and cerrados as
forests are protected.
9.4 Ecotones
9.4.1 Savanna-Forest Ecotone Dynamics
Savannas and forests are separated by water and nutrient relations. A comparison
between two types of savannas and a rainforest in Table 9.2 suggests that the main
differences between the savannas and the forest are the very high free evaporation
and the deep infiltration, respectively. Comparing tree species in a savanna-forest
tension-zone (ecotone) where both savanna and forest trees had a high light demand
showed that the forest trees invested more in tall growth at the expense of root
system development, while conversely the savanna trees developed more extended
root systems and grew less tall (Hoffmann and Franco 2003). If nutrient supply
is taken as a second dimension in addition to water supply, the two-dimensional
separation of various types of tropical forests and savannas emerges as shown in
Fig. 9.11. The forests require high nutrient supply, or at least high availability of
water when nutrient supply is small, as in sclerophyll forest and in low-productivity
rainforest. Conversely, savannas occupy areas with medium to low supply of the two
resources.
Generally it is seen that the savanna-forest ecotone is moving towards the forest
and that especially due to anthropogenic influences forest is destroyed and replaced
by secondary savanna (Sect. 10.1.1.2). However, the opposite may also occur as
shown for a site in the Western Ghâts, India, where Mariotti and Peterschmitt (1994)
have performed historical studies of the dynamics of savanna-forest interfaces based
on the different carbon isotope ratios (δ 13C) in C3 - and C4 -plants (see Sect. 2.5). Sa-
Table 9.2 Water relations of two types of savannas compared to a tropical rainforest. (Sarmiento
1984)
Precipitation Evapotranspiration Free evaporation
(mm p.a.) (mm p.a.) (mm p.a.)
Trachypogon savanna 1839 1440 2406
(Calabozo, Venezuela)
Leptocoryphium savanna 1170 1115 2156
(Barinas, Venezuela)
Rainforest (Ivory Coast) 1800 – 1950 965 – 1000 a
a Infiltration deeper than 2.3 m: > 600 mm p.a. (in savannas = 0)

9.4 Ecotones 307
vannas are dominated by C4 -grasses (Sect. 10.1.1.2). Thus, the soil-organic matter
beneath, fed by the decomposing litter of savanna plants, should be much less neg-
ative than the soil under forest, where C3 -trees dominate. On this basis horizontal
and vertical δ 13C-analyses in soils have allowed to unravel historical savanna-forest
ecotone dynamics. Figure 9.12 shows that at the study site deeper soil layers at
80 cm and below had more negative δ 13C values indicating that the whole area once
Fig. 9.11 Separation of various types of savannas and tropical forests based on nutrient and water
availability. (Medina 1987)
Fig. 9.12 Vertical and horizontal profiles of δ 13 C values in the soil organic matter along a savanna-
forest topological gradient in Kattinkar, Western Ghâts (13◦ 57 N, 77◦ 44 E), India. (After Mariotti
and Peterschmitt 1994)
was dominated by forest. The soil directly under the present savanna shows the less
negative δ 13C values expected from the predominant C4 -photosynthesis by the veg-
etation and that under the present forest corroborates prevailing C3 -photosynthesis
by the forest trees. However, in the zone between forest and savanna, less nega-
tive δ 13C values extend deep under the present forest, indicating that the savanna
must have had a larger extension in the past, and that the forest must be currently
expanding.
9.4.2 Savanna-Desert Ecotone Dynamics:

The Sahel Problem as a Case Story
In Sect. 9.4.1 we have considered the savanna ecotone to the wetter side, i.e. the
forest. There is another important interface, to the drier side, namely that between
savanna and desert. The large deserts of the world lie mainly outside the tropics,
and thus, it is not within the scope of this book to treat the ecophysiology of desert
plants. It is interesting, however, to consider briefly the ecotone between savanna
and desert in addition to that between savanna and forest.
To do this, desertification in the Sahel region of Africa offers itself as an appro-
priate case story as it has caused much public concern due to the dramatic economic
and social implications. The Arab word “sahel” in fact means coast or shore, re-
ferring to the southern delineation of the sand-ocean of the Sahara, a pertinent way
to portray the ecotone. The region is characterized by summer rain with an 8–10
months long drought period. According to the annual precipitation a north-south
zonation is given as follows:
• saharo-sahelian transition zone 100 – 200 mm,
• sahelian zone 200 – 400 mm,
• sudano-sahelian transition zone 400 – 600 mm
(Fig. 9.13; Walter and Breckle 1984). The latitudinal position of the saharo-sahelian
transition zone, the savanna-desert ecotone, may vary due to climatic oscillations.
The case of the sahel is particularly interesting because very strong climatic oscil-
lations have been documented both for the extended period of the last 30,000 years
and for much shorter intervals in the last century. Figure 9.14 illustrates the large
changes of the area occupied by the Sahara over the ages. During the last ice age
(18,000–12,000 years B.C.) the Sahara had enlarged considerably and then, in the
post-glacial period (9,500–4,500 years B.C.) contracted again. At that time the river
Niger near Timbuctu had a large inland delta with a flooding plain of 20,000 km2.
At present the Sahara again occupies a large area similar to that in the last ice
age (Petit-Maire 1984). The stochastic appearance of drought periods interchanging
with wetter intervals or the movement of the saharo-sahelian transition zone more
to the south and more to the north respectively, during the last century is shown
in Fig. 9.15. Remote-sensing of the vegetation density (Sect. 2.3) resolves differ-
ences for individual years. The long wet period between 1942 and 1966 led to the
9.4 Ecotones 309
Fig. 9.13 Klimadiagramm graphs of four stations in the Sahel zone. (Walter and Breckle 1984,
extension of savannas and was followed by an increase of the human population

and the herds of the nomads. However, this prepared the ground for the catastro-
phes of the years after 1966. During the increasingly frequent and extended drought
periods the land could not sustain the population growth any longer. The exam-
ple is both tragic and illustrative. It shows that it is impossible to make long term
prognoses on the basis of few singular events during periods of short or medium
duration.
Fig. 9.14 Oscillations of the area occupied by the Sahara desert during the past 18,000 years.
(Petit-Maire 1984, with kind permission of La Recherche)
Fig. 9.15 Annual means of precipitation in the Sahel zone from 1922 to 1984 related to the conven-
tional Saharo-Sahelian transition line given here at 150 mm (dashed horizontal line); wetter years
hatched vertically, drier years dotted. (Petit-Maire 1984, with kind permission of La Recherche)
9.5 Productivity
Productivity of savannas is largely determined by nutrients. Generally, conditions

required for intensive and productive agriculture in the tropics can be listed after
Eiten (1972):
i) above rocks, which lead to soils rich in minerals, e.g. limestone and volcanic
rocks,
ii) in flood-plains with periodically rising water-tables, leading to repeated re-
newal of the mineral content of the soils, e.g. in the Amazonas, the Nile, the
Ganges,
References 311
iii) strong organic fertilization,

iv) use of chemical fertilizers.
Limitations to the third and fourth possibilities are the economic costs, and with
only small areas available for organic fertilization. Table 9.3 gives some data for
productivity of seasonal and hyperseasonal savannas.
Table 9.3 Productivity of savannas, biomass in g m−2 (Sarmiento 1984)
Savannas Above ground Below ground Total

Seasonal 800 – 1300 600 – 800 1400 – 2100
Hyperseasonal 800 – 1400 900 – 1100 1700 – 2500
References
Baruch Z (2005) Vegetation-environment relationships and classification of the seasonal savannas

in Venezuela. Flora 200:49–64
Baruch Z, Belsky JA, Bulla L, Franco AC, Garay I, Haridasan M, Lavelle P, Medina E, Sarmiento G
(1996) Biodiversity as regulator of energy flow, water use, and nutrient cycling in savannas. In:
Solbrig OT, Medina E, Silva J (eds) Biodiversity and savanna ecosystem processes. Springer,
Eiten G (1972) The cerrado vegetation of Brazil. Bot Rev 38:201–341
Eiten G (1986) The use of the term “savanna.” Trop Ecol 27:10–23
and amphibious plants in the neotropics. Oecologia Brasiliensis vol. IV, Universidade Federal
Gottsberger G, Silberbauer-Gottsberger I (2006a) Life in the cerrado a South American tropical
seasonal ecosystem. I. Origin, structure, dynamics and plant use. Reta-Verlag, Ulm
Gottsberger G, Silberbauer-Gottsberger I (2006b) Life in the cerrado a South American tropical
seasonal ecosystem. II. Pollination and seed dispersal. Reta-Verlag, Ulm
Higgins SI, Bond WJ, Trollope WSW (2000) Fire, resprouting and variability: a recipe for grass-
tree coexistence in savanna. J Ecol 88:213–229
Hoffmann WA, Franco AC (2003) Comparative growth analysis of tropical forest and savanna
woody plants using phylogenetically independent contrasts. J Ecol 91:475–484
Huber O (1982) Significance of savanna vegetation in the Amazon territory of Venezuela. In:
Prance GT (ed) Biological diversification in the tropics. Proc Vth Int Symp Assoc Trop Biol,
Venezuela 1979, Columbia Press, New York, pp 221–244
Huber O (1987) Neotropical savannas:their flora and vegetation. Trends Ecol Evol 2:67–71
Huber O (1990) Savannas and related vegetation types of the Guayana shield region in Venezuela.
In: Sarmiento G (ed) Las sabanas americanas, aspecto de su biogeografia, ecologia y uti-
lizacion. Centro de Investigaciones Ecológicas de los Andes Tropicales, Universidad de Los
Andes, Mérida, Venezuela, pp 57–97
Humboldt A von (1982) Südamerikanische Reise. 1808, quoted after the edition of Reinhard
Jaspert, Ullstein, Berlin
Humboldt A von (1986) Ansichten der Natur JG Cotta, Stuttgart 1849, quoted after the edition of
Greno Verlagsgesellschaft, Nördlingen
Mariotti A, Peterschmitt E (1994) Forest savanna ecotone dynamics in India as revealed by carbon
isotope ratios of soil organic matter. Oecologia 97:475–480
Medina E (1987) Nutrients: requirements, conservation and cycles in the herbaceous layer. In:
Walker B (ed) Determinants of savannas, IUBS Monographs Series No 3, Chap. 3. IRL Press,
Oxford, pp 39–65
Oliveira PS, Marquis RJ (eds) (2002) The cerrados of Brazil: ecology and natural history of
a neotropical savanna. Columbia University Press, New York
Petit-Maire N (1984) Le Sahara, de la steppe au désert. Recherche 15:1372–1382
Resck DVS, Vasconellos CA, Vilela L, Macedo MCM (2000) Impact of conversion of Brazilian
cerrados to cropland and pastureland on soil carbon pool and dynamics. In: Lal R, Kimble
JM, Stewart BA (eds) Global climate change and tropical ecosystems. CRC Press, Boca Raton
London, pp 169–196
Ribeiro JF, Bridgewater S, Ratter JA, Sousa-Silva JC (2005) Ocupação do bioma Cerrado e Con-
servação da sua diversidade vegetal. In: Scariot A, Sousa-Silva JC, Felfili JM (eds) Cerrado:
ecologia, biodiversidade e conservação. Brasília: Ministério do Meio Ambiente, pp 385–399
Skole DL, Chomentowiski WH, Salas WA, Nobre AD (1994) Physical and human dimensions of
deforestation in Amazonia. Biosci 44: 314–322
Walter H, Breckle S-W (1984) Ökologie der Erde. Bd 2. Spezielle Ökologie der tropischen und
Chapter 10
Savannas. II. The Environmental Factors Water,
Mineral Nutrients and Fire
Very important anthropogenic factors determining the existence of savannas are her-
bivory by cattle (see anthropogenic hypothesis in Sect. 9.3) and fire. The major nat-
ural environmental factors are:
• water (hydrological hypothesis, Sect. 9.3),
• mineral nutrients (edaphic hypothesis, Sect. 9.3),
• fires caused by natural events
(Högberg 1986a). These are discussed in this chapter.
10.1 The Water Factor
A model of the water budget of savannas explaining the various inputs and losses is
shown in Box 10.1. The key elements, of course, are precipitation for the input and
evapotranspiration by the vegetation plus free evaporation for the losses. The strong
seasonality often encountered in savannas is mainly determined by the water fac-
tor (Fig. 9.10). Woody plants and grasses in savannas have different requirements
from the annual precipitation, dependent on the distribution of rainfall, soil avail-
ability of water over the year and the water-capacity of the soil (Table 10.1). As
a result the phenological behaviour of savanna trees and grasses is also different
(Sects. 10.1.1.1 and 10.1.2.1). This strongly affects overall fluxes of carbon, water
and energy in the seasons as documented by an important comprehensive investi-
gation in a Brazilian cerrado (Miranda et al. 1997) some comparisons of which are
depicted in Fig. 10.1. The leaf area index (see Sect. 3.4.1) was 1.0 and 0.4 in the
wet and dry season, respectively, and during the light periods of the days loss of
H2 O-vapour (transpiration rate), downward fluxes of CO2 and ecosystem surface
conductance were much higher in the wet season than at the end of the dry season.
With the different hydraulic requirements of grasses and trees it is appropriate to
consider them separately in the following two sections.
314 10 Savannas. II. The Environmental Factors Water, Mineral Nutrients and Fire
Fig. 10.1 Canopy fluxes of

water vapour (JH2 O ) (A),
carbon (JCO2 ) (B) and sur-
face conductance (C) over
a Brazilian cerrado towards
the end of the wet season
(April 1993, solid lines)
and at the end of the dry
season (September 1993, dot-
ted lines). Negative values
of JCO2 and JH2 O give the
downward fluxes from the
atmosphere to the canopy.
(Drawn after data of Miranda
et al. 1997; with kind permis-
sion of Blackwell Science,
Oxford, UK)
Table 10.1 Different requirements of grasses and woody plants in savannas with respect to the
water factor. (After Walter and Breckle 1984)
Grasses Woody plants

(i) Amount of precipitation
Lower amounts of annual precipitation Larger amounts of annual precipitation
(ii) Annual distribution of precipitation
Precipitation must occur Precipitation may occur
during the growth period during the rest period
(iii) Annual distribution of water uptake
During the rest period no water The soil must provide enough water
is taken up from the soil for a minimal water-uptake also
in the dry season
(iv) Soil-water capacity
The water capacity of the soil The water capacity of the soil does not
must be high: need to be high:
the plants do not limit their transpiration as the crumb size of the soil may be large,
long as the soil provides enough water, and the soil may be stony and rocky; the root
then the leaves dry rapidly system develops far in both horizontal and
vertical direction
Box 10.1
Model of the water budget of savannas with a continuous vegetation-soil com-
partment separated from the atmosphere. Rom. = losses, ital. = input.
(Sarmiento 1984; reprinted by permission of Harvard University Press).
10.1 The Water Factor 315
Box 10.1 (Continued)
10.1.1 Grasses
10.1.1.1 Phenology
Due to the mostly only superficial rooting of grasses phenological differentiations

are highly important for adaptations of grasses relating to the water factor (Table
10.1). The following phenological groups are observed among savanna grasses:
• perennial with a seasonal semi-dormant-period,
• annual, ephemeral with a short cycle,
• annual with a long cycle,
• perennial with a seasonal dormant period,
• continuous growth and flowering.
The first group is most frequent. It is represented, for example, by Trachypogon

plumosus and Leptocoryphium lanatum (Fig. 10.2). The phenological diagrams
for the two species, in contrast to the tree Curatella americana, Dilleniaceae (see
Sect. 10.1.2.1), show flowering and new shoot production during the rainy season,
but closer to the end of the rainy season in T. plumosus than at the beginning as
Fig. 10.2A–C Semiquantitative phenograms (relative units on the ordinates) of two tropical grasses
with C4 -photosynthesis (A, B) and a tropical savanna tree (C). (Sarmiento 1984; reprinted by
permission of Harvard University Press)
for L. lanatum (Fig. 10.2). The phenological diagram of L. lanatum also shows the
stimulation by fire of shoot production at the end of the dry season (see Sect. 10.3.3
for more details). Such phenological diagrams, as discussed by Solbrig (1993; for
many phenological diagrams of cerrado plants see Gottsberger and Silberbauer-
Gottsberger 2006) allow the separation of functional groups of savanna grasses.
Thus, there are grasses that grow early and reproduce quickly in the rainy season
(more like L. lanatum in Fig. 10.2) as opposed to grasses that grow gradually and
develop shoots slowly and reproduce in the middle or towards the end of the rainy
season (more like T. plumosus in Fig. 10.2). In general terms, the former (i.e. early
growers and reproducers):
• are more drought resistant,
• have higher turgor pressures during the dry season,
• have higher water use efficiencies (WUE),
• partition more of their photosynthetic products to roots and below-ground organs,
• are more competitive under dry conditions and have increased importance along
wet to dry gradients.
Grasses need less water than savanna trees, but the water must be available during
the growth period.
10.1.1.2 Metabolic Adaptation: C4 -Photosynthesis
A major ecophysiological aspect of savanna grasses is the dominance of C4 -

photosynthesis (Box 10.2). C4 -photosynthesis is occurring among a few shrubs
and woody plants but it is absent in real trees. Although it is frequent in some
dicotyledonous families, such as the Chenopodiaceae, C4 -photosynthesis is a typ-
ical biochemical trait of sub-tropical and tropical grasses. Somewhat similar to
CAM (see Sect. 5.2.2.2 and Box 5.1), during C4 -photosynthesis CO2 is first cycled
through malate, before it is assimilated in the Calvin cycle. However, during C4 -
photosynthesis there is simultaneously CO2 -fixation via PEP-carboxylase (PEPC),
together with CO2 -remobilisation, refixation via RuBISCO and reduction via the
Calvin cycle. The PEPC- and RuBISCO-functions are localized in different cell
types and hence separated spatially, in contrast to CAM, where they occur in the
same cells but are separated in time. Leaves of most C4 -plants have two distinct
photosynthetic tissues, first the outer spongy mesophyll where CO2 is fixed to pro-
duce malate and in many cases also aspartate via oxaloacetate, and second, the inner
bundle sheath where malate or aspartate are decarboxylated and the CO2 is refixed.
Depending on the enzymic mechanism of decarboxylation, three different types of
C4 -photosynthesis can be distinguished (Box 10.2). Because the affinity of PEPC
for CO2 is about 60 times higher than that of RuBISCO, fixation of atmospheric
CO2 in the mesophyll, which tightly surrounds the bundle sheath, is highly effec-
tive. The malate and aspartate so produced are transported symplastically to the
bundle sheath, via plasmodesmata connecting the two tissues. Frequently, there is
suberization of the cell walls between the two tissues to prevent leakage to the
Box 10.2 Biochemical pathways of C4 plants

Primary CO2 fixation via phosphoenolpyruvate carboxylase (PEPC) and refix-

ation of CO2 via ribulose-bis-phosphate carboxylase (RuBISCO) occur simul-
taneously in time and are separated in space.
The biochemical reactions in the mesophyll are basically similar in all types
of C4 plants. The first CO2 fixation product is the C4 acid anion oxaloacetate,
which is subsequently transformed to malate and/or aspartate. Three types of
C4 plants are distinguished by the mode of decarboxylation of these C4 acids
after their transport to the bundle-sheath cells:
• the NADP-malic enzyme type [reaction (3)],
• the NAD-malic enzyme type [reaction (8)],
• the PEP-carboxykinase type [reaction (10)].
Enzymatic reactions
(1) PEP-carboxylase (PEPC),
(2) NADP-dependent malate dehydrogenase,
(3) NADP-dependent malic enzyme,
(4) Pyruvate, Pi dikinase,
(5) 3-PGA kinase, NADP-dependent glyceraldehyde-3-P dehydrogenase and
triose-P isomerase,
(6) Aspartate aminotransferase,
(7) NAD-dependent malate dehydrogenase,
(8) NAD-dependent malic enzyme,
(9) Alanine aminotransferase,
(10) PEP-carboxykinase,
(11) Mitochondrial NADH oxidation systems.
Metabolites and cofactors
AMP, ADP, ATP: adenosine mono-, di- and tri-phosphate;
DHAP: dihydroxyacetone phosphate;
NAD: nicotine-adenine-dinucleotide;
NADP: nicotine-adenine-dinucleotide phosphate;
OAA: oxaloacetic acid;
P: phosphate;
PCR: photosynthetic carbon reduction;
Pi : inorganic phosphate;
PPi : inorganic pyrophosphate;
PEP: phosphoenolpyruvate;
PGA: phosphoglyceric acid;
RubP: ribulose-bis-phosphate.
(Hatch and Osmond 1976; Hatch 1987)
apoplast, and decarboxylation in the bundle sheath leads to a 6- to 10-fold increase

of CO2 -concentration as compared to atmospheric CO2 . This has several ecophys-
iological advantages which are important for savanna grasses in dry open habitats
with high irradiation:
• Under water stress the high CO2 -affinity of the first step of CO2 -fixation (PEPC)
draws down CO2 -concentration inside the leaf, providing a steeper gradient for
inward diffusion of CO2 , and allows operation of photosynthesis with partially
closed stomata, which reduces transpiratory loss of H2 O. Hence, the water use
efficiency of C4 -plants is much higher than that of C3 -plants, although still lower
than that of nocturnal CO2 -fixation in CAM plants (Box 10.3).
• The high CO2 -affinity of PEPC, together with the simultaneous use of light in
refixation of CO2 via RuBISCO, also allows high maximum rates of photo-
synthesis and high productivity, which in C4 -plants are the highest of the three
modes of photosynthesis (Box 10.3).
• The high CO2 -concentration in the bundle sheath cells reduces photorespi-
ration and the plants are less susceptible to the danger of photoinhibition and
photodamage.
Box 10.3
Some major ecophysiological characteristics of the three major modes of pho-

tosynthesis in terrestrial plants, viz. CAM (D dark; L light period), C4 - and
C3 -photosynthesis. (Black 1973)
CAM C4 C3
Water-use efficiency (WUE) (6 – 30) ×10−3 (D) (1.7 – 2.4) ×10−3 (0.6 – 1.3) ×10−3
(mol [CH2 O]: mol H2 O) (1 – 4) ×10−3 (L)
Maximum net CO2 0.5 – 2.5 (D) 25 – 50 10 – 25
uptake (µmol m−2 s−1 ) 7 – 8 (L)
Maximum productivity 1.5 – 1.8 400 – 500 50 – 200
(g DW m−2 day−1 )
Therefore, it is not surprising that C4 -grasses dominate in tropical savannas and

C3 -grasses are more scarce. The most frequent tribes and genera are listed in
Table 10.2. Table 10.3 gives a comparison of average daily fluxes of carbon over
the canopy of a C4 -pasture and a forest in SW-Amazonia, Brazil, and Table 10.4
lists representative rates of photosynthesis with the highest maximum rates obtained
in the C4 -grasses. In addition to pastures some tropical C4 -grasses have been devel-
oped into major agricultural crops of mankind, e.g. sorghum, millet, maize and sugar
cane; the Ethiopian tef is also a C4 -grass, Agrostis tef.
Although evolution of C4 -photosynthesis dates back to much earlier periods of
lowered atmospheric CO2 levels in the Mesozoic, at the geological times when sa-
vannas and cerrados began to develop in the neotropics in the late Oligocene 30
to 25 ×106 years ago they were dominated by C3 -grasses. In the Miocene the first
C4 -grasses began to appear 10 ×106 years ago. C4 -grasses became dominant over
C3 -grasses 7.6 ×106 years ago and reached absolute dominance in the Pliocene
3.7 ×106 years ago (Cerling et al. 1997, 1998; Jacobs et al. 1999; Fig. 12 in Gotts-
berger and Silberbauer-Gottsberger 2006). African C4 -grasses are also being intro-
duced anthropogenically in the neotropics and are invading savannas displacing the
native herbaceous vegetation. In relation to the advantages of C4 -photosynthesis
listed above the competitive superiority of the African grasses is due to:
• higher net-photosynthesis rates,
• more efficient use of soil nutrients,
• higher proportion of assimilates allocated to new leaves,
• higher tolerance to defoliation.
Many grasses in the Llanos of Venezuela are also C4 -grasses. In the Llanos, the
native grass Trachypogon plumosus and the successful invader Hyparrhenia rufa,
both C4 -plants, are distinguished as follows:
• H. rufa, has higher transpiration and stomatal conductance for water vapour, us-
ing water opportunistically when available; it shows earlier leaf-senescence dur-
ing the dry season, i.e. drought avoidance; however, it needs relatively deep soils;
• T. plumosus, uses water conservatively; it is more drought tolerant and can with-
stand poorer nutrient and water status on shallower soils (Baruch and Fernandez
1993).
Table 10.2 Genera of neotropical savanna grasses according to their photosynthetic pathway. (Ref-
erence to Venezuelan tribes, Montaldo 1977; Medina 1982)
Tribe Genera
C4 -grasses
Eragrosteae Eragrostis, Leptochloa
Chlorideae Microchloa, Bouteloua, Chloris, Gymnopogon
Sporoboleae Sporobolus
Paniceae Digitaria, Eriochloa, Paspalum, Echinochloa, Axonopus
Andropogoneae Imperata, Andropogon, Trachypogon, Diectomis
Aristideae Aristida
Arundinelleae Tristachya
C3 -grasses
Paniceae Lasiacis, Oplismenus
Table 10.3 Assessments of average daily fluxes of carbon (mol m−2 day−1 ) over the canopy of
a C4 -pasture of the dominant introduced C4 -grass Brachiaria brizantha and a forest in comparison.
(Grace et al. 1998)
Photosynthesis Respiration Net

C4 -pasture 0.67 0.51 0.16
Forest 0.57 0.55 0.02
Table 10.4 Representative rates of photosynthesis (µ mol CO2 m−2 s−1 ) of savanna grasses with
C3 - and C4 -photosynthesis. (Medina 1986)
South American grasses African grasses

C3 -grasses: 11.1 – 13.2 –
C4 -grasses:
Field data 3.2 – 16.4 2.2 – 7.9
Laboratory data 28.1 14.7 – 43.9
In the tropics also altitude determines the relative abundance of C4 - and C3 -

grasses. The former only dominate at lower altitudes and C3 grasses strongly prevail
at higher altitudes (Fig. 10.3). The essential factors are:
• temperature,
• water availability,
• irradiance, and
• to a more limited extent nutrient availability.
Although irradiance increases with altitude, temperatures are lower, water availabil-
ity is larger, and the altitudinal distribution of C4 - and C3 -grasses in Kenya has been
related to the soil-moisture index (Table 10.5). Moreover, cold nights in tropical high
altitude sites play an important role (Chap. 12). A study in Papua New Guinea has
shown that C4 -grasses at altitudes of 1550 and 2,600 m a.s.l. were chilling sensitive,
while C3 -grasses at 3280, 3580 and 4,350 m a.s.l. were chilling resistant (Earnshaw
et al. 1990).
CAM plants are rare in typical savannas. Stem succulent CAM species, i.e. Cac-
taceae in America and Euphorbiaceae in Africa, are important in dry thornbush-
Fig. 10.3 Relative species 100 0

diversity of C3 - and C4 -
90 10
grasses along an altitudinal
gradient in Ecuador based 80 20
on 220 total species. (After
% C3 -species
% C4 -species
Boom et al. 2001) 70 30
60 40
50 50
40 60
30 70
20 80
10 90
0 100
0 1000 2000 3000 4000 5000
Altitude ( m )
Table 10.5 Altitudinal distribution of C4 - and C3 -grasses in Kenya. The soil moisture index is in
arbitrary units increasing approximately linearly with altitude (10 at 600 m; 100 at > 3,500 m).
(Tieszen et al. 1979)
Altitude (m) Soil moisture index C4 - vs C3 -photosynthesis
< 2000 50 C4 -grasses dominate
C3 -grasses only in the
shade
2,000 – 3,000 60 – 70 Transition zone
> 3000 80 C3 -grasses dominate
forests of tropical lowlands (Sect. 3.2.4); CAM-epiphytes are abundant in wet

forests of medium altitudes (Sects. 6.6.1 and 6.6.2.2); and at higher altitudes, terres-
trial CAM species may also be frequent in dry sites of open habitats (Sect. 12.3.3).
Towards the wet side of the spectrum of savanna types (Fig. 9.10), i.e. in wet-
land ecosystems, C4 -photosynthesis has also proved to be highly successful. For
example in fertile flood plains of nutrient rich rivers and lakes (white waters) of the
Amazon region in South America (Sect. 3.2.3) the perennial C4 -grass Echinochloa
polystachya may form monotypic stands over 5,000 km2 and displays extraordinar-
ily high rates of net-CO2 uptake in photosynthesis of 30 – 40 µ mol m−2 s−1 (com-
pare Table 10.4) with fast growth and high productivity during the wet season of
108 tonnes ha−1 year−1 (comparable to the productivity of sugar cane plantations)
when the flood plains are submerged. During the shorter dry period CO2 -uptake
rates are 17 µ mol m−2 s−1 and photosynthesis shows a midday depression (Piedade
et al. 1992, 1994; Esteves 1998).
10.1.2 Trees
10.1.2.1 Phenology
The trees can develop large root systems reaching far into the soil in both verti-
cal and horizontal directions. Nevertheless, due to the strong seasonality of wa-
ter supply in most savannas, appropriate phenological behaviour remains important
(Table 10.1). Trees need larger amounts of precipitation than grasses. The requisite
rain can also fall during dormant periods, and so a small amount of water uptake
into the trunk and branches must be possible even during drought periods.
One of the most striking phenological aspects is the flowering of savanna trees
which is most attractive when it occurs in the leaf less period of deciduous trees
(Fig. 10.4). An intriguing physiological question is how these flowers are supplied
with water in the absence of leaf transpiration and substantial xylem water flow. In
some species – although not in all of them – flowers can be supplied with water via
the phloem. This requires a sink for phloem flow, which could be nectar secretion
(Chapotin et al. 2003). In Africa and Australia a few trees are evergreen and most
are drought-deciduous, whereas in South America evergreen trees prevail (Medina
Fig. 10.4 See next page for details

Fig. 10.4 Flowering sa-

vanna trees of the Llanos
in Venezuela: Tabebuia
chrysantha (A), Tabebuia
orinocensis (B), Yacaranda
filicifolia (C), Pseudobombax
sp. (D), Palicourea rigida (E),
Byrsonima crassifolia (F)
Fig. 10.5 Curatella ameri-

cana; Llanos, Venezuela
1993). Curatella americana (Fig. 10.5), a dominant tree of the Venezuelan Llanos,
is evergreen with seasonal growth. Production of flowers and fruits, new leaves and
shoots begins in the middle to the last third of the dry season, so that the plants
are “ready” when the rainy season comes, as shown in the phenological diagram of
Fig. 10.2C.
Among the deciduous trees the length of the leafless periods may be different:
• trees with a short leafless period afford low (highly negative) water potentials
and high respiration rates, e.g. Bursera simaruba (Burseraceae), Spondias lutea
(Anacardiaceae), Pereskia guamacho (Cactaceae);
• trees with a long leafless period have high (less negative) water potentials and
low respiration rates, e.g. Tabebuia chrysantha (Bignoniaceae).
The precise phenological behaviour of plants in the tropical savannas (see also
grasses, Sect. 10.1.1.1) is a reliable indicator of season. For example, using up
the last water reserves for formation of reproductive and vegetative biomass in the
last third of the dry season, as shown for the savanna tree Curatella americana
(Fig. 10.5) in the phenogram of Fig. 10.2C, would be dangerous if the rainy season
were not close. This raises the question which signals are sensed in phenological
timing. It has been observed that several phenological phenomena, including wa-
ter budget, leaf abscission and flowering are related to phytochrome equilibria (see
Box 4.7) (Reich and Borchert 1984). Several suggestions have been made to explain
phenological timing on a hormonal basis. It was also proposed that water stress
itself is involved. However, accurate phenological timing is also observed in water
storing stem succulent savanna tress (Sect. 10.1.2.2), and moreover, with variations
of only 10 to 15 days year after year, its precision is much better than it would
be given by the stronger inter annual variations of rain fall (Borchert and Rivera
2001; Rivera and Borchert 2001; Rivera et al. 2002). Directly at and very near to
the equator endogenous regulation of annual rhythmicity must play a role (Wright
1991). However, even at very low latitudes away from the equator, photoperiod is
the decisive signal.
The proposal that the decisive external signal is photoperiod has long been re-
jected a priori, because near the equator the differences between the longest and the
shortest days are rather small (Reich and Borchert 1984). However, empirical facts
demonstrate very precise photoperiod sensing of plants. In the forest tree Hildegar-
dia barteri in Nigeria at 7◦ N where photoperiod changes by 53 min a similar de-
crease of photoperiod inhibits seedling leaf production (Njoku 1963; Wright 1996).
An experiment with Hyptis suaveolens (Lamiaceae), although an annual species and
not a tree, showed, that in principle plants can even sense differences in photoperiod
close to 20 min (Fig. 10.6; Medina 1982). Seeds were germinated at the beginning of
Fig. 10.6A, B Experiments with the annual tropical short-day plant Hyptis suaveolens showing that
plants can sense very small differences of photoperiod (daylength). Note that, by definition, short-
day plants require daylengths below a certain species-specific threshold for induction of flowering.
A The daylengths from May to September at the site of the experiment. B The columns give the
number of days passed after sowing and the height of the plants attained after sowing by the end
of September. Independent of both time passed after sowing and height attained, all plants flower
at the end of September, i.e. the photoperiod of 12 h 24 min in August was still too long (above the
threshold) but the only 18 min shorter photoperiod of 12 h 06 min in September was short enough
to induce flowering in H. suaveolens at the tropical site. (After Medina 1982)
each month at a location north of the equator starting in May and ending in Septem-
ber. At the end of September, all plants were flowering irrespective of the age and
biomass they had attained during growth, such that the 180 cm tall, ∼150 day old
plants, germinated in May flowered simultaneously with the 12 cm high, ∼30 day
old plants, only germinated in early September. Thus, flowering was not related to
age or biomass. The photoperiod, which did not lead to flowering (plants germi-
nated in August with no flowering in August photoperiod 12 h 24 min), and that
which elicited flowering in September (photoperiod ∼12 h 06 min) differed by only
18 min. Borchert and coworkers have compiled a large amount of data from field
observations made at frequent intervals over several consecutive years and from
herbarium collections which now provide strong evidence for photoperiod sensing
with a precision of at least 30 minutes in phenological synchronizations year after
year. Moreover, phenological phase shifts between the northern and the southern
hemisphere are found to be six months (Borchert 2000; Borchert and Rivera 2001;
Rivera and Borchert 2001; Rivera et al. 2002). Thus, clearly many of the phenolog-
ical phenomena observed in savannas can be regulated by photoperiod.
10.1.2.2 Morphological and Anatomical Traits
For the trees the water capacity of the soil does not need to be high (Table 10.1).
In the Brazilian cerrados the soil is always deep and well drained. The ground-
water table therefore is low, i.e. from 3 to 6 m down to 30 – 50 m. Hence, the trees
develop deep roots, which reach water even during the dry season, as shown by high
transpiration rates. Woody cerrado plants have substantially higher root-to-shoot
ratios than trees in nearby forests (Hoffmann et al. 2004). In the Llanos in central
Venezuela there is frequently a hard pan – “arecife” – of lateritic iron-oxide (see
Sect. 10.2.4.1) above the ground water-table (Fig. 10.7). The roots of savanna trees
must penetrate this layer to reach the ground water, which also varies on a seasonal
basis.
Leaf xeromorphy is another structural feature frequently observed among sa-
vanna trees. It is very important in Australia and South America, but less so in
Africa (Medina 1993). It has already been noted in Sect. 3.4.4.3 that the formation
of small and longlived leathery leaves may be considered as a strategy which gives
the best return for investment of resources when nutrient supply is low. In addition
such leaves also offer ways to economise on water use by some of the following
traits:
• dense venation,
• water storage tissues (see also Fig. 6.22C),
• thick and water tight cuticle, which reduces water loss via cuticular transpira-
tion,
• dead hairs on the surface,
• sunken stomata,
• prevailingly hypostomatic distribution of stomata, i.e. stomata only on the lower
surface;
Fig. 10.7 Relations between the vegetation, the hard lateritic ferrous-oxide layer (“arecife”) and
the seasonally shifted groundwater table in the Llanos of central Venezuela. (Walter and Breckle
1984, with kind permission of S.-W. Breckle and G. Fischer-Verlag)
(the latter three properties are generally assumed to reduce evapotranspiration by

supporting the built up of unstirred layers although in detail leaf boundary-layer
relations are very complex; Schuepp 1993);
• thick cell walls,
• lignification of cell walls,
• formation of idioblasts and sclereids,
(these three properties help to stiffen the leaves, so that the trees are sclerophyllous
and leaf-shape is maintained even when turgor pressure is low);
• production of etheric oils, which due to their hydrophobic nature may also assist
in preventing water loss into the gas phase around the leaves.
Many cerrado trees were found to store water in their sapwood which may play
a dominant role in the regulation of diurnal water deficits (Scholz et al. 2007).
However, a particularly conspicuous peculiarity of some savanna trees is real stem
succulence. With particularly thickened stems these trees really look very succu-
lent, such as the Bombacaceae Pseudobombax in South America and Adansonia,
the baobab (Fig. 10.8). The latter is a most spectacular plant. There is only one
species on the African continent, A. digitata, which may be up to 9 m in diameter
(Fig. 10.8D,E) and has a geographical distribution clearly correlated with the occur-
rence of savanna (Fig. 10.9). There are seven species of Adansonia in Madagascar
and two in Australia. Woody stem succulent plants are abundant in seasonally dry
tropical environments. A list of families and species is given in Table 10.6. They are
all deciduous C3 -plants, in contrast to fleshy stem succulent plants which mostly
perform CAM (Sects. 5.2.2.2 and 8.2.3.2.1). Stable isotope ratios of Adansonia
Table 10.6 Families and species of tropical woody stem-succulent plants. (Borchert and Rivera
2001, and personal observations)
Family Species Occurrence

