Ecology of Phytoplankton

The Ecology of Phytoplankton
C. S. Reynolds
cambridge university press
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo
Cambridge University Press

The Edinburgh Building, Cambridge cb2 2ru, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521844130
© Cambridge University Press 2006
This publication is in copyright. Subject to statutory exception and to the provision of

relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
First published in print format 2006
isbn-13 978-0-511-19094-0 eBook (EBL)

isbn-10 0-511-19094-8 eBook (EBL)
isbn-13 978-0-521-84413-0 hardback

isbn-10 0-521-84413-4 hardback
isbn-13 978-0-521-60519-9 paperback

isbn-10 0-521-60519-9 paperback
Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
This book is dedicated to
my wife, JEAN, to whom its writing
represented an intrusion into
domestic life, and to Charles Sinker,
John Lund and Ramón Margalef. Each is
a constant source of inspiration to me.
Contents
Preface page ix
Acknowledgements xii
Chapter 1. Phytoplankton 1
1.1 Definitions and terminology 1
1.2 Historical context of phytoplankton studies 3
1.3 The diversification of phytoplankton 4
1.4 General features of phytoplankton 15
1.5 The construction and composition of freshwater
phytoplankton 24
1.6 Marine phytoplankton 34
1.7 Summary 36
Chapter 2. Entrainment and distribution in the pelagic 38

2.1 Introduction 38
2.2 Motion in aquatic environments 39
2.3 Turbulence 42
2.4 Phytoplankton sinking and floating 49
2.5 Adaptive and evolutionary mechanisms for
regulating ws 53
2.6 Sinking and entrainment in natural turbulence 67
2.7 The spatial distribution of phytoplankton 77
2.8 Summary 90
Chapter 3. Photosynthesis and carbon acquisition in

phytoplankton 93
3.1 Introduction 93
3.2 Essential biochemistry of photosynthesis 94
3.3 Light-dependent environmental sensitivity of
photosynthesis 101
3.4 Sensitivity of aquatic photosynthesis to carbon
sources 124
3.5 Capacity, achievement and fate of primary
production at the ecosystem scale 131
3.6 Summary 143
Chapter 4. Nutrient uptake and assimilation in

phytoplankton 145
4.1 Introduction 145
4.2 Cell uptake and intracellular transport of
nutrients 146
4.3 Phosphorus: requirements, uptake, deployment in
phytoplankton 151
viii CONTENTS
4.4 Nitrogen: requirements, sources, uptake and

metabolism in phytoplankton 161
4.5 The role of micronutrients 166
4.6 Major ions 171
4.7 Silicon: requirements, uptake, deployment in
phytoplankton 173
4.8 Summary 175
Chapter 5. Growth and replication of phytoplankton 178

5.1 Introduction: characterising growth 178
5.2 The mechanics and control of growth 179
5.3 The dynamics of phytoplankton growth and
replication in controlled conditions 183
5.4 Replication rates under sub-ideal conditions 189
5.5 Growth of phytoplankton in natural
environments 217
5.6 Summary 236
Chapter 6. Mortality and loss processes in phytoplankton 239

6.2 Wash-out and dilution 240
6.3 Sedimentation 243
6.4 Consumption by herbivores 250
6.5 Susceptibility to pathogens and parasites 292
6.6 Death and decomposition 296
6.7 Aggregated impacts of loss processes on
phytoplankton composition 297
6.8 Summary 300
Chapter 7. Community assembly in the plankton: pattern,

process and dynamics 302
7.2 Patterns of species composition and temporal
change in phytoplankton assemblages 302
7.3 Assembly processes in the phytoplankton 350
7.4 Summary 385
Chapter 8. Phytoplankton ecology and aquatic ecosystems:

mechanisms and management 387
8.2 Material transfers and energy flow in pelagic
systems 387
8.3 Anthropogenic change in pelagic environments 395
8.4 Summary 432
8.5 A last word 435
Glossary 437
Units, symbols and abbreviations 440
CONTENTS ix
References 447
Index to lakes, rivers and seas 508
Index to genera and species of
phytoplankton 511
Index to genera and species of other
organisms 520
General index 524
Preface
This is the third book I have written on the sub- the present volume to address more overtly the
ject of phytoplankton ecology. When I finished marine phytoplankton, I have set out to construct
the first, The Ecology of Freshwater Phytoplankton a new perspective on the expanded knowledge
(Reynolds, 1984a), I vowed that it would also be base. I have to say at once that the omission of
my last. I felt better about it once it was pub- ‘freshwater’ from the new title does not imply
lished but, as I recognised that science was mov- that the book covers the ecology of marine plank-
ing on, I became increasingly frustrated about ton in equivalent detail. It does, however, signify
the growing datedness of its information. When a genuine attempt to bridge the deep but wholly
an opportunity was presented to me, in the form artificial chasm that exists between marine and
of the 1994 Ecology Institute Prize, to write my freshwater science, which political organisation
second book on the ecology of plankton, Vege- and science funding have perpetuated.
tation Processes in the Pelagic (Reynolds, 1997a), I At a personal level, this wider view is a satisfy-
was able to draw on the enormous strides that ing thing to develop, being almost a plea for abso-
were being made towards understanding the part lution – ‘I am sorry for getting it wrong before,
played by the biochemistry, physiology and pop- this is what I should have said!’ At a wider level, I
ulation dynamics of plankton in the overall func- am conscious that many people still use and fre-
tioning of the great aquatic ecosystems. Any feel- quently cite my 1984 book; I would like them to
ing of satisfaction that that exercise brought to know that I no longer believe everything, or even
me has also been overtaken by events of the last very much, of what I wrote then. As if to empha-
decade, which have seen new tools deployed to sise this, I have adopted a very similar approach
the greater amplification of knowledge and new to the subject, again using eight chapters (albeit
facts uncovered to be threaded into the web of with altered titles). These are developed accord-
understanding of how the world works. ing to a similar sequence of topics, through mor-
Of course, this is the way of science. There phology, suspension, ecophysiology and dynam-
is no scientific text that can be closed with a ics to the structuring of communities and their
sigh, ‘So that’s it, then’. There are always more functions within ecosystems. This arrangement
questions. I actually have rather more now than allows me to contrast directly the new knowl-
I had at the same stage of finishing the 1984 vol- edge and the understanding it has rendered
ume. No, the best that can be expected, or even redundant.
hoped for, is a periodic stocktake: ‘This is what So just what are these mould-breaking
we have learned, this is how we think we can findings? In truth, they impinge upon the sub-
explain things and this is where it fits into what ject matter in each of the chapters. Advances in
we thought we knew already; this will stand until microscopy have allowed ultrastructural details
we learn something else.’ This is truly the way of planktic organisms to be revealed for the first
of science. Taking observations, verifying them time. The advances in molecular biology, in par-
by experimentation, moving from hypothesis to ticular the introduction of techniques for iso-
fact, we are able to formulate progressively closer lating chromosomes and ribosomes, fragmenting
approximations to the truth. them by restriction enzymes and reading genetic
In fact, the second violation of my 1984 vow sequences, have totally altered perceptions about
has a more powerful and less high-principled phyletic relationships among planktic taxa and
driver. It is just that the progress in plankton suppositions about their evolution. The classifica-
ecology since 1984 has been astounding, turning tion of organisms is undergoing change of revolu-
almost each one of the first book’s basic assump- tionary proportions, while morphological varia-
tions on its head. Besides widening the scope of tion among (supposedly) homogeneous genotypes
xii PREFACE
questions the very concept of putting names phytoplankton photosynthesis that is leaked or
to individual organisms. At the scale of cells, actively discharged into the water. Far from hold-
the whole concept of how they are moved in ing to the traditional view of the pelagic food
the water has been addressed mathematically. chain – algae, zooplankton, fish – plankton ecol-
It is now appreciated that planktic cells experi- ogists now have to acknowledge that marine
ence critical physical forces that are very differ- food webs are regulated ‘by a sea of microbes’
ent from those affecting (say) fish: viscosity and (Karl, 1999), through the muliple interactions of
small-scale turbulence determine the immediate organic and inorganic resources and by the lock
environment of microorganisms; surface tension of protistan predators and acellular pathogens
is a lethal and inescapable spectre; while shear (Smetacek, 2002). Even in lakes, where the case
forces dominate dispersion and the spatial dis- for the top–down control of phytoplankton by
tributions of populations. These discoveries flow herbivorous grazers is championed, the other-
from the giant leaps in quantification and mea- wise dominant microbially mediated supply of
surements made by physical limnologists and resources to higher trophic levels is demonstra-
oceanographers since the early 1980s. These have bly subsidised by components from the littoral
also impinged on the revision of how sinking (Schindler et al., 1996; Vadeboncoeur et al., 2002).
and settlement of phytoplankton are viewed and There have been many other revolutions. One
they have helped to consolidate a robust theory more to mention here is the progress in ecosys-
of filter-feeding by zooplankton. tem ecology, or more particularly, the bridge
The way in which nutrients are sequestered between the organismic and population ecology
from dilute and dispersed sources in the water and the behaviour of entire systems. How ecosys-
and then deployed in the assembly and replica- tems behave, how their structure is maintained
tion of new generations of phytoplankton has and what is critical to that maintenance, what
been intensively investigated by physiologists. the biogeochemical consequences might be and
Recent findings have greatly modified percep- how they respond to human exploitation and
tions about what is meant by ‘limiting nutrients’ management, have all become quantifiable. The
and what happens when one or other is in short linking threads are based upon thermodynamic
supply. As Sommer (1996) commented, past sup- rules of energy capture, exergy storage and struc-
positions about the repercussions on community tural emergence, applied through to the systems
structure have had to be revised, both through level (Link, 2002; Odum, 2002).
the direct implications for interspecific compe- In the later chapters in this volume, I attempt
tition for resources and, indirectly, through the to apply these concepts to phytoplankton-based
effects of variable nutritional value of potential systems, where the opportunity is again taken
foods to the web of dependent consumers. to emphasise the value to the science of ecol-
Arguably, the greatest shift in understanding ogy of studying the dynamics of microorganisms
concerns the way in which the pelagic ecosys- in the pursuit of high-order pattern and assem-
tem works. Although the abundance of plank- bly rules (Reynolds, 1997, 2002b). The dual chal-
tic bacteria and the relatively vast reserve of lenge remains, to convince students of forests
dissolved organic carbon (DOC) had long been and other terrestrial ecosystems that microbial
recognised, the microorganismic turnover of car- systems do conform to analogous rules, albeit
bon has only been investigated intensively dur- at very truncated real-time scales, and to per-
ing the last two decades. It was soon recog- suade microbiologists to look up from the micro-
nised that the metazoan food web of the open scope for long enough to see how their knowl-
oceans is linked to the producer network via edge might be applied to ecological issues.
the turnover of the microbes and that this state- I am proud to acknowledge the many people
ment applies to many larger freshwater systems who have influenced or contributed to the sub-
as well. The metabolism of the variety of subject matter of this book. I thank Charles Sinker
stances embraced by ‘DOC’ varies with source and for inspiring a deep appreciation of ecology and
chain length but a labile fraction originates from its mechanisms. I am grateful to John Lund, CBE,
PREFACE xiii
FRS for the opportunity to work on phytoplank- among whom special thanks are due to Tony
ton as a postgraduate and for the constant inspi- Irish, Sheila Wiseman, George Jaworski and Brian
ration and access to his knowledge that he has Godfrey. Peter Allen, Christine Butterwick, Julie
given me. Of the many practising theoretical ecol- Corry (later Parker), Mitzi De Ville, Joy Elsworth,
ogists whose works I have read, I have felt the Alastair Ferguson, Mark Glaister, David Gouldney,
greatest affinity to the ideas and logic of Ramón Matthew Rogers, Stephen Thackeray and Julie
Margalef; I greatly enjoyed the opportunities to Thompson also worked with me at particular
discuss these with him and regret that there will times. Throughout this period, I was privileged
be no more of them. to work in a ‘well-found’ laboratory with abun-
I gratefully acknowledge the various scien- dant technical and practical support. I freely
tists whose work has profoundly influenced par- acknowledge use of the world’s finest collection
ticular parts of this book and my thinking gen- of the freshwater literature and the assistance
erally. They include (in alphabetical order) Sal- provided at various times by John Horne, Ian
lie Chisholm, Paul Falkowski, Maciej Gliwicz, Pettman, Ian McCullough, Olive Jolly and Mari-
Phil Grime, Alan Hildrew, G. E. Hutchinson, Jörg lyn Moore. Secretarial assistance has come from
Imberger, Petur Jónasson, Sven-Erik Jørgensen, Margaret Thompson, Elisabeth Evans and Joyce
Dave Karl, Winfried Lampert, John Lawton, John Hawksworth. Trevor Furnass has provided abun-
Raven, Marten Scheffer, Ted Smayda, Milan dant reprographic assistance over many years. I
Straškraba, Reinhold Tüxen, Anthony Walsby and am forever in the debt of Hilda Canter-Lund, FRPS
Thomas Weisse. I have also been most fortu- for the use of her internationally renowned pho-
nate in having been able, at various times, to tomicrographs.
work with and discuss many ideas with col- A special word is due to the doctoral students
leagues who include Keith Beven, Sylvia Bonilla, whom I have supervised. The thirst for knowl-
Odécio Cáceres, Paul Carling, Jean-Pierre Descy, edge and understanding of a good pupil gener-
Mónica Diaz, Graham Harris, Vera Huszar, Dieter ally provide a foil and focus in the other direc-
Imboden, Kana Ishikawa, Medina Kadiri, Susan tion. I owe much to the diligent curiosity of Chris
Kilham, Michio Kumagai, Bill Li, Vivian Monte- van Vlymen, Helena Cmiech, Karen Saxby (now
cino, Mohi Munawar, Masami Nakanishi, Shin- Rouen), Siân Davies, Alex Elliott, Carla Kruk and
Ichi Nakano, Luigi Naselli-Flores, Pat Neale, Søren Phil Davis.
Nielsen, Judit Padisák, Fernando Pedrozo, Victor My final word of appreciation is reserved for
Smetaček, Ulrich Sommer, José Tundisi and acknowledgement of the tolerance and forbear-
Peter Tyler. I am especially grateful to Cather- ance of my wife and family. I cheered through
ine Legrand who generously allowed me to use many juvenile football matches and dutifully
and interpret her experimental data on Alexan- attended a host of ballet and choir performances
drium. Nearer to home, I have similarly benefited and, yes, it was quite fun to relive three more
from long and helpful discussions with such erst- school curricula. Nevertheless, my children had
while Windermere colleagues as Hilda Canter- less of my time than they were entitled to expect.
Lund, Bill Davison, Malcolm Elliott, Bland Finlay, Jean has generously shared with my science the
Glen George, Ivan Heaney, Stephen Maberly, Jack full focus of my attention. Yet, in 35 years of mar-
Talling and Ed Tipping. riage, she has never once complained, nor done
During my years at The Ferry House, I was less than encourage the pursuit of my work. I am
ably and closely supported by several co-workers, proud to dedicate this book to her.
Acknowledgements
Except where stated, the illustrations in this book the copyright of Blackwell Science (the specific
are reproduced, redrawn or otherwise slightly sources are noted in the figure captions) and are
modified from sources noted in the individual redrawn by permission.
captions. The author and the publisher are grate-
Figures 3.7, 3.15, 4.6 and 7.2.3 (or parts thereof)
ful to the various copyright holders, listed below,
are redrawn from Freshwater Biology by permission
who have given permission to use copyright mate-
of Blackwell Science.
rial in this volume. While every effort has been
made to clear permissions as appropriate, the Figure 3.7 incorporates items redrawn from Bio-
publisher would appreciate notification of any logical Reviews with acknowledgement to the Cam-
omission. bridge Philosophical Society.
Figures 1.1 to 1.8, 1.10, 2.8 to 2.13, 2.17, 2.20 to Figure 5.9 is redrawn by permission of John Wiley
2.31, 3.3 to 3.9, 3.16 and 3.17, 5.20, 6.2, 6.7, 6.11. & Sons Ltd.
7.6 and 7.18 are already copyrighted to Cambridge
University Press. Figure 5.14 is redrawn by permission of Springer-
Verlag GmbH.
Figure 1.9 is redrawn by permission of Oxford
University Press. Figures 5.15 to 5.17, 5.19, 6.8 and 6.9 are redrawn
by permission of SpringerScience+Business BV.
Figure 1.11 is the copyright of the American Soci-
ety of Limnology and Oceanography. Figures 6.12, 6.15, 7.1 to 7.4, 7.9 and 8.6 are repro-
duced from Journal of Plankton Research by permis-
Figures 2.1 and 2.2, 2.5 to 2.7, 2.15 and 2.16, 2.18 sion of Oxford University Press. Dr K. Bruning also
and 2.19, 3.12, 3.14, 3.19, 4.1, 4.3 to 4.5, 5.1 to gave permission to produce Fig. 6.12.
5.5, 5.8, 5.10, 5.12 and 5.13, 5.20 and 5.21, 6.1,
6.2, 6.4, 6.14, 7.8, 7.10 and 7.11, 7.14, 7.16 and 7.17, Figure 7.7 is redrawn by permission of the Direc-
7.20 and 7.22 are redrawn by permission of The tor, Marine Biological Association.
Ecology Institute, Oldendorf. Figures 7.12 to 7.14, 7.24 and 7.25 are redrawn
Figures 2.3 and 4.7 are redrawn from the source from Verhandlungen der internationale Vereini-
noted in the captions, with acknowledgement to gung f ür theoretische und angewandte Limnolo-
Artemis Press. gie by permission of Dr E. Nägele (Publisher)
(http://www.schwezerbart.de).
Figures 2.4, 3.18, 5.11, 5.18, 7.5, 7.15, 8.2 and 8.3
are redrawn from the various sources noted in Figure 7.19 is redrawn with acknowledgement to
the respective captions and with acknowledge- the Athlone Press of the University of London.
ment to Elsevier Science, B.V.
Figure 7.21 is redrawn from Aquatic Ecosystems
Figure 2.14 is redrawn from the British Phycologi- Health and Management by permission of Taylor
cal Journal by permission of Taylor & Francis Ltd & Francis, Inc. (http://www.taylorandfrancis.com).
(http://www.tandf.co.uk/journals).
Figure 8.1 is redrawn from Scientia Maritima by
Figure 3.1 is redrawn by permission of Nature permission of Institut de Ciències del Mar.
Publishing Group.
Figures 8.5, 8.7 and 8.8 are redrawn by permis-
Figures 3.2, 3.11, 3.13, 4.2, 5.6, 6.4, 6.6, 6.9, sion of the Chief Executive, Freshwater Biological
6.10 and 6.13 come from various titles that are Association.
Chapter 1
Phytoplankton
and are not necessarily contested. Thus, ‘plank-

1.1 Definitions and terminology ton’ excludes other suspensoids that are either
non-living, such as clay particles and precipitated
The correct place to begin any exposition of a chemicals, or are fragments or cadavers derived
major component in biospheric functioning is from biogenic sources. Despite the existence of
with precise definitions and crisp discrimination. the now largely redundant subdivision tychoplank-
This should be a relatively simple exercise but for ton (see Box 1.1), ‘plankton’ normally comprises
the need to satisfy a consensus of understand- those living organisms that are only fortuitously
ing and usage. Particularly among the biological and temporarily present, imported from adjacent
sciences, scientific knowledge is evolving rapidly habitats but which neither grew in this habitat
and, as it does so, it often modifies and outgrows nor are suitably adapted to survive in the truly
the constraints of the previously acceptable ter- open water, ostensibly independent of shore and
minology. I recognised this problem for plank- bottom. Such locations support distinct suites of
ton science in an earlier monograph (Reynolds, surface-adhering organisms with their own dis-
1984a). Since then, the difficulty has worsened tinctive survival adaptations.
and it impinges on many sections of the present ‘Suspension’ has been more problematic, hav-
book. The best means of dealing with it is to ing quite rigid physical qualifications of dens-
accept the issue as a symptom of the good health ity and movement relative to water. As will be
and dynamism of the science and to avoid con- rehearsed in Chapter 2, only rarely can plank-
straining future philosophical development by a ton be isopycnic (having the same density) with
redundant terminological framework. the medium and will have a tendency to float
The need for definitions is not subverted, how- upwards or sink downwards relative to it. The
ever, but it transforms to an insistence that those rate of movement is also size dependent, so
that are ventured are provisional and, thus, open that ‘apparent suspension’ is most consistently
to challenge and change. To be able to reveal achieved by organisms of small (<1 mm) size.
something also of the historical context of the Crucially, this feature is mirrored in the fact
usage is to give some indication of the limitations that the intrinsic movements of small organisms
of the terminology and of the areas of conjecture are frequently too feeble to overcome the veloc-
impinging upon it. ity and direction of much of the spectrum of
So it is with ‘plankton’. The general underwater movements. The inability to control hori-
standing of this term is that it refers to the col- zontal position or to swim against significant cur-
lective of organisms that are adapted to spend part rents in open waters separates ‘plankton’ from
or all of their lives in apparent suspension in the the ‘nekton’ of active swimmers, which include
open water of the sea, of lakes, ponds and rivers. adult fish, large cephalopods, aquatic reptiles,
The italicised words are crucial to the concept birds and mammals.
2 PHYTOPLANKTON
Box 1.1 Some definitions used in the literaure

on plankton
seston the totality of particulate matter in water; all material not

in solution
tripton non-living seston
plankton living seston, adapted for a life spent wholly or partly in
quasi-suspension in open water, and whose powers of
motility do not exceed turbulent entrainment (see
Chapter 2)
nekton animals adapted to living all or part of their lives in open
water but whose intrinsic movements are almost
independent of turbulence
euplankton redundant term to distinguish fully adapted, truly planktic
organisms from other living organisms fortuitously
present in the water
tychoplankton non-adapted organisms from adjacent habitats and
present in the water mainly by chance
meroplankton planktic organisms passing a major part of the life history
out of the plankton (e.g. on the bottom sediments)
limnoplankton plankton of lakes
heleoplankton plankton of ponds
potamoplankton plankton of rivers
phytoplankton planktic photoautotrophs and major producer of the
pelagic
bacterioplankton planktic prokaryotes
mycoplankton planktic fungi
zooplankton planktic metazoa and heterotrophic protistans
Some more, now redundant, terms