Anacardiaceae Spondias purpurea Costa Rica, Mexico
Apocynaceae Plumeria rubra Costa Rica, Mexico, Nigeria
Plumeria acuminata Hawaii
Bombacaceae Adansonia spp. Africa, Madagascar
Bombax malabaricum India, Ceylon, Singapore, Java
Bombacopsis quinta Costa Rica
Chorisia insignis Argentina
Burseraceae Bursera simaruba Costa Rica, Mexico, Venezuela
Commiphora spp. Africa
Cochlospermaceae Cochlospermum vitifolium Costa Rica
Moringaceae Moringa ovalifolia Namibia, Africa
Vitaceae Cyphostemma currori Namibia, Africa
gregorii in Australia were found to be δ 13C = − 29.06‰ and δD = − 90.10‰

(H. Ziegler, unpubl.).
These succulent stems are water-stores and below ground lignotubers or xy-
lopodia (see Sect. 10.3.3) may also serve water storage. A vast quantity of wa-
Fig. 10.8A–G “Stem-succulent” trees. A Caatinga in Brazil (Martius’ Flora Brasiliensis, 1840–
1906). B Savanna in Queensland, Australia. C Pseudobombax pilosus, Paraguana Peninsula,
Venezuela. D, E Adansonia digitata (baobab), Okawango Delta, March 1982, Botswana, Africa. F
Commiphora sp., Namibia, Africa. G Moringa ovalifolia, Namibia, Africa. (D, E courtesy Helga
and Bodo Lüttge, Munich)
ter is present in the succulent stems. This water is mainly stored in a proliferate
parenchyma and water storage is highly correlated with wood density (Meinzer
2003; Meinzer et al. 2003), so that the wood density is low (< 0.5 g cm−3 ) and
the stem water potential in the dry season may be high (> − 0.5 MPa) (Borchert
and Rivera 2001). The water storing parenchyma is separated from the transpiration
stream by a high-resistance pathway, which applies to both transport directions, i.e.
filling and emptying. Thus, this water storage does not function as a buffer against
Fig. 10.9 Geographical dis-

tribution of the baobab in
Africa. (Walter and Breckle
S.-W. Breckle and G. Fischer-
Verlag)
daily water deficits in the rainy season because refilling of stem water via the roots
is slow (Goldstein et al. 1998) and rather is a long term commodity (Chapotin et
al. 2006a). It is the most important resource for phenological leaf flushing before
the onset of the rainy season (Sect. 10.1.2.1). In baobab trees stem water content
declined by up to 12% during this period and the water was almost exclusively used
for leaf growth. Stomatal opening and transpiration only started with the rainy sea-
son after considerable rainfall and was associated with the onset of sap flow from
the soil at the base of the trunks. High transpiratory water flow can not be supported
by the water stored in the tree trunk parenchyma but must come from uptake via the
roots (Chapotin et al. 2006b).
10.1.2.3 Hydraulic Architecture, Water Use and Photosynthesis
Some extraordinarily high maximum photosynthetic rates of the mature leaves of

savanna trees have been reported, i.e. around 40 µ mol CO2 m−2 s−1 for dominant
trees, Curatella americana (Fig. 10.5) and Byrsonima crassifolia, of the Llanos of
Venezuela (Medina 1982). However, in trees of the cerrados of Brazil rates near light
saturation ranged between 4 and 18 µ mol CO2 m−2 s−1 (Moraes and Prado 1998;
Franco and Lüttge 2002) comparable to average rates of 10 – 25 µ mol m−2 s−1 for
C3 -photosynthesis (see Box 10.3). Savanna trees are always C3 -plants. Water use ef-
ficiency may vary considerably among tropical trees, e.g. between 1.6 mmol carbon
mol−1 H2 O in Tectona grandis and 4.0 mmol carbon mol−1 H2 O in Platymiscium
pinnatum which is not directly correlated with relative growth rate (Cernusak et al.
2006).
Water relations and a very high irradiance load are conspicuous stressors for pho-
tosynthesis in savannas. Life long acclimation to drought basically can involve three
contrasting changes in the water transport capacity per unit leaf area for given plant
size, namely a decrease or constancy or an increase (Maseda and Fernández 2006).
It is interesting to note therefore, that savanna trees are isohydric with respect to
their minimum leaf water potential which is highly regulated at similar levels be-
tween dry and wet seasons. It is also noted for the cerrados that in relation to the
water factor diurnal limitations are more important than seasonal ones (Gottsberger
and Silberbauer-Gottsberger 2006). Among other adaptations a reduction in total
transpiring leaf surface area in the dry season is contributing to this, but a major
physiological mechanism involved is strong stomatal control of evaporative wa-
ter loss (Bucci et al. 2003, 2005). A midday depression (see Sect. 5.2.2.1) may
assist in regulating the water economy on hot days both during the rainy and the dry
season. It is expressed in:
• daily courses of root and leaf petiole hydraulic conductivity,
• photosynthetic gas exchange, and
• photoinhibition.
Often there is root limitation of water movement in the soil-leaf continuum, and in
cerrado trees a linear correlation is given between stomatal conductance and loss of
water conductivity of the roots in the afternoon (Domec et al. 2006). Daily courses
of cavitation and embolism in the afternoon and refilling during the night were ob-
served in both roots (Domec et al. 2006) and petioles of leaves of cerrado trees
Fig. 10.10 Daily courses 1.2

of specific hydraulic con- JAN
ductivity of leaf petioles 0.8
of two cerrado trees, Cary-
ocar brasiliense (circles) and
Schefflera macrocarpa (tri- 0.4
angles) in the wet season
ks ( kg m-1 s-1 MPa-1 )

(January), in the dry season 0
(August) and at the end of AUG
the dry season (September). 1.2
(After data of Bucci et al.
2003) 0.8
0.4
0
SEP
1.2
0.8
0.4
0
4 8 12 16 20 0 4 8
Time of day (h )
(Bucci et al. 2003). Figure 10.10 shows such midday depressions of specific hy-
draulic conductivity, ks , of petioles in the dry and wet seasons. The degree of
embolism in petioles is a function of tension in the xylem water stream and the rate
of refilling is determined by internal pressure imbalances, where effects of starch
remobilization may be involved (Bucci et al. 2003). Nevertheless, transpiration dur-
ing the night may also be high and can amount to 15 – 25% of total daily water loss
(Bucci et al. 2005).
Midday depressions of net CO2 -exchange, JCO2 , are frequently found in cer-
rado trees (Moraes and Prado 1998; de Mattos 1998; Franco and Lüttge 2002; de
Mattos et al. 2002) where JCO2 considerably drops at the times of highest irradiance
(Fig. 10.11). In the examples of Fig. 10.11 there was no recovery of JCO2 in the
afternoon. Figure 10.12 shows measurements revealing rapid responses to changing
weather conditions in the rainy season. After a dry spell amidst the rainy season
there was a pronounced midday depression of net CO2 uptake, JCO2 , and transpira-
tion, JH2 O , which was reversible in the afternoon, and immediately after a rainfall
on the following day no midday depression was expressed any more (de Mattos et
al. 2002).
The midday depression of gas exchange is a reflection of the strong stomatal con-
trol involved in the isohydric performance over the seasons as noted above (Franco
and Lüttge 2002), but also creates the danger of photoinhibition as discussed in
Sects. 4.1.6 and 5.2.2.1 (Fig. 5.3). Protective mechanisms, such as photorespira-
tion (Sect. 4.1.3) and harmless thermal energy dissipation via the xanthophyll cycle
10.2 The Nutrient Factor 335
Fig. 10.11 Daily courses of

12 2400
net CO2 -exchange (JCO2 ,
closed circles) and photosyn-
thetically active photon flux 8 1600
density (PPFD, open trian-
gles) of three cerrado tree
species. (From Franco and 4 800
Lüttge 2002) Roupala
0 montana
0
12 2400
PPFD ( μmol m-2 s-1 )

8 1600
4 800
2
Qualea
0 grandiflora 0
12 2400
8 1600
4 800
Ouratea
0 hexasperma 0
5 7 9 11 13 15 17 19
time of day ( h )
(Sect. 4.1.4), are expressed in the cerrado trees (de Mattos 1998; Franco and Lüttge
2002). Protective acute photoinhibition is reflected in a pronounced midday depres-
sion of potential quantum yield of photosystem II as indicated by Fv /Fm -values
below 0.83 (see Sect. 4.1.7). Some species show a quite severe photoinhibition at
midday with Fv /Fm -values well below 0.6 while other species are less strongly in-
hibited (Fig. 10.13). In many cases this is only acute photoinhibition reversible over
night but in others (especially in the examples of Fig. 10.13B) Fv /Fm remains below
0.83 over night so that there is also chronic photoinhibition.
10.2 The Nutrient Factor
Savanna soils are mostly very poor and infertile. Mineral-nutrient relations of sa-
vannas are mainly determined by the physicochemical properties of the upper
soil layers, such as:
• texture,
• pH,
12
10
8
4
2
0
6
J H O( mmol m-2 s-1 )
2
2
0
7 9 11 13 15 17 19
Time of the day ( h )
Fig. 10.12 Daily courses of net CO2 -exchange, JCO2 , and transpiration, JH2 O , of the cer-
rado tree Miconia albicans, after a dry spell amidst the rainy season (open symbols) and
on the subsequent day after a rain (closed symbols). (From de Mattos et al. 2002, with
access at http://www.publish.csiro.au/nid/66/issue/501.htm; thanking the authors and CSIRO-
PUBLISHING, Australia)
• cation exchange capacity,

• extractable bases (K + + Ca2+ + Mg2+ + Na+ ),
• content of potentially mineralizeable N,
• availability of N and P
(Medina 1993). Nutrient categories of savannas depend on the nutrient levels in the
soil and the recirculation rates (Table 10.7). Nutritional fertility of savannas may be
estimated and ranked by the sum of extractable bases

(K+ + Ca2+ + Mg2+ + Na+ )[cmol(+)/kg(soil)] ,
where values lower 5 cmol (+) kg−1 mark dystrophic and values higher than
20 cmol (+) kg−1 eutrophic savannas, with mesotrophic savannas in between (Med-
ina 1993). Generally, however, nitrogen and phosphorus are the most strongly lim-
iting elements, although even potassium can become limiting in situations of rapid
recirculation. A particular mineral stress factor in savannas is aluminium.
0.83
0.8
0.7
Fv / Fm
0.6
0.5
A
0.83
0.8
Fv / Fm
0.7
0.6
B
0.5
5 7 9 11 13 15 17 19
Time of the day (h)
Fig. 10.13A, B Daily courses of potential quantum yield of photosystem II, Fv /Fm , of 12 cerrado
tree species as follows: (A) closed circles: Miconia albicans, closed triangles: Miconia ferrugi-
nata, closed squares: Roupala montana, open circles: Sclerolobium paniculatum, open triangles:
Eryotheca pubescens, open squares: Vochysia eliptica, inverted open triangles: Syagrus comosa
(from de Mattos 1998); (B) open squares: Didymopanax macrocarpum, open diamonds: Roupala
montana, inverted open triangles: Qualea grandiflora, open triangles: Miconia fallax, open circles:
Ouratea hexasperma (from Franco and Lüttge 2002)
Nutrient availability in the various soil horizons of savannas is much lower than
in tropical forests (Table 10.8). The height and density of the woody layer in the
cerrados (see Table 9.2) depends on the fertility, depth and drainage of the soil and
not on rainfall (Eiten 1972, 1986).The typical distribution of some mineral elements
between the vegetation and the soil in tropical forests and savannas is schematically
summarized in Table 10.9. In the savanna most of the K, N, Ca and Mg is in the soil,
Table 10.7 Nutrient categories of savannas. (After Sarmiento 1984)

←−−−−−−−−−→ ←−−−−−−−−−−−−−−−−−−−−−− Amount in the soil −−−−−−−−−−−−−−−−−−−−−−→
Rate of recirculation
Large Small
Slow Ca, Mg, Na Various elements
Rapid K can be limiting P most strongly limiting
Table 10.8 Comparison of the nutrients in a forest and a savanna soil in Nicaragua, both profiles
on the same piedmont deposit. (Data from Alexander 1973)
Soil horizon Organic C Total N Cation exchange capacity Ca Mg K
(%) (%) (meq/100 g) (meq/100 g)
Rainforest
A11 5.3 0.53 36.7 4.29 3.53 0.68
A12 4.1 0.39 26.2 1.90 1.91 0.43
A3 2.7 0.26 19.1 0.26 0.79 0.15
B1 0.9 0.12 11.0 Trace 0.56 0.16
B21 0.5 0.08 14.0 0.05 1.07 0.07
B22 0.2 0.07 12.3 0.11 0.89 0.05
B23 0.2 0.06 17.9 0.11 1.06 0.05
Savanna
A1 2.1 0.14 10.3 Trace 0.29 0.05
A2 1.0 0.07 6.2 Trace 0.23 0.04
B21 0.9 0.08 8.1 Trace 0.25 0.04
B22 0.5 0.05 9.3 Trace 0.29 0.03
B23 0.3 0.02 8.4 Trace 0.25 0.03
B24 0.1 0.01 8.1 Trace 0.29 0.04
Table 10.9 Nutrient distribution between vegetation and soil in a tropical forest and a savanna.
(After Sarmiento 1984)
Relation of vegetation vs soil

Forest:
K >
P >
N ∼
=
Ca, Mg ≶
Savanna:
K
P =
N
Ca, Mg
whereas in the forest most K and P is contained in the vegetation. For forests N is
about equally distributed between vegetation and soil, similarly to P in savannas.
10.2.1 Nitrogen
10.2.1.1 Nitrogen Cycles
Nitrogen is one of the most critical elements for plant growth in savannas. The
nitrogen cycle in general is determined by assimilatory processes in microorganisms
and plants, and the use of this primary production by consumers and decomposition
by microorganisms (Box 10.4).
Figure 10.14 gives a comparison of the annual nitrogen-cycles and the nitro-
gen levels in various compartments of the ecosystem of a non-tropical prairie and
a seasonal tropical savanna (a similar presentation for N-cycles in tropical forests is
presented in Fig. 3.33).
Fig. 10.14A, B Compartmentation and annual turnover of nitrogen in two grassland ecosystems.
A Andropogon gerardi – Andropogon scoparius prairie in Missouri (USA). B Seasonal Axonopus
purpusii-Leptocoryphium lanatum savanna in Barinas (Venezuela); strongly modified and simpli-
fied from Sarmiento (1984) (reprinted by permission of Harvard University Press). Sizes of N pools
in the different compartments (boxes) and transfer rates between the compartments (arrows) were
drawn to scale to allow direct comparisons of pools and rates both within and between the two
ecosystems
Box 10.4 Nitrogen cycles in ecosystems
Pools of N in various groups of organisms of the ecosystem and in various

N-compounds (boxes) and transfer between the pools (arrows).
-NH2 , -N-, organic N;

NH+ 4 , ammonia;
NO− 2 , nitrite;
NO− 3 , nitrate;
N2 , atmospheric dinitrogen gas.
(After Lüttge et al. 2005)
The compartments distinguished are:

• soil, organic N,
• soil, mineralized N,
• roots,
• living epigeous biomass,
• dead litter,
• atmosphere.
The largest amount of N in either case is in the organic matter of the soil, and it
is similar in the prairie and the savanna. The amount of mineral N in the soil is
much smaller, and it is somewhat larger in the prairie as compared to the savanna.
The amount of N in the roots and in the living epigeous biomass, as well as in the
litter, is not very different in the two systems. The rates of N-transfer between the
individual compartments, namely absorption of mineral N from the soil, root-shoot
translocation, mortality, decomposition/humification and mineralization, as well as
precipitation, throughfall and drainage are similar within the two systems.
It is noticeable that the rate of absorption of mineralized N from the soil is lower
in the savanna and that the root/shoot recycling of N (translocation between roots
and epigeous biomass) is higher than in the prairie. More important, however, is
the presence of two additional transfer processes in the savanna as compared to the
prairie, namely:
• volatilization,
• atmospheric N2 -fixation.
Volatilization is largely due to fire in the savannas (Sect. 10.3). However, it is also
known that tropical soils are significant natural sources of gaseous N-compounds,
e.g. in the savannas of the Venezuelan Llanos:
nitrogen oxides 3 – 13 kg N ha−1 year−1
(NO-nitrogen 2.6 kg N ha−1 year−1 )
(N2 O-nitrogen 0.65 kg N ha−1 year−1 )
ammonia 11 kg N ha−1 during the reproductive period
(Garcia-Méndez et al. 1991; Medina 1993). Atmospheric N2 -fixation is an important
activity of mats of cyanobacteria between the tussocks of savanna grasses and of free
living soil bacteria and root nodule symbioses (Sect. 10.2.3.2).
In summary, it is surprising how small are the differences between the two grass-
lands, namely the mesic prairie and the tropical savanna. It should be recalled,
however, that diagrams similar to Fig. 10.14 have been drawn for tropical forests
(Fig. 3.33), giving a very different picture. Although the soil organic N is simi-
lar, amounts of N in the roots and in the epigeous living biomass are very much
larger, and N in the dead litter is somewhat larger in the forests than in the grassland
systems. Mineral N in the soil is smaller in the forest. With the exception of pre-
cipitation, throughfall and drainage, the rates of N-transfer between the individual
compartments are considerably larger in the forests than in the grassland systems.
This relates to absorption of mineral N from the soil, root-shoot translocation, mor-
tality, decomposition/humification and mineralization, such that the cycling of N in
the forest is much more rapid than in the prairie and the savanna.
10.2.1.2 Nitrogen-Use Efficiency
In Sect. 4.1.2 we have already discussed the nitrogen-use-efficiency (NUE) of pho-

tosynthesis in relation to the light climate in tropical forests. Again in savannas and
cerrados, for both grasses (Fig. 10.15) and trees (Fig. 10.16A), we observe gener-
ally linear relationships between levels of N in biomass and rates of photosynthesis.
As mentioned above, there are often differences between species (Sect. 4.1.2). In
savannas, the slope of the line for the tropical C4 -grass is much steeper than for
the two C3 -grasses of the temperate zone given for comparison (Fig. 10.15). For
crops the ratio of photosynthetic CO2 -fixation to leaf-N also was found to be higher
in the C4 -plant maize (1,056 µ mol CO2 m−2 s−1 /mol N) than in the C3 -plant rice
(640 µ mol CO2 m−2 s−1 /mol N). On the other hand, C4 -plants do not necessar-
ily have a competitive advantage over C3 -plants under conditions of low N-supply.
Experiments with C4 - and C3 -grasses under natural conditions of a Central Euro-
Fig. 10.15 Relations between rates of photosynthesis and nitrogen content in the biomass of a trop-
ical C4 -grass (filled circles: Panicum maximum) and two C3 -grasses of the temperate zone (open
triangles: Lolium perenne; filled squares: Festuca arundinacea). (Medina 1986)
140
A ( μmol kg -1 s-1 ) A
100
60
20
4 12 20 28
Leaf N ( mg g-1 )
100
B
80
A ( μmol kg -1 s-1 )
60
40
20
400 500 600 700 800
-1
Leaf P ( μg g )
Fig. 10.16A, B Relationships between maximum rates of photosynthetic CO2 -acquisition (ordi-
nates) and levels of nitrogen (A) and phosphorus (B) in leaves of various evergreen (closed trian-
gles) and deciduous (open circles) cerrado species in Brazil. (Franco et al. 2005)
pean summer have shown that the C3 -grasses tended to be more successful at low
N-supply (Gebauer et al. 1987). The authors suggest that this could result from
lower transpiration in the C4 -grasses because of the water-saving functions of C4 -
photosynthesis (see Sect. 10.1.1.2). This would imply lower N input via the transpi-
ration stream from the soil to the shoots. However, some of this effect may be offset
by higher temperatures, and the situation may be very different in tropical savannas,
and indeed, a higher NUE may be important for C4 -grasses which dominate in the
nutrient-poor savannas.
There are even differences between C4 -groups of grasses forming malate and
aspartate as the primary CO2 -fixation product. Those species which synthesise
the amino-acid aspartate from the oxaloacetate (following CO2 -fixation via PEP-
carboxylase) have a higher N-requirement than the malate-forming NADP-malic
enzyme group (see Box 10.2). In S-Africa it was observed that NADP-malic enzyme
C4 -plants with their lower N-requirement are characteristic of particularly nutrient-
poor, moist savannas while the more N-demanding aspartate-formers (NAD-malic
enzyme and PEP-carboxykinase groups) are more frequent in arid savannas.
10.2.2 Phosphorus
In addition to nitrogen, phosphorus is one of the most critical nutritional elements in

savannas. Both elements are so important because they are the two mineral elements
most abundantly and most directly involved in the metabolic machinery of cells.
Like for N there is a linear relationship between leaf P levels and maximum rate of
photosynthetic carbon gain on a leaf mass basis (and not a leaf area basis in contrast
to the observations of Figs. 4.6 and 4.7) in cerrado trees (Fig. 10.16B), and P and
N levels in the leaves of these trees are linearly related to each other (Fig. 10.17).
Ratios of P/N have also been used to describe the state of P nutrition. In semi-
arid grasslands of the Sahel region of Africa P/N-ratios (mol/mol) of 17 ×10−3
to 68 ×10−3 are considered to mark the range within which there is response to
P-fertilization, with the lower value characterizing P-deficient and the higher one
P-sufficient plants. A more detailed comparison is presented in Fig. 10.18. Within
the “P-responsive” range N-nutrition alone only slightly lowers P/N-ratios, whereas
P-nutrition highly stimulates P/N ratios, which are somewhat lower when N + P
40
Leaf N ( mg g-1 )
30
20
10
0 500 1000 1500 2000

Leaf P ( μg g-1 )
Fig. 10.17 Correlation between N and P levels in the leaves of various evergreen (closed triangles)
and deciduous (open circles) cerrado trees in Brazil. (Franco et al. 2005)
Fig. 10.18 P/N-ratios in the above-ground biomass of different grasslands in the Llanos of
Venezuela (cut or burned at the end of the dry season), Africa and Australia (C controls), and
effects of fertilization with N, P or N + P. (After Medina 1993)
fertilization was applied. Special metabolic features may make some plant species
very phosphorus-efficient which are much searched for agricultural applications in
the tropics, such as the forage grass Brachiaria which covers 10 ×106 ha of pastures
in Brazil (Nanamori et al. 2004).
10.2.3 Biotic Interactions
10.2.3.1 General Overview
Biotic interactions, when considered in relation to the nutrient “stress factor”, deter-
mine growth, development and productivity of plants in savannas. This includes:
• plant-plant interactions,
• fungi- or microorganism-plant interactions,
• animal-plant interactions.
Among the nutritional plant-plant interactions the grass-tree relations are of partic-
ular interest in savannas. With the extended root systems of trees, tree-biomass may
concentrate nutrients from large soil volumes. Due to litter fall and decomposition,
the availability of K, Ca and Mg is often higher under tree canopies, and soil im-
provement due to litter may be even more important than by biological N2 -fixation
(Campa et al. 2000; Sect. 10.2.3.2). The droppings of perching birds may also add
to improved nutrient availability in the vicinity of trees (Medina 1993).
For the animal-plant interactions termite savannas (Sect. 9.1) are interesting
because mound-building termites excavate and explore large volumes of soil reach-
ing depths of 0.5 – 1 m. In this way they affect soil texture, but in addition they may
also enrich nutrients like Ca, K and Mg and to some extent also P. Termites ac-
celerate nutrient recycling and in termite dominated savannas (Sect. 9.1, Fig. 9.5)
this may make a very considerable contribution to nutrient turnover. In Australia it
was observed that termite-mediated mineralization of organic matter may amount
to 250 kg ha−1 year−1 (Medina 1993). As described in Sect. 3.4.4.1, leaf-cutter
ants (Fig. 3.34) concentrate nutrients in a similar way to mound-building termites.
Leaves of grasses and trees are carried into complicated underground chamber sys-
tems, where the ants cultivate fungi and play a significant role in nutrient cycling,
particularly for deep rooted trees. For the animal-plant interactions carnivory de-
serves a separate section (Sect. 10.2.3.3).
Plant-fungi interactions are of basic importance as mycorrhiza is a very widely
expressed symbiosis of plants and fungi facilitating plant’s mineral nutrient acqui-
sition. With respect to P-supply it was observed that genes which are active in P-
deficiency are down-regulated independent of each other by internal phosphorus lev-
els and mycorrhiza, where signalling involves inorganic phosphate transport from
the root to the shoot and a transportable shoot factor signalling back to the root
(Burleigh and Harrison 1999). Mycorrhiza is also important in interactions with at-
mospheric dinitrogen fixation and root nodule symbioses which are treated in a sep-
arate section (Sect. 10.2.3.2.2).
10.2.3.2 Fixation of Atmospheric Dinitrogen (N2 )
Fixation of N2 is mediated by the enzyme-complex nitrogenase. It is restricted

to procaryotic microorganisms, bacteria and cyanobacteria (“blue-green algae”),
which however, can make important contributions to the N-supply of eucaryotic
plants in associations and symbioses. The overall process is highly energy demand-
ing, i.e.
N2 + 4 [2 H] + 16 ATP −→ 2 NH3 + H2 + 16 (ADP + Pi ) .
Nitrogenase is very oxygen sensitive and requires hypoxia for its operation.
10.2.3.2.1 Plant associations with Free Living Dinitrogen Fixing Microorganisms
Free living N2 -fixing bacteria are in the genera Acetobacter, Azoarcus, Azospirillum,
Azotobacter, Beijerinckia, Clostridium, Herbaspirillum and Paenibacillus (Gotts-
berger and Silberbauer-Gottsberger 2006). All cyanobacteria (“blue-green algae”)
which have heterocytes are N2 -fixing. As the nitrogenase is O2 -sensitive it is lo-

cated in special cells in the photosynthezising filamentous cyanobacteria, i.e. the
heterocytes, which have thick cell walls limiting O2 diffusion into these cells, and
lack photosystem II and hence photosynthetic O2 -evolution. From the possession
of heterocytes, most of the cyanobacteria in the savannas are shown to be N2 -fixers
(see also Sect. 11.2.1.2).
Associations of plants with N2 -fixing soil bacteria in the rhizosphere may be of
mutualistic benefit, where exudates from the plant roots provide substrates and vi-
tamins and other regulatory compounds to the microorganisms and plants receive
N-compounds. The contribution of rhizosphere associations to total N-input to sa-
vanna soils may be quite significant (Table 10.10), and possibly in many cases their
overall contribution may even be larger than that of root nodules. In fact, there have
been considerable efforts to improve agricultural productivity of tropical grasslands
with such associations (Baldani et al. 2002). Some of the N2 -fixing bacteria can
even live endophytically in the host plants (Baldani et al. 2002). Attempts have also
been made to use genetic engineering for the introduction of the nitrogenase-genes
(nif+ -genes) into some rhizosphere bacteria, which occur more abundantly in the
soil than natural N2 -fixing organisms (Hess 1992).
The contribution of cyanobacteria in the examples of Table 10.10 is shown to
be rather modest. However, in places cyanobacteria are extraordinarily abundant in
savannas, often forming dense, continuous mats between the tussocks of grasses
(Fig. 10.19). In an example from savannas in Nigeria, where the ground coverage
with cyanobacterial mats and crusts was 30%, a much higher value of cyanobac-
terial N2 fixation is reported, namely 23 g ha−1 day−1 in the rainy season, and
60 g ha−1 day−1 have also been recorded in savannas corresponding to several kilo-
grams per ha over the year (Medina 1993, Sect. 11.2.1.2).
A step between associations of plants with free living prokaryotes and N2 -fixing
endosymbioses may be exosymbioses such as the endophytic bacteria mentioned
above. Another example which has become important in tropical agriculture is
the mutualism between the fern Azolla and the cyanobacterium Anabaena, which
Table 10.10 Nitrogen balances in two humid tropical savannas in South America, Central
Venezuela (Trachypogon savanna) and in Africa, Ivory Coast, and values of bacterial N2 fixation
associations in grasslands of Brazil and Zimbabwe. (Medina 1987, 1993)
Venezuela Brazil Ivory Coast Zimbabwe
(kg N ha−1 year−1 )
Input through rain 19 2.6
(inorganic 4.5)
Biological fixation
Blue-green algae 1.4 – 2.5 0.7
Rhizosphere association 9 – 12 5 – 18 6.7 78
Losses through fire 17 – 23 8.5
Percolation and leaching 5.6 0.5
Balance + 4.9 to + 6.8 + 1.0

Fig. 10.19A, B Mats of cyanobacteria between tussocks of grasses in the Llanos of Venezuela after
a longer rainless period (A) (February 1989) and a few days after rain (B) (March 1991)
lives extracellularly in special intercellular air spaces of the fern fronds. Azolla
grows equally well on fresh-water surfaces and on mud and is successfully used
for mulching in tropical rice culture.
10.2.3.2.2 N2 -fixing Endosymbioses: Root and Stem Nodules
In the neotropics and palaeotropics there are 40 – 45 species of Gunnera which form
a fascinating symbiosis with cyanobacteria of the genus Nostoc. Various Gunnera
species range from small creeping stoloniferous herbs up to tall plants of a height
reaching 6 m. They inhabit nutrient poor super-humid areas and bogs, leached soils
with high rainfall and are also pioneers on bare lands. There is a preference of high
altitudes in the genus, which also can inhabit sites with mesic climate. Gunnera
is the only genus of angiosperms that forms an endosymbiosis with cyanobacteria.
Nostoc is incorporated via peculiar stem glands secreting a mucilage. These glands
are induced by nitrogen deprivation of the plants but independent of the cyanobacte-
rial symbiont. Inside the host tissue internal nodules are formed which are invested
with vascular tissue. This is required for the supply of the nodules with carbohy-
drates. Light does not penetrate the host organs towards the cyanobacterial sym-
bionts which are photosynthetically inactive. Venation of the internal nodules is also
necessary for the export of combined nitrogen. The N2 -fixing cyanobacterial sym-
bionts provide ammonia to the host which has been shown to be sufficient for ful-
filling the entire N-demand even of the larger plants of Gunnera species (Bergman
et al. 1992; Osborne et al. 1992; Johansson and Bergman 1994; Stock and Silvester
1994; Silvester et al. 1996; Rai et al 2000; Parsons and Sunley 2001; Chiu et al.
2005).
Of a much more general importance are the external root and stem nodules of
higher plants with N2 -fixing bacterial endosymbionts. Nodule eliciting bacterial en-
dosymbionts are rhizobia of the genera Azorhizobium, Bradyrhizobium, Rhizobium
and Sinorhizobium. Nodules (Fig. 10.20) are of a structurally high complexity pro-
viding the anatomical basis for various physiological requirements, especially for
oxygen compartmentation and venation. Oxygen compartmentation is required to
solve the oxygen dilemma of nodules, where on the one hand the oxygen sensi-
tive nitrogenase needs a hypoxic environment and on the other hand the high en-
ergy demand of N2 -reduction as well as other essential metabolic activities in the
nodules require high respiratory activity which needs oxygen. In addition to appro-
priate anatomical differentiations keeping these functions spatially separated leghe-
moglobin, which binds oxygen and diffuses between the different compartments of
the nodules, effectively lowers and increases oxygen levels at the sites of nitrogenase
and respiration, respectively (Werner 1992; Pimenta et al. 1998). Venation is essen-
tial for import of assimilates needed as substrate for respiratory energy supply as
well as acceptors of reduced nitrogen (ammonia) and for export of organic nitrogen
compounds to the nitrogen sinks of the plants. Here, an interesting detail of eco-
logical biochemistry is the nature of the organic compounds functioning in nitrogen
export from the nodules. These are often amides or nitrogen-rich ureide molecules.
Amides are more water soluble than ureides and therefore their transport requires
less water. Thus, among the Sahelian Acaciae, amide transporting plants may colo-
nize more arid regions and ureide transporting plants inhabit areas of greater water
availability (Campa et al. 2000).
The symbiotic formation of root nodules with N2 -fixing bacteria is best docu-
mented in the Leguminosae. With N-supply limiting so much the productivity of sa-
Fig. 10.20 Nodulated roots of the legume shrub Andira legalis in a Brazilian Restinga
vannas, one might expect that plants capable of fixing atmospheric dinitrogen would
be particularly frequent. An important leguminous savanna tree in South America
is Bowdichia and in Africa various species of Acacia (Figs. 1.8A and 9.9) play an
equivalent role. Open woodlands tend to contain more nodulated trees than adja-
cent forests. There is a progressive increase in the proportion of nodulated trees
along a gradient from humid to arid areas, which is negatively correlated with the
N-content of the soils (Högberg 1986b) and N2 -fixation carried out by N2 -fixing
trees is more important in woodland than in rainforest (Högberg and Alexander
1995). Such a negative correlation is expected. N2 -fixation requires much input of
energy, of carbon skeletons for binding of reduced N, and also special morphologi-
cal differentiation (nodules), and in view of these costs N2 -fixation should not give
a competitive advantage when sufficient N is available in the soil. Alternatively, the
symbiotic association may be more susceptible to drought stress although nodulated
Leguminosae such as Prosopis and Acacia are phreatophytes (see Fig. 10.7).
In Africa Fabaceae/Leguminosae trees are important and often dominant ele-
ments of savannas (Fig. 9.9). In the Etosha National Park in Namibia there are often
sharp separations between Acacia nebrownii and Colophospermum mopane domi-
nated savannas, which are determined edaphically and where the former appears to
be somewhat more salinity-tolerant (Fig. 10.21; Berry and Loutit 2000). It has been
noted by Ethiopian scientists that in agro-forestry systems it should be sufficient
Fig. 10.21A, B Acacia nebrownii savanna (A) and Colophospermum mopane savanna (B) in the
Etosha National Park, Namibia, Africa
to keep about 40 Acacia trees per ha to have sufficient N-fertilization. However,

although in Africa numerous tree species contribute significantly to the N-budget
of savanna-woodland ecosystems through their N2 -fixation (Högberg 1986b), it is
often observed elsewhere that Leguminosae comprise a surprisingly low proportion
of biomass in other savanna grasslands. In the Llanos of Venezuela, legumes rarely
make up more than 1% of the biomass (Medina 1993). This limitation does not nec-
essarily hold for the number of species of Leguminosae in such savannas. In one
case among 127 species 109 were found to be nodulated. The frequency of legu-
minous species in Venezuela is related to low levels of exchangeable aluminium in
the soil (see Sect. 10.2.4) and high levels of exchangeable calcium, i.e. two fac-
tors which are inversely related to each other. Overall root-nodule symbioses appear
to contribute little to the productivity of this particular savanna system (Medina
1993).
Other factors limit the growth of legume species, and in particular the balance
between the supply of nitrogen and other elements must be crucial. Nodulation it-
self is nutrient limited (Souza Moreira et al. 1992), and the most important limiting
nutrient factor in savannas frequently is phosphorus (Högberg 1986a; Sect. 10.2.2).
The particular demand of phosphorus for root nodules is a well known general phe-
nomenon (Almeida et al. 2000). In soybean plants the total response of symbiotic
N2 -fixation to altered P-supply is a function of both indirect effects on growth of
the host plant and more direct effects on the metabolic functions of the nodules (Is-
rael 1993). In Africa, the low availability of phosphorus was found to be a severe
restriction for nitrogen-fixing species in moist savannas, and this can explain their
low abundance in such ecosystems. The increase in the number of nodulated trees
towards drier sites already mentioned above is correlated with a decline in soil-N
and an increase in available soil-P so that in African, and perhaps also in Australian
savannas, one may distinguish between:
• moist/dystrophic and
• arid/eutrophic svannas
(Högberg 1986a).
Phosphorus limitations of nodulation can at least be partially alleviated when the
leguminous plants develop a second symbiosis in addition to root nodules, namely
mycorrhiza. The fungal hyphae of mycorrhiza enhance nutrient acquisition and may
have positive effects, particularly due to increased supply of P to the host plants
(Högberg and Alexander 1995). In the experiments presented in Table 10.11, nodu-
lation was much more effective when mycorrhiza was present than if production of
mycorrhiza was prevented. However, mycorrhiza could be readily replaced by phos-
phate supply in these experiments. The higher nodulation in plants with mycorrhiza
or additional P-nutrition was accompanied by a considerably larger production of
total fresh weight and a reduced root/shoot ratio, showing a lower investment in nu-
trient allocating roots foraging for nutrients and hence larger resource allocation to
photosynthezising shoots. In another experiment, where nodulated soybean plants
and maize plants were connected by a common mycorrhizal mycelium, it was even
observed that the source-sink relations led to P- and N-flows in opposite directions.
Table 10.11 Effect of mycorrhiza and additional phosphorus supply on growth and nodulation of
legumes. (Crush 1974)
Genus Total Root/ Nodules Mycorrhizal P and N content
FW shoot (1 – 5) infection (mg/g)
g ratio % P N
Centrosema Mycorrhiza 3.88 0.86 5 86 2.0 28.5
Non-mycorrhizal 1.67 1.70 1 0 0.5 40.2
+ phosphate 4.95 0.68 5 0 2.2 30.8
Stylosanthes Mycorrhiza 1.63 0.54 5 74 4.4 39.0
Non-mycorrhizal 0.47 1.12 0 0 2.0 43.1
+ phosphate 0.91 0.34 5 0 5.8 38.1
There was a P-flow from maize to soybean, while soybean could provide maize with
N (Bethlenfalvay et al. 1991).
Although tropical wetlands and floodplains are not only characteristic of some
savanna types (Fig. 9.10) but also typical elements of tropical forests (Sect. 3.2.3) it
may be the best place here to mention N2 -fixing nodule symbioses in these habitats,
some of which are heavily leached, and where under flooded conditions mineral-
ization of organic matter is slow so that they are nutrient poor and biological N2 -
fixation is of great importance. Many nodulated Leguminosae also grow on floating
mats of vegetation in the flooded areas. Flooding and especially cycles of fluctuating
water levels highly amplify the dilemma of the oxygen metabolism of nodules. Oxy-
gen levels are low in inundated and flooded soils (see also in mangroves, Sect. 7.3.1).
This may appear beneficial for the oxygen sensitive nitrogenase but on the other
hand it hampers the required respiratory activity. Hence, adaptations are important
which include nodule formation on more superficial adventitious roots and even on
stems to overcome the oxygen constraints by the submerged main root system, struc-
tural facilitation of oxygen diffusion pathways from stems via particularly devel-
oped lenticels (see Sects. 3.2.3 and 7.3.1) and aerenchymas to the root and the root
nodules, and special structural features of the root nodules themselves (Loureiro et
al. 1998). However, these adaptations facilitating oxygen supply of nodules under
flooded conditions prove disastrous when nodules are exposed to the air in flooding
and emergence cycles and emergent nodules deteriorate due to rapidly increasing
oxygen levels (Loureiro et al. 1998; James et al. 2001). Stem nodules primarily may
have evolved in response to flooding. Stem nodule forming species are in the genera
of Aeschimone, Discolobium, Sesbania and Vigna (Loureiro et al. 1998). Nitrogen
accumulation by stem-nodulated Leguminosae can be very high, ranging from 41
to 532 kg N ha−1 in 1.5 – 2 months (Loureiro et al. 1998). These plants can also
be used successfully in green-manuring to improve tropical agriculture. Sesbania
rostrata fixing up to 200 kg N ha−1 season−1 contributes 50 – 150 kg N ha−1 year−1
to the soil nitrogen. In this way rice production can be increased two- to threefold
(Clarkson et al. 1986).
10.2.3.3 Carnivory
Carnivory has already been mentioned in relation to lianas and epiphytes, as a po-
tential strategy for nutrient acquisition (Sect. 6.6.3). In the temperate climate, car-
nivorous plants are particularly frequent in moist and acidic sites and especially in
peat bogs, which are very poor in nutrients. Similarly, in the tropics, carnivorous
plants of the genus Drosera are frequently found in great numbers in the wet and
often peaty soils of upland herbaceous vegetation types with savanna-like mead-
ows at 1,000 – 2,800 m a.s.l. (Fig. 10.22; see Huber 1988 for site description). The
carnivorous genus Heliamphora (Sarraceniaceae) is endemic to the Tepuis, the char-
acteristic table mountains of the Guayana highlands in tropical South America.
Drosera attracts its prey by the numerous brilliant droplets of mucilage (“sun
dew”) secreted via special glands on the surface of colourful, often reddish, tentacles
Fig. 10.22A, B Drosera roraimae. A Gran Sabana, Venezuela. B Sierra Maigualida, Venezuela,
with droplets of mucilage on the leaf tentacles
(Fig. 10.23). The sticky mucilage usually prevents the escape of small insects once
they have touched it. The tentacles move in response to mechanical and chemical
stimuli caused by the captured animals making thigmotropic and chemotropic as
well as thigmo- and chemo-nastic movements. The prey is thus enveloped and then
digested by proteases secreted from the tentacle glands. Mineral elements like N, S,
Fig. 10.23 Scheme of a tentacle of Drosera after Gilchrist and Juniper (1974; from Lüttge 1983).
The centre of the tentacle is served by tracheids. It is separated from the peripheral gland epithelium
by an endodermis, whose radial walls are suberized so that apoplastic transport is blocked and
transport between the periphery and the interior must use a symplastic route
P, Mg2+ , K+ from the prey then stimulate growth and productivity of the Drosera
plants (Lüttge 1983).
Heliamphora is a genus of pitcher plants with several species (H. nutans, H. het-
erodoxa, H. minor, H. ionasii, H. tatei). The pitchers are formed of single leaves.
They are morphogenetically derived from peltate leaves, so that the interior of the
pitcher wall corresponds to the upper leaf surface and the exterior to the lower leaf
surface. In the middle of the pitchers there is a small opening, which allows wa-
ter to flow out and thus prevents over-filling in the high rainfall habitats of He-
liamphora. Animals are attracted by coloration and nectar secretion at the pitcher
orifice. Escape is hindered by hairs and trichomes directed downwards to the bot-
tom. Most of the criteria of true carnivory are fulfilled by all Heliamphora species,
such as:
• attraction of prey through special visual and chemical signals,
• trapping and killing of prey,
• presence of wax scales and other structures preventing escape of prey,
• absorption of nutrients.
Most of the Heliamphora species, however, lack one important trait of true car-
nivory, i.e. digestive glands and enzyme secretion. In these cases digestion of prey
is mediated by bacterial commensals (Schmucker and Linnemann 1959). There is
one noticeable exception though, which is H. tatei. In this species there is enzymatic
activity in closed pitchers just as they maturate and open. Since microbes have no
access to the closed pitchers, this proves to be genuine enzyme secretion by the
pitcher tissue. Capture of small animals is very effective in Heliamphora species
in their natural habitat. The carnivorous traits are lost, however, in low light condi-
tions, which indicates that nutrient supply is limiting only under conditions of higher
growth rates, and in terms of cost-benefit optimization the sophisticated carnivorous
traits are not affordable under limited light (Jaffe et al. 1992). The occurrence of
enzyme secretion in only one of the species of Heliamphora also suggests evolu-
tionary trends in carnivory within the genus, with enzyme secretion being the most
advanced trait in carnivory.
The expression of true carnivory is more dubious in the tanks formed by the
leaf rosettes of bromeliads. Jolivet and Vasconcellos-Neto (1993) note that in gen-
eral, in contrast to dicotyledonous carnivorous plants, among the monocotyle-
dons there is only “protocarnivory” (see Sect. 6.6.3). Examples include Catop-
sis berteroniana, Brocchinia reducta and Brocchinia hechtioides among bromeli-
ads or Paepalanthus bromelioides (Eriocaulaceae) of upland plateaus in northern
Brazil. In moist upland savannas of Venezuela the terrestrial bromeliad Brocchinia
reducta shows such extensive developments in some areas, that one may speak
of a “Brocchinia-savanna”. It catches many animals and has a waxy inner sur-
face to prevent escape (Fig. 10.24). There is breakdown of the bodies of small
animals and absorption of solutes via the bromeliad scales. The outer walls of
the scale cells have an unusual structure. They have a labyrinthine-like appear-
ance and particularly large pores (6.6 nm) allowing the passage of rather large
molecules, which possibly is followed by cellular uptake via endocytosis-vesicles
(Owen and Thomson 1991). The species has been considered as a true carnivorous
plant (Givnish et al. 1984), although glands and enzyme secretion are totally ab-
sent.
10.2.4 The Aluminium Problem
High levels of aluminium in soils are a particular problem in the tropics (Sect.
10.2.4.1) but also globally develop severe adverse effects in agriculture and forestry.
Therefore the literature on aluminium/plant interactions is immense. To develop
a background here for the aluminium relations of plants in the tropics key points
were extracted from ca. 300 references to illustrate potential damage (Sect. 10.2.4.2)
and responses of defence (Sect. 10.2.4.3). (These references cannot be cited here;
a few reviews are the following: Lüttge and Clarkson 1992; Delhaize and Ryan
1995; Kochian 1995; Rengel 1996; Ma 2000; Čiamporová 2002.)
Fig. 10.24A–D Brocchinia reducta in a wet marshy savanna (A), with the typical bromeliad inflo-
rescence (B), with the wax on the adaxial leaf surface that should prevent the escape of animals
fallen into the bromeliad tank (C), and a tank cut open to show the putrefying mass of animals at
the bottom (D). (Gran Sabana, Venezuela; February 1989)
10.2.4.1 The Aluminium Load in Tropical Soils
Clay minerals typical for savanna soils are:

• kaolinite, Al2 O3 · SiO2 · 2 H2 O,
• gibbsite, Al2 O3 · 3 H2 O,
• goethite, FeO(OH).
They have low cation-exchange capacity (CEC) and low water storage capacity.
Ferralization is a frequent process where bases and silicious acid are leached, leav-
ing aluminium and iron oxides (Al2 O3 , Fe2 O3 ). Thus, ferralitic soils always have
very high concentrations of Al3+ . A special formation is the “arecife” in the Llanos
of Venezuela (see Sect. 10.1.2.2 and Fig. 10.7). Iron oxide is precipitated at high
groundwater table level in the young alluvial sediments forming these soils, such
that gravel, sand and clay are solidified to a hard crust of a thickness of 1 – 3 m. It
generally lies at a depth of 30 – 80 cm but may also be lower or closer to the surface
(Fig. 10.7). Soils of the Brazilian cerrados contain between 75 and 360 ppm Al3+
(Eiten 1972).
By comparison, high Al-load in acidifying soils has also been observed in the
temperate zone. It is thought to be one of the possible reasons for forest decline, and
here, the equilibrium soil solution contains up to 20 – 40 ppm Al.
10.2.4.2 Potential Damage to Plants by Aluminium
Damage of plants by high aluminium levels in the substratum occurs at multiple

levels and is due to both extracellular effects on the surface of roots and cells and
intracellular effects after uptake and translocation of aluminium in the plants. The
major interactions can be listed as follows.
• Ionic interactions:
– Phosphate. Al precipitates phosphate at surfaces in the apoplast in the form
of the hardly soluble Al2 (PO4 )3 salt and thus reduces P-availability.
– Iron. High Al levels in the medium are associated with high acidity, which
at the same time leads to increased mobility of iron and thus may induce Fe-
stress. Conversely Al may also induce Fe deficiency and chlorosis because
it inhibits the biosynthesis of phytosiderophores, Fe-complexing agents func-
tioning in iron uptake of plants.
– Divalent cations, Ca2+ , Mg2+ , Mn2+ , Zn2+ . Al occupies important cation-
exchange sites in the apoplast and thus prevents access of essential divalent
cations to these sites, which adversely reduces their availability to the plants.
– Various other nutrients. Al may inhibit nitrate uptake and induce boron de-
ficiency.
• Cell wall interactions:
– Al modifies cell wall components making the cell wall thick and rigid, thus
inhibiting growth.
– Al binds to pectins (depending on the degree of methylation of the pectins)

occupying ion exchange sites.
• Membrane interactions:
– Al binds to proteins and phospholipids of membranes. Thus, Al affects struc-
ture and fluidity of membranes increasing their rigidity and reducing perme-
ability and it occupies cation exchange sites.
– Al blocks K+ - and Ca2+ -channels in membranes.
– Al inhibits proton-pumping ATPases of membranes (plasma membrane and
tonoplast) and thereby reduces electric membrane polarization.
• Metabolism interactions:
– Al generally affects metabolism. It binds to ATP making it metabolically
unavailable.
– Al elicits oxidative stress indirectly via enhancement of Fe-mediated peroxi-
dation which affects membrane structure and can cause DNA-damage.
• Cytoskeleton interactions:
– Non-hydrolysable Al3+ -ADP and Al3+ -ATP complexes bind to actin/myosin
and prevent cytoskeleton function.
• Interactions with intracellular messenger networks:
– Al-induced callose formation blocks apoplastic and symplastic transport
routes, and thus, inhibits basipetal transport of the phytohormone indole
acetic acid (IAA).
– Al may become involved in Ca2+ /calmodulin interactions, which are impor-
tant in intracellular regulatory processes and signalling at the molecular level.
– Binding to phosphorylated proteins associated with DNA Al may interfere
with transcription.
10.2.4.3 Protective Plant Responses
There are multicomponent tolerance and resistance mechanisms, a summarising list-

ing of which is as follows.
• Aluminium exclusion:
– Alkalinization of the rhizosphere.
– Secretion of Al-chelators, such as organic acids (malate, citrate, oxalate) and
flavonoid-type phenolics (catechin), supports Al-exclusion from the cells.
– Excretion of phosphate diminishes Al-mobility.
– Al-elicited formation of reactive oxygen species (e.g. H2 O2 ) may result in
kind of a hypersensitive reaction killing small areas of tissue and excluding
Al from the remaining tissue.
• Aluminium inclusion:
– Organic acid complexes (malate, citrate, oxalate) of Al are not toxic and serve
as transport forms for sequestration, especially in the central cell sap vac-
uoles, and thus, provide internal tolerance.
– Complexes with inorganic phosphate bind Al, but the disadvantage is that
this locks up significant amounts of phosphate.
– Al-silicon complexes can form in the shoot tissue.
Gene regulation is involved in protective plant responses, which affects various
metabolic functions involved in organic acid secretion. We have seen above that the
organic acid anions malate, citrate and oxalate are involved in both exclusion and
inclusion and internal tolerance mechanisms of aluminium. Aluminium signalling
elicits upregulation of de novo synthesis of organic acids and transport mechanisms
and activation of anion channels and the proton transporting ATPase in the plas-
mamembrane.
10.2.4.4 Al-relations of Tropical Plants
Aluminium has different consequences for tropical forests as compared to savan-

nas. In the forests, even when Al-concentrations in the soil are high the effect on
plants is smaller, because the nutrient cycle is tightly coupled between the decom-
posing litter and the vegetation and tends not to involve the mineral soil very much.
In savannas, however, plants take up minerals from the soil solution, which is in
equilibrium with an Al-enriched exchange complex. Since most forest-tree species
are more sensitive to aluminium than savanna plants, the Al-load of soils can in
part explain the competition between forest, cerrados and savannas (Eiten 1972)
and this may be one important determinant of the complex ecological regulation
leading to the co-occurrence of forest and savanna in the tropics (Medina 1982).
Under certain circumstances Al may even stimulate growth, e.g. by alleviating H+ -
toxicity at low pH and by attenuation of excess phosphorus toxicity (Watanabe et al.
2006). For Miconia albicans (Melastomataceae) from the cerrados of Brazil, condi-
tions which lead to a degree of aluminium accumulation, such as non-calcareous
acid soils, are even favourable for growth (Haridasan 1988). In another tropical
Melastomataceae, Melastoma malabathricum, growing on acid sulphate soils, Al
is a nearly essential mineral reducing toxic iron accumulation in roots and shoots
(Watanabe et al. 2006).
Among trees in a cloud forest of Northern Venezuela there are Al-accumulators
and Al-excluders, reflected in the xylem sap concentration in Al. In the Al-accumu-
lator Richeria grandis (Euphorbiaceae), Al-levels in the leaves increased with age
to levels of about 15,000 ppm (Cuenca et al. 1990, 1991). The gallery-forest tree
Vochysia venezolensis (Vochysiaceae) in South America also accumulates up to
25,000 ppm Al related to dry matter (Eiten 1972; Sarmiento 1984). In a savanna
in Trinidad, Al-levels in the grass Panicum stenoides were on average 910 ppm
with maximum levels over 4,000 ppm, and the herbaceous Melastomataceae Acisan-
10.3 The Fire Factor 361
thera uniflora contained over 20,000 ppm (Sarmiento 1984). Haridasan (1982)
lists Al-levels between 4000 and 14,000 ppm for various Al-accumulating cer-
rado species of central Brazil, but the highest levels of Al in plants quoted from
the literature are 66,100 ppm for the Melastomataceae Miconia acinodendron and
72,300 ppm for the Symplocaceae Symplocos spicata. For comparison, in areas
of forest decline in the temperate zone, Al-levels in the root dry mass range
from 20 to 14,000 ppm depending on sites and soil depths (Lüttge and Clark-
son 1992).
10.3 The Fire Factor
10.3.1 The Causes of Fire: Anthropogenic and Natural
Fires play an important role in tropical biota (Goldammer 1990; Fig. 10.25). Dy-
namical global vegetation models (Bond et al. 2005) impressively illustrate the role
of fires (Fig. 10.26). There would be more forest and less savanna coverage with-
out fire (compare the South-American and the African continents with the actual
tree cover in Fig. 10.26A, the tree cover simulated with fire on in Fig. 10.26B and
the simulated tree cover with fire off in Fig. 10.26C). Fires can originate naturally
(see below) but currently the major cause of fires is man. Alexander von Humboldt
(1808/1982) (Humboldt 1982) recognized this in his “Journey to South America”
and mentions benefits and even the pleasure of fires, but also suggests the draw-
backs.
Fig. 10.25 Man-made fires from Martius’ Flora Brasiliensis (1840–1906)

“The pastoral people burn the grassland to obtain fresher and finer grass by new growth . . .
Thus, if one relaxes on a magnificent tropical evening at the shore of the lake1 and enjoys the
delightful coolness, one observes with pleasure the picture of the fires along the horizon,
reflected in the waves beating the shore. . . . The savanna is frequently burnt to improve
the pasture ever since the Llanos were inhabited. Together with the grasses by chance the
scattered groups of trees are also destroyed. No doubt, these plains in the 15th century were
not as bare as now. Nevertheless, even the first conquerors coming from Coro describe the
savannas, in which one sees nothing but sky and grass, widely tree-less and difficult to pass
because of the heat reflected by the ground.”2
Fires ignited by man have not only been used in slash-and-burn agriculture
(Sects. 1.3 and 3.3.3) or in the management of pastures (Sect. 10.3.3) but also by
very early hunter/gatherer societies to drive game out of forest thickets (Kern 1994).
However, fires in savannas, as well as in other tropical and non-tropical ecosystems,
may also be caused naturally (Overbeck and Pfadenhauer 2007). In certain dry sa-
vannas of Africa during storms there is often lightning with little rain or before the
rain sets in. In Australia, fire has been long considered as a natural environmental
stress factor (Walter and Breckle 1984). Palaeontological findings show that fires
destroying vegetation must have occurred since the Devonian (376 ×106 years ago).
The prerequisites for ignition of such fires are:
• a certain minimal atmospheric concentration of oxygen,
• the presence of combustible material.
The minimal O2 -concentration required was shown experimentally to be 13%, i.e.
slightly less than two-thirds of the present level. Since it is assumed that atmospheric
oxygen has resulted from photosynthetic O2 -evolution, we may conclude that 13%
must have been reached by the time of the Devonian. Terrestrial vegetation had
also developed to a stage that the second of the above criteria was fulfilled. As the
possible causes we may list:
• lightning,
• sparks formed during rockfalls,
• vulcanism, and
• self-ignition of fermenting material
(Walter and Breckle 1984; Jones and Chaloner 1991). Thus, plants have been ex-
posed to fire for long enough to allow evolution of special adaptations with stress
avoidance and resistance in an ecophysiologically defined group of plants called
pyrophytes.
1 Lake of Valencia, Venezuela.
2 Südamerikanische Reise “. . . brennt das Landvolk die Weiden ab, um ein frischeres, feineres
Gras als Nachwuchs zu bekommen. . . Wenn man so an einem herrlichen tropischen Abend am
Seeufer1 ausruht und die angenehme Kühle genießt, betrachtet man mit Lust in den Wellen, die an
das Gestade schlagen, das Bild des roten Feuerrings am Horizont. . . Seit die Llanos bewohnt. . .
sind, zündet man häufig die Savanne an, um die Weide zu verbessern. Mit den Gräsern werden
dabei zufällig auch die zerstreuten Baumgruppen zerstört. Die Ebenen waren ohne Zweifel im 15.
Jahrhundert nicht so kahl wie gegenwärtig; indessen schon die ersten Eroberer, die von Coro herka-
men, beschrieben die Savannen, in denen man nichts sieht als Himmel und Rasen, im allgemeinen
baumlos und beschwerlich zu durchziehen wegen der Wärmestrahlung des Bodens.”
Fig. 10.26A–C Tree cover in the South-American and African continents as actually observed (A)
and simulated by Bond et al. (2005) with fire on (B) and fire off (C). The graphs have been redrawn
and strongly simplified from Fig. 6 in Bond et al. (2005)
10.3.2 Pyrophytes
Discussing adaptations of plants to fire as an important natural ecological factor it

is useful to distinguish between:
• pyrophilous plants, which obtain an advantage in the competition with other
plants, and
• true pyrophytes, which essentially need fire at least at some stage in their life
cycle.
Smaller plants may survive fires since the temperature at the soil surface may reach
values of ∼ 75 ◦ C only for a few minutes, and 1 – 5 cm below the surface temper-
atures may already be much lower (Walter and Breckle 1984). Thus the terminal
buds in the center of tussock grasses are well enough protected, and regeneration
can also occur from below-ground plant organs. Taller fire resistant plants, apart
from specialized savanna trees with thick bark and dormant buds (see Sect. 10.3.3),
are often tree ferns or monocotyledonous plants (like palms, Yucca or Xanthorrhoea,
Fig. 10.27). These plants do not have a cambium at the periphery of their stems, as
found in dicotyledonous shrubs and trees. In some Eucalyptus species in Australia,
survival is guaranteed by formation of below-ground stem-thickenings (“lignotu-
bers”), and reproduction by seeds is facilitated by removal of dry litter during fires
and by killing the predators of young seedlings (Walter and Breckle 1984).
The genuine pyrophytes are literally dependent on fire. In the Cyperaceae Bul-
bostylis spadicea in the Brazilian cerrados flowering is stimulated by fire (Gotts-
berger and Silberbauer-Gottsberger 2006). A member of the Australian Liliaceae
(or Xanthorrhoeaceae), Xanthorrhoea, only flowers after a fire. Among the woody
plants of the cerrados in Brazil, Coutinho (1976) (see also Gottsberger and Silber-
bauer-Gottsberger 2006) distinguished the following responses of flowering to
fires:
• species which quantitatively and qualitatively depend on fire and where fire elic-
its flowering at any time during the seasons,
• species where fire elicits flowering only during the dry season or in combination
with short days,
• species which do not react to fire and flower during the dry season or after induc-
tion by short days,
• species which are damaged by fire and normally flower during the rainy season
or after induction by long days.
Among the Australian Proteaceae there are many species where the fruits can only
open and disperse seeds after a fire, e.g. Banksia ornata, Hakea platysperma and
Xylomelum pyriforme, as well as the conifer Actinostrobus.
The evolution of such fire resistant and fire demanding plants also implies that fire
is necessary to stabilize the ecological equilibria in ecosystems which have always
been regularly subject to fire. In fact, it has been noted in some conservation areas
and national parks that total prevention of fires has had adverse effects (Walter and
Breckle 1984).
Fig. 10.27A, B Pyrophilous plants of Australia. A Xanthorrhoea (Liliaceae). B Cycas media (Cy-
cadaceae)
10.3.3 Burning by Man: Losses and Gains
During severe drought periods in savannas the decomposition of above-ground dead

organic matter by microorganisms is very much reduced. This cover prevents new
growth. Perennial grasses die back and seedling mortality under such a dense layer
of dead plant material is high. Eventually the whole grass layer may die, as shown
in a long-term experiment over 20 years, where specific areas were protected from
fire and grazing at the biological field station at Calabozo in the Venezuelan Llanos
(Medina and Silva 1990).
Rapid mineralization by fire removes the dead biomass and also has nutritional
effects. Burning decreases soil acidity, and promotes mineralization of nitrogen.
After an episode of fire, rates of nitrification increase for several years, followed
by a decline of nitrification and increase in ammonium availability (Stewart et al.
1993). However, fire not only enriches the soil with minerals, it may also lead to
losses especially of N and S in the form of volatile oxides (Table 10.12). Most of
these losses are from vegetation rather than soil (Stewart et al. 1993). For N the
range of such losses is:
• 4.5 – 5.6 kg ha−1 year−1 in Australia,
• 8 – 10 kg ha−1 year−1 in Africa, and
• 8 kg ha−1 year−1 in Venezuela
Table 10.12 N- and S-input and losses and biomass production in a Trachypogon savanna in cen-
tral Venezuela. (Medina 1982)
Biomass production Losses as volatile gases Input by rain
N S N S
(t ha−1 ) (kg ha−1 year−1 ) (kg ha−1 year−1 )
Minimum 2.9 9.3 2.6 1.3 6.8
Maximum 10.0 32.0 9.0 4.7 9.1
(Medina 1993). It may much exceed the import via rain (Tables 10.10 and 10.12).
The global role of fires has been surveyed by Fontan (1993). Table 10.13 shows
the annual turnover of biomass in forests and savannas in the tropical regions of
the world. The contribution of savanna fires to the total biomass burnt per year
is seen to be high in America but particularly so in Africa. Fires not only cause
losses of minerals but also make a significant contribution to atmospheric loading
of infrared-active gases which cause the greenhouse effect (CO2 , CO, CH4 , O3 ) and
with straightforward pollutants (like CO, N-oxides, O3 ) (Table 10.14).
Frequent man-made fires also open the soil surface to solar radiation, which leads
to oxidation and burning of humus (Eiten 1972) and leaching following rainfall.
Therefore in dry savannas fire is always detrimental.
Moreover, of course, fire always damages the forests unless it is wet gallery for-
est with permanently inundated soil (Fig. 10.28). The fires intrude from the edges
Fig. 10.28 Savanna in the Llanos of Venezuela near Puerto Ayacucho with scattered islands of
semi-evergreen forest and wet gallery forest (background to the right)
Table 10.13 Burnt biomass in the tropical regions of the world from tropical forests, savannas and
other sources (firewood plus agricultural waste) in Tg dry matter per year. (After Fontan 1993)
Tropical region Forests Savannas Other sources

America 590 (34%) 770 (44%) 370 (22%)
Africa 390 (12%) 2430 (76%) 400 (12%)
Asia 280 (13%) 70 (3%) 1840 (84%)
Oceania – (0%) 420 (94%) 25 (6%)
Total 1260 (17%) 3690 (49%) 2635 (34%)
Table 10.14 Global gas emissions form bushfires in Tg year−1 and in % of total global emissions.
(Fontan 1993)
Compound Emissions from bush fires

Carbon dioxide (CO2 ) 3500 40%
Carbon monoxide (CO) 350 32%
Methane (CH4 ) 38 10%
Hydrocarbons other than methane 24 24%
N-oxides 9.3 27%
Ammonia (NH3 ) 5.3 12%
Chloromethane (CH3 Cl) 0.5 22%
Hydrogen gas (H2) 19 25%
Ozone (O3 ) 420 38%
into the forests, and where the trees are not fire-resistant, year by year savanna grad-
ually encroaches into the area previously occupied by forests (Fig. 10.29).
Savanna trees are particularly fire resistant (Figs. 10.30 and 10.31), being pre-
dominantly evergreen, with a thick corky bark and dormant buds, and sprout after
fires before the beginning of the rainy season (see Sect. 10.1.2.1). It is estimated
that only trees with a bark less than 5 mm thick can be killed by fires in the cerrado
(Miranda et al. 1993; Gottsberger and Silberbauer-Gottsberger 2006). Some trees,
in areas where regular burning occurs, may restrict formation of their woody stems
to below the ground surface in the form of lignotubers or xylopodia, where dormant
buds are protected and can readily produce new growth after a fire (Fig. 10.31B).
Xylopodia may also serve storage of water and minerals, and the plants forming
such organs can be considered to represent the typical life form of xylohemicrypto-
phytes (Gottsberger and Silberbauer-Gottsberger 2006; for definitions of life forms
see Sects. 3.3.4 and 11.3.1, Table 11.2). Deciduous trees, with phenological cycles
related to the seasonality of rainfall, are more sensitive to fire, and they are ex-
cluded from regularly burned savannas (Medina and Silva 1990). In the experiment
at Calabozo mentioned above, during 20 years of protection, total tree density in-
creased considerably, both of fire-resistant savanna trees and fire-sensitive species
from the surrounding semideciduous forest (Table 10.15), and Table 10.16 shows
similar findings for a Brazilian cerrado. In dry savannas of Namibia where fires are
disastrous and burning is now strictly avoided one can observe Acacia mellifera
ssp. detinens spreading out and occupying vast areas, which has become a curse for
cattle farmers.
Fig. 10.29 A Remains of montane forest near Akanzobe in Madagascar. B In the centre is shown
a zone of common brake fern (Pteridium aquilinum) between the grassland and the forest, which
burns very readily and gradually progresses towards the forest. (Photographs courtesy M. Kluge)
Fig. 10.30 Palicourea rigida in the Llanos of Venezuela with a thick corky bark coloured black
from fire (January 1989)
Table 10.15 Number of tree stems per ha in a fire-protected savanna plot over 21 years at Cal-
abozo, Venezuela. (Medina and Silva 1990)
Species 1962 1969 1977 1983

Savanna trees (evergreen) 92 174 270 1010
Forest species (evergreen) 1 3 11 32
Forest species (deciduous) 0 77 229 1319
93 254 510 2361
In wet savannas fire can be beneficial, but only when the timing is correct. If
a fire occurs before the start of the rainy season, the trees are protected (see above)
and safe sprouting of grasses is obtained. The centers of tussocks of grasses support-
ing the meristems for regrowth are protected from the heat of the fire by an insulat-
Fig. 10.31A, B Byrsonima verbascifolia. A In the cerrados of Brazil with bark coloured black from
fire (August 1993). B In the Llanos of Venezuela with fresh leaves and flowers sprouting from an
underground stem (February 1989)
ing layer of old leaves. If burning occurs too early in the dry season, subsequent
new growth uses up water reserves and dies before the rainy season sets in, and the
whole plant may dry out totally. If burning is done too late in the dry season, new
Table 10.16 Comparison of adjacent hectares of low-tree and scrub cerrado of central Brazil peri-
odically burned each 2 – 3 years and not burned for over 20 years, respectively. (Density of stems
is larger than density of individuals since some individuals produce more than one stem.) (G. Eiten
and R. H. R. Sambuichi, pers. comm.)
Not burned Burned

Stems (number ha−1 ) 6677 1765
Individuals (number ha−1 ) 5788 1663
Species 92 57
Fig. 10.32 Variations in the epigeous biomass (T total; G green; D dry) of a savanna at Cal-
abozo, Venezuela, after it was burned and until it reached a steady state (dotted line) in five years.
(Sarmiento 1984; reprinted by permission of Harvard University Press)
growth is induced when the water reserves are already exhausted and growth is very
limited. If burning occurs in the middle of the dry season, green biomass is pro-
duced, which is maintained until the beginning of the rainy season (Medina 1982;
Medina and Silva 1990). The experiments at Calabozo have shown that maximum
above-ground biomass in a protected savanna increased during four years after the
last fire (Table 10.17) and then stabilized at a certain level (Fig. 10.32). Fire later
at the beginning of the rainy season led to lower biomass production than fire be-
fore the middle of the dry season (Table 10.18). The seasonal development of the
grass Trachypogon plumosus shows that the green biomass after a fire is somewhat
increased as compared with the control (Fig. 10.33) and new dry biomass and total
above-ground biomass increase rapidly over the year after most of the old biomass
was destroyed by the fire.
Table 10.17 Maximum epigeous biomass in a fire-protected Trachypogon savanna at Calabozo,

Venezuela. (Sarmiento 1984)
Time since last fire (years) Maximum epigeous biomass (g m−2 )

1 230 – 730
2 520 – 850
3 980
4 1200
5 1200
Fig. 10.33 Variations in the epigeous biomass of a savanna at Calabozo, Venezuela, during the first
year after a fire (F) and in a non-burnt control plot (C). (Sarmiento 1984; reprinted by permission
of Harvard University Press)
Table 10.18 The effect of fire given at different times during the year on daily biomass production
in a savanna at Calabozo, Venezuela. (Medina 1982)
Biomass production (g m−2 day−1 )