The terms nannoplankton, ultraplankton, µ-algae are older names for various smaller
size categories of phytoplankton, eclipsed by the classification of Sieburth et al.
(1978) (see Box 1.2).
In this way, plankton comprises organisms isms, their phyletic affinities and physiological
that range in size from that of viruses (a few tens capabilities has expanded, it has become clear
of nanometres) to those of large jellyfish (a metre that the divisions used hitherto do not pre-
or more). Representative organisms include bac- cisely coincide: there are photosynthetic bac-
teria, protistans, fungi and metazoans. In the teria, phagotrophic algae and flagellates that take
past, it has seemed relatively straightforward to up organic carbon from solution. Here, as in gen-
separate the organisms of the plankton, both eral, precision will be considered relevant and
into broad phyletic categories (e.g. bacterioplank- important in the context of organismic prop-
ton, mycoplankton) or into similarly broad func- erties (their names, phylogenies, their morpho-
tional categories (photosynthetic algae of the logical and physiological characteristics). On the
phytoplankton, phagotrophic animals of the zoo- other hand, the generic contributions to sys-
plankton). Again, as knowledge of the organ- tems (at the habitat or ecosystem scales) of the
HISTORICAL CONTEXT OF PHYTOPLANKTON STUDIES 3
photosynthetic primary producers, phagotrophic tion to English is ‘up drive’, approximately ‘buoy-
consumers and heterotrophic decomposers may ancy’ or ‘flotation’, a clear reference to Müller’s
be attributed reasonably but imprecisely to phyto- assumption that the material floated up to the
plankton, zooplankton and bacterioplankton. surface waters – like so much oceanic dirt! It
The defintion of phytoplankton adopted for took one of Müller’s students, Ernst Haeckel, to
this book is the collective of photosynthetic champion the beauty of planktic protistans and
microorganisms, adapted to live partly or contin- metazoans. His monograph on the Radiolaria
uously in open water. As such, it is the photoau- was also one of the first to embrace Darwin’s
totrophic part of the plankton and a major pri- (1859) evolutionary theory in order to show
mary producer of organic carbon in the pelagic structural affinities and divergences. Haeckel, of
of the seas and of inland waters. The distinction course, became best known for his work on
of phytoplankton from other categories of plank- morphology, ontogeny and phylogeny. According
ton and suspended matter are listed in Box 1.1. to Smetacek et al. (2002), his interest and skills
It may be added that it is correct to refer to as a draughtsman advanced scientific awareness
phytoplankton as a singular term (‘phytoplank- of the range of planktic form (most significantly,
ton is’ rather than ‘phytoplankton are’). A single Haeckel, 1904) but to the detriment of any real
organism is a phytoplanktont or (more ususally) progress in understanding of functional differen-
phytoplankter. Incidentally, the adjective ‘plank- tiation. Until the late 1880s, it was not appreci-
tic’ is etymologically preferable to the more com- ated that the organisms of the Auftrieb, even the
monly used ‘planktonic’. algae among them, could contribute much to the
nutrition of the larger animals of the sea. Instead,
it seems to have been supposed that organic mat-
1.2 Historical context of ter in the fluvial discharge from the land was the
major nutritive input. It is thus rather interest-
phytoplankton studies ing to note that, a century or so later, this pos-
sibility has enjoyed something of a revival (see
The first use of the term ‘plankton’ is attributed Chapters 3 and 8).
in several texts (Ruttner, 1953; Hutchinson, 1967) If Haeckel had conveyed the beauty of the
to Viktor Hensen, who, in the latter half of the pelagic protistans, it was certainly Viktor Hensen
nineteenth century, began to apply quantitative who had been more concerned about their role
methods to gauge the distribution, abundance in a functional ecosystem. Hensen was a phys-
and productivity of the microscopic organisms iologist who brought a degree of empiricism
of the open sea. The monograph that is usually to his study of the perplexing fluctuations in
cited (Hensen, 1887) is, in fact, rather obscure North Sea fish stocks. He had reasoned that
and probably not well read in recent times but fish stocks and yields were related to the pro-
Smetacek et al. (2002) have provided a probing duction and distribution of the juvenile stages.
and engaging review of the original, within the Through devising techniques for sampling, quan-
context of early development of plankton science. tification and assessing distribution patterns,
Most of the present section is based on their always carefully verified by microscopic exami-
article. nation, Hensen recognised both the ubiquity of
The existence of a planktic community of phytoplankton and its superior abundance and
organisms in open water had been demonstrated quality over coastal inputs of terrestrial detritus.
many years previously by Johannes Müller. Knowl- He saw the connection between phytoplankton
edge of some of the organisms themselves and the light in the near-surface layer, the nutri-
stretches further back, to the earliest days tive resource it provided to copepods and other
of microscopy. From the 1840s, Müller would small animals, and the value of these as a food
demonstrate net collections to his students, using source to fish.
the word Auftrieb to characterise the commu- Thus, in addition to bequeathing a new
nity (Smetacek et al., 2002). The literal transla- name for the basal biotic component in pelagic
4 PHYTOPLANKTON
ecosystems, Hensen may be regarded justifiably latter quest occupies most of the rest of the book.
as the first quantitative plankton ecologist and However, it is not giving away too much to antici-
as the person who established a formal method- pate that systematics provides an important foun-
ology for its study. Deducing the relative contri- dation for species-specific physiology and which
butions of Hensen and Haeckel to the founda- is itself part-related to morphology. Accordingly,
tion of modern plankton science, Smetacek et al. great attention is paid here to the differentia-
(2002) concluded that it is the work of the lat- tion of individualistic properties of representa-
ter that has been the more influential. This is an tive species of phytoplankton.
opinion with which not everyone will agree but However, there is value in being able simul-
this is of little consequence. However, Smetacek taneously to distinguish among functional cate-
et al. (2002) offered a most profound and resonant gories (trees from herbs!). The scaling system and
observation in suggesting that Hensen’s general nomenclature proposed by Sieburth et al. (1978)
understanding of the role of plankton (‘the big has been widely adopted in phytoplankton ecol-
picture’) was essentially correct but erroneous in ogy to distinguish functional separations within
its details, whereas in Haeckel’s case, it was the the phytoplankton. It has also eclipsed the use of
other way round. Nevertheless, both have good such terms as µ-algae and ultraplankton to separate
claim to fatherhood of plankton science! the lower size range of planktic organisms from
those (netplankton) large enough to be retained
by the meshes of a standard phytoplankton net.
1.3 The diversification of The scheme of prefixes has been applied to size
categories of zooplankton, with equal success.
phytoplankton The size-based categories are set out in Box 1.2.
At the level of phyla, the classification of
Current estimates suggest that between 4000 and the phytoplankton is based on long-standing cri-
5000 legitimate species of marine phytoplank- teria, distinguished by microscopists and bio-
ton have been described (Sournia et al. 1991; chemists over the last 150 years or so, from
Tett and Barton, 1995). I have not seen a com- which there is little dissent. In contrast, subdi-
parable estimate for the number of species in vision within classes, orders etc., and the tracing
inland waters, beyond the extrapolation I made of intraphyletic relationships, affinities within
(Reynolds, 1996a) that the number is unlikely to and among families, even the validity of suppos-
be substantially smaller. In both lists, there is edly well-characterised species, has become sub-
not just a large number of mutually distinct taxa ject to massive reappraisal. The new factor that
of photosynthetic microorganisms but there is a has come into play is the powerful armoury of
wide variety of shape, size and phylogenetic affin- the molecular biologists, including the methods
ity. As has also been pointed out before (Reynolds, for reading gene sequences and for the statisti-
1994a), the morphological range is comparable to cal matching of these to measure the closeness
the one spanning forest trees and the herbs that to other species.
grow at their base. The phyletic divergence of the Of course, the potential outcome is a much
representatives is yet wider. It would be surpris- more robust, genetically verified family tree of
ing if the species of the phytoplankton were uni- authentic species of phytoplankton. This may be
form in their requirements, dynamics and sus- some years away. For the present, it seems point-
ceptibilities to loss processes. Once again, there less to reproduce a detailed classification of the
is a strong case for attempting to categorise the phytoplankton that will soon be made redun-
phytoplankton both on the phylogeny of organ- dant. Even the evolutionary connectivities among
isms and on the functional basis of their roles in the phyla and their relationship to the geochem-
aquatic ecosystems. Both objectives are adopted ical development of the planetary structures
for the writing of this volume. Whereas the for- are undergoing deep re-evaluation (Delwiche,
mer is addressed only in the present chapter, the 2000; Falkowski, 2002). For these reasons, the
THE DIVERSIFICATION OF PHYTOPLANKTON 5
Box 1.2 The classification of phytoplankton according to

the scaling nomenclature of Sieburth et al. (1978)
Maximum linear dimension Namea

0.2–2 µm picophytoplankton
2–20 µm nanophytoplankton
20–200 µm microphytoplankton
200 µm–2 mm mesophytoplankton
>2 mm macrophytoplankton
a
The prefixes denote the same size categories when used with ‘-zooplankton’, ‘-algae’, ‘-cyanobacteria’,
‘flagellates’, etc.
taxonomic listings in Table 1.1 are deliberately modern plants, water is the source of reductant
conservative. electrons and oxygen is liberated as a by-product
Although the life forms of the plankton (oxygenic photosynthesis). Despite their phyletic
include acellular microorganisms (viruses) and a proximity to the photoheterotrophs and shar-
range of well-characterised Archaea (the halobac- ing a similar complement of bacteriochloro-
teria, methanogens and sulphur-reducing bac- phylls (Béjà et al., 2002), the Anoxyphotobac-
teria, formerly comprising the Archaebacteria), teria use alternative sources of electrons and,
the most basic photosynthetic organisms of the in consequence, generate oxidation products
phytoplankton belong to the Bacteria (formerly, other than oxygen (anoxygenic photosynthesis).
Eubacteria). The separation of the ancestral bac- Their modern-day representatives are the purple
teria from the archaeans (distinguished by the and green sulphur bacteria of anoxic sediments.
possession of membranes formed of branched Some of these are planktic in the sense that
hydrocarbons and ether linkages, as opposed to they inhabit anoxic, intensively stratified layers
the straight-chain fatty acids and ester linkages deep in small and suitably stable lakes. The trait
found in the membranes of all other organisms: might be seen as a legacy of having evolved in a
Atlas and Bartha, 1993) occurred early in micro- wholly anoxic world. However, aerobic, anoxy-
bial evolution (Woese, 1987; Woese et al., 1990). genic phototrophic bacteria, containing bac-
The appearance of phototrophic forms, dis- terichlorophyll a, have been isolated from oxic
tinguished by their crucial ability to use light marine environments (Shiba et al., 1979); it has
energy in order to synthesise adenosine triphos- also become clear that their contribution to the
phate (ATP) (see Chapter 3), was also an ancient oceanic carbon cycle is not necessarily insignifi-
event that took place some 3000 million years ago cant (Kolber et al., 2001; Goericke, 2002).
(3 Ga BP (before present)). Some of these organ- Nevertheless, the oxygenic photosynthesis pio-
isms were photoheterotrophs, requiring organic neered by the Cyanobacteria from about 2.8 Ga
precursors for the synthesis of their own cells. before present has proved to be a crucial step in
Modern forms include green flexibacteria (Chlo- the evolution of life in water and, subsequently,
roflexaceae) and purple non-sulphur bacteria on land. Moreover, the composition of the atmos-
(Rhodospirillaceae), which contain pigments sim- phere was eventually changed through the biolo-
ilar to chlorophyll (bacteriochlorophyll a, b or gical oxidation of water and the simultaneous
c). Others were true photoautotrophs, capable removal and burial of carbon in marine sedi-
of reducing carbon dioxide as a source of cell ments (Falkowski, 2002). Cyanobacterial photo-
carbon (photosynthesis). Light energy is used to synthesis is mediated primarily by chlorophyll
strip electrons from a donor substance. In most a, borne on thylakoid membranes. Accessory
6 PHYTOPLANKTON
Table 1.1 Survey of the organisms in the phytoplankton
Domain: BACTERIA
Division: Cyanobacteria (blue-green algae)
Unicellular and colonial bacteria, lacking membrane bound plastids. Primary
photosynthetic pigment is chlorophyll a, with accessory phycobilins (phycocyanin,
phycoerythrin). Assimilation products, glycogen, cyanophycin. Four main sub-groups,
of which three have planktic representatives.
Order: CHROOCOCCALES
Unicellular or coenobial Cyanobacteria but never filamentous. Most planktic genera
form mucilaginous colonies, and these are mainly in fresh water. Picophytoplanktic
forms abundant in the oceans.
Includes: Aphanocapsa, Aphanothece, Chroococcus, Cyanodictyon,
Gomphosphaeria, Merismopedia, Microcystis, Snowella, Synechococcus,
Synechocystis, Woronichinia
Order: OSCILLATORIALES
Uniseriate–filamentous Cyanobacteria whose cells all undergo division in the same
plane. Marine and freshwater genera.
Includes: Arthrospira, Limnothrix, Lyngbya, Planktothrix, Pseudanabaena, Spirulina,
Trichodesmium, Tychonema
Order: NOSTOCALES
Unbranched–filamentous Cyanobacteria whose cells all undergo division in the same
plane and certain of which may be facultatively differentiated into heterocysts. In the
plankton of fresh waters and dilute seas.
Includes: Anabaena, Anabaenopsis, Aphanizomenon, Cylindrospermopsis,
Gloeotrichia, Nodularia
Exempt Division: Prochlorobacteria
Order: PROCHLORALES
Unicellular and colonial bacteria, lacking membrane-bound plastids. Photosynthetic
pigments are chlorophyll a and b, but lack phycobilins.
Includes: Prochloroccus, Prochloron, Prochlorothrix
Division: Anoxyphotobacteria
Mostly unicellular bacteria whose (anaerobic) photosynythesis depends upon an
electron donor other than water and so do not generate oxygen. Inhabit anaerobic
sediments and (where appropriate) water layers where light penetrates sufficiently.
Two main groups:
Family: Chromatiaceae (purple sulphur bacteria) Cells able to photosynthesise
with sulphide as sole electron donor. Cells contain bacteriochlorophyll a, b or c.
Includes: Chromatium, Thiocystis, Thiopedia.
Family: Chlorobiaceae (green sulphur bacteria) Cells able to photosynthesise
with sulphide as sole electron donor. Cells contain bacteriochlorophyll a, b or c.
Includes: Chlorobium, Clathrocystis, Pelodictyon.
Domain: EUCARYA
Phylum: Glaucophyta
Cyanelle-bearing organisms, with freshwater planktic representatives.
Includes: Cyanophora, Glaucocystis.
Phylum: Prasinophyta
Unicellular, mostly motile green algae with 1–16 laterally or apically placed flagella,
cell walls covered with fine scales and plastids containing chlorophyll a and b.
Assimilatory products mannitol, starch.
(cont.)
Table 1.1 (cont.)
CLASS: Pedinophyceae
Order: PEDINOMONADALES
Small cells, with single lateral flagellum.
Includes: Pedinomonas
CLASS: Prasinophyceae
Order: CHLORODENDRALES
Flattened, 4-flagellated cells.
Includes: Nephroselmis, Scherffelia (freshwater); Mantoniella, Micromonas
(marine)
Order: PYRAMIMONADALES
Cells with 4 or 8 (rarely 16) flagella arising from an anterior depression. Marine
and freshwater.
Includes: Pyramimonas
Order: SCOURFIELDIALES
Cells with two, sometimes unequal, flagella. Known from freshwater ponds.
Includes: Scourfieldia
Phylum: Chlorophyta (green algae)
Green-pigmented, unicellular, colonial, filamentous, siphonaceous and thalloid
algae. One or more chloroplasts containing chlorophyll a and b. Assimilation
product, starch (rarely, lipid).
CLASS: Chlorophyceae
Several orders of which the following have planktic representatives:
Order: TETRASPORALES
Non-flagellate cells embedded in mucilaginous or palmelloid colonies, but with
motile propagules.
Includes: Paulschulzia, Pseudosphaerocystis
Order: VOLVOCALES
Unicellular or colonial biflagellates, cells with cup-shaped chloroplasts.
Includes: Chlamydomonas, Eudorina, Pandorina, Phacotus, Volvox (in fresh
waters); Dunaliella, Nannochloris (marine)
Order: CHLOROCOCCALES
Non-flagellate, unicellular or coenobial (sometimes mucilaginous) algae, with
many planktic genera.
Includes: Ankistrodesmus, Ankyra, Botryococcus, Chlorella,
Coelastrum, Coenochloris, Crucigena, Choricystis, Dictyosphaerium,
Elakatothrix, Kirchneriella, Monorophidium, Oocystis, Pediastrum,
Scenedesmus, Tetrastrum
Order: ULOTRICHALES
Unicellular or mostly unbranched filamentous with band-shaped chloroplasts.
Includes: Geminella, Koliella, Stichococcus
Order: ZYGNEMATALES
Unicellular or filamentous green algae, reproducing isogamously by conjugation.
Planktic genera are mostly members of the Desmidaceae, mostly unicellular or
(rarely) filmentous coenobia with cells more or less constricted into two
semi-cells linked by an interconnecting isthmus. Exclusively freshwater genera.
Includes: Arthrodesmus, Closterium, Cosmarium, Euastrum, Spondylosium,
Staurastrum, Staurodesmus, Xanthidium
(cont.)
8 PHYTOPLANKTON
Table 1.1 (cont.)
Phylum: Euglenophyta
Green-pigmented unicellular biflagellates. Plastids numerous and irregular,
containing chlorophyll a and b. Reproduction by longitudinal fission. Assimilation
product, paramylon, oil. One Class, Euglenophyceae, with two orders.
Order: EUTREPTIALES
Cells having two emergent flagella, of approximately equal length. Marine and
freshwater species.
Includes: Eutreptia
Order: EUGLENALES
Cells having two flagella, one very short, one long and emergent.
Includes: Euglena, Lepocinclis, Phacus, Trachelmonas
Phylum: Cryptophyta
Order: CRYPTOMONADALES
Naked, unequally biflagellates with one or two large plastids, containing
chlorophyll a and c2 (but not chlorophyll b); accessory phycobiliproteins or other
pigments colour cells brown, blue, blue-green or red; assimilatory product,
starch. Freshwater and marine species.
Includes: Chilomonas, Chroomonas, Cryptomonas, Plagioselmis, Pyrenomonas,
Rhodomonas
Phylum: Raphidophyta
Order: RAPHIDOMONADALES (syn. CHLOROMONADALES)
Biflagellate, cellulose-walled cells; two or more plastids containing chlorophyll a;
cells yellow-green due to predominant accessory pigment, diatoxanthin;
assimilatory product, lipid. Freshwater.
Includes: Gonyostomum
Phylum: Xanthophyta (yellow-green algae)
Unicellular, colonial, filamentous and coenocytic algae. Motile species generally
subapically and unequally biflagellated; two or many more discoid plastids per cell
containing chlorophyll a. Cells mostly yellow-green due to predominant
accessory pigment, diatoxanthin; assimilation product, lipid. Several orders, two
with freshwater planktic representatives.
Order: MISCHOCOCCALES
Rigid-walled, unicellular, sometimes colonial xanthophytes.
Includes: Goniochloris, Nephrodiella, Ophiocytium
Order: TRIBONEMATALES
Simple or branched uniseriate filamentous xanthophytes.
Includes: Tribonema
Phylum: Eustigmatophyta
Coccoid unicellular, flagellated or unequally biflagellated yellow-green algae with
masking of chlorophyll a by accessory pigment violaxanthin. Assimilation product,
probably lipid.
Includes: Chlorobotrys, Monodus
Phylum: Chrysophyta (golden algae)
Unicellular, colonial and filamentous. often uniflagellate, or unequally biflagellate
algae. Contain chlorophyll a, c1 and c2 , generally masked by abundant accessory
pigment, fucoxanthin, imparting distinctive golden colour to cells. Cells
sometimes naked or or enclosed in an urn-shaped lorica, sometimes with
siliceous scales. Assimilation products, lipid, leucosin. Much reclassified group, has
several classes and orders in the plankton.
(cont.)
Table 1.1 (cont.)
CLASS: Chrysophyceae
Order: CHROMULINALES
Mostly planktic, unicellular or colony-forming flagellates with one or two
unequal flagella, occasionally naked, often in a hyaline lorica or gelatinous
envelope.
Includes: Chromulina, Chrysococcus, Chrysolykos, Chrysosphaerella, Dinobryon,
Kephyrion, Ochromonas, Uroglena
Order: HIBBERDIALES
Unicellular or colony-forming epiphytic gold algae but some planktic
representatives.
Includes: Bitrichia
CLASS: Dictyochophyceae
Order: PEDINELLALES
Radially symmetrical, very unequally biflagellate unicells or coenobia.
Includes: Pedinella (freshwater); Apedinella, Pelagococcus, Pelagomonas,
Pseudopedinella (marine)
CLASS: Synurophyceae
Order: SYNURALES
Unicellular or colony-forming flagellates, bearing distinctive siliceous scales.
Includes: Mallomonas, Synura
Phylum: Bacillariophyta (diatoms)
Unicellular and coenobial yellow-brown, non-motile algae with numerous discoid
plastids, containing chlorophyll a, c1 and c2 , masked by accessory pigment,
fucoxanthin. Cell walls pectinaceous, in two distinct and overlapping halves, and
impregnated with cryptocrystalline silica. Assimilatory products, chrysose, lipids.
Two large orders, both conspicuously represented in the marine and freshwater
phytoplankton.
CLASS: Bacillariophyceae
Order: BIDDULPHIALES (centric diatoms)
Diatoms with cylindrical halves, sometimes well separated by girdle bands. Some
species form (pseudo-)filaments by adhesion of cells at their valve ends.
Includes: Aulacoseira, Cyclotella, Stephanodiscus, Urosolenia (freshwater);
Cerataulina, Chaetoceros, Detonula, Rhizosolenia, Skeletonema, Thalassiosira
(marine)
Order: BACILLARIALES (pennate diatoms)
Diatoms with boat-like halves, no girdle bands. Some species form coenobia by
adhesion of cells on their girdle edges.
Includes: Asterionella, Diatoma, Fragilaria, Synedra, Tabellaria (freshwater);
Achnanthes, Fragilariopsis, Nitzschia (marine)
Phylum: Haptophyta
CLASS: Haptophyceae
Gold or yellow-brown algae, usually unicellular, with two subequal flagella and a
coiled haptonema, but with amoeboid, coccoid or palmelloid stages. Pigments,
chlorophyll a, c1 and c2 , masked by accessory pigment (usually fucoxanthin).
Assimilatory product, chrysolaminarin. Cell walls with scales, sometimes more or
less calcified.
Order: PAVLOVALES
Cells with haired flagella and small haptonema. Marine and freshwater species.
Includes: Diacronema, Pavlova
(cont.)
10 PHYTOPLANKTON
Table 1.1 (cont.)
Order: PRYMNESIALES
Cells with smooth flagella, haptonema usually small. Mainly marine or brackish
but some common in freshwater plankton.
Includes: Chrysochromulina, Isochrysis, Phaeocystis, Prymnesium
Order: COCCOLITHOPHORIDALES
Cell suface covered by small, often complex, flat calcified scales (coccoliths).
Exclusively marine.
Include: Coccolithus, Emiliana, Florisphaera, Gephyrocapsa, Umbellosphaera
Phylum: Dinophyta
Mostly unicellular, sometimes colonial, algae with two flagella of unequal length
and orientation. Complex plastids containing chlorophyll a, c1 and c2 , generally
masked by accessory pigments. Cell walls firm, or reinforced with polygonal
plates. Assimilation products: starch, oil. Conspicuously represented in marine
and freshwater plankton. Two classes and (according to some authorities) up to
11 orders.
CLASS: Dinophyceae
Biflagellates, with one transverse flagellum encircling the cell, the other directed
posteriorly.
Order: GYMNODINIALES
Free-living, free-swimming with flagella located in well-developed transverse and
sulcal grooves, without thecal plates. Mostly marine.
Includes: Amphidinium, Gymnodinium, Woloszynskia
Order: GONYAULACALES
Armoured, plated, free-living unicells, the apical plates being asymmetrical.
Marine and freshwater.
Includes: Ceratium, Lingulodinium
Order: PERIDINIALES
Armoured, plated, free-living unicells, with symmetrical apical plates. Marine and
freshwater.
Includes: Glenodinium, Gyrodinium, Peridinium
Order: PHYTODINIALES
Coccoid dinoflagellates with thick cell walls but lacking thecal plates. Many
epiphytic for part of life history. Some in plankton of humic fresh waters.
Includes: Hemidinium
CLASS: Adinophyceae
Order: PROROCENTRALES
Naked or cellulose-covered cells comprising two watchglass-shaped halves.
Marine and freshwater species.
Includes: Exuviella, Prorocentrum
pigments, called phycobilins, are associated with are recognised, three of which (the chroococ-
these membranes, where they are carried in calean, the oscillatorialean and the nostocalean;
granular phycobilisomes. Life forms among the the stigonematalean line is the exception) have
Cyanobacteria have diversified from simple coc- major planktic representatives that have diversi-
coids and rods into loose mucilaginous colonies, fied greatly among marine and freshwater sys-
called coenobia, into filamentous and to pseu- tems. The most ancient group of the surviv-
dotissued forms. Four main evolutionary lines ing groups of photosynthetic organisms is, in
terms of individuals, the most abundant on the origin of eukaruote plastids (Bhattacharya and
planet. Medlin, 1998; Douglas and Raven, 2003). Prag-
Links to eukaryotic protists, plants and ani- matically, we may judge this to have been a
mals from the Cyanobacteria had been sup- highly successful combination. There may well
posed explicitly and sought implicitly. The dis- have been others of which nothing is known,
covery of a prokaryote containing chlorophyll a apart from the small group of glaucophytes that
and b but lacking phycobilins, thus resembling carry cyanelles rather than plastids. The cyanelles
the pigmentation of green plants, seemed to are supposed to be an evolutionary interme-
fit the bill (Lewin, 1981). Prochloron, a symbiont diate between cyanobacterial cells and chloro-
of salps, is not itself planktic but is recover- plasts (admittedly, much closer to the latter).
able in collections of marine plankton. The first Neither cyanelles nor plastids can grow inde-
description of Prochlorothrix from the freshwa- pendently of the eukaryote host and they are
ter phytoplankton in the Netherlands (Burger- apportioned among daughters when the host cell
Wiersma et al., 1989) helped to consolidate the divides. There is no evidence that the handful
impression of an evolutionary ‘missing link’ of of genera ascribed to this phylum are closely
chlorophyll-a- and -b-containing bacteria. Then related to each other, so it may well be an arti-
came another remarkable finding: the most ficial grouping. Cyanophora is known from the
abundant picoplankter in the low-latitude ocean plankton of shallow, productive calcareous lakes
was not a Synechococcus, as had been thitherto sup- (Whitton in John et al., 2002).
posed, but another oxyphototrophic prokaryote Molecular investigation has revealed that the
containing divinyl chlorophyll-a and -b pigments seemingly disparate algal phyla conform to one
but no bilins (Chisholm et al., 1988, 1992); it was or other of two main lineages. The ‘green line’
named Prochlorococcus. The elucidation of a bio- of eukaryotes with endosymbiotic Cyanobacteria
spheric role of a previously unrecognised organ- reflects the development of the chlorophyte and
ism is achievement enough by itself (Pinevich euglenophyte phyla and to the important off-
et al., 2000); for the organisms apparently to shoots to the bryophytes and the vascular plant
occupy this transitional position in the evolu- phyla. The ‘red line’, with its secondary and even
tion of plant life doubles the sense of scientific tertiary endosymbioses, embraces the evolution
satisfaction. Nevertheless, subsequent investiga- of the rhodophytes, the chrysophytes and the
tions of the phylogenetic relationships of the haptophytes, is of equal or perhaps greater fas-
newly defined Prochlorobacteria, using immuno- cination to the plankton ecologist interested in
logical and molecular techniques, failed to group diversity.
Prochlorococcus with the other Prochlorales or even A key distinguishing feature of the algae of
to separate it distinctly from Synechococcus (Moore the green line is the inclusion of chlorophyll
et al., 1998; Urbach et al., 1998). The present view b among the photosynthetic pigments and, typ-
is that it is expedient to regard the Prochlorales ically, the accumulation of glucose polymers
as aberrent Cyanobacteria (Lewin, 2002). (such as starch, paramylon) as the main prod-
The common root of all eukaryotic algae and uct of carbon assimilation. The subdivision of
higher plants is now understood to be based the green algae between the prasinophyte and
upon original primary endosymbioses involv- the chlorophyte phyla reflects the evolutionary
ing early eukaryote protistans and Cyanobacteria development and anatomic diversification within
(Margulis, 1970, 1981). As more is learned about the line, although both are believed to have
the genomes and gene sequences of microorgan- a long history on the planet (∼1.5 Ga). Both
isms, so the role of ‘lateral’ gene transfers in are also well represented by modern genera, in
shaping them is increasingly appreciated (Doolit- water generally and in the freshwater phyto-
tle et al., 2003). For instance, in terms of ultraplankton in particular. Of the modern prasino-
structure, the similarity of 16S rRNA sequences, phyte orders, the Pedinomonadales, the Chloro-
several common genes and the identical pho- dendrales and the Pyramimonadales each have
tosynthetic proteins, all point to cyanobacterial significant planktic representation, in the sense
12 PHYTOPLANKTON
of producing populations of common occurrence cells, there is a striking variety of planktic