Three to four years protected from fire 2.5 – 2.6
Fire before the middle of the dry period 2.9 – 3.7
Fire at the beginning of the wet period 1.8 – 2.1
References
Alexander EB (1973) A comparison of forest and savanna soils in north-eastern Nicaragua. Turri-
alba 23:181–191
Almeida JPF, Hartwig UA, Frehner M, Nösberger J, Lüscher A (2000) Evidence that P deficiency
induces N feedback regulation of symbiotic N2 fixation in white clover (Trifolium repens L.).
J Exp Bot 51:1289-1297
References 373
Baldani JI, Reis VM, Baldani VLD, Döbereiner J (2002) A brief story of nitrogen fixation in
sugarcane – reasons for success in Brazil. Funct Plant Biol 29:417–423
Baruch Z, Fernández DS (1993) Water relations of native and introduced C4 grasses in a neotropi-
cal savanna. Oecologia 96:179–185
Bergman B, Johansson C, Söderbäck E (1992) The Nostoc-Gunnera symbiosis. New Phytol
122:379–400
Berry C, Loutit B (2000) Bäume und Sträucher im Etoscha Nationalpark und in Nord- und Zentral-
Namibia, 2nd edn. Namibia Wiss. Ges., Windhoek
Bethlenfalvay GJ, Reyes-Solis MG, Camel SB, Ferrera-Cerrato R (1991) Nutrient transfer between
the root zones of soybean and maize plants connected by a common mycorrhizal mycelium.
Black CC (1973) Photosynthetic carbon fixation in relation to net CO2 uptake. Annu Rev Plant
Physiol 24:253–286
Bond WJ, Woodward FI, Midgly GF (2005) The global distribution of ecosystems in a world
without fire. New Phytol 165:525–538
Boom A, Mora G, Cleef AM, Hooghiemstra H (2001) High altitude C4 grasslands in the northern
Andes: relicts from glacial conditions? Rev Palaeobot Palynol 115:147–160
Borchert R (2000) Organismic and environmental controls of bud growth in tropical trees. In:
Viemont JD, Crabbè J (eds) Dormancy in plants: from whole plant behaviour to cellular con-
trol. CAB International, Wallingford, pp 87–107
Borchert R, Rivera G (2001) Photoperiodic control of seasonal development and dormancy in
tropical stem-succulent trees. Tree Physiol 21:213–221
Bucci SJ, Scholz FG, Goldstein G, Meinzer FC, Sternberg L da SL (2003) Dynamic changes
in hydraulic conductivity in petioles of two savanna tree species: factors and mechanisms
contributing to the refilling of embolized vessels. Plant Cell Environ 26:1633–1645
Bucci SJ, Goldstein G, Meinzer FC, Franco AC, Campanello P, Scholz FG (2005) Mechanisms
contributing to seasonal homeostasis of minimum leaf water potential and predawn disequi-
librium between soil and plant water potential in neotropical savanna trees. Trees 19:296–304
Burleigh SH, Harrison MJ (1999) The down-regulation of Mt4-like genes by phosphate fertilization
occurs systemically and involves phosphate translocation to the shoots. Plant Physiol 119:241–
248
Campa C, Diouf D, Ndoye I, Dreyfus B (2000) Differences in nitrogen metabolism of Faidher-
bia albida and other N2 -fixing tropical woody acacias reflects habitat water availability. New
Phytol 147:571–578
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
Cerling TE, Harris JM, MacFadden BJ, Quade J, Leakey MG, Eisenmann V, Ehleringer JR (1998)
Miocene/Pliocene shift: one step or several. Nature 393:126–127
Cernusak LA, Aranda J, Marshall JD, Winter K (2006) Large variation in whole-plant water-use
efficiency among tropical tree species. New Phytol 173:294–305
Chapotin SM, Holbrook NM, Morse SR, Gutiérrez MV (2003) Water relations of tropical dry
forest flowers: pathways for water entry and the role of extracellular polysaccharides. Plant
Chapotin SM, Razanameharizaka JH, Holbrook NM (2006a) Water relations of baobab (Adanso-
nia spp. L.) during the rainy season: does stem water buffer daily water deficits? Plant, Cell
Environ 29:1021–1032
Chapotin SM, Razanameharizaka JH, Holbrook NM (2006b) Baobab trees (Adansonia) in Mada-
gascar use stored water to flush new leaves but not to support stomatal opening before the
rainy season. New Phytol 169:549–559
Chiu W-L, Peters GA, Levieille G, Still PC, Cousins S, Osborne B, Elhai J (2005) Nitrogen depri-
vation stimulates symbiotic gland development in Gunnera manicata. Plant Physiol 139:224–
230
Čiamporová M (2002) Morphological and structural responses of plant roots to aluminum at organ,
tissue, and cellular levels. Biol Plant 45:161–171
Clarkson DT, Kuiper PJC, Lüttge U (1986) Mineral nutrition: sources of nutrients for land plants
from outside the pedosphere. Prog Bot 48:80–96
Coutinho LM (1976) Contribuiçao ao conheciamento do papel ecologico das queimadas na flo-
raçao do especias do cerrado. Tesc de livre-docente en Ecologia Vegetal. Universidade de Sao
Paulo, Sao Paulo
Crush JR (1974) Plant growth response to vesicular-arbuscular mycorrhiza. VII. Growth and nodu-
lation of some herbage legumes. New Phytol 73:743–752
Cuenca G, Herrera R, Medina E (1990) Aluminium tolerance in trees of a tropical cloud forest.
Plant Soil 125:169–175
Cuenca G, Herrera R, Mérida T (1991) Distribution of aluminium in accumulator plants by X-ray
microanalysis in Richeria grandis Vahl leaves from a cloud forest in Venezuela. Plant Cell
Environ 14:437–441
Delhaize E, Ryan PR (1995) Aluminum toxicity and tolerance in plants. Plant Physiol 107:315–
321
Domec J-C, Scholz FG, Bucci SJ, Meinzer FC, Goldstein G, Villalobos-Vega R (2006) Diurnal
and seasonal variation in root xylem embolism in neotropical savanna woody species: impact
on stomatal control of plant water status. Plant Cell Environ 29:26–35
Earnshaw MJ, Carver KA, Gunn TC, Kerenga K, Harvey V, Griffiths H, Broadmeadow MSJ (1990)
Photosynthetic pathway, chilling tolerance and cell sap osmotic potential values of grasses
along an altitudinal gradient in Papua New Guinea. Oecologia 84:280–288
Eiten G (1972) The cerrado vegetation of Brazil. Bot Rev 38:201–341
Eiten G (1986) The use of the term “savanna.” Trop Ecol 27:10–23
and amphibious plants in the neotropics. Oecologia Brasiliensis, vol IV, Universidade Federal
Fontan J (1993) La pollution atmosphérique sous les tropiques. Recherche 24:400–408
Franco AC, Lüttge U (2002) Midday depression in savanna trees: coordinated adjustments in pho-
tochemical efficiency, photorespiration, CO2 assimilation and water use efficiency. Oecologia
131:356–365
Franco AC, Bustamente M, Caldas LS, Goldstein G, Meinzer FC, Kozovits AR, Rundel P, Coradin
VTR (2005) Leaf functional traits of neotropical savanna trees in relation to seasonal water
deficit. Trees 19:326–335
Garcia-Méndez G, Maass JM, Matson PA, Vitousek PP (1991) Nitrogen transformations and ni-
trous oxide flux in a tropical deciduous forest in Mexico. Oecologia 88:362–366
Gebauer G, Schubert B, Schumacher MI, Rehder H, Ziegler H (1987) Biomass production and ni-
trogen content of C3 - and C4 -grasses in pure and mixed culture with different nitrogen supply.
Gilchrist AJ, Juniper BE (1974) An excitable membrane in the stalked glands of Drosera capensis
L. Planta 119:143–147
Givnish TJ, Burkhardt EL, Happel RE, Weintraub JD (1984) Carnivory in the bromeliad Broc-
chinia reducta, with a cost/benefit model for the general restriction of carnivorous plants to
sunny, moist, nutrient-poor habitats. Am Nat 124:479–497
Goldammer JG (ed) (1990) Fire in the tropical biota. Ecosystem processes and global challenges.
Ecological studies vol 41. Springer, Berlin Heidelberg New York
Goldstein G, Andrade JL, Meinzer FC, Holbrook NM, Cavelier J, Jackson P, Cells A (1998) Stem
water storage and diurnal patterns of water use in tropical forest canopy trees. Plant Cell Env-
iron 21:397–406
Gottsberger G, Silberbauer-Gottsberger I (2006) Life in the cerrado a South-American tropical
seasonal ecosystem I. Origin, structure, dynamics and plant use. Reta-Verlag, Ulm
Grace J, Lloyd J, Miranda AC, Miranda H, Gash JHC (1998) Fluxes of carbon dioxide and water
vapour over a C4 pasture in south-western Amazonia (Brazil). Aust J Plant Physiol 25:519–
530
References 375
Haridasan M (1982) Aluminium accumulation by some cerrado native species of central Brazil.
Plant Soil 65:265–273
Haridasan M (1988) Performance of Miconia albicans (Sw.) Triana, an aluminium-accumulating
species, in acidic and calcareous soils. Commun Soil Sci Plant Anal 19:1091–1103
Hatch MD (1987) C4 photosynthesis: a unique blend of modified biochemistry, anatomy and ul-
trastructure. Biochim Biophys Acta 895:81–106
Hatch MD, Osmond CB (1976) Compartmentation and transport in C4 photosynthesis. In: Stocking
CR, Heber U (eds) Transport in plants III. Intracellular interactions and transport processes.
Encyclopedia of plant physiology NS. vol 3. Springer, Berlin Heidelberg New York 1976, pp
144–184
Hess D (1992) Biotechnologie der Pflanzen. Ulmer, Stuttgart
Hoffmann WA, Orthen B, Franco AC (2004) Constraints to seedling success of savanna and forest
trees across the savanna-forest boundary. Oecologia 140:252–260
Högberg P (1986a) Soil nutrient availability, root symbioses and tree species composition in trop-
ical Africa: a review. J Trop Ecol 2:359–372
Högberg P (1986b) Nitrogen-fixation and nutrient relations in savanna woodland trees (Tanzania).
J Appl Ecol 23:675–688
Högberg P, Alexander IJ (1995) Roles of root symbioses in African woodland and forest: evidence
from 15 N abundance and foliar analysis. J Ecol 83:217–224
Huber O (1988) Guayana highlands versus Guayana lowlands, a reappraisal. Taxon 37:595–614
Humboldt A von (1982) Südamerikanische Reise. 1808, quoted after the edition of Reinhard
Jaspert, Ullstein, Berlin
Israel DW (1993) Symbiotic dinitrogen fixation and host-plant growth during development of and
recovery from phosphorus deficiency. Physiol Plant 88:294–300
Jacobs BF, Kingston JD, Jacobs LL (1999) The origin of grass-dominated ecosystems. Ann Mis-
souri Bot Gard 86:590–643
Jaffe K, Michelangeli F, Gonzalez JM, Miras B, Ruiz MC (1992) Carnivory in pitcher plants of the
genus Heliamphora (Sarraceniaceae). New Phytol 122:733–744
James EK, Loureiro M de F, Pott A, Pott VJ, Martins CM, Franco AA, Sprent JI (2001) Flooding-
tolerant legume symbioses from the Brazilian Pantanal. New Phytol 150:723–738
Johansson C, Bergman B (1994) Reconstitution of the symbiosis of Gunnera manicata Linden:
cyanobacterial specificity. New Phytol 126:643–652
Jolivet P, Vasconcellos-Neto J (1993) Convergence chez les plantes carnivores. Recherche 24:456–
458
Jones TP, Chaloner WG (1991) Les feux du passé. Recherche 22:1148–1156
Kern AA (1994) Antecedentes indígenas. Editora da Universidade, Porto, Alegre
Kochian LV (1995) Cellular mechanisms of aluminum toxicity and resistance in plants. Annu Rev
Plant Physiol Plant Mol Biol 46:237–260
Loureiro MF, James EK, Franco AA (1998) Nitrogen fixation by legumes of flooded regions.
In: Scarano FR, Franco AC (eds) Ecophysiologicdal strategies of xerophytic and amphibious
plants in the neotropics. Oecologia Brasiliensis, Universidade Federal do Rio de Janeiro, Rio
de Janeiro, pp 195–233
Lüttge U (1983) Ecophysiology of carnivorous plants. In: Lange OL, Nobel PS, Osmond CB,
Ziegler H (eds) Physiological plant ecology. III. Responses to the chemical and biological
environment. Encyclopedia of plant physiology NS, vol 12C. Springer, Berlin Heidelberg New
York, pp 489–517
Lüttge U, Clarkson DT (1992) Mineral nutrition: aluminium. Prog Bot 53:63–77
Lüttge U, Kluge M, Bauer G (2005) Botanik, 5. Aufl. Wiley-VCH, Weinheim
Ma JF (2000) Role of organic acids in detoxification of aluminum in higher plants. Plant Cell
Physiol 41:383–390
Martius CFP (1840-1906) Flora brasiliensis, vol 1–15. München and Leipzig
Maseda PH, Fernández R (2006) Stay wet or else: three ways in which plants can adjust hydrauli-
cally to their environment. J Exp Bot 57:3963–3977
Mattos EA de (1998) Perspectives in comparative ecophysiology of some Brazilian vegetation

types: leaf CO2 and H2 O gas exchange, chlorophyll a fluorescence and carbon isotope dis-
crimination. In: Scarano FR, Franco AC (eds) Ecophysiological strategies of xerophytic and
amphibious plants in the neotropics. Oecologia Brasiliensis, vol IV, Universidade Federal do
Rio de Janeiro, Rio de Janeiro, pp 1–22
Mattos EA de, Lobo PC, Joly CA (2002) Overnight rainfall inducing rapid changes in photosyn-
thetic behaviour in a cerrado woody species during a dry spell amidst the rainy season. Aust J
Bot 50:241–246
Medina E (1982) Physiological ecology of neotropical Savanna plants. In: Huntles BJ, Walker BH
(eds) Ecological studies, vol 42: Ecology of tropical savannas. Springer, Berlin Heidelberg
Medina E (1986) Forests, savannas and montane tropical environments. In: Baker NR, Long SP
(eds) Photosynthesis in contrasting environments. Elsevier, Amsterdam, pp 139–171
Medina E (1987) Nutrients: requirements, conservation and cycles in the herbaceous layer. In:
Walker B (ed) Determinants of savannas, IUBS Monographs Series No 3, Chap. 3. IRL Press,
Oxford, pp 39–65
Medina E (1993) Mineral nutrition: tropical savannas. Prog Bot 54:237–253
Medina E, Silva JF (1990) Savannas of northern South America: a steady state regulated by water-
fire interactions on a background of low nutrient availability. J Biogeogr 17:403–413
Meinzer FC (2003) Functional convergence in plant responses to the environment. Oecologia
134:1–11
Meinzer FC, James SA, Goldstein G, Woodruff D (2003) Whole-tree water transport scales with
sapwood capacitance in tropical forest canopy trees. Plant Cell Environ 26:1147–1155
Miranda AC, Miranda HS, Oliveira Dias I de, Souza Dias BF de (1993) Soil and air temperatures
during prescribed cerrado fires in Central Brazil. J Trop Ecol 9:313–320
Miranda AC, Miranda HS, Lloyd J, Grace J, Francey RJ, McIntyre JA, Meir P, Riggan P, Lockwood
R, Brass J (1997) Fluxes of carbon, water and energy over Brazilian cerrado: an analysis using
eddy covariance and stable isotopes. Plant Cell Environ 20:315–328
Montaldo P (1977) El espectro de las tribus de gramineas de los Llanos venezolanos. Turrialba
27:175–177
Moraes JAPV, Prado CHBA (1998) Photosynthesis and water relations in cerrado vegetation. In:
Scarano FR, Franco AC (eds) Ecophysiological strategies of xerophytic and amphibious plants
in the neotropics. Oecologia Brasiliensis, vol IV, Universidade Federal do Rio de Janeiro, Rio
de Janeiro, pp 45–63
Nanamori M, Shinano T, Wasaki J, Yamamura T, Rao IM, Osaki M (2004) Low phosphorus toler-
ance mechanisms: phosphorus recycling and photosynthate partitioning in the tropical forage
grass, Brachiaria hybrid cultivar Mulato compared with rice. Plant Cell Physiol 45:460–469
Njoku E (1963) Seasonal periodicity in the growth and development of some forest trees in Nigeria.
J Ecol 51:617–624
Osborne BA, Cullen A, Jones PW, Campbell GJ (1992) Use of nitrogen by the Nostoc-Gunnera
tinctoria (Molina) Mirbel symbiosis. New Phytol 120:481–487
Overbeck GE, Pfadenhauer J (2007) Adaptive strategies in burned subtropical grassland in southern
Brazil. Flora 202:27–49
Owen TP, Thomson WW (1991) Structure and function of a specialized cell wall in the trichomes
of the carnivorous bromeliad Brocchinia reducta. Can J Bot 69:1700–1706
Parsons R, Sunley RJ (2001) Nitrogen nutrition and the role of root-shoot nitrogen signalling par-
ticularly in symbiotic systems. J Exp Bot 52:435–443
Piedade MTF, Junk WJ, Mello JAN (1992) A flooded grassland of the Central Amazon. In: Long
SP, Jones MB, Robert MJ (eds) Primary productivity of grass ecosystems of the tropics and
sub-tropics. Springer, Berlin Heidelberg New York, pp 127–153
Piedade MTF, Long SP, Junk WJ (1994) Leaf and canopy photosynthetic CO2 -uptake of a stand
of Echinochloa polystachya on the Central Amazon floodplain. Are the high potential rates
associated with the C4 -syndrome realized under the near-optimal conditions provided by this
exceptional natural habitat? Oecologia 97:193–201
References 377
Pimenta JA, Bianchini E, Medri ME (1998) Adaptations to flooding by tropical trees: morpholog-
ical and anatomical modifications. In: Scarano FR, Franco AC (eds) Ecophysiological strate-
gies of xerophytic and amphibious plants in the neotropics. Oecologia Brasiliensis, vol IV,
Universidade Federal do Rio de Janeiro, Rio de Janeiro, pp 157–176
Rai AN, Söderbäck E, Bergman B (2000) Cyanobacterium-plant symbioses. New Phytol 147:449–
481
Reich PB, Borchert R (1984) Water stress and tree phenology in a tropical dry forest in the lowlands
of Costa Rica. J Ecol 72:61–74
Rengel Z (1996) Uptake of aluminium by plant cells. New Phytol 134:389–406
Rivera G, Borchert R (2001) Induction of flowering in tropical trees by a 30-min reduction in pho-
toperiod: evidence from field observations and herbarium collections. Tree Physiol 21:201–
212
Rivera G, Elliott S, Caldas LS, Nicolossi G, Coradin VTR, Borchert R (2002) Increasing day-
length induces spring flushing of tropical dry forest trees in the absence of rain. Trees 16:445–
456
Schmucker T, Linnemann G (1959) Carnivorie. In: Handbuch der Pflanzenphysiologie, vol XI.
Springer, Berlin Göttingen Heidelberg, pp 198–283
Scholz FG, Bucci SJ, Goldstein G, Meinzer FC, Franco AC, Miralles-Wilhelm F (2007) Biophysi-
cal properties and functional significance of stem water storage tissues in Neotropical savanna
trees. Plant Cell Environ 30:236–248
Schuepp PH (1993) Leaf boundary layers. New Phytol 125:477–507
Silvester WB, Parsons R, Watt PW (1996) Direct measurement of release and assimilation of am-
monia in the Gunnera-Nostoc symbiosis. New Phytol 132:617–625
Solbrig OT (1993) Plant traits and adaptive strategies: their role in ecosystem function. In: Schulze
E-D, Mooney HA (eds) Biodiversity and ecosystem function. Ecological studies, vol 99.
Souza Moreira FM de, Silva MF da, Faria SM de (1992) Occurrence of nodulation in legume
species in the Amazon region of Brazil. New Phytol 121:563–570
Stewart GR, Pate JS, Unkovich M (1993) Characteristics of inorganic nitrogen assimilation of
plants in fire-prone mediterranean-type vegetation. Plant Cell Environ 16:351–363
Stock PA, Silvester WB (1994) Phloem transport of recently-fixed nitrogen in the Gunnera-Nostoc
symbiosis. New Phytol 126:259–266
Tieszen LL, Senyimba MM, Imbamba SK, Troughton JH (1979) The distribution of C3 and C4
grasses and carbon isotope discrimination along an altitudinal and moisture gradient in Kenya.
Walter H, Breckle S-W (1984) Ökologie der Erde. Bd 2. Spezielle Ökologie der tropischen und
Watanabe T, Jansen S, Osaki, M (2006) Al-Fe interactions and growth enhancement in Melastoma
malabathricum and Miscanthus sinensis dominating acid sulphate soils. Plant Cell Environ
29:2124–2132
Werner D (1992) Symbioses of plants and microbes. Chapman and Hall
Wright JS (1991) Seasonal drought and the phenology of understory shrubs in a tropical moist
forest. Ecology 72:1643–1657
Wright JS (1996) Phenological responses to seasonality in tropical forest plants. In: Mulkey SS,
Chazdon RC, Smith AP (eds) Tropical forest plant ecophysiology. Chapman and Hall, New
York, pp 440–460
Chapter 11
Inselbergs
11.1 Physiognomy
Inselbergs1 are isolated rock outcrops in the palaeotropics and the neotropics com-
ing up out of different vegetation of savannas or cerrados (Fig. 11.1A) or rain-
forests (Fig. 11.1B). In savannas with a certain savanna affinity of their flora they
also have been described as “rock savanna”. They consist of monolithic blocks,
mostly of granite or gneiss, of a considerable geological age, i.e. 10 ×106 years at
least and 40 – 50 ×106 years on average. More rarely inselbergs may also consist
of sandstone. They range from several tens or hundreds of meters high, the highest
one found in French Guiana being 740 m high (Schnell 1987). “Shield-type” insel-
bergs may have extensions of several square kilometers. In arid regions and deserts
they may have been eroded to heaps of rather small rocks (Fig. 11.1D). Inselbergs
evolve by deep and intensive weathering (Fig. 11.2A). Thus, inselbergs are really
islands separated from the surrounding savanna or forest vegetation (Figs. 11.1A, B
and 11.2B; Barthlott et al. 1993; Porembski and Barthlott 2000a). They provide very
different ecological conditions.
In the analysis of inselberg vegetation, one tries to determine whether the occu-
pation of these isolated habitats is deterministic or stochastic, and whether there are
climax states or dynamic oscillations and if there is a coexistence of equilibrium
and non-equilibrium communities (Barthlott et al. 1993; Porembski et al. 2000b).
In fact, the inselbergs themselves are again fragmented in a number of sub-islands
and smaller ecological units (Porembski et al. 2000a; Fig. 11.3), and therefore their
β-diversity (see Sect. 3.3.1). is high. Besides a large spectrum of higher plant life
forms, cryptogamic life forms are particularly important. The rock itself is rarely
free of life, usually being covered by dense cryptogamic crusts of cyanobacteria,
lichens and mosses. Interestingly, depending on the respective dominance of lichens
or cyanobacteria, one may distinguish lichen and cyanobacteria inselbergs. The for-
mer are more abundant in Africa, the latter in South America (Barthlott et al. 1993).
1 The term inselberg from the German Insel = island and Berg = mountain, coined by Bornhardt
(1900), is somewhat more specific than “rock outcrop” and has been widely adopted in scientific
writing.
380 11 Inselbergs
Fig. 11.1A–D Inselbergs coming up out of a savanna along the Rio Orinoco near Puerto Ayacucho
at the Southern rim of the Llanos, Territoria Federal Amazonas, Venezuela (A) and out of a rain-
forest with a rain storm moving up at Les Nouragues, French Guiana (B). Individual inselberg near
Puerto Ayacucho (C), eroded inselbergs, fragmented to heaps of small rocks, in the savanna, near
Caicara del Orinoco, Venezuela (D)
Sub-islands on the rock are formed by individual plants or small groups of plants
growing in humus in cracks, gaps and hollows (Fig. 11.3). Xerophilic plants, such
as succulents and epiphytes, grow saxicolously. Patches of shrubbery originate from
the vegetation islands (Fig. 11.4A,B) and small forests often form on top of the in-
selbergs (Fig. 11.4C). In South America these forests comprise the only deciduous
vegetation units occurring in the area around the Guayana shield, with trees like
Pseudobombax chrysati, Tabebuia orinocensis and Yacaranda filicifolia (Fig. 10.4),
and also palms (Syagrus orinocensis, Fig. 11.4C). While most sites on inselbergs
are extremely dry and hot in the dry season (see Sects. 11.2 and 11.3), there are also
wetter sections (Fig. 11.5) and even rock pools (Porembski et al. 2000a), which sea-
sonally keep small ponds of water and harbour aquatic plants. Floristic diversity
of inselbergs in various regions of the world is documented in detail in Porembski
11.1 Physiognomy 381
and Barthlott (2000a). It is very high and inselbergs are extraordinarily rich in en-
demic species. Huber (1980) listed the natural habitats of the inselbergs along the
Orinoco in Venezuela as belonging to the “flower paradises and botanical gardens
of the earth”.
Phytogeography, plant sociology and vegetation analysis are advancing to ask
the pertinent questions for providing a deeper understanding of these exciting sites
(see Barthlott et al. 1993). Ecophysiology of this fascinating vegetation is still less
developed. One exception are studies of desiccation tolerance in inselberg plants
(Sect. 11.4). Where there is strong seasonality, pools will dry out in the dry season.
The rock surface during midday may easily heat up to temperatures around 60 ◦ C
(Fig. 11.6). Not only cryptogams but also many angiosperms of the inselbergs have
been shown to be able to overcome such dry periods by ecophysiological adapta-
tions, particularly by desiccation tolerance (Sect. 11.4).
382 11 Inselbergs
Fig. 11.2 A Deep and intensive weathering leading to the formation of inselbergs (after Bremer
and Sander 2000). B Scheme of the vegetation of a granitic inselberg in South America (French
Guiana) (Schnell 1987, with kind permission of Masson S. A. Paris)
11.2 Cryptogams
11.2.1 Cyanobacteria
11.2.1.1 Ubiquity
A conspicuous feature of the inselberg rocks is their superficial appearance of dark

coloration. This was first noted in Venezuela by A LEXANDER VON H UMBOLDT
(Humboldt von 1849) who described the black surface of the rocks in the riverbed
of the Orinoco and also of the rock outcrops further away:
“In the Orinoco, especially in the cataracts of Maypures and Atures, all granite blocks and
even white pieces of quartz to the extent they are touched by the water of the Orinoco
11.2 Cryptogams 383
Fig. 11.3 Isolated plants and

sub islands on inselbergs:
The bromeliad Pitcairnia
pruinosa on the inselberg
at Galipero near Puerto Ay-
acucho, Venezuela (A), the
bromeliad Tillandsia arau-
jei (B) and the orchid Maxil-
laria sp. (C) on the inselberg
Pedra Grande at Atibaia, S.P.,
Brazil, and the Velloziaceae
Pleurostima gounelleana (D)
on rock outcrops at Itatiaia,
R.J., Brazil
develop a greyish-black cover which does not penetrate more than 0.01 line2 into the interior
of the rocks. One might think to see basalt or fossils stained by granite. In fact the sheath
appears to contain manganese oxide and carbon.”3
2 1 line = 2.0 to 2.5 mm; Fig. 11.7.

3 “Im Orinoco, besonders in den Cataracten von Maypures und Atures, . . . nehmen alle Gran-
itblöcke, ja selbst weiße Quarzstücke, so weit sie das Orinoco-Wasser berührt, einen graulich-
schwarzen Überzug an, der nicht 0,01 Linie ins Innere des Gesteins eindringt. Man glaubt Basalt
oder mit Granit gefärbte Fossilien zu sehen. Auch scheint die Rinde in der That braunstein- und
kohlenstoffhaltig zu sein.”
384 11 Inselbergs
Regarding the dark colour of the rock outcrops at a greater distance from the
current river bed of the Orinoco he assumed that the river had extended much further
in earlier times:
“This assumption is supported by several observations. One sees black caves 150 to 180
feet above the present water level. Their existence teaches . . . that the streams, whose size
presently excites our admiration, are only humble remains of the enormous amounts of
water in archaic times . . . These simple observations even did not escape the rough natives.
Everywhere they drew our attention to the old waterlevel.”4
The finding of prehistoric petroglyphs in the rocks of the inselbergs (Fig. 11.8A)
also played a role in H UMBOLDT’s discussion, since he asked how the Indians might
have found access to the steep walls of rock for carving them:
“Between Encaramada and Caycara on the banks of the Orinoco one frequently finds . . .
hieroglyphic pictures in considerable height on the rock faces, which now would only be
accessible by means of extraordinarily high scaffolding. If one asks the natives how these
pictures could have been carved, they reply with a smile, as if they were telling a story which
4 “Diese Vermutung wird durch mehrere Umstände bestätigt . . . man sieht schwarze Höhlun-
gen 150 bis 180 Fuß über dem heutigen Wasserstand erhaben. Ihre Existenz lehrt . . . , daß die
Ströme, deren Größe jetzt unsere Bewunderung erregt, nur schwache Überreste von der unge-
heuren Wassermenge der Vorzeit sind.”
“Selbst den rohen Eingeborenen . . . sind diese einfachen Bemerkungen nicht entgangen. Überall
machten uns die Indianer auf die Spuren des alten Wasserstandes aufmerksam.”
11.2 Cryptogams 385
Fig. 11.4 Shrubs (A, B) with

Clusia criuva (B) on the insel-
berg Pedra Grande at Atibaia,
S.P., Brazil and deciduous
forest (C) with the palm Sya-
grus orinocensis on top of
the inselberg at Galipero near
Puerto Ayacucho, Venezuela
386 11 Inselbergs
Fig. 11.5 Wetter sections and pools on the inselberg Galipero near Puerto Ayacucho, Venezuela,
where seasonally water is flowing and small ponds with water are formed in hollows
Fig. 11.6 Temperature of the air above the inselberg at Galipero (◦) and the rock surface (•) during
two consecutive days (15 – 16 March 1991) (Pto. Ayacucho, Venezuela)
only the white man may not know: that in the days of the extended waters their forefathers
were boating in canoes in such a height. This is a geological dream for the solution of the
problem of a long vanished civilization.”5
5 “Noch mehr: zwischen Encaramada und Caycara an den Ufern des Orinoco befinden sich häufig
hieroglyphische Figuren in bedeutender Höhe auf Felsenwänden, die jetzt nur mittels außeror-
dentlich hoher Gerüste zugänglich sein würden. Fragt man die Eingebornen, wie diese Figuren
haben eingehauen werden können, dann antworten sie lächelnd, als erzählten sie eine Sache, die
nur ein Weißer nicht wissen könne: ‘daß in den Tagen der großen Wasser ihre Väter auf Canots in
solcher Höhe gefahren seien’. Dies ist ein geologischer Traum, der zur Lösung des Problems von
einer längst vergangenen Civilisation dient.”
11.2 Cryptogams 387
Fig. 11.7 Black surface of the rock of the inselberg at Galipero. A small section was removed to
show that only the very surface is coloured
Furthermore, regarding the extensive distribution of the petroglyphs, H UM -

BOLDT alludes to his correspondence with Sir ROBERT S CHOMBURGK:
“I may be permitted to include a remark, which I take from a letter of the distinguished
traveller Sir Robert Schomburgk: ‘The hieroglyphic pictures have a much wider distribution
than you might have assumed’ . . . the symbolic figures, which ROBERT S CHOMBURGK
found engraved in the river valley of the Essequibo at the rapids of Waraputa according to
his observation resemble the truly Caribbean ones on one of the small Virgin Islands (St.
John)6 , however, notwithstanding the wide expansion of the invasions of Caribbean tribes
and the ancient power of this beautiful human race, I cannot believe that this whole immense
belt of carved rocks which cuts across a large part of South America from west to east is
the work of the Caribs. They are rather traces of an ancient civilization, which possibly
belongs to an epoch, when the tribes which we distinguish nowadays were still unknown
by name and relationship. Even the reverence, which everywhere is deferred to these rough
sculptures of the forefathers, proves that the present Indians have no idea of the creation of
such works.”7
6 See Fig. 11.8B,C.
7 “Es sei mir erlaubt, hier noch eine Bemerkung einzuschalten, welche ich einem Briefe des aus-
gezeichneten Reisenden Sir Robert Schomburgk an mich entnehme: ‘Die hieroglyphischen Figuren
haben eine viel größere Ausbreitung, als Sie vielleicht vermutet haben’. . . . Die symbolischen Zei-
chen, welche Robert Schomburgk in dem Flußtal des Essequibo bei den Stromschnellen . . . von
Waraputa eingegraben fand, gleichen zwar nach seiner Bemerkung den ächt caraibischen auf einer
der kleinen Jungferninseln (St. John); aber ungeachtet der weiten Ausdehnung, welche die Einfälle
der Caraiben-Stämme erlangten, und der alten Macht dieses schönen Menschenschlages, kann ich
doch nicht glauben, daß dieser ganze ungeheure Gürtel von eingehauenen Felsen, der einen großen
Theil Südamerikas von Westen nach Osten durchschneidet, das Werk der Caraiben sein sollte. Es
sind vielmehr Spuren einer alten Civilisation, die vielleicht einer Epoche angehört, wo die Racen,
die wir heut zu Tage unterscheiden, nach Namen und Verwandtschaft noch unbekannt[cont. p.389]
388 11 Inselbergs
11.2 Cryptogams 389
Fig. 11.8A–C Petroglyphs of an inselberg at Pintado near Puerto Ayacucho, Venezuela (A) and
on granite rocks near a small pond in the secondary forest covering the US Virgin Island St. John,
Lesser Antilles (B,C)
However, there are also already some slight reservations in H UMBOLDT’s writ-
ings regarding the inorganic nature of the black sheets on the rocks. He stressed that
it appears to be manganese oxide and carbon:
“I say, it appears; because the phenomenon has not been investigated diligently enough. At
the Orinoco these leadlike coloured rocks if wetted emit harmful emanations. One believes
their proximity to be a cause of fevers.”8
A biological cause is suggested by the observation that the black coloration is

associated with only the organically rich white-water rivers and not the more sterile
black-water rivers:
“It is also noteworthy that the rivers with black water, aguas negras, the coffee-brown or
wine-yellow waters, in South America do not stain the granite rocks black.”9
Indeed, an elemental analysis of the black cover of the inselbergs along the
Orinoco using X-ray fluorescence spectroscopy shows that there are traces of man-
ganese only (Table 11.1). There is only one exception, and these are the rocks di-
rectly in the riverbed of the Orinoco (Table 11.1), where the analysis actually in-
dicates the dominant presence of manganese oxide. The high levels of Al and Si
in all cases, of course, are due to the bed rock. The dominance of elements like S,
K and Ca is consistent with the occurrence of life in these crusts. Indeed, lichens
and small mosses are often readily discerned on the rock surfaces (Sect. 11.2.2).
However, microscopic inspection shows that even the smooth black covers of these
rocks result from living organisms. They are mainly composed of epilithic and en-
dolithic cyanobacteria, predominantly of the genera Gloeocapsa, Stigonema and
Scytonema (Figs. 11.9 and 11.10). Similar coverings of cyanobacteria are found on
rocks throughout the tropics and the diversity of species is quite large (Büdel 1999).
The phenomenon has also been described for sandstone rocks, e.g. near Cumana
in eastern Venezuela (Golubic 1967), and examples are given in Fig. 11.11 of the
granite rocks of Sierra Maigualida and sandstone rocks of Tepuis (Sierrania Parú) in
the Guayana highlands of Venezuela. They resemble the “Tintenstrich” (“ink strip”)
formation frequently found in the European Alps, particularly on calcareous rocks
(Jaag 1945).
In the tropics almost every free surface on rocks is covered by cyanobacterial
mats and crusts. Based on the large extension of supporting rocks these cyanobac-
teria overall must constitute an enormous biomass in the tropics.
waren. Selbst die Ehrfurcht, welche man überall gegen diese rohen Sculpturen der Altvorderen
hegt, beweist, daß die heutigen Indianer keinen Begriff von der Ausführung solcher Werke haben.”
8 “Ich sage: sie scheint; denn das Phänomen ist noch nicht fleißig genug untersucht . . . Am Orinoco
geben diese bleifarbigen Steine, befeuchtet, schädliche Ausdünstungen. Man hält ihre Nähe für
eine fiebererregende Ursache.”
9 “Auffallend ist es auch, daß die Flüsse mit schwarzen Wassern, aguas negras, die caffeebraunen
oder weingelben, in Südamerika die Granitfelsen nicht schwarz färben.”