and forming ‘blooms’ on occasions. Several mod- forms.
ern chlorophyte orders (including Oedogoniales, Closest to the ancestral root are the cryp-
Chaetophorales, Cladophorales, Coleochaetales, tophytes. These contain chlorophyll c2 , as well
Prasiolales, Charales, Ulvales a.o.) are without as chlorophyll a and phycobilins, in plastid thy-
modern planktic representation. In contrast, lakoids that are usually paired. Living cells are
there are large numbers of volvocalean, chloro- generally green but with characteristic, species-
coccalean and zygnematalean species in lakes specific tendencies to be bluish, reddish or
and ponds and the Tetrasporales and Ulotrichales olive-tinged. The modern planktic representatives
are also well represented. These show a very wide are exclusively unicellular; they remain poorly
span of cell size and organisation, with flagel- known, partly because thay are not easy to
lated and non-motile cells, unicells and filamen- identify by conventional means. However, about
tous or ball-like coenobia, with varying degrees of 100 species each have been named for marine
mucilaginous investment and of varying consis- and fresh waters, where, collectively, they occur
tency. The highest level of colonial development widely in terms of latitude, trophic state and
is arguably in Volvox, in which hundreds of net- season.
worked biflagellate cells are coordinated to bring Next comes the small group of single-
about the controlled movement of the whole. celled flagellates which, despite showing similar-
Colonies also reproduce by the budding off and ities with the cryptophytes, dinoflagellates and
release of near-fully formed daughter colonies. euglenophytes, are presently distinguished in the
The desmid members of the Zygnematales are phylum Raphidophyta. One genus, Gonyostomum,
amongst the best-studied green plankters. Mostly is cosmopolitan and is found, sometimes in abun-
unicellular, the often elaborate and beautiful dance, in acidic, humic lakes. The green colour
architecture of the semi-cells invite the gaze and imparted to these algae by chlorophyll a is, to
curiosity of the microscopist. some extent, masked by a xanthophyll (in this
The euglenoids are unicellular flagellates. case, diatoxanthin) to yield the rather yellowish
A majority of the 800 or so known species pigmentation. This statement applies even more
are colourless heterotrophs or phagotrophs and to the yellow-green algae making up the phyla
are placed by zoologists in the protist order Xanthophyta and Eustigmatophyta. The xantho-
Euglenida. Molecular investigations reveal them phytes are varied in form and habit with a
to be a single, if disparate group, some of which number of familiar unicellular non-flagellate or
acquired the phototrophic capability through biflagellate genera in the freshwater plankton, as
secondary symbioses. It appears that even the well as the filamentous Tribonema of hard-water
phototrophic euglenoids are capable of absorb- lakes. The eustigmatophytes are unicellular coc-
ing and assimilating particular simple organic coid flagellates of uncertain affinities that take
solutes. Many of the extant species are associ- their name from the prominent orange eye-spots.
ated with organically rich habitats (ponds and The golden algae (Chrysophyta) represent a
lagoons, lake margins, sediments). further recombination along the red line, giv-
The ‘red line’ of eukaryotic evolution is based ing rise to a diverse selection of modern unicel-
on rhodophyte plastids that contain phycobilins lular, colonial or filamentous algae. With a dis-
and chlorophyll a, and whose single thylakoids tinctive blend of chlorophyll a, c1 and c2 , and the
lie separately and regularly spaced in the plastid major presence of the xanthophyll fucoxanthin,
stroma (see, e.g., Kirk, 1994). The modern phy- the chrysophytes are presumed to be close to
lum Rhodophyta is well represented in marine the Phaeophyta, which includes all the macro-
(especially; mainly as red seaweeds) and fresh- phytic brown seaweeds but no planktic vege-
water habitats but no modern or extinct plank- tative forms. Most of the chrysophytes have,
tic forms are known. However, among the inter- in contrast, remained microphytic, with numer-
esting derivative groups that are believed to ous planktic genera. A majority of these come
owe to secondary endosymbioses of rhodophyte from fresh water, where they are traditionally
supposed to indicate low nutrient status and pro- is a prominent thread, as long as the cell; in oth-
ductivity (but see Section 3.4.3: they may simply ers it is smaller or even vestigial but, in most
be unable to use carbon sources other than car- instances, can be bent or coiled. Most of the
bon dioxide). Mostly unicellular or coenobial flag- known extant haptophyte species are marine;
ellates, many species are enclosed in smooth some genera, such as Chrysochromulina, are rep-
protective loricae, or they may be beset with resented by species that are relatively frequent
numerous delicate siliceous scales. The group has members of the plankton of continental shelves
been subject to considerable taxonomic revision and of mesotrophic lakes. Phaeocystis is another
and reinterpretation of its phylogenies in recent haptophyte common in enriched coastal waters,
years. The choanoflagellates (formerly Craspedo- where it may impart a visible yellow-green colour
phyceae, Order Monosigales) are no longer con- to the water at times, and give a notoriously slimy
sidered to be allied to the Chrysophytes. texture to the water (Hardy, 1964).
The last three phyla named in Table 1.1, The coccolithophorids are exclusively marine
each conspicuously represented in both limnetic haptophytes and among the most distinctive
and marine plankton – indeed, they are the microorganisms of the sea. They have a charac-
main pelagic eukaryotes in the oceans – are teristic surface covering of coccoliths – flattened,
also remarkable in having relatively recent ori- often delicately fenestrated, scales impregnated
gins, in the mesozoic period. The Bacillariophyta with calcium carbonate. They fossilise particu-
(the diatoms) is a highly distinctive phylum of larly well and it is their accumulation which
single cells, filaments and coenobia. The char- mainly gave rise to the massive deposits of chalk
acteristics are the possession of golden-brown that gave its name to the Cretaceous (from Greek
plastids containing the chlorophylls a, c1 and kreta, chalk) period, 120–65 Ma BP. Modern coc-
c2 and the accessory pigment fucoxanthin, and colithophorids still occur locally in sufficient pro-
the well-known presence of a siliceous frustule fusion to generate ‘white water’ events. One of
or exoskeleton. Generally, the latter takes the the best-studied of the modern coccolithophorids
form of a sort of lidded glass box, with one of is Emiliana.
two valves fitting in to the other, and bound by The final group in this brief survey is the
one or more girdle bands. The valves are often dinoflagellates. These are mostly unicellular,
patterned with grooves, perforations and callosi- rarely colonial biflagellated cells; some are rel-
ties in ways that greatly facilitate identification. atively large (up to 200 to 300 µm across) and
Species are ascribed to one or other of the two have complex morphology. Pigmentation gener-
main diatom classes. In the Biddulphiales, or ally, but not wholly, reflects a red-line ancestry,
centric diatoms, the valves are usually cylindri- the complex plastids containing chlorophyll a,
cal, making a frustule resembling a traditional c1 and c2 and either fucoxanthin or peridinin
pill box; in the Bacillariales, or pennate diatoms, as accessory pigments, possibly testifying to ter-
the valves are elongate but the girdles are short, tiary endosymbioses (Delwiche, 2000). The group
having the appearance of the halves of a date shows an impressive degree of adaptive radia-
box. While much is known and has been writ- tion, with naked gymnodinioid nanoplankters
ten on their morphology and evolution (see, for through to large, migratory gonyaulacoid swim-
instance, Round et al., 1990), the origin of the mers armoured with sculpted plates and to deep-
siliceous frustule remains obscure. water shade forms with smooth cellulose walls
The Haptophyta are typically unicellular gold such as Pyrocystis. Some genera are non-planktic
or yellow-brown algae, though having amoeboid, and even pass part of the life cycle as epiphytes.
coccoid or palmelloid stages in some cases. The Freshwater species of Ceratium and larger species
pigment blend of chlorophylls a, c1 and c2 , of Peridinium are conspicuous in the plankton
with accessory fucoxanthin, resembles that of of certain types of lakes during summer strati-
other gold-brown phyla. The haptophytes are dis- fication, while smaller species of Peridinium and
tinguished by the possession of a haptonema, other genera (e.g. Glenodinium) are associated with
located between the flagella. In some species it mixed water columns of shallow ponds.
14 PHYTOPLANKTON
Figure 1.1 Non-motile

unicellular phytoplankters.
(a) Synechococcus sp.; (b) Ankyra
judayi; (c) Stephanodiscus rotula;
(d) Closterium cf. acutum. Scale bar,
10 µm. Original photomicrographs
by Dr H. M. Canter-Lund,
reproduced from Reynolds (1984a).
The relatively recent appearance of diatoms, 10–11 ◦ C). Life on Earth suffered a severe set-
coccolithophorids and dinoflagellates in the back, perhaps as close as it has ever come
fossil record provides a clear illustration of to total eradication. In a period of less than
how evolutionary diversification comes about. 0.1 Ma, many species fell extinct and the sur-
Although it cannot be certain that any of these vivors were severely curtailed. As the planet
three groups did not exist beforehand, there cooled over the next 20 or so million years,
is no doubt about their extraordinary rise dur- the rump biota, on land as in water, were able
ing the Mesozoic. The trigger may well have to expand and radiate into habitats and niches
been the massive extinctions towards the end of that were otherwise unoccupied (Falkowski,
the Permian period about 250 Ma BP, when a 2002).
huge release of volcanic lava, ash and shroud- Dinoflagellate fossils are found in the early
ing dust from what is now northern Siberia Triassic, the coccolithophorids from the late Tri-
brought about a world-wide cooling. The trend assic (around 180 Ma BP). Together with the
was quickly reversed by accumulating atmo- diatoms, many new species appeared in the Juras-
spheric carbon dioxide and a period of severe sic and Cretaceous periods. In the sea, these three
global warming (which, with positive feedback groups assumed a dominance over most other
of methane mobilisation from marine sediments, forms, the picocyanobacteria excluded, which
raised ambient temperatures by as much as persists to the present day.
GENERAL FEATURES OF PHYTOPLANKTON 15
Figure 1.2 Planktic unicellular

flagellates. (a) Two variants of
Ceratium hirundinella; (b)
overwintering cyst of Ceratium
hirundinella, with vegetative cell for
comparison; (c) empty case of
Peridinium willei to show exoskeletal
plates and flagellar grooves; (d)
Mallomonas caudata; (e) Plagioselmis
nannoplanctica; (f) two cells of
Cryptomonas ovata; (g) Phacus
longicauda; (h) Euglena sp.; (j)
Trachelomonas hispida. Scale bar, 10
µm. Original photomicrographs by
Dr H. M. Canter-Lund, reproduced
from Reynolds (1984a).
whether they occurred among other precursors

1.4 General features of that subsequently established new lines of plank-
phytoplankton tic invaders.
It is not a problem that can yet be answered
satisfactorily. However, it does not detract from
Despite being drawn from a diverse range of the fact that to function and survive in the
what appear to be distantly related phyloge- plankton does require some specialised adapta-
netic groups (Table 1.1), there are features that tions. It is worth emphasising again that just as
phytoplankton share in common. In an earlier phytoplankton comprises organisms other than
book (Reynolds, 1984a), I suggested that these algae, so not all algae (or even very many of
features reflected powerful convergent forces in them) are necessarily planktic. Moreover, neither
evolution, implying that the adaptive require- the shortness of the supposed step to a plank-
ments for a planktic existence had risen inde- tic existence nor the generally low level of struc-
pendently within each of the major phyla repre- tural complexity of planktic unicells and coeno-
sented. This may have been a correct deduction, bia should deceive us that they are necessarily
although there is no compelling evidence that simple organisms. Indeed, much of this book
it is so. On the other hand, for small, unicellu- deals with the problems of life conducted in a
lar microorganisms to live freely in suspension fluid environment, often in complete isolation
in water is an ancient trait, while the transition from solid boundaries, and the often sophisti-
to a full planktic existence is seen to be a rel- cated means by which planktic organisms over-
atively short step. It remains an open question come them. Thus, in spite of the diversity of phy-
whether the supposed endosymbiotic recombina- logeny (Table 1.1), even a cursory consideration
tions could have occurred in the plankton, or of the range of planktic algae (see Figs. 1.1–1.5)
16 PHYTOPLANKTON
Figure 1.3 Coenobial

phytoplankters. Colonies of the
diatoms (a) Asterionella formosa, (b)
Fragilaria crotonensis and (d) Tabellaria
flocculosa var. asterionelloides. The
fenestrated colony of the
chlorophyte Pediastrum duplex is
shown in (c). Scale bar, 10 µm.
Original photomicrographs by Dr
H. M. Canter-Lund, reproduced
reveals a commensurate diversity of form, func- of immediate respiratory needs (Chapter 3). How-
tion and adaptive strategies. ever, radiant energy of suitable wavelengths
What features, then, are characteristic and (photosynthetically active radiation, or PAR) is nei-
common to phytoplankton, and how have they ther universally or uniformly available in water
been selected? The overriding requirements of but is sharply and hyperbolically attenuated with
any organism are to increase and multiply its depth, through its absorption by the water and
kind and for a sufficient number of the progeny scattering by particulate matter (to be discussed
to survive for long enough to be able to invest in Chapter 3). The consequence is that for a
in the next generation. For the photoautotroph, given phytoplankter at anything more than a
this translates to being able to fix sufficient car- few meters in depth, there is likely to be a crit-
bon and build sufficient biomass to form the ical depth (the compensation point) below which
next generation, before it is lost to consumers net photosynthetic accumulation is impossible.
or to any of the several other potential fates that It follows that the survival of the phytoplankter
await it. For the photoautotroph living in water, depends upon its ability to enter or remain in
the important advantages of archimedean sup- the upper, insolated part of the water mass for
port and the temperature buffering afforded by at least part of its life.
the high specific heat of water (for more, see This much is well understood and the point
Chapter 2) must be balanced against the diffi- has been emphasised in many other texts. These
cuties of absorbing sufficient nutriment from have also proffered the view that the essential
often very dilute solution (the subject of Chapter characteristic of a planktic photoautotroph is to
4) and of intercepting sufficient light energy to minimise its rate of sinking. This might be liter-
sustain photosynthetic carbon fixation in excess ally true if the water was static (in which case,
Figure 1.4 Filamentous

phytoplankters. Filamentous
coenobia of the diatom Aulacoseira
subarctica (a, b; b also shows a
spherical auxospore) and of the
Cyanobacteria (c) Gloeotrichia
echinulata, (d) Planktothrix mougeotii,
(e) Limnothrix redekei (note polar gas
vacuoles), (f) Aphanizomenon
flos-aquae (with one akinete formed
and another differentiating) and
Anabaena flos-aquae (g) in India ink,
to show the extent of mucilage, and
(h) enlarged, to show two
heterocysts and one akinete. Scale
bar, 10 µm. Original
photomicrographs by Dr H. M.
Canter-Lund, reproduced from
Reynolds (1984a).
neutral buoyancy would provide the only ideal To a greater or lesser degree, these move-
adaptation). However, natural water bodies are ments of the medium overwhelm the sinking tra-
almost never still. Movement is generated as a jectories of phytoplankton. The traditional view
consequence of the water being warmed or cool- of planktic adaptations as mechanisms to slow
ing, causing convection with vertical and hori- sinking rate needs to be adjusted. The essential
zontal displacements. It is enhanced or modified requirement of phytoplankton is to maximise the
by gravitation, by wind stress on the water sur- opportunities for suspension in the various parts
face and by the inertia due to the Earth’s rotation of the eddy spectrum. In many instances, the
(Coriolis’ force). Major flows are compensated by adaptations manifestly enhance the entrainabil-
return currents at depth and by a wide spectrum ity of planktic organisms by turbulent eddies.
of intermediate eddies of diminishing size and These include small size and low excess den-
of progressively smaller scales of turbulent diffu- sity (i.e. organismic density is close to that of
sivity, culminating in molecular viscosity (these water, ∼1000 kg m−3 ), which features do con-
motions are characterised in Chapter 2). tribute to a slow rate of sinking. They also include
18 PHYTOPLANKTON
Figure 1.5 Colonial

phytoplankters. Motile colonies of
(a) Volvox aureus, with (b) detail of
cells, (c) Eudorina elegans, (d)
Uroglena sp. and (e) Dinobryon
divergens; and non-motile colonies,
all mounted in India ink to show the
extent of mucilage, of (f) Microcystis
aeruginosa, (g) Pseudosphaerocystis
lacustris and (h) Dictyosphaerium
pulchellum. Scale bar, 10 µm.
Original photomicrographs by Dr
H. M. Canter-Lund, reproduced
mechanisms for increasing frictional resistance isms of the nekton, – cephalopods, fish, reptiles
with the water, independently of size and dens- and mammals – which are able to direct their
ity. At the same time, other phytoplankters show own movements to overcome a still broader range
adaptations that favour disentrainment, at least of the pelagic eddy spectrum.
from weak turbulence, coupled with relatively All these aspects of turbulent entrainment
large size (often achieved by colony formation), and disentrainment are explored more deeply
streamlining and an ability to propel themselves and more empirically in Chapter 2. For the
rapidly through water. Such organisms exploit a moment, it is important to understand how they
different part of the eddy spectrum from the first impinge upon phytoplankton morphology in a
group. The principle extends to the larger organ- general sense.
1.4.1 Size and shape These traits are represented and sometimes
Apart from the issue of suspension, there is blended in the morphological adaptations of spe-
a further set of constraints that resists large cific plankters. They can be best illustrated by
size among phytoplankters. Autotrophy implies a the plankters themselves and by examining how
requirement for inorganic nutrients that must be they influence their lives and ecologies. The wide
absorbed from the surrounding medium. These ranges of form, size, volume and surface area
are generally so dilute and so much much less are illustrated by the data for freshwater plank-
concentrated than they have to be inside the ton presented in Table 1.2. The list is an edited,
plankter’s cell that uptake is generally against a simplified and updated version of a similar table
very steep concentration gradient that requires in Reynolds (1984a) which drew on the author’s
the expenditure of energy to counter it. Once own measurements but quoted from other com-
inside the cell, the nutrient must be translo- pilations (Pavoni, 1963; Nalewajko, 1966; Besch et
cated to the site of its deployment, invoking dif- al., 1972; Bellinger, 1974; Findenegg, Nauwerck
fusion and transport along internal molecular in Vollenweider, 1974; Willén, 1976; Bailey-Watts,
pathways. Together, these twin constraints place 1978; Trevisan, 1978). The sizes are not precise
a high premium on short internal distances: and are often variable within an order of mag-
cells that are absolutely small or, otherwise, nitude. However, the listing spans nearly eight
have one or two linear dimensions truncated (so orders, from the smallest cyanobacterial unicells
that cells are flattened or are slender) benefit of ∼1 µm3 or less, the composite structures of
from this adaptation. Conversely, simply increas- multicellular coenobia and filaments with vol-
ing the diameter (d) of a spherical cell is to umes ranging between 103 and 105 µm3 , through
increase the constraint for, though the surface to units of >106 µm3 in which cells are embedded
area increases in proportion to d2 , the volume within a mucilaginous matrix. Indeed, the list is
increases with d3 . However, distortion from the conservative in so far as colonies of Microcystis of
spherical form, together with surface convolu- >1 mm in diameter have been observed in nature
tion, provides a way of increasing surface in (author’s observations; i.e. up to 109 µm3 in vol-
closer proportion to increasing volume, so that ume). Because all phytoplankters are ‘small’ in
the latter is enclosed by relatively more sur- human terms, requiring good microscopes to see
face than the geometrical minimum required them, it is not always appreciated that the nine
to bound the same volume (a sphere). In this or more orders of magnitude over which their
respect, the adaptive requirements for maximis- sizes range is comparable to that spanning forest
ing entrainability and for enhancing the assim- trees to the herbs growing at their bases. Like the
ilation of nutrients taken up across the surface example, the biologies and ecologies of the indi-
coincide. vidual organisms vary considerably through the
It is worth adding, however, that nutrient spectrum of sizes.
uptake from the dilute solution is enhanced
if the medium flows over the cell surface, dis- 1.4.2 Regulating surface-to-volume ratio
placing that which may have already become Dwelling on the issue of size and shape, we will
depleted. Movement of the cell relative to the find, as already hinted above, that a good deal
adjacent water achieves the similar effect, with of plankton physiology is correlated to the ratio
measurable benefit to uptake rate (Pasciak and of the surface area of a unit (s) to its volume
Gavis, 1974; but see discussion in Section 4.2.1). (v). ‘Unit’ in this context refers to the live habit
It may be hypothesised that it is advantageous of the plankter: where the vegetative form is
for the plankter not to achieve isopycnic suspen- unicellular (exemplified by the species listed in
sion in the water but to retain an ability to sink Table 1.2A), it is only the single cell that interacts
or float relative to the immediate surroundings, with its environment and is, plainly, synonymous
regardless of the rate and direction of travel of with the ‘unit’. If cells are joined together to
the latter, just to improve the sequestration of comprise a larger single structure, for whatever
nutrients. advantage, then the individual cells are no longer
20 PHYTOPLANKTON
Table 1.2 Nominal mean maximum linear dimensions (MLD), approximate volumes (v) and
surface areas (s) of some freshwater phytoplankton
MLD v s s/v
Species Shape (µm) (µm3 ) (µm2 ) (µm−1 )
(A) Unicells
Synechoccoccus ell ≤4 18 35 1.94
(1–20)
Ankyra judayi bicon 16 24 60 2.50
(3–67)
Monoraphidium griffithsii cyl 35 30 110 3.67
Chlorella pyrenoidosa sph 4 33 50 1.52
(8–40)
Kephyrion littorale sph 5 65 78 1.20
Plagioselmis ell 11 72 108 1.50
nannoplanctica (39–134)
Chrysochromulina parva cyl 6 85 113 1.33
Monodus sp. ell 8 105 113 1.09
Chromulina sp. ell 15 440 315 0.716
Chrysococcus sp. sph 10 520 315 0.596
Stephanodiscus cyl 11 600 404 0.673
hantzschii (180–1 200)
Cyclotella praeterissima cyl 10 760 460 0.605
(540–980)
Cyclotella meneghiniana cyl 15 1 600 780 0.488
Cryptomonas ovata ell 21 2710 1 030 0.381
(1 950–3 750)
Mallomonas caudata ell 40 4 200 3 490 0.831
(3 420–10 000)
Closterium aciculare bicon 360 4 520 4 550 1.01
Stephanodiscus rotula cyl 26 5 930 1 980 0.334
(2 220–18 870)
Cosmarium depressum (a) 24 7 780 2 770 0.356
(400–30 000)
Synedra ulna bicon 110 7 900 4 100 0.519
Staurastrum pingue (b) 90 9 450 6 150 0.651
(4 920–16 020)
Ceratium hirundinella (c) 201 43 740 9 600 0.219
(19 080–62 670)
Peridinium cinctum ell 55 65 500 7 070 0.108
(33 500–73 100)
(B) Coenobia
Dictyosphaerium (d ) 40 900 1 540 1.71
pulchellum (40 cells)
Scenedesmus (e) 80 1 000 908 0.908
quadricauda (4 cells)
Asterionella formosa (f ) 130 5 160 6 690 1.30
(8 cells) (4 430–6 000)
(cont.)
Table 1.2 (cont.)
MLD v s s/v
Species Shape (µm) (µm3 ) (µm2 ) (µm−1 )
Fragilaria crotonensis (g) 70 6 230 9 190 1.48
(10 cells) (4 970–7 490)
Dinobryon divergens (h) 145 7 000 5 350 0.764
(10 cells) (6 000–8 500)
Tabellaria flocculosa var. (f ) 96 13 800 9 800 0.710
asterionelloides (6 520–13 600)
(8 cells)
Pediastrum boryanum ( j) 100 16 000 18 200 1.14
(32 cells)
(C) Filaments
Aulacoseira subarctica cyl (k) 240 5 930 4 350 0.734
(10 cells) (4 740–7 310)
Planktothrix mougeotii cyl (k) 1000 46 600 24 300 0.521
(1 mm length)
Anabaena circinalis (m) 60 2 040 2 110 1.03
(20 cells) (n) 75 29 000 6 200 0.214
Aphanizomenon (p) 125 610 990 1.62
flos-aquae (50 cells) (q) 125 15 400 5 200 0.338
(D) Mucilaginous colonies
Coenochloris fottii sph 46 51 × 103 6.65 × 103 0.13
(cells 80–1200 µm3 )
Eudorina unicocca sph 130 1.15 × 106 53.1 × 103 0.046
(cells 120–1200 µm3 )
Uroglena lindii sph 160 2.2 × 106 81 × 103 0.037
(cells 100 µm3 )
Microcystis aeruginosa sph 200 4.2 × 106 126 × 103 0.030
(cells 30–100 µm3 )
Volvox globator sph (r) 450 47.7 × 106 636 × 103 0.013
(cells ∼ 60 µm3 ) (s) 450 6.4 × 106 636 × 103 0.099
Notes: The volumes and surface areas are necessarily approximate. The values cited are those adopted
and presented in Reynolds (1984a); some later additions taken from Reynolds (1993a), mostly based
on his own measurements. The volumes given in brackets cover the ranges quoted elsewhere in the
literature (see text). Note that the volumes and surface areas are calculated by analogy to the nearest
geometrical shape. Surface sculpturing is mostly ignored. Shapes considered include: sph (for a sphere),
cyl (cylinder), ell (ellipsoid), bicon (two cones fused at their bases, area of contact ignored from surface
area calculation). Other adjustments noted as follows:
a
Cell visualised as two adjacent ellipsoids, area of contact ignored.
b
Cell visualised as two prisms and six cuboidal arms, area of contact ignored.
c
Cell visualised as two frusta on elliptical bases, two cylindrical (apical) and two conical (lateral) horns.
d
Coenobium envisaged as 40 contiguous spheres, area of contact ignored.
e
Coenobium envisaged as four adjacent cuboids, volume of spines ignored
f
Coenobium envisaged as eight cuboids, area of contact ignored.
22 PHYTOPLANKTON
Notes to Table 1.2 (cont.)

g
Each cell visualised as four trapezoids; area of contact between cells ignored.
h
Coenobium envisaged as a seies of cones, area of contact ignored
j
Coenobium envisaged as a discus-shaped sphaeroid.
k
Coenobium envisaged as a chain of cylinders, area of contact between cells ignored.
l
Coenobium envisaged as a single cylinder, terminal taper ignored.
m
Filament visualised as a chain of spheres, area of contact between them ignored
n
Filament visualised as it appears in life, enveloped in mucilage, turned into a complete ‘doughnut’ ring,
with a cross-sectional diameter of 21 µm.
p
Filament visualised as a chain of ovoids, area of contact between them ignored.
q
For the typical ‘raft’ habit of this plankter, a bundle of filaments is envisaged, having an overall length
of 125 µm and a diameter of 12.5 µm.
r
The volume calculation is based on the external dimensions.
s
In fact the cells in the vegetative stage are located exclusively on the wall of a hollow sphere. This
second volume calculation supposes an average wall thickness of 10 µm and subtracts the hollow volume.
independent but rather constitute a multicellu-

lar ‘unit’ whose behaviour and experienced envi-
107
ronment is simultaneously shared by all the oth-
ers in the unit. Such larger structures may deploy
cells either in a plate- or ball-like coenobium (exem-
plified by the species listed in Table 1.2B) or, end- 106
to-end, to make a uniseriate filament (Table 1.2C).
Generally, added complexity brings increased size
but, as volume increases as the cube but surface
105
as the square of the linear dimension, there is
natural tendency to sacrifice a high surface-to-
Nominal s / µm2
volume ratio.
However, a counteractive tendency is found 104
among the coenobial and filamentous units (and
also among larger unicells), in which increased
size is accompanied by increased departure from
the spherical form. This means, in large-volume 103
units, more surface bounds the volume than

the strict geometrical minimum provided by the
sphere. In addition to distortion, surface fold- Geometrically excluded
102
ing, the development of protuberances, lobes and
horns all contribute to providing more surface
for not much more volume. The trend is shown
in Fig. 1.6, in which the surface areas of the
0 10 103 105 107
species listed in Tables 1.2A, B and C are plot-
Nominal v (µm3)
ted against the corresponding, central-value vol-
umes, in log/log format. The smaller spherical Figure 1.6 Log/log relationship between the surface areas
and ovoid plankters are seen to lie close to the (s) and volumes (v) of selected freshwater phytoplankters
(lower) slope representing the geometric mini- shown in Table 1.2. The lower line is fitted to the points
mum of surface on volume, s ∝ v, but progres- () referring to species forming quasi-spherical mucilaginous
sively larger units drift away from it. The sec- coenobia (log s = 0.67 log v + 0.7); the upper line is the
ond regression, fitted to the plotted data, has a regression of coenobia (log s = 0.82 log v + 0.49) fitted to all
other points ( t). Redrawn from Reynolds (1984a).
steeper gradient of v0.82 . The individual values of
s/v entered in Table 1.2B and C show several exam-

ples of algal units having volumes of between 103
1000
and 105 µm3 but maintaining surface-to-volume
ratios of >1. In the case of Asterionella, the slen-
der individual cuboidal cells are distorted in one C
plane and are attached to their immediate coeno-
MLD / µm−1
100
bial neighbours by terminal pads representing S
a very small part of the total unit surface. The
resultant pseudostellate coenobium is actually
a flattened spiral. In Fragilaria crotonensis, a still
10 Geometrically
greater s/v ratio is achieved. Though superficially excluded
similar to those of Asterionella, its cells are widest
in their mid-region and where they link, mid-
valve to mid-valve, to their neighbours to form
the ‘double comb’ appearance that distinguishes 0 0.01 0.1 1 10
this species from others (mostly non-planktic) of Nominal s/v ( µm−1)
the same genus. Among the filamentous forms,
there is a widespread tendency for cells to be Figure 1.7 The shapes of phytoplankters: log/log plot of
elongated in one plane and to be attached to maximum linear dimension (MLD) versus nominal s/v of
their neighbours at their polar ends so that individual phytoplankters (data in Table 1.2). Similar
morphologies are grouped together (I, spherical cells; II,
the long axes are cumulated (e.g. in Aulacoseira,
spherical colonies; III, squat ellipsoids and cylinders, of which
Fig. 1.4).
IV are exclusively centric diatoms; V, attenuate cells and
Again confining the argument to the planktic filaments; VI, coenobia; VII, bundles of filaments of Anabaena
forms covered in Tables 2.2A, B and C, whose vol- or Aphanizomenon); C (for Ceratium) and S (for Staurastrum)
umes, together, cover five orders of magnitude, identify shapes with large protuberances). The vertical dotted
these show surface-to-volume ratios that fall in lines define the distributions of most marine phytoplankters
scarcely more than one-and-a-half orders (3.6 to according to Lewis (1976). Redrawn from Reynolds (1984a).
0.1). This evident conservatism of the surface-to-
volume ratio among phytoplankton was noted in
a memorable paper by Lewis (1976). He argued
that this is not a geometric coincidence but an represented these modifications by plotting the
evolutionary outcome of natural selection of the maximum linear dimension (MLD) of the unit
adaptations for a planktic existence. Thus, how- against its surface-to-volume ratio. His approach
ever strong is the selective pressure favouring is followed in the construction of Fig. 1.7, in
increased size and complexity, the necessity to which the relevant data from Table 1.2 are plot-
maintain high s/v, whether for entrainment or ted. The diagonal line is a geometric boundary,
nutrient exchange or both, remains of overriding representing the diminution of s/v of spheres
importance. In other words, the relatively rigid against the increment in diameter and, indeed,
constraints imposed by maintenance of an opti- upon which the spherical unicells (marked I)
mum surface-to-volume ratio constitute the most and colonies (II) are located. All other shapes fall
influential single factor governing the shape of above this line, the further above it being the
planktic algae. more distorted with respect to the sphere of the
Lewis (1976) developed this hypothesis same MLD. The broken dotted lines bound the
through an empirical analysis of phytoplankton s/v ratios of non-spherical forms. All the ellipsoid
shapes. To a greater or lesser extent, departure shapes (III, such as Mallomonas, Rhodomonas),
from the spherical form through the provision squat cylinders (IV, including Cyclotella and
of additional surface area is achieved by shape Stephanodiscus spp.), attenuated needle-like cells
attenuation in one or, perhaps, two planes, and filaments (V: Monoraphidium, Closterium,
respectively resulting in slender, needle-like Aulacoseira, Planktothrix) fall within this area. So
forms or flattened, plate-like structures. Lewis do the coenobial forms comprising individual
24 PHYTOPLANKTON
attenuated cells (VI, e.g. Asterionella, Fragilaria) Cyanobacteria included in Table 1.2 (Anabaena
and the unicells with significant horn-like circinalis, Aphanizomenon flos-aquae), the supposed
or arm-like distortions (Ceratium, Staurastrum, advantage of the filamentous habit is sacrificed
individually identified). The plot backs the asser- through a combination of aggregation, coiling
tion that the attractive and sometimes bizarre and mucilage production to the attainment of
forms adopted by planktic freshwater algae are rapid rates of migration (Booker and Walsby,
functionally selected. 1979).
Provided colonies have a simultaneous capac-
1.4.3 Low surface-to-volume ratio: ity for controlled motility, there are good tele-
mucilaginous forms ological grounds for deducing circumstances
The principle of morphological conservation of a when massive provision of mucilage represents
favourable surface-to-volume ratio, which holds a discrete and alternative adaptation to a plank-
equally for the phytoplankton of marine and tic existence. However, the idea that streamlin-
inland waters, might be more strongly coming is more than a fortuitous benefit is chal-
pelling were it not for the fact that another lenged by the many non-motile species that exist
common evolutionary trend – that of embedding as mucilaginous colonies. There are other demon-
vegetative cells in swathes of mucilage – repre- strable benefits from a mucilaginous exterior,
sents a total antithesis. The formation of globu- including defence against fungal attack, grazers,
lar colonies is prevalent among the freshwater digestion or metal toxicity, and there are circum-
Cyanobacteria, Chlorophyta and the Chryso- stances in which it might assist in the sequestra-
phyta. It is also observed in the vegetative tion or storage of nutrients or in protecting cells
life-history stages of the haptophyte, Phaeocys- from an excessively oxidative environment (see
tis, though, generally, the trait is not com- Box 6.1, p. 271). Even mucilage itself, essentially
mon among the marine phytoplankton. In many a matrix of hygroscopic carbohydrate polymers
instances, the secondary structures are predom- immobilising relatively large amounts of water,
inantly mucilaginous and the live cells may is highly variable in its consistency, intraspecifi-
occupy as little as 2% of the total unit volume cally as well as interspecifically.
in Coenochloris and Uroglena and scarcely exceeds Thus, doubts persist about the true function
20% in Microcystis or Eudorina. It was originally of mucilage investment. However, a consistent
supposed to provide a low-density buoyancy aid geometric consequence of mucilage investment
but it has since been shown that any advantage is that the planktic unit is left with an excep-
is quickly lost to increased size (see Chapter 2). In tionally low surface-to-volume ratio (i.e. area II,
some instances, the individual cells are flagellate towards the left in Fig. 1.7).
(as in Uroglena and Eudorina) and the flagella pass
through the mucilage to the exterior, where their
coordinated beating propels the whole colony 1.5 The construction and
through the medium. Because the surface offers composition of freshwater
little friction, the mucilage is said to be helpful
in assisting rapid passage and migration through
phytoplankton
weakly turbulent water. Certainly, in the case
of the colonial gas-vacuolate Cyanobacteria that The architecture of the cells of planktic algae con-
are able to regulate their buoyancy (e.g. Micro- forms to a basic model, common to the major-
cystis, Snowella, Woronichinia), larger colonies float ity of eukaryotic plants. A series of differenti-
more rapidly than smaller ones of the same den- ated protoplasmic structures are enclosed within
sity (Reynolds, 1987a). Merely adjusting buoyancy a vital membrane, the plasmalemma. This mem-
then becomes a potentialy effective means of brane is complex, comprising three or four dis-
recovering or controlling vertical position in the tinct layers. In a majority of algae there is a fur-
water (Ganf, 1974a). It is interesting that, in the ther, non-living cell wall, made of cellulose or
two buoyancy-regulating filamentous species of other, relatively pure, condensed carbohydrate
THE CONSTRUCTION AND COMPOSITION OF FRESHWATER PHYTOPLANKTON 25
polymer, such as pecten. Among some algal too, is multilayered but has the distinctive bac-
groups, the wall may be more or less impreg- terial configuration. The cells lack a membrane-
nated with inorganic deposits of calcium car- bound nucleus and plastids, the genetic material
bonate or silica. High-power scanning electron and photosynthetic thylakoid membranes being
microscopy reveals that these deposits can form unconfined through the main body (stroma) of
a more or less continuous but variably thick- the cell. The pigments, chlorophyll and accessory
ened and fenestrated surface (as in the siliceous phycobilins, colour the whole cell. Glycogen is
frustule wall of diatoms) or can comprise an the principal photosynthetic condensate and pro-
investment of individual scales (like those of the teinaceous structured granules may also be accu-
Synurophyceae, made of silica, or of the coccol- mulated. Many planktic genera contain, poten-
ithophorids, made of carbonate). These exoskele- tially or actually, specialised intracellular pro-
tons are distinctive and species-diagnostic. Some teinaceous gas-filled vacuoles which may impart
algae lack a polymer wall and are described as buoyancy to the cell.
‘naked’. Both naked and walled cells may carry From an ecological point of view, the ultra-
an additional layer of secreted mucilage. structural properties of phytoplankton cells
The intracellular protoplasm (cytoplasm) is assume considerable relevance to the resource
generally a viscous, gel-like suspension in which requirements of their assembly, as well as to
the nucleus, one or more plastids and vari- adaptive behaviour, productivity and dynamics
ous other organelles, including the endoplas- of populations. Thus, it is important to establish
mic reticulum and the mitochondria, and some a number of empirical criteria of cell composi-
condensed storage products are maintained. The tion that impinge upon the fitness of individ-
plastids vary hugely and interspecifically in ual plankters and the stress thresholds relative
shape – from a solitary axial cup (as in the to light, temperature and nutrient availability.
Volvocales), numerous discoids (typical of cen- These include methods for assessing the biomass
tric diatoms), one or two broad parietal or of phytoplankton populations and the environ-
axial plates (as in Cryptophytes) or more com- mental capacity to support them.
plex shapes (many desmids). All take on the
intense coloration of the dominant photosyn- 1.5.1 Dry weight
thetic pigments they contain – chlorophyll a Characteristically, the major constituent of the
and β-carotene and, variously, other chlorophylls live plankter is water. If the organism is air-dried
and/or accessory xanthophylls. The stored con- to remove all uncombined water, the residue
densates of anabolism are also conspicuously will comprise both organic (mainly protoplasm
variable among the algae: starch in the chloro- and storage condensates) and inorganic (such as
phytes and cryptophytes, other carbohydrates the carbonate or silica impregnated into the cell
in the euglenoids (paramylon) and the Chryso- walls) fractions. Oxidation of the organic frac-
phyceae, oils in the Xanthophyceae). Many also tion, by further heating in air to ∼500 ◦ C, yields
store protein in the cytoplasm. The quantities of an ash approximating to the original inorganic
all storage products vary with metabolism and constituents. The relative masses of the ash and
environmental circumstances. ash-free (i.e. organic) fractions of the original
Intracellular vacuoles are to be found in most material may then be back-calculated.
planktic algae but the large sap-filled spaces char- The dry weights and the ash contents of a
acteristic of higher-plant cells are relatively rare, selection of freshwater phytoplankton are pre-
other than in the diatoms. Osmoregulatory con- sented in Table 1.3. Generally, cell dry mass
tractile vacuoles occur widely, though not univer- (Wc ) increases with increasing cell volume (v),
sally, among the planktic algae, varying in num- as shown in Fig. 1.8. The regression, fitted to
ber and distribution among the phylogenetic all data points, has the equation Wc = 0.47
groups. v0.99 , with a high coefficient of correlation (0.97).
The prokaryotic cell of a planktic Cyanobac- At first sight, the relationship yields the useful
terium is also bounded by a plasmalemma. It, general prediction that the dry weights of live
26 PHYTOPLANKTON
Table 1.3 Air-dry weights, ash-free free dry weights, chlorophyll content and volume of individual cellsa
from natural populations (all values are means of collected data having considerable ranges of variability)
Ash-free dry
Dry weight weight Chla Volume
Species (pg cell−1 ) (pg cell−1 ) Ash % dry (pg cell−1 ) (µm3 cell−1 )
Cyanobacteria
Anabaena circinalisb 45 – – 0.72 99
Aphanizomenon flos-aquaeb 3.9 – – 0.04 8.2
Microcystis aeruginosab 32 – – 0.36 73
Planktothrix mougeotiia,b 28000 – – 243 46 600
Chlorophytes
Ankyra judayib – – – 0.45 24
Chlorella pyrenoidosab 15 – – 0.15 33
Chlorella pyrenoidosac 5.1–6.4 4.5–5.7 11.4 – 20
Closterium aciculareb – – – 89 4 520
Eudorina elegansb 273 – – 5.5 320
Eudorina elegansc 251–292 233–268 7.9 – 320
Eudorina unicoccab – – – 9.5 586
Monoraphidium contortumc 5.2–5.7 4.7–5.0 10.4 – 30
Scenedesmus quadricaudac 99–104 91–95 8.5 200
Staurastrum pingueb – – – 57 9 450
Staurastrum sp.c 4680–4940 4480–4620 5.3 – 20 500
Volvox aureusb 99 – – 1.1 60
Diatoms
Asterionella formosab 292–349 – – 1.8 554–736
Asterionella formosac 243–291 104–136 55 – 650
Asterionella formosad 318 171 46 1.7 645
Aulacoseira binderanac 247–281 137–159 44 – 1 380
Aulacoseira granulatab 519 – – 4–5 847
Fragilaria capucinac 197–215 99–109 50 – 350
Fragilaria crotonensisb 272 – – 2 623
Stephanodiscus sp.c 115–122 62–66 47 – 310
Stephanodiscus hantzschiib 58 – – 0.9 600
Stephanodiscus rotulab 2770 – – 41 5 930
Tabellaria flocculosa 525 279 47 2.5 1 725
v. asterionelloidesb
Tabellaria flocculosa v. 383–407 205–210 47 – 820
asterionelloidesc
Cryptophytes
Cryptomonas ovatab 2090 – – 33 2 710
Dinoflagellates
Ceratium hirundinellab 18790 – – 237 43 740
a
Data for Planktothrix is per 1 mm length of filament.
b
From compilation of previously unpublished measurements of field-collected material, as specified in
Reynolds (1984a).
c
From Nalewajko (1966).
d
From later data of Reynolds (see 1997a).
Source: List assembled by Reynolds (1984a) from thitherto unpublished field data, and from data of
Nalewajko (1966).
planktic cells will be equivalent to between 0.41

and 0.47 pg µm−3 . In fact, the data suggest an
order-of-magnitude range, from 0.10 to 1.65 pg 105
µm−3 . In part, this reveals an inherent danger
Wc (cell dry weight / pg)

in the interpretation of a wide spread of raw 104
data through log/log representations. Caution is
required in interpolating species-specific deriva-
103
tions from a general statistical relationship. How-
ever, the additional difficulty must also be recog-
nised that the regression is fitted to phytoplank- 102
ton of considerable structural variation (vacuole
space, carbonate or silica impregnation of walls). 10
Accumulated dry mass is also influenced by the
physiological state of cells and the environmen-
tal conditions obtaining immediately prior to 10 102 103 104 105
the harvesting of the analysed material. It is, v (µm3)
perhaps, surprising that the data used to con-
struct Table 1.3 and Fig. 1.8, based on analyses Figure 1.8 Log/log plot of cell dry mass (Wc ) against cell
of material drawn largely from wild populations, volume (v) for various freshwater phytoplankers (data in Table
should show any consistency at all. Thus, the 3: t, Cyanobacteria; , diatoms; d, chlorophytes; others).
patterns detected are worthy of slightly deeper The equation of the regression is Wc = 0.47 v0.99 . Redrawn
investigation.
For instance, the variation in the percentage
ash content (where known) appears to be consid-
erable. Nalewajko (1966; see also Table 1.3) found silicon in lakes and oceans are frequently inad-
that ash accounted for between 5.3% and 19.9% equate to meet the potential demands of unfet-
(mean 10.2%) of the dry weights of 16 species of tered diatom development. On the other hand,
planktic chlorophytes but for between 27% and the requirement is obligate and within relatively
55% (mean 41.4%) of those of 11 silicified diatoms. narrow, species-specific ranges and uptake is sub-
ject to physiologically definable levels. The biolog-
1.5.2 Skeletal silica ical availability of silicon, its consumption and
Much of the high percentage ash content of the deployment, as well as the fate of its biogenic
diatoms is attributable to the extent of silicifi- polymers, are of special relevance to planktic
cation of the cell walls. Besides their value to ecology, as they may well determine the envi-
taxonomic diagnosis, the ornate and often del- ronmental carrying capacity of new diatom pro-
icate exoskeletal structures, celebrated in such duction. Thus, they have some selective value for
photographic collections as that of Round et particular types of diatom, or for other types
al. (1990), command wonder at the evolution- of non-siliceous plankter, when external supplies
ary trait and at the genetic control of frustule are substantially deficient.
assembly. There are ecological ramifications, too, The cells of all living organisms have a
arising from the relatively high density of the requirement for the small amounts of silicon
deposited silicon polymers, seemingly quite oppo- involved in the synthesis of nucleic acids and
site to the adaptive requirements for a plank- proteins (generally <0.1% of dry mass: Sullivan
tic existence. Moreover, the amounts of silicon and Volcani, 1981). However, it is the demands
consumed in the development of each individ- of those groups of protistans and poriferans that
ual diatom cell have to be met by uptake of sili- characteristically employ silicon in skeletal struc-
con dissolved in the medium. Whatever may have tures – notably diatoms, other chrysophytes radi-
been the situation at the time of their evolu- olarians and sponges – that impinge most on
tion, the present-day concentrations of dissolved the geochemical cycling of silicon (Simpson and
28 PHYTOPLANKTON
Volcani, 1981). In passing, it should be noted as fenestration, strengthening ribs, bracing struts
that skeletal silica also makes up some 10% of and spines. Later data on some of the species of
the dry weight of the grasses, whose co-evolution diatom considered by Einsele and Grim (1938)
with the mammals and relative abundance dur- suggest that the area-specific silicon content is
ing the tertiary period may have been responsi- particularly responsive to increasing size (see
ble for the long-term fluctuations in the export Table 1.5).
and availability of the main soluble source of
silicon (monosilicic acid: Siever, 1962; Stumm 1.5.3 Organic composition
and Morgan, 1996) in the aquatic environments Discounting the typical 5–12% ash (possibly up to
(Falkowski, 2002). 80% in the case of some diatoms) content of air-
For the moment, our concern is with the dried plankters, the balancing mass is supposed
cell content of silicon. Most is deposited as to be the organic components of the cell, derived
a cryptocrystalline polymer of silica ((SiO2 )n ), from the living protoplasm. In comprising mainly
resembling opal (Volcani, 1981). The silica con- proteins, lipids and condensed carbohydrates,
tents of several species of planktic diatoms have albeit in variable proportions, the elemental com-
been derived, either by direct analysis or, indi- position of the ash-free dry material is dominated
rectly, from the depletion of dissolved silicon by carbon (C), hydrogen (H), oxygen (O) and nitro-
by a known specific recruitment of cells by gen (N), together with smaller amounts of phos-
growth. Unlike other elements critical to their phorus (P) and sulphur (S). At least 14 other ele-
survival, diatoms take up scarcely more silicon ments (Ca, Mg, Na, Cl, K, Si, Fe, Mn, Mo, Cu, Co,
than is immediately required to form the frus- Zn, B, Va) are consistently recoverable if sufficient
tules of the next generation (Paasche, 1980; Sulli- analytical rigour is applied (Lund, 1965). It is
van and Volcani, 1981). As already indicated, the impressive that, whilst alive, every planktic cell
amounts deposited are generally quite species- had not only the capability of taking up these
specific (Reynolds, 1984a; see also Table 1.4), elements from extremely dilute media but also
at least when variability in cell size is taken the success in doing so. This should be borne
into account (Lund, 1965; Jaworski et al., 1988). in mind when interpreting the relative quanti-
Cell-specific silicon requirements differ consid- ties in which these elements occur in the dry
erably among planktic species, reportedly rang- matter of cells, because not all were necessar-
ing between 0.5% (in the marine Phaeodactylum ily equally available relative to demand. Besides
tricornutum: Lewin et al., 1958) and 37% of dry falling deficient in one element that is relatively
weight (in some freshwater Aulacoseira spp.: Lund, scarce, others that are relatively abundant may
1965; Sicko-Goad et al., 1984). tend to accumulate in the cell. In this way, the
In terms of mass of silicon per cell, broad rela- ratios in which the elements make up the ash-
tionships with the mean volume and with the free air-dried algal tissue often give a reliable
mean surface area are demonstrable (Fig. 1.9). reflection of the conditions of nutrient availabil-
The regressions reflect the increasing silicon ity in the growth medium. Compounded by the
deployment with increasing cell size but their special mechanisms that cells may have for tak-
slopes, within the interpretative limits of log/log ing up and retaining elements from unreliable
relationship, suggest that increased cell size environmental sources (to be explored in Chap-
is accompanied by a decreasing ratio of sili- ter 4), the absolute quantities of several of the
con : enclosed volume and an increasing ratio of component elements are liable to wide variation.
silicon : area. This is in accord with the expecta- Not surprisingly, the elemental ratios in
tion of Einsele and Grim (1938), among the ear- natural phytoplankton plankton have for long
liest investigators of the silicon requirements of aroused the interest of physiological ecologists
diatoms, that interspecific variations in deploy- and of biogeochemists and they have been
ment are related to differences in shape (surface much studied and reported. Absolute quantities
area-to-volume effects), together with the rela- do vary substantially, as do the ratios among
tive investment in such species-specific features them. Yet, remarkably, the same data indicate
Table 1.4 The silicon content of some freshwater planktic diatoms
Species Si (pg cell−1 ) Si nSiO2 Referencesa