390 11 Inselbergs
Fig. 11.9A, B Closeup photographs of rocks covered with cyanobacteria on the inselberg at
Galipero (A) and on the granite in the Sierra Maigualida, Guayana Highlands (05◦ 30 N, 65◦ 15 W,
2,040 m a.s.l.) Venezuela (B)
Fig. 11.10A–C Cyanobacteria composing the black crusts of the inselbergs along the Orinoco,
Venezuela. A Stigonema ocellatum. B Scytonema crassum. C Gloeocapsa sanguinea. (Courtesy
B. Büdel, Kaiserslautern; see Büdel et al. 1994)
11.2 Cryptogams 391
392 11 Inselbergs
Fig. 11.11A–C Black granite rocks of the Sierra Maigualida (at 05◦ 30 N, 65◦ 15 W, 2,040 m
a.s.l.; A and B, in B shortly after a rain storm) and sandstone rock of the Serrania Parú (at 04◦ 25 N,
64◦ 32 W, 1,200 m a.s.l.; C), Guayana Highlands Venezuela, with cyanobacteria
11.2 Cryptogams 393
Table 11.1 Elemental analysis (elements of order-number 11 – 80) by X-ray fluorescence spec-
troscopy of samples from sandstone rocks of the Serriania Parú (04◦ 25 N, 65◦ 32 W, 1,200 m
a.s.l.), inselbergs along the Orinoco and the riverbed of the Orinoco near Puerto Ayacucho. (See
Büdel et al. 1994)
Element % Serrania Parúa Inselbergs Orinocob Riverbed Orinococ

Al 37.9 ± 9.5 16.8 ± 1.3 7.8 ± 0.1
Si 46.0 ± 10.0 55.1 ± 4.2 18.9 ± 3.1
S 3.2 ± 2.4 1.1 ± 0.4 0
K 4.3 ± 0.7 7.5 ± 1.2 1.9 ± 0.3
Ca 1.1 ± 0.7 2.7 ± 1.0 2.8 ± 0.2
Mn 0.2 ± 0.1 0.5 ± 0.1 49.7 ± 4.0
Fe 0.5 ± 0.3 6.3 ± 1.3 15.5 ± 0.9
Others 6.8 ± 0.5 10.0 ± 2.4 3.4 ± 0.3
a Serrania Parú 3 samples, 1 – 2 analyses each.
b Inselbergs Orinoco 8 samples, 4 – 13 analyses each.
c Riverbed Orinoco 1 sample, 10 analyses.
Values are x ± SE for the averages of the individual analyses of the three and eight samples in
a and b respectively, and for the ten analyses of the sample in c . One sample of b had 82% Os.
11.2.1.2 Success on Bare Substratum
The role of cyanobacteria in cryptogamic soil crusts of deserts and other dry habi-
tats has recently received much attention (Lange et al. 1992; Evans and Ehlehringer
1993; Jeffries et al. 1993a,b; Belnap and Lange 2001). Very little is known, however,
about the ecophysiology of cyanobacteria on rocks in the tropics, and in view of the
enormous distribution of terrestrial cyanobacteria, which cover almost any surface
not occupied by other vegetation, this is quite astonishing. The major reasons for
the success of cyanobacteria on the bare substratum of rocks appear to be:
• a potential to adapt to high light intensities,
• the ability to fix atmospheric dinitrogen, and
• desiccation tolerance.
High light intensities bring about considerable heating up of the rock surfaces with
cyanobacterial crusts. In response various heat shock proteins (hsp) are synthesised
in the cyanobacteria (Adhikary 2003). Survival of high light intensities in the ex-
tremely sun exposed habitat of the inselbergs is also sustained by the production
of effective sunblocking pigments, such as the indol-alkaloid scytonemin (Garcia-
Pichel and Castenholz 1991), which occurs in cyanobacteria in light-exposed en-
vironments (Büdel et al. 1997b; Büdel 1999). In exposed epilithic cyanobacteria of
Venezuela and French Guiana, scytonemin, with its absorption maximum at 380 nm,
is found in such high concentrations that irradiance up to 500 nm is reduced in-
side the cells. Furthermore, intracellular carotenoids such as zeaxanthin (Demmig-
Adams et al. 1990) and canthaxanthin (Albrecht et al. 2001; Lakatos et al. 2001)
may prevent photodamage. The formation of these pigments is slow and takes place
over days adapting the cyanobacterial cells to high irradiance.
394 11 Inselbergs
Thus, the responses of photosynthetis of cyanobacteria to high light intensi-

ties depend greatly on the irradiance experienced during growth. Cells grown at
50 µ mol photonsm−2 s−1 (at λ = 400 − 700 nm) or below are already photoin-
hibited at 250 µ mol m−2 s−1 and strongly affected at still higher light intensities
(Samuelsson et al. 1985; Lüttge et al. 1995). However, inselberg rocks usually re-
ceive full sunlight unless clouds and rain quench exposure. Figure 11.12 shows
that in an inselberg sample from Ivory Coast, a transfer from 480 to 1,200 µ mol
photons m−2 s−1 did not affect fluorescence yield and photochemical fluorescence
quenching, indicating unimpaired photochemical and carbon-assimilatory activity,
Fig. 11.12 Optimal quantum yield of photosystem II (Fv /Fm ), effective quantum yield (F/Fm )
and photochemical quenching (qp ) of cyanobacterial crusts of an inselberg near Seguéla, Ivory
Coast (07◦ 42 N, 06◦ 43 W) in drying (arrow drying above the graphs) and rewetting (arrow
H2 O above the graphs) cycles and during transfers between lower and higher light intensities
(arrows with numbers above the graphs giving light intensities at λ = 400 − 700 nm in µ mol pho-
tons m−2 s−1 ). Dark and white bars above the graphs indicate dark and light periods respectively
(Lüttge et al. 1995)
11.2 Cryptogams 395
while a transfer from 240 to 1,280 µ mol photons m−2 s−1 resulted in a slight and
rapidly reversible inhibition. (The drying and wetting cycles shown in Fig. 11.12
will be discussed in Sect. 11.4.) As in higher plants (Schreiber and Bilger 1993),
potential quantum yield of photosystem II after dark adaptation (Fv /Fm ), effec-
tive quantum yield (F/Fm ) and photochemical fluorescence quenching (qp ) (see
Sect. 4.1.7, Box 4.6) decrease with increasing light intensity in cyanobacterial crusts
(Fig. 11.13). (In higher plants Fv /Fm is close to 0.8 in non-photoinhibited samples
(Sect. 4.1.7) and we note that in comparison in the cyanobacteria in Figs. 11.12 and
11.13 Fv /Fm is rather low even in the early morning. This is an intrinsic property
Fig. 11.13 Potential quantum

yield of photosystem II after
dark adaptation (Fv /Fm ),
effective quantum yield
(F/Fm ), photochemical
quenching (qp ) and relative
photosynthetic electron trans-
port rates (F/Fm × PPFD)
related to photosynthetic pho-
ton fluence density (PPFD) as
determined in the laboratory
with tropical rock samples.
A and closed symbols in C
sample of a granitic rock
from the eastern slope of
the coastal mountain range
in the SE of Madagascar
(24◦ 49 S, 46◦ 57 E, 100 m
a.s.l.) with Stigonema minu-
tum and Gloeocapsa magma
B and open symbols in C
sample of rock outcrops
in Menagesha State Forest
near Addis Ababa, Ehtiopia
(09◦ 04 N, 38◦ 22 E, 2,800 m
a.s.l.) with Gloeocapsa san-
guinea (Lüttge et al. 1995)
396 11 Inselbergs
of the prokaryotic cells due to the particular structure of their photosynthetic mem-
branes (Lüttge 1997) and does not imply that these cyanobacteria were under very
severe chronic photoinhibition.) The multiplication of (F/Fm ) by photosynthetic
photon fluence density (PPFD) gives relative photosynthetic electron transport rates,
which saturated at the high intensities of 1,000 µ mol photons m−2 s−1 or above in
the inselberg sample from Madagascar and the rock outcrop sample from Ethiopia
measured in the experiments of Fig. 11.13. Thus, although even cyanobacteria crusts
grown under full sun exposure may be subject to partial photoinhibition, it is quite
clear, that cyanobacteria can adapt very well to very high irradiance (Lüttge et al.
1995).
Cyanobacterial communities show conspicuous zonations and niche occupa-
tion across furrows running down the rocks of inselbergs (Fig. 11.14). On an in-
selberg at Les Nouragues in French Guiana a community in the centre of such
furrows was found to be dominated by compact growth forms, like the unicellu-
lar, colony-building Gloeocapsa sanguinea (Fig. 11.10C) and the short branching
species Stigonema mamillosum. The growth form appears important because during
and after rainfall the cyanobacterial mats are covered by water up to a few centime-
tres in depth, occasionally with strong current. The lateral slopes of the furrows are
covered by a different community dominated by a thick layer of Stigonema ocel-
latum (Fig. 11.10A). A third community dominated by Scytonema myochrous was
found at some distance (normally > 50 cm) from the furrows in the more or less
horizontal rock areas covering more than 80% of the whole rock surface (Rascher
Fig. 11.14 Drainage furrow

running down from a small
vegetation island on the insel-
berg Pedra Grande at Atibaia,
SP, Brazil
11.2 Cryptogams 397
et al. 2003). After rainfall before drying again the mats in the centre of the furrows
and at the lateral slopes are covered for longer times by water and films of water
than the mats on the horizontal rocks, and therefore, their photosynthesis is more
pronouncedly limited by diffusion of CO2 and HCO− 3 in the liquid phase. To coun-
terbalance the liquid diffusion limited carbon supply cyanobacteria have evolved
an inorganic carbon concentrating mechanism or a CO2 /HCO− 3 pump, where trans-
port mechanisms and carbonic anhydrase catalyzing the CO2 /HCO− 3 equilibrium
are involved (Skleryk et al. 1997; Sültemeyer et al. 1997). Diffusion limitation is
reflected in the stable isotope ratios δ 13C of the cyanobacteria. They are performing
C3 -photosynthesis and the enzyme of primary CO2 -fixation ribulose-bis-phosphate
carboxylase/oxygenase (RuBISCO) has a 13 C-discrimination of +27‰. The δ 13 C-
values of cyanobacteria of inselbergs, however, are much less negative than −27‰
Fig. 11.15A, B Carbon isotope ratios (δ 13 C, ‰) of cyanobacteria in a transect across a seepage

furrow of the inselberg at Galipero (A) and distribution of δ 13 C-values among 17 samples of
cyanobacterial crusts and mats from the area around Caicara – Puerto Ayacucho along the Orinoco
river, Venezuela (B). (Lüttge 1997; Ziegler and Lüttge 1998)
398 11 Inselbergs
which would be obtained if RuBISCO were maily determining 13 C-discrimination

during photosynthesis (Ziegler and Lüttge 1998; Fig. 11.15). This is due to the
lower 13 C-discrimination of dissolution of CO2 ( − 0.9‰) and diffusion of CO2
and HCO− 3 in water (0.0‰) determining CO2 delivery to RuBISCO, and as ex-
pected then, there is also a gradient from less negative to more negative δ 13 C val-
ues from the centre via the lateral slopes to the open rock surfaces of the inselbergs
(Fig. 11.15). The mats in the centre of the furrows also have a somewhat lower effec-
tive quantum use efficiency (F/Fm ’) and apparent electrom transport rate (ETR)
of photosynthesis than the mats at the slope and outside the furrows (Fig. 11.16).
The second important trait of cyanobacterial rock crusts as highlighted above
is N2 -fixation. By the possession of heterocytes they are characterized as N2 -
fixing organisms, because heterocytes are special cells in the coenobial colonies
0.5 A
0.4
∆ F / Fm´
0.3
0.2
400
B
300
ETR ( rel. )
200
100
0
0 500 1000 1500 2000 2500
–2 –1
Irradiance ( μmol s )
Fig. 11.16A, B Apparent effective quantum yield (F/Fm ) (A) and apparent rates of electron
transport (ETR) (B) recorded over several days while the cyanobacteria communities across fur-
rows of the inselberg at Les Nouragues, French Guiana, were wetted after rainfall. Lines indicate
linear regressions (A) or fitted exponential growth to maximum (B). Filled circles and continuous
lines, cyanobacterial community in the centre of the furrows, open squares and broken lines with
long dashes, cyanobacterial community at the sides of the furrows, open triangles and broken lines
with short dashes, cyanobacterial community on the vertical rocks beside the furrows. (Rascher et
al. 2003)
11.2 Cryptogams 399
Fig. 11.17 Distribution of in-

selbergs at the upper Orinoco
river. Black areas, largely soil
free granitic inselbergs with
a slope of at least 15% (from
Gröger 1995; see Lüttge
1997)
or filaments bearing the enzymatic machiney for the reduction of atmospheric

N2 (Sects. 10.2.3.2.1 and 10.2.3.2.2). Evidently N2 -fixation of the cyanobacteria
is important primarily for their own nutrition and growth. However, by leachates
cyanobacterial crusts and mats on the rock habitat probably provide an essential
starting point for succession and possibly also contribute considerably to the N-input
into the ecosystems of the inselbergs themselves and via run-off into the surround-
ing savannas or forests. In fact the nitrogen content of the soil 10 m from the base
400 11 Inselbergs
of inselbergs in savannas was three times (1.40 g N/kg soil) that measured more
than 30 m from the base (0.45 g N/kg soil) (E. Medina in Büdel et al. 1997a,b).
On the basis of a map of the inselbergs along the upper Orinoco river (Fig. 11.17)
presented by Gröger (1995) one can estimate the total savanna area with inselbergs
as ca. 3,425 km2 (342.5 ×103 ha) and the area of the inselbergs as ca. 480 km2
(48 ×103 ha) or 14%. The N2 -fixation of cyanobacterial mats in tropical savannas
was estimated as 60 g N/ha per day (Medina 1993) so that with assuming 300 good
days per year as Gröger (1995) gives the number of dry months in the area at only
one to three and a total coverage of the inselbergs with cyanobacteria, the fixation
of the inselbergs would be ca. 18 kg N/ha per year. This appears small in compar-
ison to N-fertilisation in high-technology agriculture (several 10s to 200 kg N/ha;
see Lüttge 1997) but is substantial for the nutrient poor savanna ecosystem.
The third of the traits highlighted above which allow cyanobacteria survival on
bare rocks, i.e. desiccation tolerance, appears to be the most important one and is
discussed in Sect. 11.4.
11.2.2 Lichens and Mosses
Lichens (Büdel et al. 2000, Fig. 11.18) and bryophytes (Frahm 2000, Fig. 11.19)
are important elements of the inselberg vegetation. Their photosynthesis like that of
the cyanobacterial crusts (Sect. 11.2.1.2) is also much determined by the frequent
Fig. 11.18 Lichens (Peltula tortuosa) on the rock surface of the granite inselberg at Galipero,
Venezuela
11.3 Vascular Plants 401
Fig. 11.19 Belts of mosses around a vegetation island on the inselberg Pedra Grande at Atibaia,
Brazil. Inner zone Campylopus savannarum, outer zone Racocarpus fontinaloides
wetting and drying cycles on the inselberg rocks. Diffusion limitation of inorganic
carbon supply for photosynthesis is modulated by the growth form of the lichens
and mosses determining the extent and duration of the build up of water films in the
thalli after precipitation (Tuba et al. 1996a: Büdel et al. 2000; Lüttge et al. 2007),
and this may effect niche occupation of different inselberg mosses (Lüttge et al.
2007). Different moss species may perform characteristic belts around small vege-
tation islands on inselbergs (Fig. 11.19) but apparently do not differ in their basic
photosynthetic capacity (Lüttge et al. 2007). A major adaptive trait of these cryp-
togams for life on inselberg rocks is their desiccation tolerance (Sect. 11.4).
11.3 Vascular Plants
11.3.1 Diversity and Life Forms
The floristic diversity of cryptogamic and phanerogamic vascular plants of insel-

bergs is large (Sect. 11.1). The diversity of life forms occurring on rock outcrops
is also quite noticeable. It varies at different locations (Table 11.2). The dominance
of different life forms is governed by the size of the inselbergs (Table 11.3) and the
richness of vascular species increases with inselberg size (Fig. 11.20).
11.3.2 Physiological Ecology
Although at least some inselbergs are readily accessible even with heavy equip-
ment (Fig. 11.21) only very little ecophysiological work has been performed on site.
From the frequency and intense development of species of Agavaceae, Bromeli-
402 11 Inselbergs
140
120
Number of species
100
80
60
40
20
700000
800000
900000
400000
600000
300000
500000
25900
46400
3412
5852
1900
655
370
Size of inselbergs in square metres
Fig. 11.20 Correlation between species richness and inselberg size in Ivory Coast (From Poremb-
ski et al. 2000b with kind permission of S. Porembski)
Table 11.2 Relative abundance (%, rounded values from the literature) of various life forms of
vascular plants on inselbergs in various regions of the world. For the description of R AUNKIAER ’s
life forms see Sect. 3.3.4. For the other life forms cf. also Sects. 6.4 and 12.3. (1) Safford and Mar-
tinelli (2000), (2) Courtesy S.T. Meirelles, A.C. da Silva and E.A. de Mattos, (3) Gröger (2000),
(4) Seine and Becker (2000)
Life form Pão de Serra State of Venezuelan Ivory Zimbabwe
Açucar do Mar Rio de Guayana Coast
Rio de Rio de Janeiro
Janeiro Janeiro
(1) (1) (2) (3) (1) (4)
Phanerophytes 39 34 29 39 18 23
Chamaephytes 42 38 1 11 22 15
Hemicryptophytes 9 17 32 13 7 13
Geophytes 5 10 5 5 8 8
Terophytes 5 1 5 16 45 35
Lianas 1 10 3
Succulents 4
Epiphytes 3 4
Hemiepiphytes 1
Tank-formers 5
Epilithic plants 6 2
Atmospheric plants 6
Hygroscopic plants 12
aceae, Cactaceae, Crassulaceae, Euphorbiaceae and Orchidaceae on inselbergs in

the palaeotropics and the neotropics (Kluge and Brulfert 2000) we may conclude
that crassulacean acid metabolism (CAM) is an important mode of photosynthesis
on these rock habitats. On the inselbergs along the Orinoco both CAM-bromeliads
(e.g. Ananas ananassoides, Bromelia goeldiana, and the epiphytic Tillandsia flexu-
osa) and C3 -bromeliads (Pitcairnia armata, P. bulbosa, P. pruinosa) occur together.
C3 /CAM-intermediate Clusiaceae (various species of Clusia and Oedematopus
11.3 Vascular Plants 403
Table 11.3 Life-form specific size classification of inselbergs in relation to the number of thero-
phytes in the Ivory Coast (Porembski et al. 2000b)
Inselberg type Therophytes (%) Minimal size of inselberg (m2 )

Lichen inselberg – ca. 1
Therophyte inselberg > 80 ca. 50
Perennial herb inselberg ca. 60 ca. 10,000
Phanerophyte inselberg ca. 45 ca. 50,000
Fig. 11.21 Negotiation of the inselberg at Galipero, Venezuela, with heavy equipment
obovatus, see Sect. 6.6.2.3) are found on South-American inselbergs. Bromeliads,

like Ananas ananassoides and Bromelia goeldiana, develop the contrasting pheno-
types of yellow-reddish fully exposed plants and dark-green shaded plants under
the canopy of shrubbery and small forests respectively, as already described for
Bromelia humilis (Sect. 4.1.2). On an inselberg in Madagascar, the distribution of
three Kalanchoë species with different expression of CAM was found to be related
to micro-habitat characteristics: K. campanulata with a very weak CAM capacity
in the shade of deep humid gaps performed largely C3 photosynthesis; K. mini-
ata in open bush formations showed pronouncded CAM performance; K. synsepala
spreading by stolons on the bare rocks showed the strongest expression of CAM
(Kluge and Brulfert 2000). On the other hand, C3 -photosynthesis is also dominant
among vascular plants of inselbergs.
Desiccation tolerance as for cryptogams is a most important adaptation of vascu-
lar plants on inselbergs (Sect. 11.4).
404 11 Inselbergs
11.4 Desiccation Tolerance
Desiccation tolerance appears to be the most outstanding particular ecophysiolog-

ical adaptation among the plants of tropical inselbergs. Drought tolerance allows
plants to overcome shorter periods of stress, such as the duration of dry seasons. In
contrast, desiccation tolerant plants can survive equilibration with ambient air hu-
midity below 50% and down to 0%, and withstand the loss of more than 90% of
their normal water content for many years (Gaff 1977, 1987). We call desiccation
tolerant plants poikilohydrous in contrast to the homoiohydrous non-desiccation
tolerant plants. Because of their recovery from dryness they are also called resur-
rection plants. Poikilohydrous plants may degrade or retain their chlorophyll dur-
ing drying and are then called poikilo-chlorophyllous and homoio-chlorophyllous,
respectively. Poikilohydry is well known of lower cryptogams, such as terrestrial
cyanobacteria, lichens and bryophytes. However, it is also found among vascular
plants and these are especially species adapted to the life on inselbergs in the trop-
ics.
11.4.1 Cyanobacteria
As it is illustrated by the loss and reappearance of chlorophyll-fluorescence sig-

nals in drying and rewetting cycles performed in the laboratory on a rock sample
with cyanobacterial crusts from an inselberg in Ivory Coast (with the cyanobac-
teria Stigonema mamillosum, Scytonema lyngbioides and Gloeocapsa sanguinea,
and traces of Stigonema ocellatum) desiccation and recovery can occur very rapidly
(Fig. 11.12). Triggered by sensing the loss of a small amount of water a rapid de-
activation of photosynthetic activities is essential to avoid oxidative stress during
desiccation (Hirai et al. 2004). Recovery from desiccation may occur within a few
minutes up to several hours after rewetting and recovery of photosystem I activ-
ity is faster than that of photosystem II (Jones 1977; Coxson and Kershaw 1983;
Lüttge et al. 1995; Satoh et al. 2002; Rascher et al. 2003). It depends on the du-
ration of the dormant state of desiccation, and frequent drying and wetting cycles
maintain stability (Scherer and Zhong 1991). The sequence of events during recov-
ery is firstly, reappearance of respiration, followed by photosynthesis and finally
N2 -fixation (Scherer et al. 1984). Rapid recovery is associated with rapid repair
mechanisms, e.g. of the D1-protein (see Sect. 4.1.5) of photosystem II (Harel et al.
2004).
11.4.2 Lichens and Bryophytes
Wetting and drying cycles have been studied in much detail for lichens in arid
habitats. Photosynthesis of lichens is related in a complex and delicate way to the
11.4 Desiccation Tolerance 405
transient water conditions of the thallus (Fig. 11.22). At very low water content,
i.e. below 20% of dry weight, the lichens are metabolically dormant showing nei-
ther photosynthesis nor respiration. Between 20% and 50% water content, photo-
synthetic net CO2 -uptake increases sharply and then reaches a plateau as optimal
water content is attained. However, when the water content increases further and
thalli are fully saturated with water, CO2 -assimilation is depressed. This is due to
increased limitation of photosynthesis by CO2 diffusion when the capillary system
of the lichen thallus is infiltrated. Thus, upon drying the assimilation rates may first
increase again and then decline as the thalli desiccate (Lange 1988).
Both protection and repair mechanisms are essential during desiccation as well
as rehydration. Reactive oxygen species can be formed during desiccation and rehy-
dration, especially when the use of exitation energy of the photosynthetic apparatus
by photochemical work is reduced. Protection mechanisms such as by the redox
state of glutathione are effective in both lichens and mosses (Kranner 2002; Mayaba
et al. 2002). However, we must assume that in the desiccated state water structures
required for enzymatic reactions including photochemical work of CO2 -reduction
and photorespiration as well as epoxidases and de-epoxidases of the xanthophyll cy-
cle (Sect. 4.1.4, Box 4.4) are not intact, and therefore, these excitation-energy using
processes are negligible in dry homoiochlorophyllous material.
Thus, by the excitation of chlorophyll in homoiochlorophyllous plants in the
dry state high irradiance causes photo-oxidative stress. An interesting feature of
Fig. 11.22 Net CO2 uptake in the light (◦) and net CO2 -release in the dark (•) of the lichen Ra-
malina maciformis at varied thallus water content related to dry weight. (Lange 1988, with kind
permission of the author and Journal of Ecology)
406 11 Inselbergs
homoiochlorophyllous desiccation tolerant lichens and bryophytes is the loss of

ground fluorescence of chlorophyll a, F (see Sect. 4.1.7) in the desiccated state
in great contrast to desiccation tolerant vascular plants. While F decreases during
desiccation and increases again during rewetting in mosses and lichens, the oppo-
site dynamics are found in vascular plants (Lange et al. 1989; Calatayud et al. 1997;
Eickmeier et al. 1993; Heber et al. 2000, 2001).
High irradiance causes photodamage in dried vascular plant leaves but not in
dried mosses (Heber et al. 2000). This shows that the reduction of F to very low lev-
els in the latter can be considered as a protection mechanism against photodamage
under full sunlight on exposed rock surfaces of inselbergs in the dehydrated stage.
Effective mechanisms of chlorophyll fluorescence quenching mediate the conver-
sion of excitation energy into heat when metabolism can no longer control it. Two
Fig. 11.23A–D Development of basic fluorescence, F, of light adapted dry cushions of Campylo-
pus savannarum (A), Racocarpus fontinaloides (B) and Ptychomitrium vaginatum upon rewater-
ing (C,D) (Lüttge et al. 2007)
processes are involved in this, namely zeaxanthin-dependent energy dissipation in

the antenna of photosystem II (see Sect. 4.1.4) and desiccation induced thermal en-
ergy dissipation in the reaction centres (RCs) of photosystem II (Deltoro et al. 1998;
Heber et al. 2000, 2006a,b; Bukhov et al. 2001). The latter is essential in desiccated
lichens and mosses because zeaxanthin does not protect RCs directly from photo-
oxidation and in the absence of photosynthetic electron transport dynamics in the
dry state dissipation must be extremely fast given the half-lives of first and second
singlet excited states of chlorophyll of 10−11 to 10−9 and 10−15 to 10−13 , respec-
tively (Box 4.3).
Homoiochlorophylly is an important prerequisite of rapid recovery upon rewa-
tering (Tuba et al. 1996b; Csintalan et al. 1999). Kinetics of recovery can vary con-
siderably among mosses (Csintalan et al. 1999; Proctor 2000; Proctor and Smirnoff
2000). In the example of Fig. 11.23 the recovery of F has a very fast initial phase
with a drastic increase within less than 1 min followed by a more gradual increase
lasting much longer. Protein synthesis is not required during the fast initial phase
(Proctor and Smirnoff 2000), but the gradual increase following the first rapid phase
suggests that in addition to an immediate reactivation slower repair mechanisms in-
volving protein synthesis, such as that of the D1 protein may be involved (Proctor
and Smirnoff 2000).
11.4.3 Vascular Plants
11.4.3.1 Evolution and Diversity of Poikilohydrous Vascular Plants
Inselbergs are the diversity centres of poikilohydrous vascular plants (Meirelles

et al 1997; Biedinger et al. 2000; Porembski and Barthlott 2000a,b; Proctor and Tuba
2002). Desiccation tolerant vascular plants, especially phanerogamic species, were
initially known especially from inselbergs of Africa (Gaff 1977) and then also from
South America, Australia and India (Gaff 1987; Meirelles et al. 1997). Although
dormant dried developmental stages, such as spores, pollen grains and seeds, are
known from the life cycles of all vascular plants, desiccation tolerance of the vege-
tative plant bodies is rare and there are only 200 (Kappen and Valladares 1999) to
350 (Proctor and Tuba 2002) desiccation tolerant species among vascular taxa. It has
been proposed to consider desiccation tolerant cryptogams such as cyanobacteria,
algae, lichens and bryophytes as constitutively poikilohydrous as they are exohy-
dric and cannot actively control their water relations as compared to the vascular
plants which are constitutively homoiohydrous with their complex regulation of
water relations. In the former desiccation tolerance is a primary trait during evolu-
tion, in the latter it is a special late development in highly advanced taxa (Kappen
and Valladares 1999; Proctor and Tuba 2002). Indeed, although among cryptogamic
vascular plants desiccation tolerance occurs in a high proportion of taxa (e.g. in the
class Lycopodiopsida, order Selaginellales; in the class Pteridopsida orders Schiza-
les and Pteridales), tolerance appears to be less pronounced than observed in an-
giosperms. Among the phanerogams desiccation tolerant species mostly are mono-
408 11 Inselbergs
cotyledons and there are fewer desiccation tolerant dicotyledons (Gaff 1977), with
the latter apparently absent from South America (Gaff 1987). Evolution of poikilo-
hydry in the angiosperms was polyphyletic and occurred at least eight times (Proctor
and Tuba 2002). The poikilohydrous vascular plants of the more basic taxa and of
the dicotyledons generally are homoiochlorophyllous and retain their photosynthetic
apparatus in a recoverable form. Poikilochlorophylly is only found among the des-
iccation tolerant monocotyledons (Kappen and Valladares 1999; Proctor and Tuba
2002) and appears as an advanced trait in evolution where plants lose all of their
chlorophyll and 70 – 80% of their carotenoids (xanthophylls and β-carotene) and
the internal structure of their chloroplasts (thylakoids) only retaining the outer en-
velope. While homoiochlorophylly has the advantage of rapid resumption of photo-
synthetic metabolism during rehydration, poikilochlorophylly provides much better
protection from oxidative stress in the dehydrated stage which appears to outweigh
the disadvantage of much slower recovery.
11.4.3.2 Dynamics of the Performance in Dehydration/Rehydration Cycles

The dynamics of dehydration/rehydration cycles are best documented considering
capacity of photosynthesis.
In homoiochlorophyllous species recovery upon rehydration is generally much
faster than in poikilochlrophyllous plants. A unique example of a homoiochloro-
phyllous angiosperm is the aquatic species Chamaegigas intrepidus (Scrophulari-
aceae) living in rock pools of granite outcrops in Namibia (Hartung et al. 1998).
Effects of dehydration on photosynthetic quantum yield are maximal between 10
to 15 h and photosynthetic quantum yield rises again rapidly to high values within
a few hours of rehydration and reaching maximum values after 10 h (Fig. 11.24).
For poikilochlorophyllous species desiccation and recovery has been studied in
much detail in the Velloziaceae Xerophyta scabrida growing in the Uluguru Moun-
tains in Tanzania at 650 m a.s.l. (about 05◦ 30 S, 35◦ 30 E), where they form a semi-
desert like bush vegetation on cliffs.
Dehydration Rehydration
2.0
0.8
ABA ( nmol g–1 DW )
Quantum yield
1.5
0.6
0.4 1.0
Fig. 11.24 Quantum yield
(circles) during a dehydra-
0.2 0.5
tion/rehydration cycle of
a single plant of Chamaegigas
intrepidus and development
0 0
of abscisic acid levels during
0 20 40 60
dehydration (Hartung et al.
1998) Time ( h )
During desiccation net photosynthesis declined to the compensation point within

8 h, while respiration continued beyond the 24th hour at water potentials much lower
than − 3.2 MPa. An active respiration, which lasts until the end of the desiccation
period, is responsible for the metabolic degradation of chlorophylls and other com-
ponents of thylakoids (Tuba et al. 1996b).
Water uptake and rehydration after rewetting initially must occur predominantly
via the surface of the leaves, because functional roots are lost during desiccation.
As in other desiccation tolerant plants (Gaff 1977), water uptake subsequently oc-
curs via the roots when new adventitious roots are developed (Tuba et al. 1993a).
Times required for recovery vary considerably between species and up to sev-
eral days may be needed (Kappen and Valladares 1999). The events occurring in
X. scabrida upon rehydration are summarized in Fig. 11.25 after the work of Tuba
et al. (1993a,b, 1994). The sequence in time is turgor → maximum leaf water
content → respiration → resynthesis of carotenoids and → chlorophylls ac-
companied by thylakoid development → chlorophyll fluorescence yield → net
CO2 -fixation. Turgor reappears after 2 h, maximum leaf water content is reached
after 10 h. Among the biochemical activities respiration generally recovers early
Fig. 11.25 Recovery of the poikilochlorophyllous desiccation tolerant Velloziaceae Xerophyta

scabrida upon rehydration. The graphs in the lower three panels indicate the relative inreases of
the structures and functions described in the left column. Compiled after data in Tuba et al. (1993a,
b, 1994)
410 11 Inselbergs
(Kappen and Valladares 1999). Mitochondrial membranes are better preserved dur-
ing desiccation than the thylakoids of chloroplasts. In X. scabrida respiration is
increased above the normal rates after 12 h. This so-called rehydration respiration
must be related to repair mechanisms and ceases after 30 h, when normal rates are
attained again. In contrast to chromoplasts and gerontoplasts of senescing leaves,
which cannot regreen, the chloroplasts of desiccation tolerant plants, which lose
their entire photosynthetic apparatus, i.e. thylakoid membranes and pigments dur-
ing desiccation, are totally rebuilt after rehydration. Tuba et al. (1993b) have named
these plastids desiccoplasts. Some of the carotenoids, i.e. 22 – 28%, are preserved
during desiccation, and they may play an essential role in reorganization. Reaccu-
mulation of carotenoids and chlorophyll a + b starts after 12 h when thylakoid de-
velopment also begins to appear, as indicated by the thylakoid frequency, thylakoid
stacking and the ratio appressed/exposed membranes. We recall that appressed thy-
lakoid regions and the membranes stacked in the grana are the sites of photosys-
tem II (see Box 4.2). Chlorophyll fluorescence reappears at about the same time,
while it takes considerably longer, i.e. about 24 h for the onset of net CO2 -uptake.
Recovery of photosystem I usually is faster than that of photosystem II (Kappen
and Valladares 1999). After 72 h all functions have reached their normal levels
again.
11.4.3.3 Structure Function Relations
Whole leaves shrink, curl, roll and fold during desiccation and the epidermis of
leaves is much wrinkling (Hartung et al. 1998; Proctor and Tuba 2002; Vicré et al.
2004). This is a consequence of water loss, but it also contributes to a much re-
duced light absorption and irradiance stress in the dry state. In addition resurrection
plants may develop accumulation of anthocyanin a sun-blocking pigment (Farrant
et al. 2003). During the reduction of the volume of leaf cells and the associated
extensive folding of cell walls a tight connection remains established between the
plasma membrane and the cell wall, a phenomenon called cytorrhysis, which is
important for recovery during rehydration (Hartung et al. 1998; Vicré et al. 1999).
The cell wall is subject to considerable mechanical stress in this process. Resur-
rection plants develop cell walls with particular tensile strength. Arabinan polymers
and arabinogalactan proteins, xyloglucans, homogalacturonan, rhamnogalacturonan
and unesterified pectins are important macromolecules building up the special cell
wall structure with the high mobility and water absorbing capacity required in dehy-
dration/rehydration cycles and cross linking Ca2+ ions may also be involved (Vicré
et al. 1999, 2004; Moore et al. 2006).
Hydraulic architecture is especially important during rehydration. Conductive
elements of the xylem suffer embolism during desiccation. Lipids lining the water
conducting elements of the xylem may be involved in preventing complete loss of
water in the dry state (Schneider et al. 2000, 2003; Wagner et al. 2000; but see also
Tyree 2001). In the initial stages of moistening before xylem emboli are repaired
and especially in the poikilochlorophyllous Vellozias, which lose their absorptive
roots in the desiccated state (Sect. 11.4.3.2) external capillary water movement is
important. In the Vellociaceae species this can occur quite effectively in the ex-
ternal space of the pseudostems built up of the retained leaf bases of dead leaves
(Proctor and Tuba 2002; Fig. 11.26). In the xylem in addition to capillary water
movement root pressure is essential for overcoming embolis and for refilling the
conductive elements with water (Kappen and Valladares 1999; Proctor and Tuba
2002; Schneider et al. 2000, 2003; Wagner et al. 2000). This may be the reason why
desiccation tolerant angiosperms do not reach larger sizes than up to 3 – 4 m tall
(Gaff 1977; Kappen and Valladares 1999; Proctor and Tuba 2002), e.g. monocotyle-
donous pseudo shrubs such as Vellozia (Fig. 11.26). The homoiohydrous resurrec-
tion plant Craterostigma plantagineum (Scrophulariaceae) has desiccation tolerant
roots which die two weeks after rehydration (Norwood et al. 2003).
Fig. 11.26A–C Vellozia gigantea (A) and detailed views of its pseudo stem (B, C). Serra do Cipó,
MG, Brazil
412 11 Inselbergs
11.4.3.4 Cell Physiology