(% dry wt) (% dry wt)
Asterionella 65.2 (45.5–80.3) – – Einsele and Grim (1938)
formosa 64.3 (46.9–82.1) 21 45 Lund (1950, 1965)
61.0 (52.1–69.9) – – Reynolds and Wiseman (1982)
64.3 20 43 Reynolds (1997a), given as Si
0.239W1.003 24 51 Regression of Jaworski et al.
(1988) fitted to measurements
made on a cultured clone of
diminishing size, data given as Si
Fragilaria 88.7 (79.8–100.9) – – Einsele and Grim (1938)
crotonensis 88.7 (88.2–89.2) 22 46 Lund (1965)
117.8 (98.0–142.7) – – Reynolds and Wiseman (1982)
49.7 (41.3–55.9) – – Reynolds (1973a)
Aulacoseira 61.0 – – Einsele and Grim (1938)
granulata 291.0 25 54 Prowse and Talling (1958)
138.0 27 57 Thitherto unpublished records
cited in Reynolds (1984a),
calculated indirectly from SiO2
uptake
Aulacoseira 111.2 (77.4–128.6) 30 63 Lund (1965)
subarctica
Stephanodiscus 3989 – – Einsele and Grim (1938)
rotula 1942 (1704–2182) – – Thitherto unpublished records
uptake
1075 – – Gibson et al. (1971)
978 32 69 Thitherto unpublished records
uptake
751 27 58 Thitherto unpublished records
uptake
Stephanodiscus 19.2 (15.0–22.1) 26 55 Lund (1965), Swale (1963)
hantzschii 16.4 28 60 Thitherto unpublished records
uptake
Tabellaria 185.4 – – Einsele and Grim (1938)
flocculosa (173.6–197.1)
var. 145.5 25 53 Lund (1965)
flocculosa (117.3–197.1)
a
Original citations quote content of SiO2 , except where stated otherwise.
30 PHYTOPLANKTON
dry weight (Ketchum and Redfield, 1949). A

slightly lower range (45–51%) was derived by
Round (1965) and Fogg (1975) from measure-
ments on freshwater phytoplankton. However,
the same sources of data showed extremes of
about 35%, in cells deprived of light or a sup-
ply of inorganic carbon, and 70%, if deficiencies
of other elements impeded the opportunities for
growth.
The importance of carbon assimilation by
photoautotrophs to system dynamics has encour-
aged interest in being able to make direct esti-
mates of organismic carbon content as a function
of biovolume. It will be obvious, from the recog-
nition of the variability in the absolute contents
of carbon, its proportion of wet or dry biomass,
and the relative fractions of ash and vacuolar
space, that any general relationship must be sub-
ject to a generous margin of error. For instance,
Mullin et al. (1966) derived an order-of-magnitude
range of 0.012–0.26 pg C µm−3 for a selection
of 14 marine phytoplankters that included large
and small diatoms. Reynolds’ (1984a) analysis of
data, pertaining exclusively to freshwater forms,
adopted simultaneous approaches to diatoms
Figure 1.9 The silicon content of selected diatoms from and non-diatoms. The relatively low ash content
the freshwater ( t) or marine () phytoplankton or other
and absence of large vacuoles among the latter
aquatic habitat (), plotted on log/log scales against (a) cell
permitted a much narrower relationship between
volume and (b) surface area. Ast refers to Asterionella, Fra to
Fragilaria and Ste to species of Stephanodiscus; Bac refers to carbon and biovolume (averaging 0.21–0.24 pg
Bacillaria, Dit to Ditylum, Nit to Nitzschia, Ske to Skeletonema C µm−3 ). Supposing carbon makes up a lit-
and Tha to Thalassosira. The equations of the least-squares tle under half of the ash-free dry mass and
regression fitted to the data in (a) is log [Si] = 0.707 log v – that dry mass averages 0.47 pg µm−3 (Fig. 1.8),
0.263 (r = 0.85); that for (b) is log [Si] = 1.197 log s – 1.634 this figure is highly plausible. For diatoms,
(r = 0.83). Redrawn, with permission, from Reynolds (1986a). there seemed little alternative but to calcu-
late carbon as a function of the silica-free dry
mass.
This approach does not satisfy the quest for
collectively that the quantities of the compo- a volume-to-carbon conversion for mixed diatom-
nents vary within generally consistent limits and, dominated assemblages, which continues to tax
though they do fluctuate, the ratios with other ecosystem ecologists. A recent re-exploration by
constituents do not vary by more than can be Gosselain et al. (2000) confirms the wisdom of sep-
reasonably explained in these terms. arating diatoms from other plankters. It provides
For instance, carbon generally makes up an evaluation of several of the available formu-
about half the dry organic mass of organic cells. laic methods for estimating the carbon contents
The normal content of phytoplankton strains of various diatoms.
cultured under ideal laboratory conditions of Of the other elements comprising biomass,
constant saturating illumination, constant tem- nitrogen accounts for some 4–9% of the ash-free
perature and an adequate supply of all nutri- dry mass of freshwater phytoplankters, depend-
ents, was found to be 51–56% of the ash-free ing on growth conditions (Ketchum and Redfield,
Table 1.5 The silicon content of some planktic diatoms relative to cell volume and surface area
Species Si v s Si Si Referencesa
(pg cell−1 ) (µm3 ) (µm2 ) (pg µm−3 ) (pg µm−2 )
Asterionella 61.0 630 860 0.097 0.071 Reynolds and Wiseman
formosa (1982)
Fragilaria 117.8 780 1080 0.151 0.109 Reynolds and Wiseman
crotonensis (1982)
Stephanodiscus 1075 8600 2574 0.125 0.418 Gibson et al. (1971)
rotula
1942 15980 4390 0.122 0.442 Thitherto unpublished records
calculated indirectly from
SiO2 uptake
SiO2 uptake
SiO2 uptake
Stephanodiscus 16.4 600 404 0.027 0.041 Thitherto unpublished records
hantzschii cited in Reynolds (1984a),
SiO2 uptake
a
All citations converted from the original published data quoted content in terms of SiO2 , by
multiplying by 0.4693.
1949; Lund, 1965, 1970; Round, 1965). Maxi- range 0.2–0.4% of ash-free dry mass. The inves-
mum growth rates are sustained by cells con- tigation of Mackereth (1953) of the phosphorus
taining nitrogen equivalent to some 7–8.5% of contents of the diatom Asterionella formosa, which
ash-free dry mass. Among freshwater algae, at reported a range of 0.06 to 1.42 pg P per cell, is
least, the phosphorus content is yet more vari- much cited to illustrate how low the cell quota
able, although again, maximum growth rate is may fall. The lower value, which is, incidentally,
attained in cells containing phosphorus equiva- corroborated by data in earlier works (Rodhe,
lent to around 1–1.2% of ash-free dry mass (Lund, 1948; Lund, 1950), corresponds to ∼0.03% of ash-
1965; Round, 1965). Growth is undoubtedly pos- free dry mass. On the other hand, cell phos-
sible at rather lower cell concentrations than phorus quotas may be considerably higher than
this but further cell divisions cannot be sus- the minimum (certainly up to 3% of ash-free
tained when the internal phosphorus content is dry mass is possible: Reynolds, 1992a), especially
too small to divide among daughters and can- when uptake rates exceed those of deployment
not be replaced by uptake. This concept of a mini- and cells retain more than their immediate needs
mum cell quota (Droop, 1973) has been much used (so-called luxury uptake). Uptake and retention of
in the understanding the dynamics of nutrient phosphorus when carbon or nitrogen supplies
limitation and algal growth: for phytoplankton, are limiting uptake (cell C or N quotas low) may
the threshold minimum seems to fall within the also result in high quotas of cell phosphorus.
32 PHYTOPLANKTON
Analogous arguments apply to the minimal the recommendation. The fourth line relates the
quota of all the other cell components. How- Redfield ratio to the base of carbon (= 100, for
ever, it is the variability in the carbon, nitrogen convenience), while further entries give elemen-
and phosphorus contents that is most used by tal ratios for specific algae, reported in the litera-
plankton ecologists to determine the physiologi- ture but cast relative to carbon. Chlorella is a fresh-
cal state of phytoplankton. Taking the ideal quo- water chlorophyte and Asterionella (formosa) is a
tas relative to the ash-free dry mass of healthy, freshwater diatom, having a siliceous frustule.
growing cells as being 50% carbon, 8.5% nitro- The ‘peridinians’ are marine. Approximations of
gen and 1.2% phosphorus, these elements occur the order of typical elemental concentrations in
in the approximate mutual relation 41C : 7N : 1P lake water are included for reference. They are
(note, C : N ∼6). Division by the respective atomic sufficiently coarse to pass as being applicable to
weights of the elements (∼12, 14, 31) and normal- the seas as well. The important point is that
ising to phosphorus yields a defining molecular plankters are faced with the problem of gath-
ratio for healthy biomass, 106C : 16 N : 1P. ering some of these essential components from
This ratio set is well known and is generally extremely dilute and often vulnerable sources.
referred to as the Redfield ratio. As a young marine As applied to phytoplankton, the Redfield
scientist, A. C. Redfield had noted that the com- ratio is not diagnostic but an approximation
position of particulate matter in the sea was sta- to a normal ideal. However, departures are real
ble and uniform in a statistical sense (Redfield, enough and they give a strong indication that the
1934) and, as he later made clear, ‘reflected . . . cell is deficient in one of the three components.
the chemistry of the water from which materials Extreme molecular ratios of 1300 C : P and 115
are withdrawn and to which they are returned’ N : P in cells of the marine haptophyte, Pavlova
(Redfield, 1958). The notion of a constant chem- lutheri, cultured to phosphorus exhaustion, and
ical condition was clearly intended to apply on of 35 C : P and 5 N : P in nitrogen-deficient strains
a geochemical scale but the less-quoted investi- of the chlorophyte Dunaliella (from Goldman et al.,
gations of Fleming (1940) and Corner and Davies 1979), illustrate the range and sensitivity of the
(1971) confirm the generality of the ratio to living C : N : P relationship to nutrient limitation.
plankton. Because the normal (Redfield) ratio is indica-
It is, of course, very close to the approxi- tive of the health and vigour that underpin rapid
mate ratio in which the same elements occur cell growth and replication, and given that depar-
in the protoplasm of growing bacteria, higher tures from the normal ratio result from the
plants and animals (Margalef, 1997). Stumm and exacting conditions of specific nutrient deficien-
Morgan (1981; see also 1996) extended the ideal cies, it is tempting to suppose that cells to which
stoichiometric representation of protoplasmic the normal ratios apply are not so constrained
composition to the other major components and must therefore be growing rapidly (Goldman,
(those comprising >1% of ash-free dry mass – 1980). It would follow that, given the stability
hydrogen, oxygen and sulphur) or some of those of the ratio in the sea, natural populations hav-
that frequently limit phytoplankton growth in ing close-to-Redfield composition are not only not
nature (silicon, iron). The top row of Table 1.6 nutrient-limited but may be growing at maximal
shows the information by atoms and the sec- rates. This may be sometimes true but there is
ond by mass, both relative to P. The third line is a possibility that biomass production in oceanic
recalculated from the second but related to sul- phytoplankton is less constrained by N or P than
phur. Unlike carbon, nitrogen or phosphorus, sul- was once thought (see Chapter 4). However, there
phur is usually superabundant relative to phyto- are other constraints on growth rate and upon
plankton requirements and plankters have no nutrient assimilation into new biomass, which
special sulphur-storage facility. Following Cuhel may tend to uncouple growth rate from nutrient
and Lean (1987a, b), sulphur is a far more stable uptake rate (see Chapter 5). Tett et al. (1985) pro-
base reference and deserving of wider use than it vided examples – of phytoplankton in continuous
receives. Unfortunately, few studies have adopted culture, of natural populations of Cyanobacteria
Table 1.6 Ideal chemical composition of phytoplankton tissue and relative abundance of major
components by mass
C H O N P S Si Fe References
Redfield atomic ratio 106 263 110 16 1 0.7 trace 0.05 Stumm and Morgan
(atomic (1981)
stoichiometry rel
to P)
Redfield ratio by mass 41 8.5 57 7 1 0.7 trace 0.1 Stumm and Morgan
(stoichiometry rel (1981)
to P)
Redfield ratio by mass 60 12 81 10 1.4 1 Stumm and Morgan
to S)
Redfield ratio by mass 100 16.6 2.4 Stumm and Morgan
to C)
Chlorella 100 15 2.5 1.6 trace Round (1965)
(dry weight rel
to C)
Peridinians 100 13.8 1.7 6.6 3.4 Sverdrup et al. (1942)
(dry weight rel
to C)
Asterionella 100 14 1.7 76 Lund (1965)
(dry weight rel
to C)
Medium (mol L−1 ) 10−3 102 102 10−4 10−6 10−3 10−2 <10−5 Author’s approxi-
mation but omitting
dissolved nitrogen
gas
stratified deep in the light gradient, and of spring distribution among all the photoautotrophic
blooms of the diatom Skeletonema costatum in a algae and cyanobacteria, the photosynthetic pig-
Scottish sea loch – where growth rates were kept ment chlorophyll a is also widely used as a con-
very low but the cell contents of carbon, nitro- venient index of phytoplankton biomass. This
gen and phosphorus stayed close to the Red- makes its contribution to cellular composition
field ideal in each instance. It is even possible extremely important to extrapolating to phyto-
that natural cells do not drift as far from the plankton abundance and to its use as a base for
ideal as it possible to force them under labora- estimating phytoplankton productivity. Although
tory conditions. How cells balance availability, there have been many values published alluding
uptake, storage and self-replication over the to the absolute chlorophyll a contents of freshwa-
period of a single generation will be explored in ter phytoplankton (some reviewed in Reynolds,
Chapter 4. 1984a), these quantities are now known to be
so variable that they have little value by them-
1.5.4 Chlorophyll content selves. The variability is most obviously linked to
Besides being a distinguishing constituent the cell’s requirement for carbon and the light
of phytoplankton and having a universal energy available to drive its fixation. In broad
34 PHYTOPLANKTON
dry weight) between its deep-stratified and free-

1000 mixed phases (Reynolds 1997a).
On the other hand and analogously with
Redfield stoichiometry, the probabilistic relation-
100 ships derived from mixed populations over peri-
Cell chlorophyll a content / pg
ods of time point to 0.003–0.007 pg µm−3 , or

roughly 0.013–0.031 pg chl (pg cell C)−1 , or
10 0.7–1.6% of the ash-free dry mass as each being
typical. For many purposes, the common approx-
imations that chlorophyll accounts for 1% of the
dry mass and about 2% of the value of the cell
1
carbon quota are not at all unreasonable aver-
age estimates. Indeed, field chlorophyll measure-
3 ments are commonly converted to approximate
0.1
producer biomass, expressed as active cell carbon
1 by the application of a ratio 50 : 1 by weight. A
4
margin of variation, from 30 : 1 to 70 : 1, should
2
nevertheless be allowed. It should be borne in
0 10 102 103 104 105
Cell volume ( µm3)
mind too that the amount of chlorophyll a is
proportionate to cell volume and not cell num-
Figure 1.10 Log/log plot of cell chlorophyll-a content ber, bigger cells carrying proportionately more
against cell volume of various freshwater phytoplankters (data chlorophyll than small ones, and that there may
in Table 1.3): t, Cyanobacteria; d, diatoms; , chlorophytes; be systematic interphyletic differences in the typ-
, others). Regression equations are fitted to the data for ical cell-specific contents. The plot of data taken
Cyanobacteria (1, log chl = 1.00 log v – 2.26), diatoms (2, log from Reynolds (1984a) emphasises both points
chl = 1.45 log v – 3.77) and chlorophytes (3, log chl = 0.88
(Fig. 1.10).
log v – 1.51) and to all points (4, log chl = 0.98 log v – 2.07).
Redrawn from Reynolds (1984a).
terms, the weaker is the photon flux, and the

1.6 Marine phytoplankton
greater is the probability of limitation of growth
rate by light, then the greater is the need for The information on the size, morphology and ele-
the light-harvesting centres of which the chloro- mental composition of phytoplankton presented
phyll is an essential component. Where appropri- in this chapter has been dominated by relation-
ate, this behaviour may be accompanied by the ships detected among freshwater species. This is
production of additional quotas of accessory pig- partly attributable to the interests and experi-
ments (for a fuller discussion, see Chapter 3). Syn- ences of the author and not to any lack of cor-
thesis of chlorophyll a is also sensitive to nutri- responding data for the sea; useful data compila-
ent supply and deployment, directly or as a con- tions are to be found in, for instance, Mullin et
sequence of altered internal resource allocation. al. (1966), Strathmann (1967), Sournia (1978) and
The measurements of biomass-specific estimates Verity et al. (1992). However, it seemed more inter-
of chlorophyll a presented in Reynolds (1984a) esting to take the data and derivations of Mon-
range over an order of magnitude, between tagnes et al. (1994), corrected for live cell volumes
0.0015 and 0.0197 pg µm−3 of live cell volume, as opposed to those of material shrunk by preser-
which corresponds to 3 to 39 mg g−1 of dry mass vatives, and to compare their findings with the
(0.3% to 3.9%). The true range is probably wider: patterns presently discerned among freshwater
later data showed that the chlorophyll-a content species.
of just one species of cyanobacterium, Planktothrix Montagnes et al. (1994) compiled a thorough
cf. mougeotii, may vary nearly ninefold (0.45–3.9% compilation of the dimensions and volumes of
MARINE PHYTOPLANKTON 35
Table 1.7 Some cell measurements of cultured marine phytoplankton
Species shapea MLD v C N Chla

(µm) (µm3 ) (pg cell−1 ) (pg cell−1 )
Prasinophytes
Micromonas pusilla ps 2 2 0.8 0.1 0.01
Mantoniella squamata sph 4 25 3.6 0.6 0.15
Chlorophytes
Nannochloris ocelata sph 3 4 1.4 0.2 0.05
Dunaliella tertiolecta sph 8 201 41.7 7.9 1.8
Chlamydomonas sp. sph 19 3 300 969.7 129.6 11.3
Diatoms
Thalassiosira pseudonana cyl 4 20 5.9 0.94 0.2
Thalassiosira weissflogii syl 19 286 64.4 11.1 1.5
Detonula pumila cyl 36 4 697 355.2 61.4 7.4
Chrysophytes
Pelagococcus sp. sph 5 18 2.8 0.7 0.1
Pseudopedinella pyriformis ps 9 80 18.0 2.9 0.7
Apedinella spinifera sph 10 222 47.6 9.9 2.3
Haptophytes
Emiliana huxleyi sph 5 25 4.6 1.0 0.1
Pavlova lutheri ps 6 25 8.5 1.2 0.2
Isochrysis galbana sph 6 38 7.0 1.2 0.2
Phaeocystis pouchetii ps 6 45 5.5 1.3 0.1
Chrysochromulina herdlensis sph 7 74 8.2 1.9 0.2
Prymnesium parvum ps 8 79 15.1 2.0 0.3
Coccolithus pelagicus sph 12 620 65.5 14.5 1.9
Cryptophytes
Rhodomonas lens ps 10 203 40.7 11.4 0.7
Chroomonas salina ps 11 167 32.4 7.9 0.9
Pyrenomonas salina ps 12 181 32.4 7.0 1.4
Cryptomonas profunda ps 17 765 104.7 21.0 2.6
Dinophytes
Gymnodinium simplex ps 11 224 38.0 8.3 0.7
Gymnodinium vitiligo ps 14 683 113.8 22.6 2.3
Gyrodinium uncatenatum ps 32 11 246 2275.3 441.0 37.3
Gymnodinium sanguineum ps 56 31 761 2913.2 688.4 57.4
Karenia mikimotoi ps 24 3 399 513.9 93.8 12.9
Raphidophytes
Heterosigma carterae ps 16 362 87.8 17.7 3.8
a
Shapes: cyl, cylindrical; ps, prolate spheroid; sph, spherical.
Source: After Montagnes et al. (1994).
36 PHYTOPLANKTON
and a statistically predictable pattern of the rela-

tive quantities of the various constituents. This
simple fact contributes to the fascination for
students of phytoplankton ecology as the sub-
ject embraces the observable wonder of the sea-
sonal replacement of one dominant among many
species by another among others, as well as the
opportunity to express the dynamics of produc-
tion and attrition and of population wax and
wane in empirical terms interlinked by powerful
and predictable statistical relationships.
The scene is set for the subsequent chapters.
1.7 Summary
The chapter provides an introduction to phy-

toplankton. The phytoplankton is defined as
a collective of photosynthetic microorganisms,
adapted to live partly or continuously in the open
Figure 1.11 Log/log relationships between (a) cell carbon, of the seas, of lakes (including reservoirs), ponds
(b) cell nitrogen and (c) chlorophyll-a content and cell volume and river waters, where they contribute part or
in the marine phytoplankters listed in Table 1.7. Redrawn, most of the organic carbon available to pelagic
with permission, from Montagnes et al. (1994).
food webs. Although their taxonomy is currently
undergoing major revision and even the phylo-
30 or so species of phytoplankton, representative genies are questioned, it is difficult to be cat-
of various phyla and covering a good range of egorical about the species representation and
cell sizes, together with measured cell quotas of phyletic make-up of phytoplankton. It is reason-
carbon, nitrogen, protein and chlorophyll a. The able to point to the description of some 4000 to
material reproduced in Table 1.7 represents but a 5000 species from the sea and, probably, a similar
fragment of the original. Live cell volumes cover order of species from inland waters. The species
a similar series of magnitudes as the freshwa- belong to what appear to be 14 legitimate phyla,
ter species listed in Table 1.2. The carbon content coming from both bacterial and eukaryotic pro-
of cells varied between 0.08 and 0.4 pg µm−3 , tist domains. In both the marine and the fresh-
with a mean value of 0.20. The ratio of carbon- water phytoplankton, there is a wide diversity of
to-nitrogen varied between 3.6 and 7.6 by mass, size, morphology, colony formation. Though gen-
with a mean of 5.43. Chlorophyll a fell within the erally microscopic, phytoplankton covers a range
range 0.001–0.009 pg µm−3 . of organism sizes comparable to that spanning
Some log/log relationships from the work forest trees and the herbs that grow at their
of Montagnes et al. (1994) are plotted in Fig. bases.
1.11. Carbon, nitrogen and chlorophyll are each The early history of phytoplankton studies
closely correlated to live cell volume and in a is recapped. Although a knowledge of some of
way which is similar to the corresponding rela- the organisms goes back to the invention of the
tionships among the freshwater phytoplankton. microscope, and many genera were well known
Thus (and, again, as is true for the freshwater to nineteenth-century microscopists, their role
phytoplankton), despite a remarkable diversity in supporting the aquatic food webs of open
of phylogeny and morphology, as well as a 5- water, culminating in commercially exploitable
orders-of-magnitude range of cell volumes, there fish populations, was not realised until the 1870s.
is an equally striking pattern of cell composition The early work by Müller, Haeckel and Hensen
SUMMARY 37
(who invented the name ‘plankton’) is briefly mineral-reinforced walls, carbon accounts for
described. Some of the terms used in plankton about 50% of the dry mass, nitrogen about 8–9%
science are noted with their meanings, while and phophorus between 1% and 1.5%. Relative to
those that appear still to be conceptually useful phosphorus, these amounts correspond to a prob-
are singled out for retention. abilistic atomic ratio of 106 C : 16 N : 1 P, close to
Despite variation of several orders of magni- the so-called Redfield ratio for particulate matter
tude in the sizes of plankters, there is a pow- in the ocean. It is also similar to the composi-
erful trend towards conservatism of the surface- tion of most living protoplasm. The amounts are
to-volume ratio, which is achieved through dis- related also to hydrogen, oxygen, silicon, sulphur
tortion and departure from the spherical form and iron. Up to 12 other elements are regularly
among the larger species. This aids exchange of present in phytoplankton in trace proportions.
gases, nutrients and other solutes across the cell Departures from the ratio are rarely systematic,
surface and it also has some role in prolongation merely indicative of one of the highly variable
of suspension. In an apparently diametrically components falling to the minimum cell quota.
opposite trend, some algae form mucilaginous The amount of chlorophyll a is also highly
coenobia that have very low surface-to-volume variable according to growth conditions but nev-
ratios. When it is combined with some other ertheless tends to average about 1% of the ash-
power of motility, the streamlining effect allows free dry mass of the cell and to represent about
the colony to move relatively quickly through 2% of the elemental carbon. A carbon:chlorophyll
water and to move to a more favourable position value of 50 : 1 is considered typical but it may
in the water column. vary routinely between about 70 : 1 (cells in high
The construction and composition of plank- light) to 30 : 1 or lower (in cells exposed to con-
ton are critically reviewed. Apart from a vari- sistently low light).
ety of scales, exoskeleta, plastid type and pig- Despite the extreme diversity of phylogeny,
ment composition, the ultrastructural compo- morphology and size, both the marine and the
nents and architecture of the living protoplasm freshwater phytoplankton are characterised by a
are comparable among the phytoplankton. Sim- striking and statistically predictable blend of ele-
ilarly, the elemental make-up of the protoplast mental constituents. This proves very helpful in
is similar among all groups of phytoplankton, quantifying production and attrition processes
ideally occurring in approximately stable rela- contributing to the dynamics of natural, func-
tive proportions. Discounting the ash from the tioning assemblages of plankton.
Chapter 2
Entrainment and distribution in the pelagic
the persistence of the clone that residence is con-

2.1 Introduction tinuous, only that individuals of any given gener-
ation spend sufficient of their life in the photic
The aims of this chapter are to develop an appre- zone to make the net autotrophic gains in syn-
ciation of the adaptive requirements of phyto- thesised carbon, over the burden of respiration,
plankton for pelagic life and to demonstrate the to be able to sustain the next cell replication. The
consequences of its embedding in the movements point here is that the essential adaptation is to
of the suspending water mass. The exploration maximise the exposure to adequate light, by any
begins by dismissing the simplistic notion that appropriate mechanism.
the essential requirement of plankton is to pre- The mechanisms for this are not self-evident,
vent or minimise the rate of sinking, in the sense unless the behaviour of the water itself is taken
that this will prolong its residence in the upper into account. For this is the feature that the clas-
part of the water column. This would be a clear sical explanation of phytoplankton adaptations
nonsense, were there no counteractive mecha- rather omits – that, at every scale, the water is
nism to ensure that organisms start out at the never a passive component. Under the influence
top of the water in the first instance. Moreover, of its warming and cooling, of the influence of
slow sinking from the upper layers is of illusory gravity, the pull of the Moon, of the work of wind
respite if the downward passage to depths beyond and even of the rotation of the Earth, water is in
the adequacy of penetrating light, whether that motion. Some of these inputs are continuously
is 50 cm or 50 m beneath the water surface, variable, and their various interactions with the
is inevitable, unless there is some mechanism internal viscous forces contribute to a spectrum
for the organism’s return. Manifestly, it is not of motion that is characteristically variable, in
enough just to reduce the rate of irreversible sink- both time and space.
ing to qualify as a phytoplankter. Thus, there is an explicit, inescapable and
Prolonged residence in the upper insolated variable velocity component to the medium.
layers of the open water of lakes and seas (the Moreover, the movement is, almost always, turbu-
photic zone) is, without doubt, a primary require- lent, so that flow tends to be in billowing eddies
ment of the individual phytoplankter, if it is to rather than along direct trajectories. Such move-
synthesise sufficient organic carbon to build the ments are capable of alternately enhancing or
tissue of the next generation. The survival of the counteracting the intrinsic velocity of the verti-
genetic stock and the seed population capable cal tendency of the settling plankter and, within
of providing the base of subsequent generations the finite bounds of the water mass, may force
may also depend upon the survival of a rela- its lateral displacement or even push it upwards.
tively small number of extant individuals. It is These possibilities are the basis of the
not a condition either for the individual or for principle of entrainment of phytoplankton in
MOTION IN AQUATIC ENVIRONMENTS 39
Table 2.1 Comparison of the physical properties of air, pure water and sea water
Air Pure water Sea water

Maximum density, ρ w (kg m−3 ) 1.2 1 × 103 ∼1.03 × 103
Absolute viscosity, η (kg m−1 s−1 ) 1.8 × 10−5 1 × 10−3 ∼1.1 × 10−3
the motion of natural water bodies. Empirical shared between lakes, rivers and the atmosphere,
description of entrainment relies on the anal- is less than 0.02%. However, even this fraction,
ogous relativity between the intrinsic sinking totalling c. 225 000 km3 , is overwhelmingly dom-
velocity of the plankter (ws ) and the turbulent inated by the volume of standing inland waters:
velocity of the motion (u∗ ) This immediately the 13 largest lakes in the world (by volume)
introduces an anomaly, which must be addressed alone hold 160 000 km3 (Herdendorf, 1990). At
at once. The idea is that the smaller is ws relative any moment of time, most of this volume is actu-
to u∗ then the more complete is the entrainment ally so inhospitable to primary producers that it
of the particle in the motion. This is another way is not conducive to phytoplankton development
of saying that the best way to ensure entrain- but, because it is fluid and in persistent motion,
ment is to minimise the rate of sinking. Isn’t all the volume is potentially available, sooner or
this just the idea that was so summarily dis- later. The global rate of the hydrological renewal,
missed in the first paragraph? No, for the com- in the cycle of precipitation, flow and evapora-
ment was directed to the redundant, notional tion, results in an estimated annual loss from
context of a slow settlement of plankters through the ocean of 353 000 km3 , made good by direct
a static water column. What is really needed, if precipitation (roughly, 324 000 km3 ) and net river
full entrainment is the goal, is a slow rate of set- run-off from the land masses (∼29 000 km3 ). The
tlement relative to the water immediately adja- theoretical replacement time for the ocean is
cent to the organism and its instantaneous tra- thus around 3800 years.
jectory and velocity.
The approach adopted in developing this
chapter is to first consider the nature, scale and
2.2.1 Physical properties of water
How this vast and enduring body of water reacts
variability of water movements and the estima-
to the forces placed upon it is related to the some-
tion of u∗ . Then the question of settling veloci-
what anomalous physical properties of water
ties, buoyant velocities and swimming rates (ws )
itself. Given its low molecular weight (18 dal-
is reviewed, before the consequences on spa-
tons), water is a relatively dense, viscous and
tial and temporal distributions are considered at
barely compressible fluid (see Table 2.1 for ref-
the end.
erence), with relatively high melting and boiling
points. This behaviour is due to the asymmetry
of the water molecule and to the fact that the
2.2 Motion in aquatic environments two hydrogen atoms, each sharing its electron
with the oxygen atom, are held at a relatively nar-
The aquatic environment is the greatest habitat row angle on one side of the molecule. In turn,
to be continuously exploited by organisms. Liquid this gives a polarity to the molecule, one side
water presently covers about 71% of planet Earth, (the ‘hydrogen side’) having a net positive charge
the sea alone occupying 361.3 × 106 km2 of it. The and the other (the ‘oxygen side’) a net negative
estimated volume of the sea (∼1 350 000 000 km3 ) one. The molecules then have a mutual attrac-
accounts for 97.4% of all the water on the planet. tion, giving rise to the formation of aquo poly-
Taking off the volume stored in the polar ice mers. It is the complexation into larger molecules
caps (27.8 × 106 km3 ) and the amount stored which raises the melting point of what is oth-
in the ground (∼8 × 106 km3 ), the balance, erwise a low-molecular-weight compound into
40 ENTRAINMENT AND DISTRIBUTION IN THE PELAGIC
Figure 2.1 Plots showing (a) the

density and (b) the absolute
viscosity of pure water as a function
of temperature. Redrawn, with
permission, from Reynolds (1997a).
the range we perceive as being normal. As tem- opposite-charged poles of the water molecules
perature is raised and the motion of molecules all contribute further modifications to the poly-
is increased, individual molecules break from merisation. Individual ions become surrounded
the complexes. In most liquids, the molecules by water molecules in a hydrated layer, disrupt-
come to occupy more space, that is the liquid ing their structure and altering the properties of
expands and the density decreases. In water, this the liquid. The salinity of sea water ranges from
effect is countered by the fact that the liberated a trace (in some estuaries and adjacent to melt-
molecules fall within the complexes, so that the ing glaciers) to a maximum of about 40 g kg−1
same number of molecules occupies less space, (the Red Sea; note that this is greatly exceeded
leading to increased density. In pure water, the latin some inland lakes). In most of the open ocean,
ter effect dominates up to 3.98 ◦ C; above this tem- salinity is generally about 35 (±3) g kg−1 , having
perature, the separation of molecules becomes, a density of 27 (±2) kg m−3 greater than pure
progressively, the dominant effect and the liquid water of the same temperature. The presence of
expands accordingly (see Fig. 2.1). salt depresses not just the freezing point but also
The molecular behaviour explains not only the temperature of maximum density. When the
why fresh water achieves its greatest density at salt content is about 25 g kg−1 , these tempera-
close to 4 ◦ C but also why, under appropriate con- tures coincide, at −1.3 ◦ C. Thus, in most of the
ditions, ice forms at the surface of a lake (where, sea, the density of water increases with lowering
incidentally, it insulates the deeper water against temperatures right down to freezing. Sea ice does
further heat loss to the atmosphere), and why, not form at the surface, as does lake ice, simply as
with every degree step above 4 ◦ C, the difference a consequence of cooling of the water. Normally,
in density also becomes greater. Limnologists are some other component (dilution by rain and or
well aware of the effect this has in enhancing the terrestrial run-off) is necessary to decrease the
mechanical-energy requirement to mix increas- density of the topmost water.
ingly warmed surface waters with the dense lay- Molecular behaviour influences the
ers below; the limnetic ecologist is familiar with temperature-dependent viscous properties of
the impact of both processes on the environ- water. Viscosity, manifest as the resistance
ments of phytoplankton. provided to one water layer to the slippage of
The same principles apply in the sea and in another across it, decreases rapidly with rising
salt lakes, except that the higher concentrations temperature (Fig. 2.1b); according to the stan-
of dissolved ionic salts, their separation into con- dard definition of viscosity, this means there is
stituent charged ions and their attraction to the a decreasing resistance of one water layer to the
MOTION IN AQUATIC ENVIRONMENTS 41
slippage of another across it, for the same given

difference in temperature. Viscosity is greater
in sea water than in pure water: an increment
about 0.1 × 10−3 kg m−1 s−1 applies to water
containing 35 g kg−1 over a normal temperature
range. High viscosity, like large differences in
density, is an effective deterrent to physical
mixing and mechanical heat transfer.
2.2.2 Generating oceanic circulation

It is also relevant to the generation of major
flows that water has a high specific heat, which
in essence means that it takes a lot of heating
to raise its temperature (4186 J to raise 1 kg by
1 ◦ C, or by 1 kelvin). However, it is just as slow to
lose it again, save that evaporation, turning water
liquid into vapour, is very consumptive of accu-
mulated heat (2.243 × 106 J kg−1 ). Nevertheless,
the exchange of incoming and outgoing heat is
a major component in the physical behaviour of
the oceans. In fact, the patterns of motion are
subject to a complex of drivers and the outcome
is usually complicated. Empirical description of
motion can only be probabilistic and, in any case,
far beyond the scope of this book. In essence, the
Figure 2.2 (a) The spectrum of the solar flux at ground
energy to drive the circulation comes from the
level compared to that of the ‘solar constant’ at the top of
Sun. Because of its relevance also to local vari- the atmosphere; the visible wavelengths (light) are shown
ability in heat exchanges and its obvious links hatched. (b) Daily integrals of undepleted solar radiation at
to the energy fluxes used in photosynthesis, a the top of the atmosphere, shown as a function of latitude
deeper consideration is given later to the solar (degrees) and time of year in the northern hemisphere. The
irradiance fluxes. For the moment, it is necessary approximate match for the southern hemisphere is gained by
to accept that the proportion of the solar energy displacing the horizontal scale by 6 months. Redrawn, with
flux that penetrates the atmosphere to heat the permission, from Reynolds (1997a).
surface of the sea or lake is first a function of
the solar constant. This is the energy income to a diminishes with latitude and not even the com-
notional surface held perpendicular to the solar bination of the tilt of the Earth’s axis and its
electromagnetic flux, before there is any reflec- annual excursion round the sun even this out.
tion, absorption or consumption in the Earth’s The plots in Fig. 2.2b show the annual varia-
atmosphere. Confusingly, it is not constant, as tion in the undepleted daily flux at each of the
the elliptical orbit of the Earth around the Sun selected northern-hemisphere latitudes.
varies around the mean distance (149.6 × 106 km) Although the highest potential daily heat flux
fluctuates during the year within ±2.5 × 106 km. is everywhere quite similar, sustained heating
Besides, the heat radiated from the Sun also fluc- through the year is always likely to be greatest
tuates. Nevertheless, there is a valuable reference in the tropics but never for such long diurnal
(∼1.36 kW m−2 ) against which the absorption, periods as occur at high latitudes in summer.
reflection and backscatter by dust, water vapour On a rotating but homogeneously water-covered
and other gases and, especially, clouds can be Earth having a continuously clear atmosphere,
scaled. Even before those losses are deducted (see there would be considerable latitudinal differ-
Fig. 2.2a), however, the heat flux per unit area ences in the heat flux directed to the surface
(other things being equal, sufficient at the equa- atmosphere; its movements are subject to analo-
tor to raise the temperature of the top metre by gous global-scale forces. It is its much lower mass,
∼1 ◦ C every daylight hour, for 12 months of the density and viscosity that gives the impression
year). Ignoring night-time losses and any convec- of a different behaviour. In fact, there is close
tional heat penetration, the expectation is that coupling between them, in the sense that strong
the now less dense surface water, heated in the winds generate waves, drive surface drift cur-
tropics, would spread out to the higher latitudes, rents and force the transfer of some of mechan-
at least until it had cooled to the temperature of ical energy to the water column. At the same
the high-latitude water. In compensation, water time, differences in inertia and in specific heat
must be drawn from the higher latitudes, via bring differential rates of warming over land
a deeper return flow. In this way, we may visu- and water, leading to differences in air pres-
alise the initiation of a convectional circulation sure and the superimposition of prevalent wind
of hemispheric proportions. conditions.
This simplified conception is complicated by The predictability of wind action on individ-
several interacting factors. The rotation of the ual water bodies is generally difficult (as are most
Earth causes everything on it, including oceanic aspects of weather forecasting), save in probabilis-
drift currents, to move eastwards. As surface tic terms, based on the statistics of experience
water moves poleward, however, the rotational and pattern recognition. However, the linkage
speed of the ground under it lessens and the iner- between wind effects and the motion of water
tia of the trajectory tends to pull it ahead of the in which phytoplankton is resident has been
solid surface, the relative motion thus drifting deeply explored. Broad flow patterns of surface
further east. This easterly deflection, known as currents in the oceans have been discerned and
Coriolis’ effect, acts like a laterally applied force. described by mariners over a period of centuries
The positions of the continental land masses, of and since committed to oceanographers’ maps
course, obstruct the free development of these (for an overview, see Fig. 2.3). Patterns of circu-
motions while the irregularity of their distribu- lation in certain large lakes have been described
tion gives rise to compensatory latitudinal flows over a rather shorter period of time (e.g. Mor-
among the major oceans (especially in the south- timer, 1974; Csanady, 1978) and those of many
ern hemisphere). The variable depth of the ocean smaller lakes have been added in recent years;
floor also interferes with the passage of deep the example in Fig. 2.4 is just one such instance.
return currents which, locally, may be forced to
deflect upwards and to ‘short-circuit’ the poten-
tial hemispheric circulation.
2.3 Turbulence
Also superimposed upon the circulatory pat-
tern are the tidal cycles exerted by the variable 2.3.1 Generation of turbulence
gravitational pull on the water exerted by the Despite the rather self-evident relationship that
rotation of the Moon around the Earth, hav- plankton mostly goes where the water takes it,
ing frequencies of ∼25 hours and ∼28 days. The the large-scale motion of water bodies tells us
effect of tides on the pattern of circulation may frustratingly little about what the conditions of
not be large in the open ocean but may dom- life are like at the spatial scales appropriate to
inate inshore circulations near blocking land- individual species of phytoplankton (generally
forms that may trap tidal surges (the Bay of <2 mm), or about the trajectories followed by the
Fundy, between Nova Scotia and New Brunswick, individual phytoplankter whose survival depends
experiences the greatest tidal extremes in the on its passing a reasonable fraction of its life
world – over 13 m – and some of the most aggres- in the insolated upper reaches of the water col-
sive tidal mixing). umn. Although it has long been appreciated that
Surface currents, especially in lakes, are prox- the energy of the major circulations is dissipated
imally influenced by wind. Wind is the motion through cascades of smaller and smaller gyra-
of air in the adjacent fluid environment, the tory structures, now called the Kolmogorov eddy
Figure 2.3 The currents at the surface of the world’s oceans in the northern winter. Redrawn, with permission, from Harvey (1976).
of turbulence have gradually given way to a

developing turbulence theory (Levich, 1962; Ten-
nekes and Lumley, 1972) but it is the remark-
able progress in instrumentation and direct sens-
ing of turbulence (see especially Imberger, 1998),
together with the rapid assimilation of its quan-
tification into physical oceanography and lim-
nology (see, for example, Denman and Gargett,
1983; Spigel and Imberger, 1987; Imberger and
Ivey, 1991; Mann and Lazier, 1991; Imboden and
Wüest, 1995; Wüest and Lorke, 2003) that have
given most to the characterisation of the envi-
ronment of the phytoplankton.
A helpful starting point is to envisage a com-
pletely static column of water (Fig. 2.5a). If its
upper surface is subjected to a mild horizontal
force, τ , then water molecules at the air–water
interface are dragged across the surface in the
direction of the force. Their movement is trans-
mitted to the layer below, which also begins to
move, albeit at a lesser velocity. Further down-
ward propagation soon leads to a configuration
envisaged in Fig. 2.5b, each layer of molecules
sliding smoothly over the one below, in what
is described as laminar flow. The structure con-
forms to a vertical gradient of horizontal veloci-
ties, u, the steepness of which is defined by the
Figure 2.4 Model reconstructions of the near-surface
currents generated in Esthwaite Water, UK, by steady winds
differential notation, du/dz (literally the incre-
of 5 m s−1 and various orientations. The longest axis of the ment or decrement of horizontal velocity for
lake is approximately 2.2 km. Redrawn, with permission, from a small increment in the vertical direction, z).
Falconer et al. (1991). While the condition of laminar flow persists, the
ratio between the applied force (per unit area)
and the velocity gradient corresponds to the abso-
spectrum after one of its most famous investiga- lute viscosity of the water, η. That is,
tors (Kolmogorov, 1941), the impact of the lower
end of the series on the behaviour of plankton η = τ (du/dz)−1 (2.1)
had remained philosophically and mathemati- Adopting the appropriate SI units for force (N =
cally obscure. My cumbersome attempts to over- newtons, being the product, mass × acceleration,
come this deficiency in an earlier text (Reynolds, may be expressed as kg m s−2 ) per unit area (m2 ),
1984a) only emphasise this frustration. Looking for velocity (m s−1 ) and for vertical distance (m),
back from the present standpoint, they serve the absolute viscosity is solved in poises (P =
as a point of reference for just how far the kg m−1 s−1 ). The values plotted in Fig. 2.1b approx-
appreciation and quantification of turbulence, imate to 10−3 kg m−1 s−1 . Caution over units is
together with their impacts on particle entrain- urged because it is common in hydrodynamics
ment, have moved on during the subsequent cou- to work with the kinematic viscosity of a fluid, ν,
ple of decades. which is equivalent to the absolute viscosity with
‘Turbulence’ testifies to the failure of the the density (ρ w ) divided out:
molecular structure of a fluid to accommodate
introduced mechanical energy. The mysteries ν = η (ρw )−1 = τ (ρw du/dz)−1 m2 s−1 (2.2)
TURBULENCE 45
Figure 2.5 The generation of turbulence by shear forces. In The transition between ordered and turbulent
(a), the water beneath the horizontal surface is unstressed flow patterns has long been supposed to depend
and at rest. In (b), a mild force, τ , is applied which that drags upon the ratio between the driving and viscous
water molecules at the surface in the direction of the force; forces; this ratio is expressed by the dimension-
(b) their movement serves to drag those immediately below, less Reynolds number, R e:
and so on, giving rise to an ordered structure of laminar flow.
In (c), the transmitted energy of the intensified force can no
R e = (ρw ul a )η−1 = ul a ν −1 (2.3)
longer be dissipated through the velocity gradient which
breaks down chaotically into turbulence. Redrawn, with
where la is the length dimension available to
the dissipation of the energy, usually the depth
of the flow. Turbulence will develop wherever
there is a sufficient depth of flow with suffi-
Given the density of water of ∼103 kg m−3 cient horizontal velocity. Solving Eq. (2.3) for a
(Fig. 2.1a), its kinematic viscosity approximates notional small stream travelling at 0.1 m s−1 in a
to 10−6 m2 s−1 . channel 0.1 m deep, R e ≈ 104 ; for a 10-mm layer
If the applied force is now increased suffi- in a well-established thermocline in a small lake
ciently, it begins to shear molecules from the subjected to a horizontal drift of 10 mm s−1 ,
upper surface of the water column. Thus, the R e ≈ 102 . The former is manifestly turbulent
smooth, ordered velocity gradient fails to accom- but the latter maintains its laminations. There
modate the applied energy; the structure breaks is no unique point at which turbulence develops
up into a complex series of swirling, recoiling or subsides; rather there is a transitional range
eddies (Fig. 2.5c). A new, turbulent motion is super- which, for water, is equivalent to Reynolds num-
imposed upon the original direction of flow. bers between 500 and 2000. The depth–velocity
The layer now assumes a net mean velocity in dependence of turbulence is sketched in
the same direction (ū m s−1 ). Now, at any given Fig. 2.6.
point within the turbulent flow, there would be The point in the spectrum where the eddies
detected a series of velocity fluctuations, acceler- are overwhelmed by molecular forces and col-
ating to (ū + u ) and decelerating to (ū − u ) lapse into viscosity is more difficult to pre-
m s−1 . Simultaneously, the displacements in the dict without information on their velocities.
vertical (z) direction introduce a velocity compo- As suggested above, it is now possible to mea-
nent which fluctuates between (0 + w ) and (0 − sure the velocity fluctuations directly with the
w ) m s−1 . This pattern is maintained for so long aid of sophisticated accoustic sensors but the
as the appropriate level of forcing persists. The quantities still need to be interpreted within a
driving energy is, as it were, extracted into the theoretical context. More significantly, the the-
largest eddies, is progressively dissipated through oretical framework can be used estimate the
smaller and smaller eddies of the Kolmogorov intensity of the turbulence from properties of
spectrum and is finally discharged as heat, as the the flow which are measurable with relative
smallest eddies are overwhelmed by viscosity. ease.
10 Figure 2.6 The onset of

turbulence as a function of velocity
and water depth. The boundary is
1 not precise, occurring at Reynolds
Depth, H / m
numbers between R e = 500 and

Re Turbulent R e = 2000. Plot based on Reynolds
0.1 =2
000 (1992b) and redrawn, with
Tra
Laminar ns itio permission, from Reynolds (1997a).
n
Re
0.01 =5
00
0.001 0.01 0.1 1

Velocity, u / m s−1
2.3.2 Quantifying turbulence coefficient of frictional drag on the surface (∼1.3

The key empirical quantification of turbulence × 10−3 ). The equation is imprecise for a number
is based upon the time-averaged velocity fluctu- of reasons, one being the interference in trans-
ations in the horizontal and vertical directions fer caused by surface waves. Nevertheless, the
(±u , ±w ). Because the summation of positive implied linear relationship, (u∗ ) ≈ U/800, is suffi-
and negative measurements must tend quickly ciently robust in the wind speed range 5–20 m s−1
to zero, the positive roots of their squares, (u∗ ∼6 × 10−3 to 2.5 × 10−2 m s−1 ) for it to stand
[(±u )2 ]1/2 , [(±w )2 ]1/2 , are cumulated instead. The as a good rule-of-thumb quantity.
turbulent intensity, (u∗ )2 , comes from the product A general derivation for water flowing down a
of their root mean squares: channel in response to gravity relates (u∗ ) to ū, as
has been developed, inter alia, by Smith (1975):
(u∗ )2 = [(±u )2 ]1/2 [(±w )2 ]1/2 m2 s−2 (2.4)
(u∗ ) = ū[2.5 ln (12H /rp )]−1 m s−1 (2.6)
The square root (u∗ ) has the dimensions of veloc-
ity and is known variously as the turbulent velocity, where H is the depth of the flow and rp is the
the friction velocity or the shear velocity. In natural roughness of the bed, as defined by the heights of
systems, both (u∗ ) and (u∗ )2 are extremely vari- projections from the bottom. The ratio between
able, in time and in space, depending upon the them is such that (u∗ ) is generally 1/30 to 1/10
energy of the mechanical forcing and the speed the value of ū. A general relationship relating (u∗ )
at which it is dissipated through the eddy spec- to channel form is:
(u∗ ) · [g(A x / p)s b ]1/2

trum. By considering the effects of forcing in con-
m s−1 (2.7)
trasted situations, relevant ranges of values for
(u∗ ) may be nominated. where g is gravitational acceleration (in m s ), sb −2
Thus, the simplest model that may be pro- is the gradient of the bed (in m m−1 ), Ax is the
posed applies to a body of open water, of infi- cross-sectional area of the channel (in m2 ) and p
nite depth and horizontal expanse and lacking is its wetted perimeter (in m). In wide channels
any gradient in temperature. We subject it to a Ax /p approximates to the mean depth, H. Thus,
(wind) stress of constant velocity and direction.
(u∗ ) · [g H s b ]1/2 m s−1 (2.8)
The momentum transferred across the surface
must balance the force applied. So we may pro- The turbulent velocity in a river 5 m deep and
pose the following equalities: falling 0.2 m in every km, approximates to (u∗ ) =
10−2 m s−1 . Turbulent intensity increases in rivers
τ = ρa c d U 2 = ρw (u∗ )2 kg m−2 s−2 (2.5)
with increasing relative roughness, with increas-
where ρ a is the density of air (∼1.2 kg m−3 ), U ing depth and increasing gradient. Theoretical
is the wind velocity (properly, measured 10 m contours of (u∗ ) are mapped in Fig. 2.7 in terms
above the water surface) and cd is a dimensionless of gradient and water depth and links them to
TURBULENCE 47
Figure 2.7 Contour map of the

approximate distribution of u∗ in
aquatic environments (including
rivers), in terms of water-column
height, bed gradient and applied
wind velocity. Original plot from
Reynolds (1994b) and redrawn, with
those driven by surface wind stress: the dog-legs ished. For the wind-stirred boundary layer, with
thus represent the ‘switch points’, where atmo- a vertical velocity gradient, (du/dz),
spheric forcing overtakes gravitational flow as the
(u∗ ) ≈ l e (du/dz) m s−1 (2.9)
main source of turbulent energy in the water.
Major aquatic habitats are noted on the map. Even in the open ocean, the wind-mixed layer
rarely extends more than 200–250 m from the
surface (Nixon, 1988; Mann and Lazier, 1991). It is
2.3.3 Turbulent dissipation clear that so long as the inputs remain steady, the
Before proceeding to the comparison of (u∗ ) with boundary-layer structure serves to dissipate the
sinking velocities of plankters (us ), it is helpful input of kinetic energy through the spectrum of
to grasp how the turbulent energy runs down subsidiary eddies. The rate of energy dissipation,
through the eddy spectrum and how this, in E, also turns out to be an important quantity in
turn, sets the environmental grain. In truly open plankton ecology. Dimensionally, it is equivalent
turbulence, the largest eddies generated should to the product of the turbulent intensity and the
propagate smoothly into smaller ones, having velocity gradient. Thus,
E = (u∗ )2 (du/dz)
progressively lesser velocities as well as lesser
m2 s−2 m s−1 m−1 (2.10)
dimensions. Momentum is lost until the resid-
ual inertia is finally overcome once again by vis- By rearranging Eq. (2.9) for (du/dz) and substitut-
cosity and order returns. In the absence of any ing for it in Eq. (2.10), it follows that:
E ≈ (u∗ )3 l e−1
constraining solid surfaces (shores or bottom) or
m2 s−3 (2.11)
density gradients, it is possible to envisage a
structure in which the largest eddies are adjacent Where the vertical dimension is constrained,
to the source of their mechanical forcing (such however, either because the basin is considerably
as wind stress on the water surface) and a layer less deep than 250 m in depth or because den-
of active, propagating turbulence (in this case, sity gradients resist the downward eddy prop-
from the surface downwards), until the turbu- agation, the smaller surface mixed layer must
lence is finally overwhelmed some distance away still dissipate the turbulent kinetic energy within
(in this case, its lower base). The entire struc- the space available. Were this not so, the motion
ture might then be regarded as a single bound- would have to spill out of the containing struc-
ary layer, separating the energy source from the ture in, for instance, breaking waves or some
non-energised water. The mechanical properties rapid erosion of the perimeter shoreline. What
of the boundary layer then relate to the sizes of happens is that the residual energy reaches into
the dimensions of the largest eddies (le ) and the smaller eddies before it is overcome by viscosity.
gradient with which their velocities are dimin- Thus it is that the most relevant feature of the
Table 2.2 Shear velocity of turbulence (u∗ ), mixed-layer thickness (hm ), dissipation rate (E) and smallest
eddy size (lm ) for various kinetic systems. Abstracted from the compilation of Reynolds (1994a).
Site (u∗ ) (m s−1 ) hm (m) E (m2 s−3 ) lm (mm)