At the cell physiological level during desiccation and rehydration the three most
important functions are the dynamics (i) of compatible solutes for the protection of
cellular components and (ii) of lipids for the maintenance of membrane structures
and (iii) the control of reactive oxygen species.
(i) Compatible solutes. Compatible solutes stabilize proteins and membranes

(Sect. 7.4, Box 7.1). In resurrection plants mainly sugars function as compatible
solutes, predominantly sucrose but also glucose, fructose and unusual sugars, such
as trehalose, glucopyranosyl-β-glycerol and arbutin (Hartung et al. 1998; Vicré et al.
2004). Increase in hexokinase activity is associated with the acquisition of desicca-
tion tolerance (Whittaker et al. 2001). Unusual sugars, such as stachyose (Norwood
et al. 2003) and in C. plantagineum the C8 -sugar 2-octulose are converted to sucrose
(Norwood et al. 2000; Bartels and Salamini 2001; Ramanjulu and Bartels 2002).
Hydrophilic protective proteins are also important (Rodrigo et al. 2004) and are for
example accumulated in plastids (Bartels and Salamini 2001). An interesting cyto-
logical consequence of the accumulation of sugars with a higher degree of polymer-
ization such as raffinose, stachyose and other galactosyl-sucrose-oligosaccharides in
addition to sucrose is the suppression of crystallisation of protoplastic constituents
and the promotion of glass formation or vitrification controlling metabolism in the
desiccated state at low water content (Hartung et al. 1998; Proctor and Tuba 2002;
Vicré et al. 2004).
(ii) Lipids. Membranes of resurrection plants appear to be well protected (Hartung

et al. 1998). Dynamics of the chemical composition of membrane lipids are impor-
tant to maintain the structure of membranes including the plasma membrane and
chloroplast membranes during desiccation/rehydration cycles, where the unsatura-
tion level of phospholipids and the level of total lipids decrease during desiccation
(Navari-Izzo et al. 2000; Quartacci et al. 2002; Ramanjulu and Bartels 2002).
(iii) Reactive oxygen species (ROS). The formation of ROS is a particular

problem during desiccation/rehydration cycles especially in homoiochlorophyllous
resurrection plants (Sect. 11.4.3.1). Antioxidative defence systems such as the ascor-
bate/glutathione cycle and superoxide dismutases play an important role in pro-
tection to keep functional sulfhydryl groups in the reduced state (Smirnoff 1993;
Hartung et al. 1998; Kappen and Valldares 1999; Proctor and Tuba 2002; Vicré
et al. 2004). Angiosperms with internal carbon concentrating mechanisms such as
C4 -plants (Sect. 10.1.1.2) and CAM-plants (Sect. 5.2.2.2) where oxidative stress is
at least partially controlled by high internal CO2 concentrations appear to be rare
among resurrection plants. A curiosity is the small desiccation tolerant cactus with
CAM Blossfeldia liliputana (Barthlott and Porembski 1996; Hartung et al. 1998).
11.4.3.5 Gene Regulation
The protection of the genome against desiccation and rehydration induced damage
and the down regulation and up regulation of genes follow longer time constants
References 413
than metabolic processes (Cooper and Farrant 2002). As many as 800 to 3000 genes
may be involved in desiccation tolerance of plants (Hartung et al. 1998). Screen-
ing and transcriptomics techniques show differential expression, up regulation and
down regulation, of a large number of genes (Ingram and Bartels 1996; Bockel et
al. 1998; Velasco et al. 1998; Garwe et al. 2003; Collett et al. 2004; Neale et al.
2000). This underlines the high complexity of molecular responses in desiccation
tolerance. Many of these genes can be related to special functions, but the roles
of very many others are not understood. Some of these genes are specific in des-
iccation tolerance many others are seen to be involved in general in various stress
responses. Gene regulation cascades and networks in desiccation tolerance (see Ra-
manjulu and Bartels 2002) are modulated by various phytohormones (Ghasempour
et al. 2001; Vicré et al. 2004). The most central messenger, however, is abscisic acid
(ABA) (Ingram and Bartels 1996; Hartung et al. 1998; Kappen and Valladares 1999;
Bartels and Salamini 2001; Proctor and Tuba 2002; Bernacchia and Furini 2004;
Vicré et al. 2004). Its levels, for example, increase rapidly in desiccating Chamaegi-
gas intrepidus (Fig. 11.24). Gene expression dynamics in relation to the various
cell physiological functions discussed in Sect. 11.4.3.4 have been revealed, namely
carbohydrate metablosim for compatible solutes (Kleines et al. 1999; Bernacchia
and Furini 2004), lipid metabolism (Bartels and Salamini 2001), photosynthesis
(Phillips et al. 2002; Collett et al. 2003), and aquaporins, channel proteins facilitat-
ing water exchange across membranes (Hartung et al. 1998; Ramanjulu and Bartels
2002). Differential gene expression also addresses protective hydrophilic proteins
such as the late embryogenesis abundant proteins (LEA) first discovered due to
their involvement in cellular protections in dehydrating seeds, dehdrins and small
heat shock proteins (Hartung et al. 1998; Bartels and Salamini 2001; Ditzer et al.
2001; Bernacchia and Furini 2004).
References
Adhikary SP (2003) Heat shock proteins in the terrestrial epilithic cyanobacterium Tolypothrix
byssoidea. Biol Plant 47:125–128
Albrecht M, Steiger S, Sandmann G (2001) Expression of a ketolase gene mediates the synthe-
sis of canthaxanthin in Synechococcus leading to tolerance against photoinhibition, pigment
degradation and UV-B sensitivity of photosynthesis. Photochem Photobiol 73:551–555
Bartels D, Salamini F (2001) Desiccation tolerance in the resurrection plant Craterostigma plan-
tagineum. A contribution to the study of drought tolerance at the molecular level. Plant Physiol
127:1346–1353
Barthlott W, Porembski S (1996) Ecology and morphology of Blossfeldia liliputana (Cactaceae):
a poikilohydric and almost astomate succulent. Bot Acta 109:161–166
Barthlott W, Gröger A, Porembski S (1993) Some remarks on the vegetation of tropical inselbergs:
diversity and ecological differentiation. Biogeographia 69:105–124
Belnap J, Lange OL (eds) (2001) Biological soil crusts: structure, function and management. Eco-
logical studies, vol 150. Springer, Berlin Heidelberg New York
Bernacchia G, Furini A (2004) Biochemical and molecular responses to water stress in resurrection
plants. Physiol Plant 121:175–181
Biedinger N, Porembski S, Barthlott W (2000) Vascular plants of inselbergs: vegetative and repro-
ductive strategies. In: Porembski S, Barthlott W (eds) Inselbergs. Ecological studies, vol 146.
414 11 Inselbergs
Bockel C, Salamini F, Bartels D (1998) Isolation and characterization of genes expressed during
early events of the dehydration process in the resurrection plant Craterostigma plantagineum.
J Plant Physiol 152:158–166
Bornhardt W (1900) Zur Oberflächengestaltung und Geologie Deutsch-Ostafrikas. Reimer, Berlin
Bremer H, Sander H (2000) Inselbergs: geomorphology and geoecology. In: Porembski S, Barthlott
W (eds) Inselbergs. Ecological studies, vol 146. Springer, Berlin Heidelberg New York, pp 7–
35
Büdel B (1999) Ecology and diversity of rock inhabiting cyanobacteria in tropaical regions. Eur J
Phycol 34:361–370
Büdel B, Lüttge U, Stelzer R, Huber O, Medina E (1994) Cyanobacteria of rocks and soils in the
Orinoco region in the Guayana highlands, Venezuela. Bot Acta 107:422–431
Büdel B, Becker U, Porembski S, Barthlott W (1997a) Cyanobacteria and cyanobacteria lichens
from inselbergs of the Ivory Coast, Africa. Bot Acta 110:458–465
Büdel B, Karsten U, Garcia-Pichel F (1997b) Ultraviolet-absorbing scytonemin and mycosporine-
like amino acid derivatives in exposed, rock-inhabiting cyanobacterial lichens. Oecologia
112:165–172
Büdel B, Becker U, Follmann G, Sterflinger K (2000) Algae, fungi, and lichens on inselbergs.
In: Porembski S, Barthlott W (eds) Inselbergs. Ecological studies, vol 146. Springer, Berlin
Bukhov NG, Kopecky J, Pfündel EE, Klughammer C, Heber U (2001) A few molecules of zeax-
anthin per reaction centre of photosytem II permit effective thermal dissipation of light energy
in photosystem II of a poikilohydric moss. Planta 212:739–748
Calatayud A, Deltoro VI, Barreno E, Valle-Tascon del S (1997) Changes in in vivo chlorophyll fluo-
rescence quenching in lichen thalli as a function of water content and suggestion of zeaxanthin-
associated photoprotection. Physiol Plant 101:93–102
Collett H, Butwot R, Smith J, Farrant J, Illing N (2003) Photosynthetic genes are differentially tran-
scribed during the dehydration-rehydration cycle in the resurrection plant Xerophyta humilis.
J Exp Bot 54:2593–2595
Collett H, Shen A, Gardner M, Farrant JM, Denby KJ, Illing N (2004) Towards transcript profil-
ing of desiccation tolerance in Xerophyta humilis: construction of a normalized 11k X. humilis
cDNA set and microarray expression analysis of 424 cDNAs in response to dehydration. Phys-
iol Plant 122:39–53
Cooper K, Farrant JM (2002) Recovery of the resurrection plant Craterostigma wilmsii from des-
iccation: protection versus repair. J Exp Bot 53:1805–1813
Coxson DS, Kershaw KA (1983) Rehydration response of nitrogenase activity and carbon fixation
in terrestrial Nostoc commune from Stipa-Bonteloa grassland. Can J Bot 61:2658–2668
Csintalan Z, Proctor MCF, Tuba Z (1999) Chlorophyll fluorescence during drying and rehydration
in the mosses Rhytidiadelphus loreus (Hedw.) Warnst., Anomodon viticulosus (Hedw.) Hook.
& Tayl. and Grimmia pulvinata (Hedw.) Sm. Annals Bot 84:235–244
Deltoro VI, Calatayud A, Gimeno C, Abadía A, Barreno E (1998) Changes in chlorophyll a fluores-
cence, protective CO2 assimilation and xanthophyll cycle interconversion during dehydration
in desiccation-tolerant and intolerant liverworts. Planta 207:224–228
Demmig-Adams B, Adams WW, Czygan F-C, Schreiber U, Lange OL (1990) Differences in the
capacity for radiation less energy dissipation in the photochemical apparatus of green and blue-
green algal lichens associated with differences in carotenoid composition. Planta 180:582–589
Ditzer A, Kirch H-H, Nair A, Bartels D (2001) Molecular characterization of two alanine-rich Lea
genes abundantly expressed in the resurrection plant C. plantagineum in response to osmotic
stress and ABA. J Plant Physiol 158:623–633
Eickmeier WG, Casper C, Osmond CB (1993) Chlorophyll fluorescence in the resurrection plant
Selaginella lepidophylla (Hook. & Grev.) Spring during high-light and desiccation stress, and
evidence for zeaxanthin-associated photoprotection. Planta 189:30–38
Evans RD, Ehleringer JR (1993) A break in the nitrogen cycle in arid lands? Evidence from δ 15 N
of soils. Oecologia 94:314–317
References 415
Farrant JM, Willingen C van der, Loffell DA, Bartsch S, Whittaker A (2003) An investigation into
the role of light during desiccation of three angiosperm resurrection plants. Plant Cell Environ
26:1275–1286
Frahm J-P (2000) Bryophytes. In: Porembski S, Barthlott W (eds) Inselbergs. Ecological studies,
vol 146. Springer, Berlin Heidelberg New York, pp 91–102
Gaff DF (1977) Desiccation tolerant vascular plants of Southern Africa. Oecologia 31:95–109
Gaff DF (1987) Desiccation tolerant plants in South America. Oecologia 74:133–136
Garcia-Pichel F, Castenholz RW (1991) Characterization and biological implications of scytone-
min, a cyanobacterial sheath pigment. J Phycol 27:395–409
Garwe D, Thomson J, Mundree SG (2003) Molecular characterization of XVSAP1, a stress-
responsive gene from the resurrection plant Xerophyta viscosa Baker. J Exp Bot 54:191–201
Ghasempour HR, Anderson EM, Gaff DF (2001) Effects of growth substances on the protoplasmic
drought tolerance of leaf cells of the resurrection grass Sporobulus stapfianus. Aust J Plant
Physiol 28:1115–1120
Golubic S (1967) Die Algenvegetation an Sandsteinfelsen Ost-Venezuelas (Cumaná). Int Rev Hy-
drobiol 52:693–699
Gröger A (1995) Die Vegetation der Granit-Inselberge Südvenezuelas: ökologische und bio-
geographische Untersuchungen. Doctoral Thesis, Bonn
Gröger A (2000) Flora and vegetation of inselbergs of Venezuelan Guayana. In: Porembski S,
Barthlott WE (eds) Inselbergs. Ecological studies, vol 146. Springer, Berlin Heidelberg New
York, pp 291–314
Harel Y, Ohad I, Kaplan A (2004) Activation of photosynthesis and resistance to photoinhibition
in cyanobacteria within biological desert crust. Plant Physiol 136:3070–3079
Hartung W, Schiller P, Dietz K-J (1998) Physiology of poikilohydric plants. Progr Bot 59:299–327
Heber U, Bilger W, Bligny R, Lange OL (2000) Phototolerance of lichens, mosses and higher
plants in an alpine environment: analysis of photoreactions. Planta 211:770–780
Heber U, Bukhov NG, Shuvalov VA, Kobayashi Y, Lange OL (2001) Protection of the photosyn-
thetic apparatus against damage by excessive illumination in homoiohydric leaves and poik-
ilohydric mosses and lichens. J Exp Bot 52:1999–2006
Heber U, Bilger W, Shuvalov VA (2006a) Thermal energy dissipation in reaction centres and in the
antenna of photosystem II protects desiccated poikilohydric mosses against photo-oxidation.
J Exp Bot 57:2993–3006
Heber U, Lange OL, Shuvalov VA (2006b) Conservation and dissipation of light energy as com-
plementary processes: homoiohydric and poikilohydric autotrophs. J Exp Bot 57:1211–1223
Hirai M, Yamakawa R, Nishio J, Yamaji T, Kachino Y, Koike H, Satoh K (2004) Deactivation
of photosynthetic activities is triggered by loss of a small amount of water in a desiccation-
tolerant cyanobacterium, Nostoc commune. Plant Cell Physiol 45:872–878
Huber O (1980) Die Felsvegetation am oberen Orinoko in Südvenezuela. In: Reisig H (ed) Blu-
menparadiese und botanische Gärten der Erde. Pinguin-Verlag, Innsbruck und Umschau Ver-
lag, Frankfurt, pp 200–203
Humboldt A von (1849) Ansichten der Natur. JG Cotta, Stuttgart. Quoted after the edition of Greno
Verlagsgesellschaft, Nördlingen 1986
Ingram J, Bartels D (1996) The molecular basis of desiccation tolerance in plants. Ann Rev Plant
Phys Plant Mol Biol 47:377–403
Jaag O (1945) Untersuchungen über die Vegetation und Biologie der Algen des nackten Gesteins
in den Alpen, im Jura und im schweizerischen Mittelland. Beitr Kryptogamenflora Schweiz
9:1–560
Jeffries DL, Link SO, Klopatek JM (1993a) CO2 fluxes of cryptogamic crusts. I. Response to
resaturation. New Phytol 125:163–173
Jeffries DL, Link SO, Klopatek JM (1993b) CO2 fluxes of cryptogamic crusts. II. Response of
dehydration. New Phytol 125:391–396
Jones K (1977) The effects of moisture on acetylene reduction by mats of blue-green algae in
sub-tropical grassland. Ann Bot 41:801–806
416 11 Inselbergs
Kappen L, Valladares F (1999) Opportunistic growth and desiccation tolerance: the ecological suc-
cess of poikilohydrous autotrophs. In: Pugnaire FI, Valladares F (eds) Handbook of functional
plant ecology. Marcel Dekker, New York Basel, pp 9–80
Kleines M, Ester R-C, Rodrigo M-J, Blervacq A-S, Salamini F, Bartels D (1999) Isolation and ex-
pression analysis of two stress-responsive sucrose-synthase genes from the resurrection plant
Craterostigma plantagineum (Hochst.). Planta 209:13–24
Kluge M, Brulfert J (2000) Ecophysiology of vascular plants on inselbergs. In: Porembski S,
Barthlott W. Ecological studies, vol 146. Springer, Berlin Heidelberg New York, pp 143–174
Kranner I (2002) Glutathione status correlates with different degrees of desiccation tolerance in
three lichens. New Phytol 154:451–460
Lakatos M, Bilger W, Büdel B (2001) Carotenoid composition of terrestrial cyanobacteria: re-
sponse to natural light conditions in open rock habitats in Venezuela. Eur J Phycol 36:367–375
Lange O (1988) Ecophysiology of photosynthesis: performance of poikilohydric lichens and ho-
moiohydric mediterranean sclerophylls. J Ecol 76:915–937
Lange OL, Bilger W, Rimke S, Schreiber U (1989) Chlorophyll fluorescence of lichens containing
green and blue-green algae during hydration by water vapor uptake and by addition of liquid
water. Bot Acta 102:306–313
Lange OL, Kidron GJ, Büdel B, Meyer A, Kilian E, Abeliovich A (1992) Taxonomic composition
and photosynthetic characteristics of the ‘biological soil crusts’ covering sand dunes in the
western Negev Desert. Funct Ecol 6:519–527
Lüttge U (1997) Cyanobacterial Tintenstrich communities and their ecology. Naturwissenschaften
84:526–534
Lüttge U, Büdel B, Ball E, Strube F, Weber P (1995) Photosynthesis of terrestrial cyanobacteria
under light and desiccation stress as expressed by chlorophyll fluorescence and gas exchange.
J Exp Bot 46:309–319
Lüttge U, Meirelles ST, de Mattos EA (2007) Strong quenching of chlorophyll fluorescence in
the desiccated state in three poikilohydric and homoiochlorophyllous moss species indicates
photo-oxidative protection on highly light exposed rocks of a tropical inselberg. J Plant Physiol
(in press)
Mayaba N, Minibayeva F, Beckett RB (2002) An oxidative burst of hydrogen peroxide during
rehydration following desiccation in the moss Atrichium androgynum. New Phytol 155:275–
289
Medina E (1993) Mineral nutrition: tropical savannas. Prog Bot 54:237–253
Meirelles ST, de Mattos EA, da Silva AC (1997) Potential desiccation tolerant vascular plants from
southeastern Brazil. Pol J Env Stud 6(4):17–21
Moore JP, Nguema-Ona E, Chevalier L, Lindsey GC, Brandt WE, Lerouge P, Farrant JM, Driouich
A (2006) Response of the leaf cell wall to desiccation in the resurrection plant Myrothamnus
flabellifolius. Plant Physiol 141:651–662
Navari-Izzo F, Quartacci MF, Pinzino C, Rascio N, Vazzana C, Sgherri CLM (2000) Protein dy-
namics in thylakoids of the desiccation tolerant plant Boea hygroscopica during dehydration
and rehydration. Plant Physiol 124:1427–1436
Neale AD, Blomstedt CK, Bronson P, Le T-N, Guthridge K, Evans J, Gaff DF, Hamill JD (2000)
The isolation of genes from the resurrection grass Sporobulus stapifianus which are induced
during severe drought stress. Plant Cell Environ 23:265–277
Norwood N, Truesdale MR, Richter A, Scott P (2000) Photosynthetic carbohydrate metabolism in
the resurrection plant Craterostigma plantagineum. J Exp Bot 51:159–165
Norwood N, Toldi O, Richter A, Scott P (2003) Investigation into the ability of roots of the poikilo-
hydric plant Craterostigma plantagineum to survive dehydration stress. J Exp Bot 54:2313–
2321
Phillips JR, Hibricht T, Salamini F, Bartels D (2002) A novel abscisic acid- and dehydration-
responsive gene family from the resurrection plant Craterostigma plantagineum encodes
a plastid-targeted protein with DNA-binding activity. Planta 215:258–266
Porembski S, Barthlott W (eds) (2000a) Inselbergs. Ecological studies, vol 146. Springer, Berlin
Heidelberg New York
References 417
Porembski S, Barthlott W (2000b) Granitic and gneissic outcrops (inselbergs) as centers of diver-
sity for desiccation-tolerant vascular plants. Plant Ecol 151:19–28
Porembski S, Becker U, Seine R (2000a) Islands on islands: habitats on inselbergs. In: Porembski
S, Barthlott W (eds) Inselbergs. Ecological studies, vol 146. Springer, Berlin Heidelberg New
York, pp 49–67
Porembski S, Seine R, Barthlott W (2000b) Factors controlling species richness of inselbergs.
In: Porembski S, Barthlott W (eds) Inselbergs. Ecological studies, vol 146. Springer, Berlin
Proctor MCF (2000) Mosses and alternative adaptation to life on land. New Phytol 148:1–6
Proctor MCF, Smirnoff N (2000) Rapid recovery of photosystems on rewetting desiccation tolerant
mosses: chlorophyll fluorescence and inhibitor experiments. J Exp Bot 51:1695–1704
Proctor MCF, Tuba Z (2002) Poikilohydry and homoiohydry: antithesis or spectrum of possibili-
ties? New Phytol 156:327–349
Quartacci MF, Glišić O, Stevanoviæ B, Navari-Izzo F (2002) Plasma membrane lipids in the res-
urrection plant Ramonda serbica following dehydration and rehydration. J Exp Bot 53:2159–
2166
Ramanjulu S, Bartels D (2002) Drought- and desiccation-induced modulation of gene expression
in plants. Plant Cell Environ 25:141–151
Rascher U, Lakatos M, Büdel B, Lüttge U (2003) Photosynthetic field capacity of cyanobacteria
of a tropical inselberg of the Guiana Highlands. Eur J Phycol 38:247–256
Rodrigo MJ, Bockel C, Blervacq A-S, Bartels D (2004) The novel gene CpEdi-9 from the resur-
rection plant C. plantagineum encodes a hydrophilic protein and is expressed in mature seeds
as well as in response to dehydration in leaf phloem tissue. Planta 219:579–589
Safford HD, Martinelli G (2000) Southeast Brazil. In: Porembski S, Barthlott W (eds) Inselbergs.
Samuelsson G, Lönneborg A, Rosenqvist E, Gustafsson P, Öquist G (1985) Photoinhibition and re-
activation of photosynthesis in the cyanobacterium Anacystis nidulans. Plant Physiol 79:992–
995
Satoh K, Hirai M, Nishio Y, Yamaji T, Kashino Y, Koike H (2002) Recovery of photosynthetic
systems during rewetting is quite rapid in a terrestrial cyanobacerium, Nostoc commune. Plant
Cell Physiol 43:170–176
Scherer S, Zhong ZP (1991) Desiccation independence of terrestrial Nostoc commune ecotypes
(Cyanobacteria). Microb Ecol 22:271–283
Scherer S, Ernst A, Chen T-W, Böger P (1984) Rewetting of drought-resistant blue-green algae:
time course of water uptake and reappearance of respiration, photosynthesis, and nitrogen
fixation. Oecologia 62:418–423
Schneider H, Wistuba N, Wagner H-J, Thürmer F, Zimmermann U (2000) Water rise kinetics in
refilling xylem after desiccation in a resurrection plant. New Phytol 148:221–238
Schneider H, Manz B, Westhoff M, Mimietz S, Szimtenings M, Neuberger T, Faber C, Krohne G,
Haase A, Volke F, Zimmermann U (2003) The impact of lipid distribution, composition and
mobility on xylem water refilling of the resurrection plant Myrothamnus flabellifolius. New
Phytol 159:487–505
Schnell R (1987) La flore et la végétation de l’Amérique tropicale. Masson, Paris
Schreiber U, Bilger W (1993) Progress in chlorophyll fluorescence research: major developments
during the past years in retrospect. Prog Bot 54:151–173
Seine R, Becker U (2000) East and southeast Africa. In: Porembski S, Barthlott W (eds) Inselbergs.
Skleryk RS, Tyrell PN, Espie GS (1997) Photosynthesis and inorganic carbon acquisition in the
cyanobacterium Chlorogoeopsis sp. ATCC 27193. Physiol Plant 99:81–88
Smirnoff N (1993) The role of active oxygen in the response of plants to water deficit and desicca-
tion. New Phytol 125:27–58
Sültemeyer D, Klughammer B, Ludwig M, Badger MR, Price GD (1997) Random insertional
mutagenesis used in the generation of mutants of the marine cyanobacterium Synechococcus
418 11 Inselbergs
sp. strain PCC 7002 with an impaired CO2 concentrating mechanism. Aust J Plant Physiol
24:317–327
Tuba Z, Lichtenthaler HK, Csintalan Z, Pócs T (1993a) Regreening of desiccated leaves of the
poikilochlorophyllous Xerophyta scabrida upon rehydration. J Plant Physiol 142:103–108
Tuba Z, Lichtenthaler HK, Maroti I, Csintalan Z (1993b) Resynthesis of thylakoids and functional
chloroplasts in the desiccated leaves of the poikilochlorophyllous plant Xerophyta scabrida
upon rehydration. J Plant Physiol 142:742–748
Tuba Z, Lichtenthaler HK, Csintalan Z, Nagy Z, Szente U (1994) Reconstitution of chlorophylls
and photosynthetic CO2 assimilation upon rehydration of the desiccated poikilochlorophyllous
plant Xerophyta scabrida (Pax) Th. Dur. et Schinz. Planta 192:414–420
Tuba Z, Csintalan Z, Proctor MCF (1996a) Photosynthetic response of a moss, Tortula ruralis, ssp.
ruralis, and the lichens Cladonia convoluta and C. furcata to water deficit and short periods
of desiccation, and their ecophysiological significance: a baseline study at present-day CO2
concentration. New Phytol 133:353–361
Tuba Z, Lichtenthaler HK, Csintalan Z, Szente K (1996b) Loss of chlorophylls, cessation of pho-
tosynthetic CO2 assimilation and respiration in the poikilochlorophyllous plant Xerophyta
scabrida during desiccation. Physiol Plant 96:383–388
Tyree M (2001) Capillary and sap ascent in a resurrection plant: does theory fit the facts? New
Phytol 150:9–11
Velasco R, Salamini F, Bartels D (1998) Gene structure and expression analysis of the drought- and
abscisic acid-responsive CDeT11-24 gene family from the resurrection plant Craterostigma
plantagineum Hochst. Planta 204:459–471
Vicré M, Sherwin HW, Driouich A, Jaffer MA, Farrant JM (1999) Cell wall characteristics and
structure of hydrated and dry leaves of the resurrection plant Craterostigma wilmsii, a micro-
scopical study. J Plant Physiol 155:719–726
Vicré M, Farrant JM, Driouich A (2004) Insights into the cellular mechanisms of desiccation tol-
erance among angiosperm resurrection plant species. Plant Cell Environ 27:1329–1340
Wagner HJ, Schneider H, Mimietz S, Wistuba N, Rokitta M, Krohne G, Haase A, Zimmermann U
(2000) Xylem conduits of a resurrection plant contain a unique lipid lining and refill following
a distinct pattern after desiccation. New Phytol 148:239–255
Whittaker A, Bochicchino A, Vazzana C, Lindsey G, Farrant J (2001) Changes in leaf hexoki-
nase activity and metabolite levels in response to drying in the desiccation-tolerant species
Sporobulus stapifianus and Xerophyta viscosa. J Exp Bot 52:961–969
Ziegler H, Lüttge U (1998) Carbon isotope discrimination in cyanobacteria of rocks of inselbergs
and soils of savannas in the neotropics. Bot Acta 111:212–215
Chapter 12
Páramos
12.1 Summer Every Day, Winter Every Night
The cold tropics (Sect. 1.2) comprise the “regions within the tropics occurring be-
tween the upper limit of continuous, closed-canopy forest (often around 3,500 –
3,900 m) and the upper limit of plant life (often around 4,600 – 4,900 m)”. In this
way Rundel et al. (1994a) define “tropical alpine environments”. They use “alpine”
as a more general term in an attempt to avoid regional terms like páramo and jalca
for the moist Andes and puna for the drier Andes in South America and Afro-
alpine and moorland in Africa. However, “alpine” is also a regional term applying
to environments outside the tropics. On the other hand, since the conditions and
the physiognomy of vegetation are similar on tropical mountains in different con-
tinents, especially in Africa and South America, we might as well choose the term
páramo. Increasingly, this is used as the general term to describe vegetation in the
cold tropics extending from somewhat above 3,000 m to nearly 5,000 m above sea
level (Fig. 12.1).
The high altitude tropical environments were succinctly described by H ED -
BERG ’s (Hedberg 1964a) aphorism “summer every day and winter every night”.
The most important feature of the tropical alpine zone is the “Frostwechselklima”1
(Troll 1943) with an extraordinarily high amplitude of day/night fluctuations of hu-
midity and especially of temperature (Fig. 12.2). Clearly, general characteristics of
tropics are strongly accentuated in these high altitudes, such that daily oscillations
of temperature are much more pronounced than the seasonal ones. Thus, nocturnal
frosts followed by high day-time temperatures represent the most conspicuous stress
to which plants are exposed in these environments. Additional, often very important
stressors, are limited water supply and mineral nutrition.
1 Day-night freezing climate.

420 12 Páramos
Fig. 12.1A–C Profiles of high tropical mountains in South America (A Chimborazo and Cotopaxi
of the Andes) and Africa (B Mount Kenya; C Mt. Kilimanjaro) with the altitudinal vegetation
zones indicated, and in C also the approximate annual precipitation. (Walter and Breckle 1984,
12.1 Summer Every Day, Winter Every Night 421
Fig. 12.2A,B Thermohygrograms obtained in September 1981 at the Shira Plateau on Mt. Kili-
manjaro at about 3,950 m (A) and in March 1985 on Mt. Kenya at 4,200 m altitude (B). Upper
part of each graph temperature ◦ C; lower part relative humidity %. (Courtesy E. Beck)
422 12 Páramos
12.2 The Stress Factor Frost
With temperatures below 0 ◦ C every night, frost is the permanently dominating en-
vironmental stress factor (“stressor”) in the tropical alpine habitat. Some relations
between cold stress and cold resistance are presented in Fig. 12.3 according to the
stress concept (see Sect. 3.3.2; Box 3.1).
First of all we need to distinguish between low-temperature stress at temper-
atures above the freezing point (0 to +6 ◦ C) and stress caused by sub-freezing
temperatures. Low-temperature stress may lead to a loss of fluidity of membrane
lipids or to an increase in membrane rigidity with many consequences for mem-
brane permeability and intracellular compartmentation. It also slows down many
metabolic reactions. It may cause injury, elastic and plastic strain, and it requires
chilling resistance. It is largely presented by the upper panel in Fig. 12.3 but will
not be discussed here further, since we are dealing with sub-freezing temperatures
Fig. 12.3 Cold stress (left part) and cold resistance (right part) with low temperatures above the
freezing point (upper panel) and subfreezing temperatures (lower panels) as stress factors. The
terminology presented in this scheme according to the stress concept (see Box 3.1) is used in the
present chapter to discuss adaptations of páramo plants
12.3 Life Forms of Páramo Plants 423
in the páramo habitat. Sub-freezing temperatures sooner or later may lead to ice for-
mation, i.e. the crystallization of cellular water. For stress resistance, the location
of ice crystal formation in the cells is critical. If it is on the outer face of the cell
walls, i.e. apoplastic, ice formation is tolerable and thus frost resistance may be
achieved through freezing tolerance. However, if ice crystals are formed in the cell
interior, i.e. intracellularly, this always leads to cell death, and frost resistance can
only be achieved by freezing avoidance. These two cases are represented by the
lower two panels in Fig. 12.3. They constitute options with different advantages and
disadvantages, and it is interesting to note that Afro-alpine species commonly tol-
erate extracellular freezing while Andean species apparently rely on the freezing
avoidance mechanism, as will be shown below in Sects. 12.4.1 and 12.4.2 respec-
tively.
12.3 Life Forms of Páramo Plants
The five major life forms of páramos are:

• giant-rosette plants,
• tussocks of grasses or sedges,
• acaulescent rosette plants,
• cushion plants, and
• sclerophyllous shrubs
(Hedberg 1964a; Beck et al. 1983) (Fig. 12.4), and, in South America:
• cacti
may also be included.
12.3.1 Giant-Rosette Plants
The most typical life form of alpine tropical regions are the giant-rosette plants
of the genera Lobelia (Lobeliaceae) and Senecio (Asteraceae) in Africa, Espele-
tia (Asteraceae) in South America and Argyroxiphium (Asteraceae) in Hawai’i
(Fig. 12.5). They may reach heights of several meters and have developed a num-
ber of morphological and anatomical features adapting them to the diurnal type of
climate in their habitat.
The tallest plant of Senecio johnstonii ssp. cottonii actually measured by Beck
et al. (1983) was 9.60 m, but an even taller one was also encountered at another
location. The plants bear giant rosettes of living leaves at the end of their branches.
Dead leaves usually cover the entire shoot in a dense layer. The dead leaves are kept
for decades and perhaps even centuries, since the larger giant-rosette plants may be
of considerable age. Estimations for Senecio keniodendron suggest that about 50
424 12 Páramos
Fig. 12.4 The five major life forms of Afro-alpine vegetation (Hedberg 1964b)
new leaves are formed annually, the stem growth rates are approximately 2.5 – 3 cm
per year giving an age of 35 years per meter of unbranched stem (Beck et al. 1980).
A 10 m tall giant-rosette plant then might have an age of 350 years.
These dead leaves provide heat insulation of the stem. However, the morphol-
ogy of the living leaf rosette, together with some physiological reactions, enables
most important acclimatory strategies (Fig. 12.6).
The conical leaf bud in the center of the rosette is protected by nyctinastic move-
ments of the adult leaves. Thus, a “night-bud” is formed in these plants, which
of course grow during the whole year and have no dormant periods, as plants in
the temperate zone have in the form of winter buds. In Espeletia schultzii the new
leaves, which had just developed from the central bud, wilted and died when the
formation of the night-bud (by nocturnal closure of the rosette) was experimentally
prevented (Smith 1974).
In African giant-rosette plants the leaf water may freeze apoplastically during
the night (see Sect. 12.4.1). The closing mechanism of the inward movement of
the adult leaves in these plants is based on water loss from the cell interior and
the associated decrease of pressure in the whole tissue as turgor of the individual
cells declines due to cellular loss of water. The opening of the night-bud is due to
instantaneous resorption of the water and restoration of turgor after melting of the
apoplastic ice in the morning.
The dense packing of the developing leaves in the central bud gives this organ
a massive structure with a considerable inherent heat-storage capacity. In addition,
Fig. 12.5A–D Tropical alpine giant-rosette plants. Espeletia timotensis (A) and E. schultzii (B),
Aguila Pass, Venezuela, 3,600 – 4,000 m a.s.l.; Lobelia keniensis (C) and Senecio kenioden-
dron (D), Teleki valley, Mt. Kenya, at 4,100 m and 4,300 m a.s.l. respectively. (C, D courtesy
E. Beck)
426 12 Páramos
Fig. 12.6A–E Giant rosettes of Espeletia schultzii (A), E. moritziana (B) and E. timotensis (C),
Aguila Pass, Venezuela, 3,600 – 4,000 m a.s.l., and of Lobelia deckenii in the day position (D) and
as the night-bud (E), Mt. Kilimanjaro, Tanzania, 3,850 m a.s.l. (D, E courtesy E. Beck)
excreted fluid and mucilage may also contribute to the heat-storage capacity. The
bases of the adult leaves form tanks or cisternae where mucilageous fluid is collected
during the day. In Lobelia keniensis, for example, this may amount to several liters
per rosette. During the nyctinastic leaf movements this fluid, which warms up during
the day, is pressed upwards to cover the meristematic part of the night-bud in the
evening (Beck et al. 1982).
Moreover, the leaves are very pubescent and have large intercellular air spaces
which add to the insulating effect (Fig. 12.7). The consequences of leaf pubescence
are not straightforward. There are complex interactions between several factors
Fig. 12.7 Leaf anatomy of giant-rosette plants, with hairs and large intercellular spaces and mu-
cilage inside and on the surface of the upper epidermis. Cross-section of an adult leaf of Lobelia
keniensis in the upper cisterna region. (Beck et al. 1982)
implying non-linear behaviour of the system. The role of pubescence has been
examined in more detail in Espeletia timotensis in relation to temperature, wind
speed and high solar radiation in the páramo habitat (Meinzer and Goldstein 1985;
Schuepp 1993). Pubescence is more effective at high wind speeds. Increased bound-
ary layer thickness due to a coat of hairs hinders exchange between leaf surface
and ambient wind, and its primary effect in cool air would be an increase in
surface temperature. This may be about 7 ◦ C in the case studied, which is asso-
ciated with a small increase in transpiration (∼ 17%) due to the effects of leaf
temperature on leaf/air water vapour pressure difference. In numerical simula-
tions it was shown however, that in contrast an increase of surface temperature of
about 7 ◦ C would result in a doubling of the transpiration rate of non-pubescent
leaves. Increased transpiration will have a feed-back on leaf temperature because
of the effect of transpirational cooling, which adds to the complexity of the sys-
tem.
428 12 Páramos
Fig. 12.8 Exponential heat decay curves and time constants, te , for inflorescences of Puya hamata
(controls and denuded, respectively), P. clava-herculis and P. aequatorialis. te is the time when
the heat decay curve crosses the line of 1e (T1 − T2 ) + T2 and is independent of the magnitude of
the T1 − T2 temperature difference. Thus, te gives the time it takes for a 63% decrease in the total
temperature difference between the plant organ (T1 ) and the atmosphere (T2 ). (After Jones 1992;
Miller 1994)
In any event, pubescence will slow down the establishment of thermal equilib-
rium of plant organs with air temperature. For the pubescent inflorescences of Puya
(Bromeliaceae) in the equatorial páramo zone of the Ecuadorian Andes this has
been quantified using time constants (te ) derived from the exponential heat decay
curves (Fig. 12.8). Clearly, in the non-pubescent species P. aequatorialis, which
grows on rocky outcrops between 1900 and 2,100 m a.s.l., te is much lower (15 s)
than in the pubescent species P. hamata (135 s) and P. clava-herculis (126 s), and
in P. hamata it drops to 30 s when inflorescences are denuded (Jones 1992; Miller
1994).
All of the features discussed above in this section are mechanisms to delay cool-
ing and provide freezing avoidance in the buds. Indeed, measurements in Kenya
showed that only the temperature of adult leaves closely follows air temperature
and is for many hours every night below the freezing point. However, nocturnal bud
temperatures in Lobelia and Senecio are significantly higher and may even remain
positive. In Senecio brassica for example, at air temperatures around −8 ◦ C, bud
temperature remains +1 ◦ C (Fig. 12.9). We shall see below (Sect. 12.4.2) that for
Andean giant-rosette plants the mechanisms of freezing avoidance based on insula-
tion, heat storage and delay of cooling, together with some supercooling effects, may
suffice for the frost-resistance of the adult leaves. However, in Afro-alpine giant-
rosette plants they are insufficient and freezing tolerance is needed (Sect. 12.4.1).
Fig. 12.9A, B Comparisons

between leaf temperature
and night-bud temperature
as related to air temperature
in Lobelia telekii (A) and
Senecio brassica (B) on Mt.
Kenya during the dry season
in March 1979. (Beck et al.
1982)
12.3.2 Other Life Forms: Tussocks, Cushions, Acaulescent

Rosettes, Sclerophylls
In general, it appears that smaller plants are less threatened by frost than taller ones,
as suggested by Fig. 12.10, which relates average plant height of cushions and small
rosettes, shrubs and perennial herbs, giant rosettes and small trees to temperature
causing injury.
In tussock and cushion plants the regenerating buds are insulated by adult
leaves and dead material. Extensive studies on alpine plants in Europe have shown
that the internal microclimate, temperature and air humidity, within such plant bod-
430 12 Páramos
Fig. 12.10 Injury temperature

of tropical alpine plants re-
lated to average plant height
of cushions and small rosettes,
shrubs and perennial herbs,
giant rosettes and small trees.
(Squeo et al. 1991)
ies and mats can be much different and highly protected from the outside (Reisigl
and Keller 1987). The acaulescent rosettes (Fig. 12.11) closely adopt the soil
temperature. In the Andes, Squeo et al. (1991) found that all ground-level plants
(cushions and acaulescent rosettes) were freezing tolerant. Day-night temperature
changes are somewhat buffered by the heat-storage capacity of the soil and stress
may be less extreme than from air temperature alone (Fig. 12.12A). Moreover, for
taller plants, a specific problem of the day-night freezing climate is not only the
freezing during the night but also the process of thawing at the beginning of the day.
If ice formed in leaves thaws more rapidly than ice in the stems, transpiration sets
in, while water transport in the shoot is still blocked. Thus, thawing every morning
after regular nocturnal freezing leads to cavitation in the conductive elements of the
xylem and the serious problems of embolism (Ball et al. 2006).
This is avoided in acaulescent rosettes, where the whole plant body thaws at
the same time. An example is given in Fig. 12.13 showing leaf temperatures and
corresponding maximum photosynthetic electron transport rates at light saturation
(ETRmax ) in Haplocarpha rueppellii (Fig. 12.11A) during a clear day at 4,100 m
a.s.l. in the Simien Mountains in Ethiopia. As irradiance (PPFD in Fig. 12.13) in-
creased after sunrise, the rosettes thawed and leaf temperatures increased to the
average value of 20 ◦ C when a PPFD of 400 µ mol m−2 s−1 was reached. Above
400 µ mol m−2 s−1 leaf temperature was independent of irradiance which clearly
indicates dissipation of heat via the soil. Maximum rates of photosynthetic electron
transport obtained from measurements of chlorophyll fluorescence (Box 4.6) at light
saturation were linearly related to leaf temperature between ∼ 5 and ∼ 20 ◦ C indi-
cating limitation of photosynthesis by temperature dependence of electron transport
in the thylakoids and/or biochemical reactions of CO2 -assimilation.
Some acaulescent rosette plants follow the CAM mode of photosynthesis, e.g.
Echeveria columbiana (Crassulaceae) in the Andes (Fig. 12.11B) (Medina and Del-
gado 1976). These plants thus must be able to maintain the intensive metabolism
during the night as it is required for dark CO2 -fixation. They may possibly achieve
this in a state of supercooling or non-ideal equilibrium freezing (see Sect. 12.4).
Fig. 12.11 Small acaules-

cent rosettes of Haplocarpha
rueppellii, Simien Moun-
tains, Ethiopia, 4,100 m
a.s.l. (A) and Echeveria
columbiana, Aguila Pass,
Venezuela, 3,600 m a.s.l. (B)
and a cushion of Helichry-
sum newii, Shira Plateau, Mt.
Kilimanjaro, Tanzania (C).
(C Courtesy E. Beck)
432 12 Páramos
Fig. 12.12A, B Air temperature and soil temperature (A) and subepidermal temperatures (B) of
the hairy and the hairless variety of Tephrocactus floccosus growing at the same site. (Keeley and
Keeley 1989)
Fig. 12.13 Relationships between leaf temperature and ambient irradiance (PPFD; dotted line)
and maximum photosynthetic electron transport rate at light saturation (ETRmax ; solid line) in
acaulescent rosettes of Haplocarpha rueppellii (see Fig. 12.11A). (Unpubl. results of the author)
In the sclerophyllous shrubs transpiration is reduced and water economy is sus-

tained by leaf-xeromorphy with reduced leaf surfaces, folded leaves and other mor-
phological/anatomical adaptations.
12.4 Frost Resistance in Giant-Rosette Plants 433
Rundel et al. (1994b) invoke two different reasons for the diversity of life forms
found in páramo ecosystems, namely:
• the complexity of environmental stresses, i.e. the diurnal freezing cycles and wa-
ter and nutrient deficiencies, and
• the occurrence of microhabitat mosaics.
However, additional research is required to determine the correlations between life
forms, habitat conditions, and physiological characteristics.
12.3.3 Cacti
The expression of CAM (see Box 5.1) in Andean plants has been studied in some
detail in the cacti Oroya peruviana and Tephrocactus floccosus in central Peru at
4,000 – 4,700 m a.s.l. (Keeley and Keeley 1989). There was nocturnal malate ac-
cumulation even at air temperatures of −8 ◦ C and subepidermal temperatures of
−3 ◦ C.
The occurrence of hairy and hairless varieties in T. floccosus offered the op-
portunity to demonstrate the effects of a cover by hairs for insulation (see also
Sect. 12.3.1). During the night the hairy form had subepidermal temperatures that
were several degrees higher than in the hairless form (Fig. 12.12). A disadvantage of
the hairs, however, is light scattering and therefore shading of the photosynthetically
active stem tissue. Hence, the cacti have a problem of optimizing the two options to
allow maximal CAM activity. It is interesting to note that at sites where one of the
two forms predominated consistently the dominant form showed the higher noctur-
nal malate accumulation (mal). Thus, where the hairless type was rare, the hairy
morph had the higher mal, and where the glabrous type was frequent, the hairy
type had the lower mal.
12.4 Frost Resistance in Giant-Rosette Plants
12.4.1 Afro-Alpine Plants: Freezing Tolerance
As already noted above (Sect. 12.2), formation of ice crystals in the cytoplasm is
always disastrous. This is predominantly due to effects of ice competing with the
ordered superficial water film of membranes. In addition ice crystals may puncture
membranes and organelles. Thus, freezing tolerance is only possible with extracel-
lular ice formation. For the Afro-alpine giant-rosette plants the phenomenon has
been studied in detail by E. B ECK and collaborators (Beck 1983).
For the formation of extracellular ice, the basic laws of cell water relations (see
Box 6.1) apply as follows:
ψcell = P − π , (12.1)
434 12 Páramos
where ψcell is the water potential of the cell, P and π are turgor and osmotic pres-
sure respectively. During freezing the cell looses water, and ice forms outside the
plasmalemma in the intercellular spaces. Thus, the protoplast shrinks and turgor
becomes zero (P = 0). Therefore
ψcell = −π , (12.2)
and at equilibrium of the protoplasts with the extracellular ice
ψice = ψcell = −π , (12.3)
a situation called “equilibrium freezing” (Fig. 12.14). Essentially, ice forms gradu-
ally as water moves out of the symplast, and not abruptly as occurs after supercool-
ing (see Sect. 12.4.3). Therefore, the occurrence of nucleating agents in the apoplast,
which may include mucilage (Goldstein and Nobel 1994), are important in eliciting
ice-crystal formation and avoiding supercooling (Krog et al. 1979).
According to (12.3), the concentrations of solutes in the protoplast, which de-
termine π, are therefore given by the water potential of the ice, which is linearly
dependent on the subfreezing temperature. The increased cytoplasmic concentra-
tions of solutes may damage or protect membranes and proteins. Often special cry-
oprotective solutes decrease the injurious effects of high ion concentrations on the
membranes. In Afro-alpine plants, sucrose is most likely to fulfil such a role (Beck
1994a). Cryoprotectants are similar to compatible solutes discussed above in rela-
tion to osmotic stress (Sect. 7.4 and Box 7.1). In fact extracellular ice formation
is nothing more than a dramatic osmotic stress, however, at low temperatures.
In this way, for example in Lobelia keniensis at −6 ◦ C, 85% of the tissue water
is frozen. When the ice thaws in the morning, the water is immediately taken up
Fig. 12.14 Equilibrium freez-

ing: water potentials of frozen
leaves (ψcell = •) of S. kenio-
dendrion and L. keniensis and
frozen expressed cellular sap
(−π = ◦) of S. keniodendron
compared to ice at subfreez-
ing temperatures (lines: ψice
calculated by two different
methods A and B; see (12.3))
give the same relationships.
(Beck et al. 1984)
Table 12.1 Frost tolerance of leaf segments of four species of Afro-alpine megaphytes. (Beck et
al. 1982)
Species Frost tolerance (◦ C) 50% damage (◦ C)

Senecio keniodendron − 8 −10
Senecio brassica −10 −15
Lobelia keniensis Lower than −20 Lower than −20
Lobelia telekii Lower than −20 Lower than −20
osmotically into the cells again, and full competence of photosynthesis is regained
rapidly. Using this mechanism, the Afro-alpine giant-rosette plants achieve frost
resistance at temperatures of −8 ◦ C down to −20 ◦ C (Table 12.1).
There may also be deviations from the ideal behaviour given by (12.3) due to the
osmotic contribution of extracellular solutes, which allow lower external water po-
tentials (non-ideal equilibrium freezing) (Goldstein and Nobel 1991; Zhu and Beck
1991). At water losses > 50% a matrix potential is also generated which prevents
intrusion of air between the cell wall and the plasmalemma, and in this way the wall
may get under tension (negative turgor).
12.4.2 Andean Plants: Freezing Avoidance
In contrast to the behaviour of Afro-alpine plants, Andean Espeletias are injured

by freezing of the cell water. Obviously ice formation is occurring intracellularly in
these plants. They need to avoid freezing of cellular water and survive the nights in
a super-cooled state. As depicted in Fig. 12.15 for all the different organs and tis-
sues of Espeletia spicata and E. timotensis in the Andes, the supercooling points, at
which ice formation would happen, are always much lower than the lowest temper-
atures actually found in these organs and tissues. Ambient temperature is modulated
in the tissues by the various strategies discussed in Sect. 12.3.1. Supercooling and
freezing avoidance for the Andean giant-rosette plants therefore must be considered
as an appropriate mode of achieving sufficient frost resistance.
12.4.3 Comparison of the Strategies

of Freezing Tolerance and Avoidance
If freezing tolerance and freezing avoidance are such successful adaptations of

giant-rosette plants to the Frostwechsel climate in the African and South-American
tropical high mountains, the question arises why the plants in the two continents
evolved two such different modes of frost resistance. As compared to freezing and
its tolerance, freezing avoidance by supercooling has one important advantage and
one important disadvantage:
436 12 Páramos
Fig. 12.15 Scheme of a caulescent giant-rosette plant with the lowest leaf temperatures (LT)
recorded in the field during the days of measurements and the supercooling points (SP) of vari-
ous organs and tissues of two Espeletia species indicated. (Rada et al. 1985)
• the advantage is that the cells always remain metabolically competent,

• the disadvantage is the very high risk inherent in the supercooling strategy, since
supercooling is a thermodynamically labile state; at the lowest supercooling
point the water freezes instantaneously, there is no time for water-export into
the apoplast, and cell death becomes unavoidable.
Continuous metabolic competence should allow higher productivity. In the Andean
páramos, where night temperatures go down to 0 ◦ C and rarely below −5 ◦ C, the risk
of reaching the crystallization point of water is small (Fig. 12.15) as compared to
the benefit of higher productivity. Conversely, in the Afro-alpine zone temperatures
are frequently below −10 ◦ C and the risk inherent in supercooling becomes too high
to be a reasonable choice.
Table 12.2 Productivity of an African and two Andean giant-rosette plants. (After Rada et al.
1985)
Species Productivity (g DW m−2 leaf surface year−1 )

Africa Senecio keniodendron 166
South America Espeletia spicata 671
Espeletia timotensis 370
Hence, it appears fairly straightforward that the different nocturnal temperature

regimes in the two regions led to evolution of the two different strategies, one taking
the risk and obtaining more productivity and the other one avoiding the risk on ac-
count of the loss of some productivity. Indeed Table 12.2 suggests that the African
Senecio keniodendron has a much lower annual productivity than the Andean Es-
peletias (Rada et al. 1985).
Interestingly, Lipp et al. (1994) note that in species growing at high elevations in
Hawai’i features of both adaptive strategies are combined and a complete suite of
Fig. 12.16A–C Plants of high altitudes on Haleakala volcano, Maui island, Hawai’i. Styphelia
tameiameiae (A), Dubautia menziesii (B), Argyroxiphium sandwicense (C)
438 12 Páramos
characteristics with either strict tolerance or avoidance of extracellular ice formation

is not expressed. Five species were studied, namely Argyroxiphium sandwicense and
Dubautia menziesii (both Asteraceae), Sophora chrysophylla (Fabaceae), Vaccinium
reticulatum (Ericaceae) and Styphelia tameiameiae (Epacridaceae) (Fig. 12.16).
Typical freezing tolerance is not fully expressed possibly due to a more recent evo-
lutionary status of these taxa. A combination of the two possible adaptive strategies
is given in that a period of supercooling occurs prior to ice nucleation. Four of the
five species could tolerate extracellular ice formation to a certain degree. For exam-
ple, in S. tameiameiae in the laboratory there was considerable supercooling prior
to ice formation at − 9.4 ◦ C, and the latter did not cause tissue injury.
12.5 Other Stress Factors
12.5.1 Water Availability
Precipitation decreases with increasing altitude in the cold tropics (Lauer 1975;
Rundel 1994). Although patterns of precipitation in páramo regions are very com-
plex and do not give as clear a picture as the temperature patterns, it is evident
that diurnal drought problems are associated with the tropical Frostwechsel climate.
Massive structures which contribute to heat-storage capacity (Sect. 12.3.1) may also
provide water-storage capacity. Thus, the pith storage capacity of Espeletias in the
Andes is thought to be involved in diurnal drought avoidance mechanisms. This
may also explain the apparent paradox that the height of giant rosette species in-
creases with higher altitudes as this implies an increase of stem water-storage capac-
ity (Meinzer et al. 1994). Water retaining gels have also been thought to contribute to
water storage, e.g. in Argyroxiphium sandwicense (Carlquist 1994). Although water
relations in high elevation tropical plants have been studied to some extent (Meinzer
et al. 1994), ecophysiological studies of drought tolerance are less advanced than for
the frost stressor.
12.5.2 Mineral Nutrition and Carbon
Nutrient relations of tropical high elevation plants have been little investigated (Reh-
der 1994).
An interesting case story is presented by Beck (1994b). Dead leaves kept on the
stems of Senecios for heat insulation (Sect. 12.3.1) obviously are withheld from
mineralization in the soil and the soil-plant nutrient cycle. However, in Senecio ke-
niodendron leaves may gradually decay in situ on the stems, and where moisture
is retained, adventitious roots emerge from the stems and produce a network in the
sheath of decaying leaf bases not unlike the tank roots of bromeliads (Sect. 6.4,
Fig. 6.15B). Mineralised nutrients are reabsorbed by living plant tissues and the
nutrient cycle is closed within the plant.
References 439
Another intriguing case of mineral nutrition is inorganic carbon acquisition from

the pedosphere by terrestrial páramo species of Isoëtes (Lycopodiopsida, Isoë-
taceae). Isoëtes andicola is a terrestrial species which has no stomata on the short
leaves of its rosettes and gains the bulk of its carbon from the peat sediment near
lakes and lagoons in the high Andes. Most of the C-uptake occurs during the day
with C-reduction via the Calvin cycle; however, there is also some CAM-type C-
acquisition during the night. The significance of this pedosphere-based carbon nu-
trition is not clear. Sediment-based carbon acquisition is well known of aquatic
Isoëtes species, which also may perform CAM. Among other hypotheses, in ter-
restrial Isoëtes species adaptation to seasonal droughts may be an explanation, as
sealing off the leaves from the atmosphere may provide an advantage under drought-
stress conditions. Other stomata-less terrestrial species with similar modes of carbon
nutrition are I. andina and I. novo-grandensis in the Andes and possibly I. hopei in
New Guinea. Aquatic Isoëtes species are nearly ubiquitous in páramo regions of
South America providing a possible ancestral pool for the terrestrial forms with
their unique strategy (Keeley et al. 1994).
12.5.3 Irradiance and Heat
Leaf hairs and scales represent an effective barrier against high-light stress, pho-
toinhibition and damage of the photosynthetic apparatus (see also Sect. 5.2.2.1) and
UV-radiation (Lang and Schindler 1994), as well as overheating (Melcher et al.
1994). Naturally the same insulating effect which reduces cooling during the night
delays heating during the day under the Frostwechsel climate. Melcher et al. (1994)
showed that in Argyroxiphium sandwicense on the Haleakala volcano on Maui is-
land, Hawai’i, temperature of expanded leaves was similar to, or even lower than,
air temperature at full solar radiation. Conversely, the apical bud in the center of the
rosette was usually 25 ◦ C warmer than air at noon. This may facilitate physiological
processes required for the maintenance of growth of new leaves in the apical bud.
However, this must also be the reason which limits A. sandwicense to altitudes of
≥ 1,900 m a.s.l. below which apical bud-temperatures might reach lethal levels.
References
Ball MC, Canny MJ, Huang CX, Egerton JJG, Wolfe J (2006) Freeze/thaw-induced embolism
depends on nadir temperature: the heterogeneous hydration hypothesis. Plant Cell Environ
29:729–745
Beck E (1983) Frost- und Feuerresistenz tropisch-alpiner Pflanzen. Naturwiss Rundsch 36:105–
109
Beck E (1994a) Cold tolerance in tropical alpine plants. In: Rundel PW, Smith AP, Meinzer FC
(eds) Tropical alpine environments. Plant form and function. Cambridge University Press,
Cambridge, pp 77–110
440 12 Páramos
Beck E (1994b) Turnover and conservation of nutrients in the pachycaul Senecio keniodendron.
In: Rundel PW, Smith AP, Meinzer FC (eds) Tropical alpine environments. Plant form and
function. Cambridge University Press, Cambridge, pp 215–221
Beck E, Scheibe R, Senser M, Müller W (1980) Estimation of leaf and stem growth of unbranched
Senecio keniodendron trees. Flora 170:68–76
Beck E, Senser M, Scheibe R, Steiger H-M, Pongratz P (1982) Frost avoidance and freezing toler-
ance in Afroalpine “giant-rosette” plants. Plant Cell Environ 5:215–222
Beck E, Scheibe R, Senser M (1983) The vegetation of the Shira Plateau and the western slopes of
Kibo (Mt. Kilimanjaro, Tanzania). Phytocoenologia 11:1–30
Beck E, Schulze E-D, Senser M, Scheibe R (1984) Equilibrium freezing of leaf water and extra-
cellular ice formation in Afroalpine “giant-rosette” plants. Planta 162:276–282
Carlquist S (1994) Anatomy of tropical alpine plants. In: Rundel PW, Smith AP, Meinzer FC
(eds) Tropical alpine environments. Plant form and function. Cambridge University Press,
Cambridge, pp 111–128
Goldstein G, Nobel PS (1991) Changes in osmotic pressure and mucilage during low-temperature
acclimation of Opuntia ficus-indica. Plant Physiol 97:954–961
Goldstein G, Nobel PS (1994) Water relations and low-temperature acclimation for cactus species
varying in freezing tolerance. Plant Physiol 104:675–681
Hedberg O (1964a) Features of Afroalpine plant ecology. Acta Phytogeogr Suec 49:1–144
Hedberg O (1964b) Etudes écologiques de la flore Afroalpine. Bull Soc R Bot Belg 97:5–18
Jones HG (1992) Plants and microclimates, 2nd edn. Cambridge University Press, Cambridge
Keeley JE, Keeley SC (1989) Crassulacean acid metabolism (CAM) in high elevation tropical
cactus. Plant Cell Environ 12:331–336
Keeley JE, DeMason DA, Gonzalez R, Markham KR (1994) Sediment-based carbon nutrition in
tropical alpine Isoëtes. In: Rundel PW, Smith AP, Meinzer FC (eds) Tropical alpine environ-
ments. Plant form and function., Cambridge University Press, Cambridge, pp 167–194
Krog JO, Zachariassen KE, Larsen B, Smidsrod O (1979) Thermal buffering in Afroalpine plants
due to nucleating agent-induced water freezing. Nature 282:300–301
Lang M, Schindler C (1994) The effect of leaf-hairs on blue and red fluorescence emission and on
zeaxanthin cycle performance of Senecio medley L. J Plant Physiol 144:680–685
Lauer W (1975) Vom Wesen der Tropen. Klimaökologische Studien zum Inhalt und zur Abgren-
zung eines irdischen Landschaftsgürtels. Akad Wiss Lit Abh Math Naturwiss Kl (Mainz)
1975, 3:5–52
Lipp CC, Goldstein G, Meinzer FC, Niemczura W (1994) Freezing tolerance and avoidance in
high-elevation Hawaiian plants. Plant Cell Environ 17:1035–1044
Medina E, Delgado M (1976) Photosynthesis and night CO2 -fixation in Echeveria columbiana
Poellnitz. Photosynthetica 10:155–163
Meinzer F, Goldstein G (1985) Some consequences of leaf pubescence in the Andean giant-rosette
plant Espeletia timotensis. Ecology 66:512–520
Meinzer FC, Goldstein G, Rundel PW (1994) Comparative water relations of tropical alpine plants.
In: Rundel PW, Smith AP, Meinzer FC (eds) Tropical alpine environments. Plant form and
function. Cambridge University Press, Cambridge, pp 61–76
Melcher PJ, Goldstein G, Meinzer FC, Minyard B, Giambelluca TW, Loope LL (1994) Deter-
minants of thermal balance in the Hawaiian giant rosette plant, Argyroxiphium sandwicense.
Miller GA (1994) Functional significance of inflorescence pubescence in tropical alpine species of
Puya. In: Rundel PW, Smith AP, Meinzer FC (eds) Tropical alpine environments. Plant form
and function. Cambridge University Press, Cambridge, pp 195–213
Rada F, Goldstein G, Azocar A, Meinzer F (1985) Freezing avoidance in Andean giant rosette
plants. Plant Cell Environ 8:501–507
Rehder H (1994) Soil nutrient dynamics in East African alpine ecosystems. In: Rundel PW, Smith
AP, Meinzer FC (eds) Tropical alpine environments. Plant form and function. Cambridge Uni-
versity Press, Cambridge, pp 223–228
Reisigl H, Keller R (1987) Alpenpflanzen im Lebensraum. G Fischer, Stuttgart
References 441
Rundel PW (1994) Tropical alpine climates. In: Rundel PW, Smith AP, Meinzer FC (eds) Tropical
alpine environments. Plant form and function. Cambridge University Press, Cambridge, pp
21–44
Rundel PW, Smith AP, Meinzer FC (eds) (1994a) Tropical alpine environments. Plant form and
function. Cambridge University Press, Cambridge
Rundel PW, Meinzer FC, Smith AP (1994b) Tropical alpine ecology: progress and priorities. In:
Rundel PW, Smith AP, Meinzer FC (eds) Tropical alpine environments. Plant form and func-
tion. Cambridge University Press, Cambridge, pp 355–363
Schuepp PH (1993) Leaf boundary layers. New Phytol 125:477–507
Smith AP (1974) Bud temperature in relation to nyctinastic leaf movement in an Andean giant-
rosette plant. Biotropica 6:263–266
Squeo FA, Rada F, Azocar A, Goldstein G (1991) Freezing tolerance and avoidance in high trop-
ical Andean plants: is it equally represented in species with different plant height? Oecologia
86:378–382
Troll C (1943) Die Frostwechselhäufigkeit in den Luft- und Bodenklimaten der Erde. Meteorol Z
60:161–171
Walter H, Breckle S-W (1984) Spezielle Ökologie der tropischen und subtropischen Zonen. G
Fischer, Stuttgart
Zhu JJ, Beck E (1991) Water relations of Pachysandra leaves during freezing and thawing. Evi-
dence for a negative pressure potential alleviating freeze-dehydration stress. Plant Physiol
97:1146–1153
Scientific Name Index
A Arbutus unedo 130

Argyroxiphium 423
Acacia 8, 298, 350, 351 Argyroxiphium sandwicense 437–439
Acacia mellifera ssp. detinens 367 Aristida 321
Acacia nebrownii 350, 351 Aspidosperma fendleri 138
Acanthocereus tetragonus 274 Asplundia 175
Acanthus ilicifolius 230 Atta colombica 93, 94
Acetobacter 346 Avicennia corniculatum 246, 247, 251, 252
Acisanthera uniflora 361 Avicennia eucalyptifolia 230
Acrostichum aureum 232, 233, 255, 256, Avicennia germinans 231, 234, 237, 241,
258 244–249, 253, 270, 290
Actinostrobus 364 Avicennia marina 230, 243, 245–247, 249,
Adansonia 330 252–254, 256
Adansonia digitata 329, 330 Axonopus 321
Adansonia gregorii 330 Axonopus purpusii 339
Aechmea aquilega 161, 162 Azoarcus 346
Aechmea fendleri 161, 162 Azolla 348
Aechmea lingulata 177, 184 Azorhizobium 349
Aechmea nudicaulis 207, 266–268 Azotobacter 346
Aegialitis annulata 230, 239
Aegiceras corniculatum 230, 256 B
Aeschimone 353
Aglaomorpha heracleum 194 Banksia ornata 364
Agrostis tef 320 Bastardia viscosa 274
Allagoptera arenaria 266, 267 Batis maritima 270, 273, 274, 277, 279–281
Alocasia macrorrhiza 110, 113, 114, 133, Beijerinckia 346
134 Biovularia 213
Anabaena 348 Blossfeldia liliputana 412
Ananas 107 Bombacopsis quinta 330
Ananas ananassoides 402, 403 Bombax malabaricum 330
Andira legalis 268, 350 Bonnetia crassa 300
Andropogon 321 Bostrychia simpliciuscula 258
Andropogon gerardi 339 Bostrychia tenuisissima 258
Andropogon scoparius 339 Bouteloua 321
Anthurium bredmeyeri 195 Bowdichia 350
Anthurium crassinervium 274 Brachiaria 345
Anthurium hookeri 194 Brachiaria brizantha 321
443
444 Scientific Name Index
Bradyrhizobium 349 Clusia alata 204

Brocchinia hechtioides 300, 356 Clusia criuva 210, 385
Brocchinia reducta 356 Clusia fluminensis 209, 210, 268
Brocchinia tatei 217, 219 Clusia hilariana 266–268
Bromelia 107 Clusia major 204
Bromelia goeldiana 402, 403 Clusia minor 160, 161, 204, 206, 207, 210
Bromelia humilis 107, 108, 114–116, 124, Clusia multiflora 59, 128, 210
157, 187, 197, 274, 287, 288, 403 Clusia parviflora 210
Brownea 96 Clusia rosea 177, 178, 204, 207, 209, 210,
Bruguiera exaristata 230 212
Bruguiera gymnorhiza 230, 250, 254, 256 Clusia uvitana 180, 181, 200
Bruguiera parviflora 253, 256, 257 Clusia venosa 204
Bulbostylis spadicea 364 Cochlospermum vitifolium 330
Bursera simaruba 326, 330 Codonophytum epiphytum 170
Byblis 213 Colophospermum mopane 350, 351
Byrsonima crassifolia 325, 333 Commiphora 330
Byrsonima verbascifolia 370 Conocarpus erectus 189, 246–249, 251, 270,
273, 274, 290
C Cora pavonia 172
Cordia alliodora 104, 106
Caesalpinia coriaria 274 Craterostigma plantagineum 411, 412
Caloglossa 258 Croton 274
Calophyllum brasiliense 268 Croton glabellus 104, 106, 107
Campylopus savannarum 401, 406 Croton heliaster 274
Capparis flexuosa 274 Curatella americana 316, 326, 333
Capparis hastata 274 Cycas media 365
Capparis linearis 274 Cyphostemma currori 330
Capparis odoratissima 274
Carapa guianensis 107 D
Carex 274
Caryocar brasiliense 334 Darlingtonia 213
Catasetum viridiflavum 181 Dendrobium 211
Catopsis berteroniana 209, 216, 356 Desmantus virgatus 274
Cecropia 6, 82, 83, 93 Dicorynia guianensis 107
Cecropia obtusifolia 135, 136 Dictyonema glabratum 172–174
Cecropia peltata 104, 106 Didymopanax macrocarpum 337
Centrosema 352 Diectomis 321
Cephalocereus moritzianus 62 Digitaria 321
Cephalotus 213 Dionaea 213
Cercidium praecox 62, 63 Diplacus 115
Cereus jamacaru 64 Diplacus aurantiacus 117
Cereus lemairei 62 Dipteryx panamensis 135, 136
Cereus validus 282, 283, 285 Dischichia major 219
Ceriops australis 256, 257 Dischidia rafflesiana 218
Ceriops tagal v. australis 230 Discolobium 353
Ceriops tagal v. tagal 230 Drosera 213, 353, 355
Chamaegigas intrepidus 408, 413 Drosera roraimae 354
Chloris 321 Drosophyllum 213
Chorisia insignis 330 Drymoglossum piloselloides 194
Cissus sicyoides 274 Dubautia menziesii 437, 438
Claoxylon sandwicense 131, 132
Clostridium 346 E
Clusia 167, 176, 178, 200, 201, 203, 206,
207, 209, 212, 268, 403 Echeveria columbiana 430, 431
Scientific Name Index 445
Echinochloa 321 Heliamphora ionasii 355

Echinochloa polystachya 62, 323 Heliamphora minor 355
Entandrophragma angolense 141, 142 Heliamphora nutans 355
Eperna falcata 107 Heliamphora tatei 355
Epiphyllum 198 Helichrysum newii 431
Eragrostis 321 Herbaspirillum 346
Eriochloa 321 Heritiera littoralis 230
Eryotheca pubescens 337 Hibiscus tiliaceus 230, 249, 250
Erythroxylon cumanense 274 Hildegardia barteri 327
Espeletia 423, 435, 437 Hormosira banksii 258
Espeletia moritziana 426 Hyparrhenia rufa 321
Espeletia schultzii 424–426 Hyptis suaveolens 327
Espeletia spicata 435, 436
Espeletia timotensis 425, 426, 435, 436 I
Eucalyptus 8, 9, 364
Imperata 321
Eucalyptus saligna 8, 9, 11, 127
Inga 124
Eugenia 274
Isoëtes andicola 439
Euphorbia forbesii 131, 132 Isoëtes andina 439
Euterpe oleracea 61 Isoëtes hopei 439
Evolvulus tenuis 274 Isoëtes novo-grandensis 439
Excoecaria agallocha 230
J
F
Jacaranda copaia 107
Festuca arundinacea 342 Jacquinia revoluta 274
Ficus 168, 179, 200–202
K
Ficus bengalensis 178
Ficus microcarpa 178, 179 Kalanchoë 39, 157, 403
Ficus pertusa 178 Kalanchoë campanulata 403
Ficus trigonata 178 Kalanchoë daigremontiana 158
Fimbristylis cymosa 274 Kalanchoë miniata 403
Fimbristylis spadicea 274 Kalanchoë uniflora 194, 195, 203
Kyllingias 293
G
L
Genlisea 213
Geonema paraguanensis 59 Laguncularia racemosa 231, 237, 241, 245
Gloeocapsa 389 Lasiacis 321
Gloeocapsa magma 395 Leptochloa 274, 321
Gloeocapsa sanguinea 390, 395, 396, 404 Leptocoryphium 306
Goethalsia meiantha 104, 107 Leptocoryphium lanatum 316, 317, 339
Goupia glabra 107 Lobelia 423, 428
Guapira 274 Lobelia deckenii 426
Gunnera 349 Lobelia keniensis 425–427, 434, 435
Guzmania monostachia 124, 203 Lobelia telekii 429, 435
Gymnopogon 321 Lolium perenne 342
Lumnitzera littorea 230
H Lumnitzera racemosa 230
Lycopodium cernum 104
Haematoxylon praecox 62 M
Hakea platysperma 364
Haplocarpha rueppellii 430–432 Mauritia 5
Heliamphora 213, 353, 356 Mauritia flexuosa 299
Heliamphora heterodoxa 355 Maytenus karstenii 274
446 Scientific Name Index
Melaleuca 230 Pinguicula 213