Bodensee winter 6.2 × 10−3 to 44 to 177 1.4 × 10−8 to 2.9 to 1.5
winds, 5–20 m s−1 2.5 × 10−2 2.2 × 10−7
summer winds, ≤1 × 10−3 ≤20 ≤1.3 × 10−7 ≥1.7
<8 m s−1
Lough Neagh winds, 1.3 × 10−2 to 8.9 5.4 × 10−7 to 1.2 to 0.7
10–20 m s−1 2.5 × 10−2 4.3 × 10−6
Ashes Hollow ≤3.1 × 10−1 0.05 ≤1.5 × 10−6 ∼0.9
(hill stream)
River Thames ≤9.4 × 10−3 4.1 5.1 × 10−7 ∼1.2
(Reading)
Open ocean ≤3.3 × 10−2 ≤233 ≤3.8 × 10−7 1.27
Shelf water (Irish Sea) ≤1.2 × 10−1 ≤100 ≤ 4 × 10−5 >0.4
Tidal estuary (Severn)
spring ≤1.3 × 10−1 ≥10 ≤5.5 × 10−4 ≥0.20
neap ≥4.3 × 10−2 ≤40 ≥5.0 × 10−6 ≤0.67
physical environment of plankton is determined rates of dissipation, and here the smallest eddies
not by the intensity of mechanical energy intro- may be in the range 200–400 µm.
duced but by the rate of its dissipation and the
sizes of the smallest eddies that it can sustain.
Simply, the greater is the rate of dissipation, the 2.3.4 Turbulent embedding of
finer is the structural grain. phytoplankton
In much the same way, we can deduce that These considerations are directly comparable
the size of the smallest eddies (lm ) in a structure with the dimensions of phytoplankton (Box 1.2,
is independent of the forcing but depends only Tables 1.2, 1.7). Microplanktic algae are smaller,
on the rate of energy dissipation (work) per unit by one or more orders of magnitude, than the
mass (E, in J kg−1 s−1 , which cancels to m2 s−3 ; smallest eddy sizes in what are arguably among
see glossary of units, symbols and abbreviations) the most aggressively mixed, fastest dissipat-
and the kinematic viscosity (ν, in m2 s−1 ): ing turbulence fields that they might inhabit.
There are some observations and the evidence of
l m = (ν 3 /E )1/4 m (2.12) some experiments (Bykovskiy, 1978) that together
suggest larger species of phytoplankton do not
Solutions of Eq. (2.12) range from the order of tolerate eddy diminution and intensified shear
millimetres in mixed layers, extending to metres implicit in enhanced, fine-grained turbulence
in stratified layers (Spigel and Imberger, 1987). fields but are, instead, readily fragmented. It is
According to the data on well-mixed systems com- an unverified hypothesis to argue that phyto-
piled by Reynolds (1994a), some of which are plankton have evolved along lines that exploited
reproduced here as Table 2.2, the smallest eddy the viscous range of the aquatic eddy spectrum,
sizes calculated to be experienced in oceans and rather than to have invested in the mechanical
deeper lakes are hardly smaller than 1.3 mm. tissue necessary to resist the collapse and frag-
In rivers and shallow lakes, the smallest eddies mentation of larger structures (Reynolds, 1997a).
may be are only half as large for the same input If the dominant vegetation of pelagic environ-
of kinetic energy. Tidal mixing of estuaries and ments is truly selected by its ability to escape the
coastal embayments powers some of the fastest smallest scales of turbulence, then the corollary
PHYTOPLANKTON SINKING AND FLOATING 49
is that the individual organisms are, in effect, to the power generation of its saturated rate of
embedded deep within the turbulence structures photosynthesis (∼1.4 kW kg−1 ). If it stops oper-
to which the water has frequently to accom- ating its flagella, however, the alga comes to a
modate. This is worth emphasising: planktic complete rest in ∼1 µs, having travelled no more
algae live most of their lives in an immedi- than another 10 nm (10−8 m) in relation to the
ate environment that is wholly viscous but, at adjacent medium (Purcell, 1977).
a slightly larger scale, one that is simultane- Embedding is directly relevant to the issues of
ously liable to be transported far and rapidly entrainment and distribution of phytoplankton,
through the turbulence field, and with vary- insofar as the behaviour of the plankter relative
ing intensity and frequency. The pelagic world to a body of water in motion is strongly influ-
of phytoplankton might be analogised to one enced by the behaviour of the plankter within
of little viscous packets being moved rapidly in its immediate viscous environment. To progress
any of three dimensions. In reality, the pack- this exploration requires us to account for buoy-
ets have no enduring integrity but it is the ancy and gravitation behaviour in relation to
behaviour of phytoplankton relative to the imme- suspension.
diate water and to the transport of the water
within the mixed layer that determines the sus-
pension and settling characteristics of the whole 2.4 Phytoplankton sinking and
population. floating
The consequences of living in a viscous
medium have been graphically recounted in
a much-celebrated paper by Purcell (1977). For The buoyant properties of non-motile plankters,
instance, it is not possible for a planktic alga having rigid walls but lacking flagella or cilia,
or bacterium to ‘swim’ through the medium as moving through a column of water, in response
does (say) a water beetle (3–20 mm), by means of to gravity (g = 9.8081 m s−2 ), are subject to the
a reciprocating, rowing movement of paddle-like same forces that govern the settlement of inert
limbs, any more than can a man floundering in a particles in viscous fluids, which were quantified
vat of treacle. The alternative options for forward over a century and a half ago (Stokes, 1851). As the
progression that are exploited by microplank- body moves, it displaces some of the fluid. Pro-
ters and smaller organisms include the serial vided the movement of the displaced fluid over
deformation of the protoplast (amoeboid move- the particle is laminar, thus causing no turbu-
ment), the spiral rotation of the body (as do lent drag, then its velocity (ws ) is related to its
many ciliates and euglenoids) and the rotating size (diameter, d) and the difference between its
of a flagellum like a corkscrew (as in the bac- density from that of the water (ρ w ). For a spheri-
terium Escherichia). The speed of self-propulsion cal particle of uniform density (ρ c ),
relative to the medium (us ) of (say) a Chlamy- w s = gd 2 (ρc − ρw )(18 η)−1 m s−1 (2.14)
domonas cell, 10 µm in diameter (d), is, at about
10 µm s−1 , trivial in absolute terms though nev- This is the well-known Stokes equation. Note
ertheless impressive in body-lengths covered per that for a buoyant particle (ρ c < ρ w ), Eq. 2.14
second. The Reynolds number of its motion per has a negative solution, representing a rate of
second, solved by analogy to Eq. (2.3), confirms flotation upwards. An empirical verification by
that the alga moves smoothly through the water, McNown and Malaika (1950), who measured the
its motion creating no turbulence: sinking rates of machined metal shapes in vis-
cous oils, is also frequently cited in the litera-
R e = (ρw us d)η−1 ture on phytoplankton. The Stokes equation is
≈ 10−4 (2.13) implicitly taken as a valid base for predicting
the sinking behaviour of phytoplankton but it is
The power required to maintain this momentum, necessary also to test all its assumptions and com-
∼0.5 W kg−1 , is also quite trivial when compared ponents if we are to grasp the many mechanisms
that distort the relationship when applied to liv- otherwise known as form resistance. The effect
ing organisms of other than spherical shapes. of departure from spherical form was clearly
demonstrated by the experiments of McNown
2.4.1 Planktic movement and laminar flow and Malaika (1950). In most instances (the
Let us take the condition of laminar flow. ‘teardrop’ shape being a notable exception), sub-
McNown and Malaika (1950) showed good adher- spherical metal shapes sank through oil more
ence to the Stokes formulation whilst R e < 0.1 slowly than the sphere of the same volume. It
and that the error was <10% for R e < 0.5. For is difficult to account quantitatively for these
comparison, Walsby and Reynolds (1980) applied results, as mathematical theory is not so well
published data for phytoplankton to solve Eq. developed that the effects of distortion can
(2.13) for various phytoplankton, approximating be readily calculated. However, McNown and
ρ w as 103 kg m−3 and η as 10−3 kg m−1 s−1 in each Malaika (1950) also published their findings on
case. For the large marine centric diatom Coscin- the sinking of spheroids, both oblate (flattened
odiscus wailseii (d ∼150 × 10−6 m), and substitut- in one axis, like a medicinal pill, or what British
ing ws = 0.1 × 10−3 m s−1 for us (from Smayda, readers will understand as ‘Smartie’-shaped) or
1970), Eq. (2.13) was balanced by R e = 0.015. prolate (shortened in two axes, towards the shape
Similarly, using measurements from Reynolds of classic airships). These have provided plank-
(1973a) for a freshwater centric diatom, Stephan- ton scientists with an important foundation on
odiscus rotula (d ∼50 × 10−6 m, ws = 25 × 10−6 the impacts of form on algal sinking. Some inter-
m s−1 ), R e = 0.00125. The deduction that esting theoretical or experimental investigations
the movements of most phytoplankton comfort- have been pursued using cylinders (Hutchinson,
ably conform to the laminar flow condition 1967; Komar, 1980), chains of spheres (Davey and
is, however, challenged by very large plankters. Walsby, 1985) and, more recently, some inge-
According to Smayda’s (1970) data, the sink- nious alga-like shapes fashioned in polyvinyl chlo-
ing of the extraordinary Ethmodiscus rex, one ride (PVC) or malleable ‘Plasticine’ (Padisák et al.,
of the largest known diatoms (d ∼ 1 mm, 2003a). These have helped to amplify an under-
ws = 6 × 10−3 m s−1 ), generates an R e ∼ 6. Work- standing of the importance of shape in the
ing with a size range of colonies of the Cyanobac- behaviour of phytoplankton.
terium Microcystis, Reynolds (1987a) showed that In the case of spheroids, the reduction in sink-
the Stokes equation (2.14) predicted velocities ing is related to the ratio of the vertical axis
well in colonies of known densities up to (d =) (say, a) to the square root of the product of the
√
200 × 10−6 m in diameter, but in larger colonies other two ( bc). The fastest-sinking spheroid is
(d up to 4 mm), velocities became significantly one in which a ≈ 2b and b = c: the horizon-
overpredicted especially when R e > 1. tal cross-sectional area is smaller than that of
the sphere of the same volume but with most
2.4.2 Departure from spherical shape: of the volume in the vertical where it offers
form resistance less drag, the spheroid actually sinks faster than
√
If the laminar-flow condition may thus be the sphere. As the ratio is increased [a/( bc) to
assumed to apply to the movements of micro- >3], drag increases and velocity falls below that
phytoplankton, probably at all times, it is not of the equivalent sphere. Analogously, making
√
at all clear that the Stokes equation can apply a < b and a/( bc) < 1, drag increases more
other than to spherical organisms, coenobia or than the horizontal cross-sectional area and to
colonies, less than 200 µm in diameter. In fact, >3. Spheroids with the most disparate diameters,
for the majority of phytoplankters that are not that is, the narrower or the flatter they are with
spherical, the shape distortion has a significant respect to the sphere of identical volume and
impact on the rate of settling. Distortion from the density, offer increased form resistance and up
sphere inevitably results in a greater surface area to a twofold reduction in sinking rate.
to an unchanged volume and density and, hence, In the case of cylinders, increasing the
a greater volume-specific frictional surface, length but keeping the cross-sectional diameter
PHYTOPLANKTON SINKING AND FLOATING 51
constant increases sinking velocity, although this the calculated rate of sinking (ws calc) with the
approaches a maximum when length exceeds observed rate of sinking by direct measurement,
diameter about five times. Cylinders at this criti- ws . Thus,
cal length have about the same sinking speed as
spheres of diameter 3.5 times the cylinder sec- ϕr = ws calc/ws (2.15)
tion (dc ). As a cylinder having a length of 7(dc )
The Stokes equation should also be modified in
has the same volume as such a sphere, we may
respect of phytoplankton (Eq. 2.16) by including a
deduce that cylinders relatively longer than this
term for form resistance, accepting that the value
will sink more slowly than the equivalent sphere,
of ϕ r may be so close to 1 that the estimate pro-
so long as all other Stokesian conditions are ful-
vided by Eq. (2.14) would have been acceptable.
filled.
It has been suggested in several earlier stud- ws = g(ds )2 (ρc − ρw )(18 η ϕr )−1 m s−1 (2.16)
ies that distortions in shape have another role in
orienting the cell, that (for instance) the cylin- This approach allows systematic variability in the
drical form makes it turn normal to the direc- coefficient to be investigated as a feature of phy-
tion of sinking and that this may be advan- toplankton morphology. Some of the interesting
tageous in presenting the maximum photosyn- findings that impinge on the evolutionary ecol-
thetic area to the penetrating light. Walsby and ogy of phytoplankton are considered in the next
Reynolds (1980) argued strongly for the counter- section but it is important first to mention the
view. In a truly turbulence-free viscous medium, practical difficulties that have been encountered
a shape should proceed to sink at any angle at in estimating form resistance in live phytoplank-
which it is set. There is an exception to this, ton and the ways in which they have been solved.
of course, which will apply if the mass distribu- As observed elsewhere (Chapter 1) precise esti-
tion is significantly non-uniform: the ‘teardrop’ mates of the volume of a plankter (whence ds
reorientates so that it sinks ‘heavy’ end first. In is calculable) are difficult to determine if the
the experiments of Padisák et al. (2003a), some of shape is less than geometrically regular. Making
the models of Tetrastrum were made deliberately a concentrated suspension and determining the
unstable by providing spines on one side only: volume of liquid it displaces offers an alterna-
these, too, reoriented themselves on release but tive to careful serial measurements of individu-
then remained in the new position throughout als. Walsby and Xypolyta (1977) gave details of a
the subsequent descent. Their observations on procedure using 14 C-labelled dextran to estimate
Staurastrum models that go on reorienting recalls the unoccupied space in a concentrated suspen-
the observations of Duthie (1965) on real algae of sion. The usefulness of the approach neverthe-
this genus, which reorientated persistently dur- less depends upon a high uniformity among the
ing descent to the extent that they rotated and organisms under consideration – cultured clones
veered away from a vertical path. At the time, the are more promising than wild material in this
influence of convection on the sinking behaviour respect.
of Staurastrum could not be certainly excluded. The densities of phytoplankton used to be dif-
The results of Padisák et al. (2003a) suggest that ficult to determine precisely, having to rely on
the form of the cells engenders the behaviour as good measurements of mass as well as of vol-
it reproduced in a viscous medium. ume. Now, it has become relatively easy to set up
For the present, the impact on sinking of dissolution gradients of high-density solutes, intro-
tortions as complex as those of Fragilaria or Pedi- duce the test organisms then centrifuge them
astrum coenobia requires more prosaic methods until they come to rest at the point of isopycny
of assessment. The most widely followed of these between the organism and solute (Walsby and
is to calculate a coefficient of form resistance (ϕr ) Reynolds, 1980). Following Conway and Trainor
by determining all the variables in the Stokes (1972), Ficoll is frequently selected as an appropri-
equation (2.14) for a sphere of equivalent vol- ate solute. Being physiologically inert and osmot-
ume (having the diameter, ds ) and comparing ically inactive improves its utility.
Even to get accurate measurements of sinking

rate (ws ) is problematic. Here, the main issue is
to be able to keep the water static, when almost
any conventional observation system involving
uninsulated light sources is beset by generated
convection. Techniques have been developed or
applied, using some combination of strict ther-
mostatic control, thin observation cells (Wise-
man and Reynolds, 1981) or non-heat-generating
measuring systems based on fluorometry (see, for
instance, Eppley et al., 1967; Tilman and Kilham,
1976; Jaworski et al., 1981). Another approach has
been to make measurements in solutions of high
viscosity: Davey and Walsby (1985) used glycerol.
Alternatively, to measure the rate of loss across
a boundary layer from an initially mixed suspen-
sion works with convection and yields acceptable
results on field-collected material when sophis-
ticated techniques may not be readily available
(Reynolds et al., 1986). This actually imitates, in
part, the way that plankton settles from natural
water columns (see Section 2.6).
Once confidence was gained in the measure- Figure 2.8 Log/log plot of the instantaneous intrinsic
ment of sinking rates, another, more tantalising settling rates (ws ) of Stephanodiscus rotula cells, collected from
the field and plotted against mean cell diameter, ds (◦). There
source of variability was detected. Several, quite
is apparently no correlation. However, when corresponding
independent investigations of the species-specific
samples are killed by heat prior to determination ( t), a strong
sinking rates of diatoms each yielded order-of- positive correlation is found. Redrawn from Reynolds (1984a).
magnitude variability. For any given species, the
sinking rates seemed least when the cells were
healthy and physiologiclly active but were as phytoplankton was to first kill the diatoms under
much as three to seven times faster in similar test. A case in point is shown in Fig. 2.8, where
cells that were naturally moribund (Eppley et al., the mean sinking rates of Stephanodiscus rotula
1967; Smayda, 1970; Reynolds, 1973a), or whose cells sampled during the increase and decrease
photosynthesis was experimentally inhibited or phases of a natural lake population seem to vary
carbon-limited (Jaworski et al., 1981), or which randomly. However, the corresponding rates of
had been exposed to sublethal doses of algi- cells killed by dipping in boiling water just before
cide (Margalef, 1957; Smayda, 1974), or had been measurement were demonstrably correlated to
otherwise freshly killed (Wiseman and Reynolds, size. It is now quite generally accepted that live,
1981). However, comparing the fastest rates from healthy diatoms have the capacity to lower their
each of the studies that had made measure- sinking rates below that of dead or moribund
ments on comparable material (eight-celled stel- ones. The mechanism of change is not obviously
late coenobia of the freshwater diatom Asteri- contributed by variability in size or shape or even
onella formosa), some conformity among the var- density. Sinking of live diatoms and, possibly,
ious results became apparent (Jaworski et al., other algae is plainly influenced by the interven-
1988). tion of further, vital components that must be
Wiseman et al. (1983) had previously estab- taken into account in any judgement on how phy-
lished that the one sure way to get the consistent, toplankton regulate their sinking rates.
inter-experimental results necessary to be able to To go on now to review some of the ana-
investigate the morphological form resistance of lytical investigations into the sinking rates of
ADAPTIVE AND EVOLUTIONARY MECHANISMS FOR REGULATING WS 53
phytoplankton and the role of form resis- by a relatively greater silicon content. However,
tance provides the simultaneous opportunity it is probable that the effect is enhanced by the
to observe the cumulative influences of the fact that a relatively greater part of the inter-
biotic components of the modified Stokes equa- nal space of larger diatoms is occupied by cell
tion (2.16). sap rather than cytoplasm (Walsby and Reynolds,
1980) and that many marine diatoms, at least,
are known to be able to vary the sap density rela-
2.5 Adaptive and evolutionary tive to that of sea water (Gross and Zeuthen, 1948;
mechanisms for regulating ws Anderson and Sweeney, 1978). This mechanism of
density regulation is not available to freshwater
algae (see Section 2.5.2) but, either way, density
Of the six variables in the modified Stokes equa- effects may be vital to the suspension of larger
tion, three (g, ρ w and η) are either constants diatoms in the sea, if the reduction of sinking
or are independent variables. The other three rates is the ultimate adaptive aim. According to
(size, density and form resistance) are organismic Smayda’s (1970) data, cells of Coscinodiscus wailesii
properties and, as such, are open to adaptation should settle at a rate of 40 m d−1 , had they the
and evolutionary modification through natural same net density as the much smaller Cyclotella
selection. It is possible that certain metabo- nana, rather than the observed 8–9 m d−1 .
lites released into the water by organisms also
affect the viscosity of the adjacent medium. The 2.5.2 Density
present section looks briefly at the influences The cytoplasm of living cells comprises compo-
of each on sinking behaviour and the extent to nents that are considerably more dense than
which a planktic existence selects for particular water (proteins, ∼1300; carbohydrates, ∼1500;
adaptive trends. nucleic acids ∼1700 kg m−3 ), so that the aver-
age density of live cells is rarely less than ∼1050
2.5.1 Size kg m−3 . Inclusions such as polyphosphate bodies
The relatively small size of planktic algae has (∼2500 kg m−3 ) and exoskeletal structures of cal-
been alluded to in Chapter 1. For instance, the cite and, especially, the opaline silica of diatom
diameters of the spheres of equivalent volumes to frustules (∼2600 kg m−3 ) increase the average
species named in Table 1.2 cover a range from 1 density still further. Some of the excess density
to 450 µm. It has been suggested that this is itself can be offset by the presence of oils and lipids
an adaptive feature for living in fine-grained tur- that are lighter than water, the lightest having a
bulence. Many species may present rather greater density in the order of 860 kg m−3 (Sargent, 1976).
maximal dimensions if (presumably) the rate of However, these oils rarely account for more than
turbulent energy dissipation allows. The effect of 20% of the cell dry mass. Without adaptation,
size on the settling velocities of centric diatoms most freshwater phytoplankters are bound to be
was demonstrated empirically by Smayda (1970). heavier than the medium and naturally sink! The
A striking feature of his regression of the loga- corresponding deduction in respect of marine
rithm of velocity ws on the average cell diameter phytoplankton suspended in sea water (ρ w gener-
(d) is that its slope lies closer to 1 than 2, as would ally ∼1030 kg m−3 ) is made with rather less con-
have been expected from the Stokes equation fidence, where the scope for regulation of ρ c can
(2.14). The regression fitted to the plot of sink- be more purposeful.
ing rates of killed Stephanodiscus cells against the The list in Table 2.3 is an abbreviated version
diameter (in Fig. 2.8) also has a slope of ∼1.1. The of one that was included in Reynolds (1984a). It
observations suggest that the larger size (and, is intended to illustrate the range of densities
hence, the larger internal space) is compensated rather than the comprehensiveness of the data.
by a lower overall unit density (cf. Section 1.5.2 In particular, it is easy to distinguish the gas-
and Fig. 1.9). The implication is that more of the vacuolate Cyanobacteria which, in life, have den-
overall density of the small diatom is explained sities of ≤1000 kg m−3 that enable them to float,
Table 2.3 Some representative determinations of phytoplankton densities
Species ρ c (range, in Methoda References