Melastoma malabathricum 360 Piper aequale 135
Miconia 105 Piper auritum 135
Miconia acinodendron 361 Pisum auritum 117
Miconia albicans 336, 337, 360 Pisum hispidum 117
Miconia fallax 337 Pisum sativum 110, 255
Miconia ferruginata 337 Pitcairnia 382
Microchloa 321 Pitcairnia armata 402
Mimosa oligocantha 274 Pitcairnia bulbosa 402
Monstera 175 Pitcairnia integrifolia 153–155, 187
Monstera acuminata 181 Pitcairnia pruinosa 383, 402
Moringa ovalifolia 330 Platycerium 170
Myrmecochista 82 Platycerium grande 194
Myrsine parviflora 268 Platymiscium pinnatum 333
Pleurostima gounelleana 383
N Ploiarium alternifolium 104
Plumeria acuminata 330
Nauclea diderrichii 141, 142 Plumeria alba 64
Nepenthes 195, 213, 214 Plumeria rubra 330
Nepenthes × hookeriana 194 Podocarpus falcatus 8, 9, 11, 127
Nepenthes gracilis 214 Polypodium 170
Nostoc 349 Polypodium crassifolium 181
Polypodium vulgare 165, 183
O Polypompholyx 213
Portulaca rubricaulis 277, 279, 280
Oedematopus obovatus 403 Prosopis 350
Oplismenus 321 Prosopis juliflora 274
Opuntia caribea 64 Pseudobombax 325, 329
Opuntia excelsea 157 Pseudobombax chrysati 380
Opuntia wentiana 64, 274 Pseudobombax pilosus 330
Oroya peruviana 433 Pseudoborina ursina 170
Osbornea octodenta 230 Psilotum 170
Ouratea hexasperma 335, 337 Pteridium aquilinum 368
Oxycarpha suaedifolia 270, 274 Ptychomitrium vaginatum 406
Puya aequatorialis 428
P Puya clava-herculis 428
Puya hamata 428
Paenibacillus 346 Pyrrosia confluens 203
Paepalanthus bromelioides 356 Pyrrosia lanceolata 194
Palicourea rigida 325, 369 Pyrrossia longifolia 194
Pandanus 79
Panicum maximum 342 Q
Panicum stenoides 360
Paspalum 293, 321 Qualea grandiflora 335, 337
Paspalum plicatulum 274
Peltula tortuosa 400 R
Peperomia 203
Pereskia guamacho 64, 274, 326 Racocarpus fontinaloides 401, 406
Phalaenopsis grandifolia 194 Ramalina maciformis 405
Phalaenopsis violacea 194 Rheedia brasiliensis 268
Philodendron 175, 274 Rhipidocladum racemiflorum 181
Phoradendron mucronatum 274 Rhizobium 349
Phthirusa ovata 190 Rhizophora 227, 256
Pilosocereus ottonis 218 Rhizophora apiculata 230, 254, 256
Scientific Name Index 447
Rhizophora lamarckii 230 T

Rhizophora mangle 231, 233, 234, 236–238,
253, 255 Tabebuia chrysantha 325, 326
Rhizophora mucronata 239 Tabebuia orinocensis 325, 380
Rhizophora stylosa 230, 253–256 Tabernaemontana juruana 60
Richeria grandis 138, 360 Tectona grandis 333
Ritterocereus griseus 62 Tephrocactus floccosus 432, 433
Roridula 213 Tillandsia 186
Roupala montana 335, 337 Tillandsia araujei 383
Tillandsia fasciculata 187
S Tillandsia flexuosa 169, 218, 274, 289, 290,
402
Salix maritima 60 Tillandsia recurvata 169, 196, 274
Sarracenia 213 Tillandsia usneoides 187, 202, 203
Schefflera macrocarpa 334 Tmesipteris 170
Schomburgkia humboldtiana 218, 289 Tococa 82, 83
Sclerolobium paniculatum 337 Tococa guianensis 93
Scyphiphora hydrophylacea 230 Tococa occidentalis 82–84
Scytonema 389 Trachypogon 306, 321, 347, 366, 371
Scytonema crassum 390 Trachypogon plumosus 316, 317, 321, 371
Scytonema lyngbioides 404 Trichilia trifoliata 274
Scytonema myochrous 396 Tristachya 321
Selenicereus inermis 198 Tristerix aphyllus 189
Senecio 423, 428
Senecio brassica 428, 429, 435 U
Senecio johnstonii spp. cottonii 423
Senecio keniodendron 423, 425, 434–438 Utricularia 213, 216
Sesbania 353 Utricularia humboldtii 217, 219
Sesbania rostrata 353
Sesuvium portulacastrum 270, 273, 274,
V
277, 279–281
Sinorhizobium 349
Vaccinium reticulatum 438
Solidago altissima 117
Vanda tricolor 211
Sonneratia 240
Vellozia 410, 411
Sonneratia alba 230, 254
Sophora chrysophylla 438 Vellozia gigantea 411
Spondias lutea 326 Vigna 353
Spondias purpurea 330 Vismia 6
Sporobolus 321 Vochysia eliptica 337
Sporobolus virginicus 270, 274 Vochysia venezolensis 360
Stegolepis hitchcockii 300 Vriesea 186
Stictosiphonia 258
Stictosiphonia arbuscula 258 X
Stigonema 389
Stigonema mamillosum 396, 404 Xanthorrhoea 364, 365
Stigonema minutum 395 Xerophyta scabrida 408–410
Stigonema ocellatum 390, 396, 404 Xylocarpus granatum 230, 254
Stylosanthes 352 Xylocarpus mekongensis 230
Styphelia tameiameiae 437, 438 Xylomelum pyriforme 364
Subpilosocereus ottonis 270, 273, 274,
281–287 Y
Syagrus comosa 337
Syagrus orinocensis 380, 385 Yacaranda filicifolia 325, 380
Symplocos spicata 361 Yucca 364
Subject Index
A architecture
hydraulic 410
abscisic acid (ABA) 408, 413 arecife 328, 329
acclimation 142 arid 19–21
photosynthetic 113 climates 51
aeration 78 aroid 212
aerenchymas 353 ascorbate 122, 123
afforestation 7, 209 aspartate 317, 318, 344
Africa 21 atmosphere
Agavaceae 157 carbon-dioxide 89
agriculture 6, 11, 76 CO2 budget 13
shifting 6, 7, 73, 90 ATPase 109, 359, 360
air autecology 2, 3
δ 13 C values 89 auxin 237
air humidity 88
Al-accumulator 360, 361 B
algae 167, 195, 407
pleurococcoid 165, 195 bacteria 170, 259
altitude 54, 322, 323, 419, 420 N2 -fixing 213, 259, 346, 347
aluminium 336, 351, 356, 358 baobab 329, 332, 333
exclusion 359 basidiolichen
inclusion 360 cyanobacterial 171
ammonia 95 baskets
ant 93, 213, 219 humus 196
ant garden 219 β-diversity 379
ant house 218, 219 biodiversity 13, 305
ant leaf-cutter 93, 94 biological stress concept 118, 422
ant nest 83, 218, 219 biomes 18
ant plants 82 Bromeliaceae 157, 170, 183, 186, 193, 194,
ant-nest garden 185 203
antheraxanthin 121, 123 bromeliad 191, 192, 195–198, 200, 202, 203,
anthocyanin 410 212, 216, 218, 219, 267, 268, 281, 283,
antioxidant 121, 123, 152, 195 287, 289, 356, 383, 402, 403
antioxidative defence 412 atmospheric 169, 203
Apocynaceae 64 evolution 183
aquaporins 413 life forms 186–188
Araceae 170 tanks 176
449
450 Subject Index
Bromelioideae 186, 193–195 channel

bryophytes 165, 171, 195, 400, 404, 406, 407 Ca2+ 359
bundle sheath 318, 320 K+ 359
buttresses 234 malate 158
chaos 73
C deterministic 13, 42, 44–46, 77
Chenopodiaceae 317
13 C-discrimination 37 chilling 322, 422
C3 -grasses 320–323, 342, 343 chlorophyll 112–115, 118, 119, 126, 404,
C3 -photosynthesis 37, 131, 153, 156, 407, 409, 410
203–206, 308, 333, 397, 403 first excited singlet state 120, 126
C3 -plants 246, 251, 320 second excited singlet state 120
C3 /CAM intermediate 180, 181, 200, 203, third excited singlet state 120
402 triplet state 120
C4 -grasses 39, 307, 320–323, 342–344 chlorophyll a/b ratio 113, 115, 118
C4 -photosynthesis 37, 62, 132, 280, 308, chloroplast 107, 108, 113, 116, 408, 410
316, 317, 321, 323, 343 movement 119
evolution 320 chlorosis 358
C4 -plants 246, 251, 280, 318, 320, 412 citrate 207, 208, 360
Caatinga 62, 66 citric acid 206, 207
cacti 64, 268, 281, 283–286, 289, 433 clearing 139, 175, 305
epiphytic 197, 198 climate 17–20, 35, 51, 227
Caesalpiniaceae 63, 82 arid 227
calcium 351 humid 227
calmodulin 359 climatic oscillations 308
Calvin cycle 317 climax 275, 379
CAM see crassulacean acid metabolism climax association 73
(CAM) climax species 106, 136, 138, 139, 142, 175
cambium 188 climax theory 46, 73
canopy fluxes 314 climax tree 135
canthaxanthin 393 climbers 165, 175, 214
Capparidaceae 63 clock
carbon dioxide 89 circadian 152
carbon isotope discrimination 36 CO2
carbon-isotope ratios (δ 13 C) 37 compensation point 251
carbonic anhydrase 397 concentrating 172
carnivory 213, 214, 353, 356 elevated 256
carotene 111 recycling 157, 158, 161, 285, 288
carotenoid 119, 408–410 sinks 13
cation-exchange capacity (CEC) 336, 358 vertical gradients 89
cations CO2 concentration 320
divalent 358 horizontal profile 32
cavitation 152, 180, 430 vertical profile 31
cell wall 182, 358, 410 cohesion-tension theory 236, 237
cell wall elasticity (ε) 201 cold
cellular automata 42 resistance 422
cerrado 155, 189, 210, 301, 302, 305, 306, stress 422
313, 314, 317, 328, 329, 333, 335–337, compartmentation 238, 422
343, 344, 358, 360, 361, 364, 367, 370, compatible solutes 242, 243, 256, 258, 259,
371, 379 280, 412, 413, 434
concept 301 complexity 46, 71
cerradão 301, 302 conductance
chablis 73–75 hydraulic 151
chamaephytes 77 cooling
Subject Index 451
delay 422, 428 domatia 82, 93

resistance 422 dormancy 139
Crassulaceae 157 drainage 315
crassulacean acid metabolism (CAM) 37, 39, drought 149, 153, 185, 195, 203, 207, 309,
153, 156, 158, 192–195, 197, 200–203, 321, 323
205, 206, 219, 246, 251, 268, 281, 285, dry forest 39
289, 317, 320, 323, 402, 403, 430, 433, dynamics
439 non-linear 42, 45, 268
cycling 203
idling 203 E
phase II 160
phases 157, 158 ecology 1
phylogenetic tree 156, 157 physiological 2
crusts ecosystem
cryptogamic 379, 393 fluxes of CO2 313
cyanobacterial 389, 395 ecotone 306
cryoprotective solutes 434 savanna-desert 308
cryptogams 382 savanna-forest 306
cryptophytes 77 ecotypes 38
cushion 423, 424, 429, 430 El Niño 149
cyanobacteria 165, 167, 170, 195, 213, 259, elaiosomes 82
341, 346, 347, 349, 379, 382, 389, 390, elasticity 178, 201
392, 396, 399, 400, 404, 407 electron transport rates 129
crusts 347, 390, 396, 398, 399, 404 embolism 152, 180, 181, 410, 411, 430
endolithic 389 endosymbioses
epilithic 389 N2 -fixing 347, 349
mats 347, 348, 397, 399, 400 energy transfer 120, 121
N2 -fixing 349 environment
cytochrome b6,f-complex 109 heterogeneity 68
cytochrome f 114 epiphylls 170
cytorrhysis 410 epiphyte gardens 176
cytoskeleton 359 epiphytes 82, 165, 168, 183, 185, 186, 191,
194, 195, 197, 200, 209, 212–214, 219,
D 267, 289, 323, 380
diversity 170
D1 -protein 124, 126, 130, 136, 407 facultative 183
deforestation 6, 13 humus 176
dehdrins 413 epiphytism 188, 195
dehydration 408, 410 evolution 170, 191, 194
δ 13 C signatures 35 Espinar 62
desert 301, 308 etheric oils 329
desiccation 150, 156, 157, 173, 195, 253, Euphorbiaceae 62, 65, 157
258, 259, 381, 393, 400, 401, 403–405, eustress 70, 71
407–413 evaporation 51, 54, 151, 306, 315
desiccoplast 410 evaporative water loss 333
Didieraceae 62, 65, 157 evapotranspiration 22, 23, 152, 313, 329
distress 70, 71 exhaustion 70
diversity 6, 8, 45, 59, 67–70, 73, 77, 78, 209, exosymbioses
260, 401, 407 N2 -fixing 347
community 72
cross 72 F
floristic 67, 69, 73, 380, 401
functional 72, 151, 152 factor
genetic 71, 72 environmental 68, 69, 73, 84
452 Subject Index
stress 69 tolerance 435

fallow 7 Frostwechsel climate 419, 435, 438, 439
feedback 14
ferns 176, 195, 200, 201 G
fire 149, 305, 313, 317, 341, 347, 361–364,
366, 367, 369, 371, 372 gallery forest 296, 297
mineralization 365 gaps 73, 74, 119, 131, 175
flexibility 199, 203 gas analysis 29
flood plains 310, 323, 353 genes
flooding 62, 265 desiccation tolerance 413
flowering 323, 325–328, 364
genotype 71
fluorescence 26–28, 120, 126–129, 160, 161,
diversity 68
189, 394, 395, 404, 406, 409, 410, 430
giant rosette plants 423–428, 430, 433, 435,
quenching 126, 128
436
food webs 39
gibbsite 358
forest 5, 6, 11, 13, 51
glands
cactus 55, 56, 62–64
digestive 355
cloud 57, 59, 60, 68, 81, 166, 171, 175,
glutathione 405, 412
195, 209, 210, 360
goethite 358
deciduous 273–275, 307
grana 109, 113, 115, 410
drought-deciduous 52, 55, 56, 63
greenhouse effect 31, 366
elfin 54, 57, 59, 209, 210
growth rings 227
evergreen 12
floodplain 60
fog 57, 166, 175, 210 H
gallery 165, 209, 210, 360, 366
low land 209, 210 hair 432
lower montane 173, 174 halophytes 242, 243, 245, 270, 273, 275,
marsh 60 277, 280
monsoon 52, 55 facultative 241
montane 172, 209, 210, 265 hardening 70, 71
moss 165, 166 haustoria 188
secondary 6, 7 heat 439
semi-evergreen 12, 307 heat shock proteins 393, 413
semideciduous 92 heat-storage 426, 430, 438
shrub 210 heating 151
strata 85 hemi-epiphytes 165, 175, 180, 181, 185,
swamp 60, 265 191, 195, 200, 203, 209
thorn scrub 62 primary 176
thorn scrub and cactus 52 secondary 175
thornbush 55, 64–67, 153, 157 hemicryptophytes 77
tide 227 herbivory 93–95, 139, 151
trade-wind 52, 55, 56, 68 heterocytes 347, 398
types 51 homeostasis 73
upper montane 166, 210 homoiochlorophyllous 404–408, 412
forestry 11 homoiohydrous 185, 404, 411
freezing 422, 423, 430, 433, 434 constitutively 407
avoidance 422, 423, 428, 435 hormonal 326
equilibrium 434, 435 Huber-values 180
tolerance 422, 423, 428, 433, 435, 438 humid 19–21
tolerant 430 climates 51
freezing point 183 humidity 51
frost 422 humus 210
resistance 422, 423, 428, 433, 435 baskets 210
Subject Index 453
hydraulic architecture 151, 152, 180, 236, latitude 51

237, 333 leaf 78, 81
hydraulic conductivity 152, 181, 237, 333, leaf area index (LAI) 23, 86, 313
334 leaf flushing 95, 96
hydrophobin 173 leaf longevity 95, 135
hygrophytes 185 leaf-conductance 188
hypoxia 233, 236, 346 leaf-cutter ants 346
leaf-senescence 321
I leaf-water potential 188
leghemoglobin 349
ice 422–424, 430, 433–435, 438 Leguminosae 92, 349–353
ice age 308 Lentibulariaceae 213
igapó 60 lenticel 78, 79, 236, 353
indole acetic acid (IAA) 359 leptophyll 78, 82
infrared gas analysis (IRGA) 29 lianas 165, 168, 175, 178–180, 191
inselberg 209, 210, 379 lichens 165, 167, 171, 195, 379, 389, 400,
insulation 424, 433 401, 404–407
iron 358 biomass production 175
irradiance 22, 85, 94, 95, 103, 113–116, 118, cyanobacterial 175
119, 124, 126, 130, 131, 134, 135, 138, green alga 175
151, 153, 160, 175, 207, 253, 256, 258, photosynthesis 173
259, 320, 322, 333, 334, 393, 394, 396, productivity 174
410, 430, 432, 439 water saturation 172
isethionic acid 258 life forms 77, 183, 195, 401–403, 423
island light 103, 191, 204, 207, 286
vegetation 265, 266, 273, 275, 277, 280, blue 136, 142
281, 289, 290, 380, 396 far-red 137, 139
isohydric 151, 333, 334 quality 136
isoprene 152 red 137
isotope effect 34 signalling 136, 137
hydrological 34 ultraviolet 136, 259
metabolism 37 light compensation 103, 105, 106, 142, 194,
isotope-ratio 34 195
isotopes light curve 268
radioactive 33 light flecks 87, 94, 119, 131–136, 139, 175
stable 33 light harvesting 110
light intensity 86, 206, 393, 394
J light quality 86
light response 103–106, 127
jalca 419 light response curves 194
light saturation 103, 105, 106, 194
K light stress 136
light trap 110
K+ /Na+ -selectivity 236 light use efficiency (LUE) 133
kaolinite 358 lightning 362
K AUTSKY -effect 127 lignotubers 82, 330, 364, 367
Klimadiagramm 18, 19, 21, 39, 51, 52, 54, Liliaceae 157
56, 64, 67, 309 lipids 412, 413
litter 92, 93, 289, 342, 345, 346
L decomposition 257
degradation 6
Lambert–Beer’s law 85 liverworts 171
landslides 5 Llanos 5, 293, 294, 296, 325, 326, 329, 333,
lateritic 328, 329 341, 345, 351, 358, 362, 366, 369, 380
454 Subject Index
logistic equation 44 mosses 165, 167, 379, 389, 400, 401,

Loranthaceae 188–190 405–407
lutëin 111, 122–124 motion
lutëin cycle 122, 124 chaotic 43
random 43
M regular 43
movement
nyctinastic 424, 426
macroalgae 258 mucilage 426, 434
macrophanerophytes 77, 82 Müller-bodies 82
macrophyll 78, 82 myco-sporine 259
Madagascar 39, 65 mycorrhiza 91, 212, 346, 352
malate 208, 285, 287, 317, 344, 360, 433 myrmecophytes 82, 219, 289
malic acid 156, 158, 161, 206, 207
management N
forestry 8
manganese 383, 389
N-nutrition 38
mangroves 25, 189, 227, 270, 290
N2 -fixation 62, 92, 113, 175, 341, 346,
associate 247, 249, 270, 290 350–353, 398–400
coastal 227 NAD-malic enzyme 318, 319
coral 227 NADP-malic enzyme 318, 319
estuarine 227 NADP-reductase 109
fern 232 nanophyll 82
global distribution 231 nectaries
productivity 259, 260 extra floral 82
mats networks 42
cyanobacterial 389 night-bud 424, 426, 429
megaphyll 78 nitrate 95
membrane 359, 412, 413, 433 nitrification 365
fluidity 422 nitrogen 92–96, 110, 113–115, 117, 118,
rigidity 422 124, 142, 171, 179, 212, 258, 336, 339,
semipermeable 182 342, 347, 349, 352, 353, 365
mesophyll 78, 82 cycles 339, 340
mesophytes 185 nitrogen fixation 170
microbial mats 257, 259 nitrogen isotope effects 38
microphanerophytes 77, 82 nitrogen use efficiency (NUE) 115, 116, 181,
microphyll 78, 82 342
midday depression 152–156, 160, 247, 280, nitrogenase 346, 347, 349, 353
323, 333, 334 nodules
mimicry 191 external 349
Mimosaceae 62, 64, 82 internal 349
mineral nutrient 90 oxygen dilemma 349
mineral nutrition 42 root 349
mineralization 93, 95 stem 349, 353
minerals 366 noise 42
minimal quadrat 59, 67, 68 non-linear dynamics 77
mistletoes 124, 165, 188–191 non-photochemical quenching (NPQ) 25
carbon gain 189 nurse
modeling plant 266
vegetation 17, 18, 22 tree 8, 10
montane forest 39 nutrient 153, 170, 176, 186, 188, 190, 196,
morichales 296, 299 210, 213, 219, 267, 305, 306, 310, 313,
mosaic 321–323, 328, 335–338, 345, 346, 349,
oscillating 73, 275 352, 355, 358, 360, 433, 438
Subject Index 455
nutrient cycling 90, 91 photoperiod 327, 328

nutrition 13, 257, 344, 419, 438 photoprotection 124–126, 255
photorespiration 121, 125, 126, 255, 320,
O 334
photosynthesis 26, 36, 38, 39, 85, 89, 94, 95,
Orchidaceae 170 103, 115, 121, 131, 132, 134, 135, 139,
orchids 191, 192, 195, 197, 200, 201, 211, 156, 157, 188, 191, 193, 195, 200, 236,
218, 219, 268 246, 268, 320–323, 333, 342, 394, 398,
osmolytes 243 400–402, 405, 408, 409, 413, 430, 435
osmotic pressure 200, 201, 207 lichens 171, 172, 174
oxalate 360 mosses 171
oxidative stress 152, 253, 359, 404, 408, 412 photosynthetic capacity 37
oxygen photosynthetic electron transport 109
singlet activated 121, 122 photosynthetic light-use efficiency (LUE) 25
oxygen concentration 161 photosystem 107, 109, 113, 119, 121, 124,
oxygen radicals 121 125, 128, 129, 404, 407, 410
phyllosphere 165, 171, 213
P physiognomy 1
phytochrome 86, 137–139, 142, 326
palaeoclimatology 35 phytosiderophores 358
palms 5 Pictarnioideae 186
parasites 185 pigmentation 107
hemi 188 pigments 110, 111
parenchyma sunblocking 393
water-storage 287 pioneer species 106, 107, 136, 139, 142
park-land 296 pioneer tree 135
patchiness Piperaceae 170
spatiotemporal 268 piracy
pattern nutritional 212
spatiotemporal 42 Pitcairnioideae 193, 194
pectin 359 pitchers 214, 216, 355
PEP-carboxykinase 318, 319 plant geography 1, 2
petroglyphs 384, 387, 389 plant size 197
phanerophytes 77 plantation
phenogram 316, 326 exotic timber 8
phenological 151, 152, 313, 328 plasmodesmata 317
phenological diagram 316, 317, 326 plasmolysis 183
phenology 315, 323 plasticity 68, 69, 72, 142, 157, 202, 209
phenotypes 71, 107, 108, 114, 118, 288, 403 phenotypic 68, 72
phloem 188–190 plastoquinone 109, 121, 124, 127
phosphate 346, 358–360 pneumatophore aerenchyma 236
phosphoenol-pyruvate carboxylase (PEPC) pneumatophores 234
37, 158, 317, 319, 320, 344 poikilochlorophyllous 404, 408, 409
phosphorus 59, 70, 71, 91, 93, 257, 336, 344, poikilohydrous 185, 195, 404, 407, 408
352, 360 constitutively 407
photo-oxidation 407 pollutants 366
photochemical work 119, 120, 126, 130 polyols 255, 256
photodamage 120, 130, 136, 137, 393, 406 population biology 45
photodestruction 126 population dynamics 44
photoinhibition 121, 126, 129, 130, 136, potassium 70, 71, 258, 336
137, 153, 155, 157, 161, 181, 207, 246, prairie 339, 341, 342
253, 333–335, 394, 396, 439 precipitation 51, 313, 314, 323
acute 130, 254, 335 predictability 43, 44
chronic 130, 254, 335 preservation 6
456 Subject Index
pressure nutrient concentration 90

hydrostatic 182 reactive oxygen species (ROS) 121, 123,
osmotic 182 255, 405, 412
turgor 182 recovery 407, 408
pressure chamber 183 red:far red ratio 86
pressure probe 183 reflectance index
productivity 5, 6, 8, 288–290, 310, 320, 323, anthocyanin 24, 25
436, 437 photochemical 24
ecosystem 13 rehydration 405, 408–410, 412
Proteaceae 364 relative air humidity (RH) 88, 196
proton pumps 158 remote sensing 17, 23
proton-electrochemical gradient 109 resistance 70, 71
pseudostems 178, 179 respiration 6, 13, 36, 89, 103, 105, 135, 174,
Psilophytatae 170 236, 321, 326, 349, 404, 405, 409, 410
Psilotatae 170 cyanide resistant pathway 105
psychrometry 183 soil 78, 89, 93
pubescence 426, 428 restingas 265
puna 419 resurrection plants 404, 412
pyrophilous 365 rhizosphere 347, 359
pyrophytes 362, 364 rhythmicity
páramos 419 annual 327
ribulosebisphosphate carboxylase/oxygenase
Q (RuBISCO) 37, 113, 114, 121, 125,
134, 157, 158, 161, 317, 319, 320, 397,
quantum yield 103, 105, 106, 113, 115, 129, 398
130, 335, 337, 395, 398, 408 rivers
effective 25 black-water 389
quenching white-water 389
non-photochemical 126, 129, 130 root 78, 79, 92, 93, 95, 186, 283–285, 317,
photochemical 126, 129, 130 328, 333, 353, 358, 409
adventitious 176, 178, 212
R aerial 168, 176, 210, 211, 234
buttress 78, 79
radiation 23 stilt 78, 79, 234
absorption 23 root nodule 39, 341, 347
infrared 30 root pressure 411
reflection 23 rosette
transmission 23 acaulescent 423, 424, 430, 431
radical scavengers 256
radiometry S
micro-wave 25
rainforest 3, 12, 13, 21, 39, 55, 58, 73, 78, Sahel 308–310, 344
166, 168, 171, 172, 175, 199, 200, 301, salinas 269, 275, 281, 283, 287
306, 307, 380 salinity 227, 233, 236–238, 241, 246–248,
Atlantic 265, 267 251, 253–259, 281, 283, 285, 289, 350
destruction 11 salt
evergreen 52, 55 accumulation 240
lower montane 54, 91 compartmentation 241, 242
lowland 54, 57 dilution 238, 241
montane 57 exclusion 237–239, 281, 283, 287
seasonal 52, 55, 56 excretion 237, 238, 245
semi-evergreen 55 glands 238, 239, 243, 245
upper montane 54, 57 inclusion 237–239, 277, 279, 280
rainwater marsh 269, 270
Subject Index 457
spray 248, 289 stemflow

succulence 241 nutrient concentration 90
sand dunes 270 stomata 2, 36, 88, 107, 119, 131, 132, 134,
Santalales 188 135, 151–153, 155, 157, 158, 160, 188,
savanna 3, 5, 6, 11, 21, 39, 55, 155, 209, 210, 195, 200, 203, 246, 252, 253, 285, 320,
219, 293, 313, 379, 380, 400 328, 333, 334, 439
concept 301 stomatal conductance 36, 246, 321
grass 299 stomatal diffusion 37
herb 299 stomatal regulation 37
hyperseasonal 60, 304, 311 strain 70
marsh 60, 299, 304 elastic 126, 153, 422
problem 305 plastic 126, 153, 422
rock 379 stranglers 176, 178, 200, 203, 209
seasonal 304, 311 strata 87, 88, 90
semi-seasonal 304 stratification 84
saxicolously 380 stress 70, 183
sclerophyllous 329 avoidance 281, 287
scytonemin 259, 393 high-irradiance 118
seasonality 13, 20, 51, 59, 149, 301, 304, light 103
313, 323, 381 osmotic 434
secondary thickening 175, 176, 188 salinity 189
secretion stress concept 69, 70
enzyme 356 stressor 69–71, 191
seed 138 stroma 109
dormancy 138 strontium 42
germination 138 succession 73, 106, 107, 138, 139, 275
seed banks 138 progressive 73
seedling 138 regressive 73
survival 138, 139 succulence 82, 197, 200, 201, 239–243, 249,
seedling bank 138 279–281, 287, 326, 380
selection 72 stem 329–331
natural 1 sulphate 280, 360
self-ignition 362 sun plants 85, 103, 105–107, 110, 114, 124,
semi-arid 21 130, 139, 142, 194, 288
senescence 151 supercooling 422, 428, 430, 434–436, 438
shade and sun plants 135 synecology 2, 3
shade plants 103, 105–107, 110, 113, 114,
118, 124, 130, 139, 142, 288 T
shade-demanding 195
shade-tolerant 195 tank 183, 186, 187, 196, 197, 212, 216–219,
silviculture 7, 8 268, 281, 287
singlet state 121 temperature 22, 88, 153, 204–206, 322, 386,
soil 90–94, 96, 151, 176, 212, 285, 315, 323, 419, 422, 423, 427, 428, 430, 432–437,
335, 337–339, 341, 342, 346, 347, 349, 439
351, 353, 356, 358, 360, 364, 366, 430 air 88
ferralitic 358 soil 88
inundated 234 tendrils 175
moisture 149 tentacles 353–355
swampy 233 termites 296, 298, 346
water capacity 313, 314 thawing 430
spill over 121, 125 thermophily 174
stable isotope analysis 33 therophytes 77
starvation 156, 157, 253 thornbush 39, 270
stem 82 throughfall
458 Subject Index
nutrient concentration 90 absorption index 24

thylakoid 107, 109, 115, 121, 122, 130, black 60, 62
408–410 climate effect 34
tides 227 white 60, 62
Tillandsioideae 186, 193, 194 water content
timing relative 200
phenological 326 water potential 152, 181, 182, 195, 200, 236,
Tintenstrich 389 326, 333, 434
tocopherol 195 leaf 150
tonoplast 158 water potential gradient 189
trade winds 59 water relations 35, 72, 181, 197, 200–202,
transpiration 35–37, 88, 119, 151, 158, 181, 204, 433
188, 189, 200, 249, 251–253, 286, 314, water storage 152, 157, 183, 287, 330, 331,
315, 321, 323, 328, 331, 333, 334, 336, 438
343, 427, 430, 432 mosses 171
cuticular 157 water storage capacity 196, 200
transpirational cooling 252, 427 water storage parenchyma 286
tree 323 water storage tissue 198
saplings 136, 175 water stress 326
tree rings 149 water transport 152
tree seedling 149 water use efficiency (WUE) 36, 158, 181,
trichomes 186, 196, 213 188, 189, 199, 207, 246, 251–253, 317,
bromeliad 155 320
epidermal 183, 186, 192, 210 wetland 60, 304, 353
triplet state 121 wilting 150
tropics 12 wood density 152
cold 1, 3, 4, 419
dry 1, 4 X
warm 1, 3, 4
wet 1, 4 xanthophyll 111, 121, 122, 124, 126, 254,
turgor pressure 200, 201, 203, 317, 424, 434 256, 408
tussock 423, 424, 429 xanthophyll cycle 25, 121, 124, 136, 255,
334, 405
V xeromorphy 328, 432
xerophytes 185, 380
vacuole 158, 161
xylem 152, 188, 237
vapour pressure deficit 156
water potentials 181
variation
xylem conductivity 236
spatiotemporal 72
xylem sap 189
vegetation index 23, 24
xylem tension 236
velamen radicum 210, 211
xylem vessels 179
vines 165, 175, 191
xylohemicryptophytes 82, 367
violaxanthin 121–123
xylopodia 82, 330, 367
Viscaceae 188
vitrification 412
Y
vivipary 237, 238
volume flow 182
yerbazal 299, 300
vulcanism 362
Young’s modulus 178
várzea 60
W Z
water 13, 195, 197, 305, 306, 313–315, zeaxanthin 121–123, 136, 195, 254, 393, 407
321–323, 328, 331, 333, 370, 401, 404, zeaxanthin cycle 121–124
405, 407, 409–412, 419, 434, 438 zono-biomes 18

Physiological Ecology of Tropical Plants

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

Physiological Ecology of Tropical Plants

Uploaded by

Copyright:

Available Formats

Physiological Ecology of Tropical Plants

Cover photo showing the fringe of a rainforest on Sierrania Parú, Guayana

ISBN-13: 978-3-540-71792-8 e-ISBN-13: 978-3-540-71793-5

Recently, in many countries, particularly in the industrialized parts of the world,

I am indebted to former students and postdoctoral fellows in my laboratory and

Darmstadt, February 1997 Ulrich Lüttge

Darmstadt, April 2007 Ulrich Lüttge

des Gedankens reflectiert. Kosmos, vol. IV, 1858

2 Large-Scale Sensing and Diagnosis in Relation

3 Tropical Forests. I. Physiognomy and Functional Structure . . . . . . . . . 51

3.3 Horizontal Structure and Diversity of Tropical Forests . . . . . . . . . . . . 67

4 Tropical Forests. II. Ecophysiological Responses to Light . . . . . . . . . . . 103

5 Tropical Forests. III. Ecophysiological Responses to Drought . . . . . . . . 149

7 Tropical Forests. V. Mangroves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

8 Ecosystems of Coastal Sand Plains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265

8.2 Salinas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269

9 Savannas. I. Physiognomy, Terminology and Ecotones:

11.4.1 Cyanobacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404

Scientific Name Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443

Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449

1.1 Historical Background of Ecophysiology

1.2 The Tropics

Applying the approach of physiological ecology to understanding ecological func-

Fig. 1.2A,B Diurnal (A) and

• phyto-geographically the tropics are indicated by the distribution of palms;

1.3 Tropical Forests and Savannas: Their Emotional,

Commercially it is important to remember that most tropical countries have to

large. However, if short-term cultivation after slash-and-burn action is followed by

Fig. 1.7 Area covered by

Table 1.1 Comparison of disadvantages and advantages of exotic tree plantations

in Ethiopia in a region where some remnants of original native forest of Podocarpus

A E. saligna plantation Native forest

1.4 The Destruction of Tropical Forest

Of particularly wide public concern is the ongoing destruction of tropical rainforest.

In the discussion of “global change” it has become increasingly important to develop

2.2 Climatic Relations and Vegetation Modeling

2.2.1 The Klimadiagramm

Box 2.1 Klimadiagramm after Walter (1973)

Box 2.1 (Continued)

For a large-scale evaluation of a certain area, or even an entire continent, one

2.2.2 Vegetation Modeling Based on Irradiance and Water Budgets

Climatic conditions like irradiance and temperature affect the water-vapour

availability in turn are modulated by other climatic factors. Thus, it is possible to

2.3 Remote Sensing Using Radiation

Remote sensing of the biosphere is based on the analysis of electromagnetic radi-

2.3.1 Reflection and Absorption

The analysis of reflection, absorption and transmission of radiation by individ-

Fig. 2.3 Relationship be-

the epoxidation-state of components of the xanthophyll cycle which is related to the

Fig. 2.4 Correlations between 0.00

In many of these examples analysis of temporal variations is essential, integrated

Remote fluorescence excitation is possible using powerful lasers. A laser-induced

Fig. 2.5 Laser-induced fluo-

2.4 Gas Analysis

Box 2.2 Infrared active and non-active gases

An IRGA measures the specific absorption of infrared radiation with a char-

Fig. 2.9 Vertical profiles of

Fig. 2.10 Infrared-radiation

2.5 Stable Isotope Analysis

Element Isotope Abundance (%)

x may be heavier or lighter than x, one can obtain the isotope-ratio δx in ‰ as

In this relationship a gives 13 C discrimination due to CO2 -diffusion in air (4.4‰)

where δa and δp are given in ‰. Determined by the contribution from respiratory