kg m−3 )
Cyanobacteria
Anabaena flos-aquae 920–1030 b 1 Reynolds (1987a)
Microcystis aeruginosa 985–1005 b 1 Reynolds (1987a)
(colonies)
Planktothrix agardhii 985–1085 b 1 Reynolds (1987a)
Chlorophyta
Chlorococcum sp. 1020–1140 2 Oliver et al. (1981)
Chlorella vulgaris 1088–1102 2 Oliver et al. (1981)
Bacillariophyta
Stephanodiscus rotula 1078–1104 3 Reynolds (1984a)
Synedra acus (culture) 1092–1138 2 Reynolds (1984a)
Thalassiosira weissflogii 1121 3 Walsby and Xypolyta (1977)
Tabellaria flocculosa 1128–1156 2 Reynolds (1984a)
Asterionella formosa 1151–1215 2 Wiseman et al. (1983)
Fragilaria crotonensis 1183–1209 2 Reynolds (1984a)
Aulacoseira subarcticac 1155–1183 2 Reynolds (1984a)
Aulacoseira subarctica d 1237–1263 2 Reynolds (1984a)
a
Methods: (1) calculations based on experimental changes in velocity and gas-vesicle content
(see text); (2) deteminations by centrifugation through artificial density gradient to isopycny;
(3) gravimetric determinations of mass and volume.
b
Range covers colonies or filaments with maximum known gas vacuolation to colonies or filaments
after subjection to pressure collapse (see text).
c
Post-auxosporal (wide filaments).
d
Pre-auxosporal (narrow filaments).
All measurements on wild material, unless otherwise stated.
but which are heavier than fresh water if the Lipid accumulation
vesicles are collapsed by pressure treatment (see Fats and oils normally account for some 2–20% of
below). Two, small unarmoured chlorophytes are the ash-free dry mass of phytoplankton cells, per-
included (data of Oliver et al., 1981); how typical haps increasing to 40% in some instances of cel-
they are of non-siliceous algae is not known. The lular senescence (Smayda, 1970; Fogg and Thake,
silica-clad diatoms have densities generally ≥1100 1987). Most lipids are lighter than water and,
kg m−3 , although there is a good deal of vari- inevitably, their presence counters the normal
ability. Interestingly, it seems likely that density excess density to some limited extent. Oil accu-
varies inversely to internal volume (Asterionella vs. mulation is responsible for the ability of colonies
Stephanodiscus, slender pre-auxosporal Aulacoseira of the green alga Botryococcus to float to the
vs wide post-auxosporal cells). surface in small lakes and at certain times of
Apart from these generalisations, average den- population senescence (Belcher, 1968). However,
sities of many planktic algae are plainly influ- it is improbable that oil or lipid storage could
enced by a number of discrete mechanisms. reverse the tendency of diatoms to sink. Reynolds
These include lipid accumulation, ionic regula- (1984a) calculated that were the entire internal
tion, mucilage production and, in Cyanobacteria, volume of an Asterionella cell to be completely
the regulation of gas-filled space. filled with the lightest known oil, its overall
density (∼1005 kg m−3 ) would still not be enough scope for ionic regulation in phytoplankton of
to make it float. Walsby and Reynolds (1980) con- freshwater (ρ w < 1002 kg m−3 ) is too narrow to
cluded that the reduction in density consequen- be advantageously exploited.
tial upon intracellular lipid accumulation would,
unquestionably, contribute to a reduced rate of Mucilage
sinking but they doubted any primary adaptive The mucilaginous investment that is such a com-
significance, neither was there evidence of its use mon feature of phytoplankton, especially of fresh-
as a buoyancy-regulating mechanism. water Cyanobacteria, chlorophytes and chrys-
ophytes, has long been supposed to function
as a buoyancy aid. Again, that mucilage does
Ionic regulation reduce overall density is, generally, indisputable.
Inherent differences in the densities of equimo- Whether this is a primary function is less cer-
lar solutions of organic ions raise the possibil- tain (see Box 6.1, p. 271) and it is mathematically
ity that selective retention of ‘light’ ions at the demonstrable that the presence of a mucilagi-
expense of heavier ones could enable organisms nous sheath does not always reduce sinking rate;
to lower their overall densities. In a classical in fact it may positively enhance it.
paper, Gross and Zeuthen (1948) calculated the The presence, relative abundance and consis-
density of the cell sap of the marine diatom Dity- tency of mucilage is highly variable among phy-
lum brightwellii to be ∼1020 kg m−3 , that is, signif- toplankton. Mucilages are gels formed of loose
icantly lower than the density of the suspending networks of hydrophilic polysaccharides which,
sea water and actually sufficient to bring overall though of high density (∼1500 kg m−3 ) them-
density of the live diatom close to neutral buoy- selves, are able to hold such large volumes of
ancy. The density difference between sap and sea water that their average density (ρ m ) approaches
water was explicable on the basis of a substantial isopycny. Reynolds et al. (1981) estimated the den-
replacement of the divalent ions (Ca2+ , Mg2+ ) by sity of the mucilage of Microcystis to average
monovalent ones (Na+ , K+ ) with respect to their ρ w + 0.7 kg m−3 . The presence of mucilage can-
concentrations in sea water. Some years later, not make the organism less dense than the sus-
Anderson and Sweeney (1978) were able to follow pending water but it can bring the the aver-
changes in the ionic composition of cell sap of age density of the cell or colony maintaining it
Ditylum cells grown under alternating light–dark much closer to that of the medium (ρ c ). How-
periods. They were able to show that density ever, the clear advantage that this might bring
may, indeed, be varied by up to ±15 kg m−3 , to reducing ws in, for instance, Eq. (2.16) must
through the selective accumulation of sodium or be set against a compensatory increase in over-
potassium ions, though interestingly, not suffi- all size (ds is increased). Thus, for mucilage to
ciently to overcome the net negative buoyancy be effective in depressing sinking rate, the den-
of the cells. Elsewhere, Kahn and Swift (1978) sity advantage must outweigh the disadvantage
were investigating the relevance of ionic regula- of increased size.
tion to the buoyancy of the dinoflagellate Pyrocys- This relationship was investigated in detail by
tis noctiluca; they showed that by selective adjust- Hutchinson (1967). If a spherical cell of density,
ment of the content of Ca2+ , Mg2+ and (SO4 )2− , ρ c , is enclosed in mucilage of density, ρ m , such
the alga could become positively buoyant. that its overall diameter is increased by a factor
The effectiveness of this mechanism is not to a, then its sinking rate will be less than that of
be doubted but its generality must be regarded the uninvested cell, provided:
with caution. For it to be effective does depend
(ρc − ρm )/(ρm − ρw ) > a (a + 1) (2.17)
upon maintaining a relatively large sap volume.
The scope of density reduction is limited, in so Because a is always >1, the density difference
far as the dominant cations in sea water are the between cell and mucilage must be at least that
lighter ones and the lowest sap density is the iso- between mucilage and water. Supposing ρ c to be
tonic solution of the lightest ions available. The ∼1016 kg m−3 , ρ w to be 999 and ρ m to be 999.7,
diameter of the sphere equivalent to the total vol-

ume of cells enclosed. The data presented appear
to fall within the range of benefit (in terms of
sinking rate), although in the case of Pseudosphae-
rocystis and some Coenochloris colonies, the ratios
appear to unfavourable, at least if the assump-
tions about the component densities apply in all
these cases.
Gas vacuoles
The surest way of lowering average density is
to maintain gas-filled space within the proto-
plast. This is precisely what some of the plank-
tic Cyanobacteria do and, as is well known, the
organisms become buoyant at times, accumulatu-
ing at the surface as a scum, constituting what
was originally called a ‘water bloom’. The term
‘bloom’ has since been applied to almost any
planktic population (not even necessarily algal)
significantly above the norm: it is another of
Figure 2.9 The effect of mucilage thickness (a, as a multiple those words that has been misused to the point of
of diameter, ds ) on the average density (ρ = ρ c + ρ m ) of a being rendered unhelpful. However, the biology
spherical alga of constant diameter and density (ρ c =
of scum-forming Cyanobacteria is a fascinating
1016 kg m−3 ). The arrow on the relative velocity plot
topic, and not only because of the almost univer-
indicates the point of maximum advantage of mucilage
investment in the context of sinking-velocity reduction.
sal contempt in which most environmental and
Redrawn from Reynolds (1984a). water-supply managers hold their unsightliness
and potential toxicity (Bartram and Chorus, 1999;
see also Section 8.3.2). Part of the remarkable
Eq. (2.17) can be solved as 23.3 > a (a + 1), or that account of the functional morphology and pop-
a must be <4.3, if the presence of mucilage is ulation dynamics concerns the ability of these
to reduce the sinking rate. The maximum advan- Cyanobacteria to regulate the buoyancy provided
tage can be solved graphically (see Fig. 2.9); with by their gas vacuoles (Reynolds et al. 1987). Nested
the nominated values, the greatest advantage within this is the unfolding appreciation of the
occurs when a ∼ 2.3. Of course, the precise opti- structure and function of the buoyancy provision
mal value of a varies from alga to alga, depend- itself. Much of the progress over the last 30 years
ing partly upon the nature of the mucilage but has been spearheaded by A. E. Walsby and his co-
mainly upon the cell density. For a diatom with workers. Walsby’s (1994) review is one of the most
a cell density of 1200 kg m−3 , the greatest value comprehensive, and it is this to which the reader
of a that would bring a net reduction in its sink- is referred for all details.
ing rate could be as high as 16.4, with maximum Here, it is sufficient to emphasise that, from
advantage at a ∼ 8.7. the time their existence was first established
In order to compare the mucilage provision (Klebahn, 1895), gas vacuoles have been assumed
among planktic algae, which are not all spheri- to have the function of providing buoyancy.
cal, it is useful first to express cell volume as a Although this may have been neither their origi-
proportion of the total unit volume (vc /vc+m , as nal nor their only function (Porter and Jost, 1973,
included in Fig. 2.9). Some examples are given in 1976), these uniquely prokaryotic organelles cer-
Table 2.4. In each instance, the range of values of tainly do reduce the average density of the cell
a is calculated as the ratio of the diameter of the in which they occur. They are not bubbles –
sphere equivalent to the full unit volume and the surface tension is much too powerful to permit
Table 2.4 Relative volumes of cell material (vc ) as a proportion of the full unit volume (vc+m ) of named
mucilage-producing phytoplankters, expressed in terms of Hutchinson’s (1967) factor a (see text)
Plankter vc /vc+m a References

Microcystis aeruginosa 0.03–0.05 1.59–3.22 Reynolds et al. (1981)
Anabaena circinalis 0.045–0.124 2.00–2.81 Previously unpublished
measurements reported in
Reynolds (1984a)
Chlamydocapsa planctonica 0.126 1.99 Previously unpublished
Reynolds (1984a)
Coenochloris fottii 0.005–0.532 1.24–9.52 Previously unpublished
Reynolds (1984a)
Eudorina unicocca 0.055–0.262 1.64–2.63 Previously unpublished
Reynolds (1984a)
Pseudosphaerocystis lacustris 0.008–0.013 4.30–4.92 Previously unpublished
Reynolds (1984a)
Staurastrum brevispinum – 2.2–2.3 From direct measurements
taken from Fig. 27 of
Ruttner (1953)
Fragilaria crotonensis 0.032–0.047 2.78–3.14 From direct measurements of
linear dimensions taken
from Plates 1c and 1d of
Canter and Jaworski (1978),
with calculation of a
their existence at the scale of micrometres – but assembles. The structures are vulnerable to exter-
rigid stacks of proteinaceous cylindrical or pris- nal pressure, including the internal turgor pres-
matic envelopes called gas vesicles (Bowen and sure of the cell. They have a certain strength but,
Jensen, 1965). In the Cyanobacteria, they gener- once a critical pressure has been exceeded, they
ally measure between 200 and 800 nm in length. collapse by implosion. They cannot be reinflated;
The diameters of isolated gas vesicles vary inter- they can only be built de novo, although the gas-
specifically between 50 and 120 nm, but are rea- vacuole protein is believed to be recyclable.
sonably constant within any given species. Each The critical pressures of isolated gas ves-
molecule of the specialised gas-vesicle protein has icles are inversely correlated to their diameters
a hydrophobic end and they are aligned in ribs in (Hayes and Walsby, 1986). The higher is the crit-
the vesicle wall so that the entry of liquid water ical pressure of the vesicles, the greater the
into the internal space is prevented. However, the hydrostatic pressure and, thus, the greater water
vesicle wall is fully permeable to gases and in depth they can withstand. Intriguingly, vesicle
no sense do the vesicles hold gas under anything size, like organism size and shape, co-varies with
but ambient pressure. The gas inside the vesicles the principal ecological ranges in which individ-
is usually dominated by nitrogen with certain ual species occur and, arguably, the habitats to
metabolic by-products but it is clear that the gas which they are best adapted. In Anabaena flos-
composition is of much less significance than is aquae, a common scum-forming species in small
the gas-filled space, which is created as the vesicle eutrophic lakes, vesicles measuring about 85 nm
in diameter have a critical pressure of 0.3 to 0.7

MPa (Dinsdale and Walsby, 1972). In Microcystis
aeruginosa, a species sometimes found in larger
and more physically variable lakes, vesicles aver-
aging 70 nm in diameter have critical pressures
in the order of 0.6–1.1 MPa (Reynolds, 1973b;
Thomas and Walsby, 1985). The species of Plank-
tothrix (formerly Oscillatoria) of deep, glaciated
lakes in mountainous regions have vesicles mea-
suring 60–65 nm that withstand pressures of
between 0.7 and 1.2 MPa (Walsby and Klemer,
1974; Walsby et al., 1983; Utkilen et al., 1985a).
Anabaena lemmermanni, a species more usually
found distributed in the deep mixed layers of
larger temperate lakes, also has much stronger
vesicles than most other Anabaena species (0.93
MPa: Walsby et al., 1991). Vesicles from the
oceanic Trichodesmium thiebautii were found to tol-
erate up to 3.7 MPa (Walsby, 1978).
It is now recognised that gas-vesicle size is Figure 2.10 The relationship between sinking velocity (ws )
subject to very strong selective pressure. Narrow or floating velocity (−ws ) of cells of Anabaena circinalis and
gas vesicles are less efficient at providing buoy- their gas-vacuole content as a proportion of cell volume.
After Reynolds (1972) and redrawn from Reynolds (1984a).
ancy and, for a given yield of gas-filled space,
they are assembled at greater energetic cost. Nar-
rower ones should only be selected if the extra tially linear relationship between density (and
strength is required (Walsby and Bleything, 1988). buoyant velocity) and the gas-vacuole content. It
Now that the genes controlling gas-vesicle assem- is ‘potential’ in so far as other items in the com-
bly can be identified relatively easily (Beard et al., plement of cell materials affect the density, and
1999, 2000), the selection by hydrographic events the velocity is sensitive to size and form resis-
(for instance, incidences of deep mixing of Plank- tance. Measurements of the gas-vesicle content
tothrix populations) for the survival of relatively required to gain neutral buoyancy vary between
more of the stronger or relatively more of the 0.7% and 2.3% of the cell volume (Reynolds and
weaker kind is one of the most elegant demon- Walsby, 1975).
strations of gene-based natural selection to have To conclude this very brief overview of the
been contrived (Bright and Walsby, 1999; Davis buoyancy provision that gas vacuoles impart, it
et al., 2003). is relevant to the ecology of these organisms to
The buoyancy-providing role of the gas vesi- refer to the mechanisms of buoyancy regulation.
cles has been studied for over 30 years. By prepar- To be continuously buoyant is arguably advan-
ing very thick suspensions of Cyanobacteria, plac- tageous in deep, continuously mixed water lay-
ing them in specific-gravity bottles and then ers. In small, possibly sheltered lake basins and
subjecting them to pressures sufficient to col- in larger ones at low latitudes, where the vari-
lapse the vesicles, the volume of gas displaced ability in convective mixing is highly responsive
can be measured very accurately. Expressed rel- to diurnal heat income and net nocturnal heat
ative to the cell volume, the percentage of gas- loss, it is biologically useful to be able to alter or
filled space is readily calculated (Walsby, 1971). even reverse buoyancy. There are at least three
Reynolds (1972) used this method to collect the ways in which the planktic Cyanobacteria do
data used to construct Fig. 2.10, which is included this. The relative content of gas vesicles is, in
to show that there is, for any given alga, a poten- the first instance, the outcome of the balance
between their assembly and their collapse. As small differences in density (as predicted in the
cells simultaneously grow and divide, gas vesicles modified Stokes equation, 2.16), allowing Micro-
will be ‘diluted out by growth’ unless the cellular cystis colonies to migrate on a diel basis in stable
resource allocation to their assembly keeps pace. water columns and to recover vertical position
There is plenty of evidence (reviewed in Reynolds, very rapidly after disruptive storms. Apart from
1987a) that the processes are not closely coupled the first detailed descriptions of this behaviour in
and that the relative content of vesicles increases shallow, tropical Lake George (Ganf, 1974a), simi-
during slow (especially light-limited) growth and lar adjustments, the ability of Microcystis to attain
decreases during rapid growth. It is also appar- this control on its buoyancy, is apparent from
ent that for the species with the strongest vesi- the studies of Okino (1973), Reynolds (1973b,
cles, this is the main mechanism of control and, 1989a), Reynolds et al. (1981) and Okada and Aiba
of course, it operates at the scale of genera- (1983).
tion times. For the species with weaker vesicles No less striking is the formation of persistent
that are vulnerable to collapse in the face of plate-like layers in the stable metalimnia of cer-
rising turgor generated by low-molecular-weight tain relatively deep lakes: the plankters may be
carbohydrates, there is a rather more respon- almost lacking from the water column but for
sive mechanism of reversing buoyancy. Cells float- a band of 1 m or rather less, where they remain
ing into higher light intensities photosynthesise poised, often at quite low light intensities. The
more rapidly, raise cell turgor, collapse vesicles, behaviour has been known for many years from
lose buoyancy, sink back where the cycle can alpine lakes in central Europe (Findenegg, 1947;
start again. The cycle of buoyant adjustments can Thomas, 1949, 1950; Ravera and Vollenweider,
operate on a diel basis and bring about daily 1968; Zimmermann, 1969; Utkilen et al., 1985a)
migratory cycles over depth ranges of 2–4 m, and generally involves the solitary filaments
cells accumulating near the surface by night and of Planktothrix of the rubescens–prolifica–mougeotii
at greater depths by the end of the daylight group but it is also known from small, stratifying
period (Reynolds, 1975; Konopka et al., 1978). Such continental lakes elsewhere (Juday, 1934; Atkin,
behaviour may be invoked to explain the diel 1949; Eberley, 1959, 1964, Lund, 1959; Brook et
migratory cycles of Anabaena spp. (reported by al., 1971, Gorlenko and Kuznetsov, 1972; Walsby
Talling, 1957a; Pushkar’, 1975; Ganf and Oliver, and Klemer, 1974) and to involve other genera
1982) and Aphanizomenon (Sirenko et al., 1968; (Lyngbya, or Planktolyngbya, Spirulina: Reynolds et
Horne, 1979), or imitated in laboratory mesocosm al., 1983a; Hino et al., 1986). The ability to main-
(Booker et al., 1976; Booker and Walsby, 1981). tain station is attributable to close regulation of
The third mechanism can also result in fairly gas-vesicle content but the very low light inten-
fine control of buoyancy trimming in Cyanobac- sities suggest that this is regulated by alloca-
teria with gas vesicles of intermediate strength, tion. Zimmermann’s (1969) study showed that,
such as those of Microcystis, beyond the scope of through the season, P. rubescens moves up and
the turgor-collapse mechanism. Here, the buoy- down in the water column of Vierwaldstättersee,
ancy provided by a coarsely variable complement mainly in response to changes in the down-
of robust gas vesicles is countered by a finely welling irradiance. The cells are able to main-
variable accumulation of photosynthetic poly- tain biomass or even to grow slowly in situ and
mers (chiefly glycogen) of high molecular weight the behaviour has been interpreted as a sort of
(Kromkamp and Mur, 1984; Thomas and Walsby, ‘aestivation’, to escape the period of minimal
1985). So long as approximate neutral buoyancy resource supply. However, the recent season-long
is maintained, relatively small differences in the investigation by Bright and Walsby (2000) of the
glycogen content take the average density of P. rubescens stratified in the Zürichsee, points to a
the colonies either side of neutral buoyancy, in sophisticated set of adaptations to gain positive
response to insolation and photosyntetic rate. growth in the only region of the lake where a
The large size attained by colonies magnifies the small nutrient base and a low light income are
simultaneously available. The ability to control strain, although there is a possibility that there
organismic density is crucial to the exploitation was a density difference between the two forms
of the opportunity. (Conway and Trainor, 1972). However, the investi-
gations of the role of the 70-nm chitin fibres that
2.5.3 Form resistance adorn the frustules of Thalassiosira weissflogii (for-
Many plankters are markedly subspherical in merly T. fluviatilis) in slowing the sinking speed
shape and theory suggests that, in a majority of cells have been carefully evaluated by Walsby
of instances, the departure results in the organ- and Xypolyta (1977). Cells from which the fibres
ism having a slower passive rate of sinking (or had been removed with a fungal chitinase sank
floating) than the sphere of equivalent volume almost twice as fast as those not so treated, even
and overall density. As has been suggested ear- though the density of the fibres (1495 kg m−3 ) was
lier, there have really been few attempts to verify rather greater than that of the fibreless cells. The
that this is true for a majority of species, and overall volume of the untreated cells was also
then mainly through resort to empirical evalu- larger (1.9-fold) than that of the fibreless cells
ation of the coefficient of form resistance, ϕr . but the surface area was 2.8 times greater. Only
Even where significant form resistance is estab- the increased form resistance could have been
lished experimentally (see entries in Table 2.5), it responsible for the reduced sinking rate.
does not prove distortion to be necessarily adap-
tive in the context of floating and sinking. Never- Chain formation
theless, the experimental demonstrations of the Joining two or more cells together obviously
impact on sinking rate made by the presence of increases the volume of the settling particle in
horns or spines, cell elongation in one (or pos- the same ratio. It also increases the surface area,
sibly two) axes and the creation of secondary but for the area of mutual contact between indi-
shapes by coenobial formations of chains, fila- vidual cells in the chain. Theory dictates that
ments and spirals make a fascinating study. In the chain must sink faster than the individual
the end, they may provide the key to how larger component cells – were sinking rate the only
plankters actually do maximise their suspension criterion, joining cells together could not be
opportunities. claimed to be an adaptation to suspension. On
the other hand, as pointed out by Walsby and
Protuberances and spines Reynolds (1980), if there is another constraint
The value of distortions to staying in suspen- favouring larger size (say, resistance to grazing),
sion goes back a long way in planktology, cer- it is equally clear that the linear arrangement
tainly to Gran’s (1912) interpretation, quoted by preserves much more surface drag than a sphere
Hardy (1964), of a verifiable tendency for Ceratium of the same volume of the aggregate of cells.
species of less viscous tropical seas to have longer Hutchinson (1967) invoked the results of some
and, often, more branched horns than the species experiments by Kunkel (1948), who had mea-
typical of colder, high-latitude seas. Yet it is only sured sinking rates of identical glass beads, either
relatively recently that effects were quantitatively singly or cemented together in linear chains of
demonstrated. Smayda and Boleyn (1966a) inves- one, two, three, four or eight. Hutchinson (1967)
tigated several aspects of the variability in sink- calculated the relative form resistance and fit-
ing rate in the marine diatom Rhizosolenia setigera, ted a linear plot against chain length with the
including the fact that spineless pre-auxospore equation:
cells settle significantly faster than the spined
ϕr = 0.837 + 0.163b (2.18)
vegetative cells that follow auxospore ‘germina-
tion’ (see p. 64). The spines that occur on the where b is the number of beads. Superficially, this
end cells of four-celled coenobia of the freshwa- supported observations of Smayda and Boleyn
ter chlorophyte Scenedesmus quadricauda are said (1965, 1966a, b) in Thalassiosira, Chaetoceros and
to reduce the sinking rate relative to a spineless other chain-forming marine diatoms that sinking
Table 2.5 Comparison of measured sinking rates (ws ) of various freshwater plankters and the rates (ws
calc) calculated from Stokes’ equation for spheres of identical volume and density
Planktera Dynamic shape ws (µm s−1 ) (ws ) calc ϕr References

Chlorella vulgaris ± spherical – – 0.98–1.07 Oliver et al.
(1981)
Chlorococcum sp. ± spherical – – 1.02–1.04 Oliver et al.
(1981)
Cyclotella Squat cylinder – – 1.03 Oliver et al.
meneghiniana (1981)
∗
Stephanodiscus Squat cylinder
rotula
(ds = 12–14 µm) 11.5 ± 0.8 13 ± 1 1.06 From data plotted
in Fig. 2.8
(ds = 24–28 µm) 27.6 ± 2.6 26 ± 2 0.94 From data plotted
in Fig. 2.8
∗
Synedra acus Attenuate
cylinder
(MLD = 17d) 7.3 ± 1.2 29.8 4.1 Reynolds (1984a)
∗
Aulacoseira
subarctica
(1–2 cells) Cylinder 7.4 ± 2.8 17.2 2.3 Reynolds (1984a)
(7–8 cells) Attenuate 11.4 ± 4.1 50.1 4.4 Reynolds (1984a)
cylinder
∗
Tabellaria Stellate, 8-armed 10.3 ± 1.0 54.1 5.5 Reynolds (1984a)
flocculosa var colony
asterionelloides
∗
Asterionella
formosa
(4 cells) Stellate colony 5.8 ± 0.2 18.2 3.2 Reynolds (1984a)
(8 cells) Stellate colony 7.3 ± 0.6 28.9 3.9 From data plotted
in Fig. 2.14
(16 cells) Stellate colony 10.7 ± 1.2 45.9 4.3 Reynolds (1984a)
(8 very short Chain of ovoids 4.0 ± 0.6 7.4–8.0 1.9 Jaworski et al.
cells) (1988)
∗
Fragilaria
crotonensis
(single cell) Cylinder 3.9 ± 0.2 10.6 2.8 Reynolds (1984a)
(11–12 cells) Plate 11.2 ± 0.5 54.1 4.8 Reynolds (1984a)
a
Asterisks indicate experiments on diatones killed prior to measurement of sinking rates, to overcome
vital interference (see Section 2.5.4).
rates increase with increasing chain length. How-

ever, Walsby and Reynolds (1980) argued that
the linear expression is misleading as increas-
ing chain length should tend to a finite maxi-
mum. The sinking velocity of the chain (say wc )
is linked to the sinking rate of the sphere of sim-
ilar volume (ws ), through wc = ws /ϕr . According
to Stokes’ equation (2.16), ws ∝ (ds /2)2 , where ds
is the diameter of the equivalent sphere. How-
ever, ds /2 ∝ b0.33 , so ws ∝ b0.67 , i.e. ws does not
increase linearly, as Hutchinson’s equation pre-
dicts, neither does it fall asymptotically to zero
as implied.
Cylinder elongation and cylindrical filaments

This theory is upheld by the data of another set
of observations presented by Reynolds (1984a) for
filaments of Melosira italica (now Aulacoseira subarc-
tica). This freshwater diatom comprises cylindri-
cal cells joined together by the valve ends, effec-
tively lengthening the cylinder in a linear way.
In Reynolds’ experiments, the mean length of the
cells was hc = 19.0 µm, the mean diameter (dc )
Figure 2.11 Plot of sinking rates (ws ) against length (as
was 6.3 µm. The external volume of the cylindical
cells per filament) of Aulacoseira subarctica filaments compared
cell was calculated from π(dc /2)2 hc ∼592 µm3 . with the sinking rates calculated for a spheres of the same
The ratio hc /dc , a sort of index of cylindricity, volume and density (ws calc). The ratio, φ r = (ws calc)/ws , is
is 3.0 and the diameter of the sphere (ds ) of the also shown against the horizontal axis. Redrawn from
same volume is ∼10.4 µm. Adding another cell Reynolds (1984a)
doubles the length, volume and cylindricity, but
the area of mutual contact between them means
that the surface is not quite doubled but the (200–220 µm, hc /dc ∼ 30, ϕr ∼ 4.5). This velocity
diameter of the equivalent sphere is increased is, moreover, about the same as that calculated
by about a third, to 13.1 µm. The sinking rates for a sphere of similar density and of diameter
(ws ) of individual filaments of a killed suspen- equivalent to only 1.2 cells. Thus, the sacrifice of
sion of an otherwise healthy, late-exponential extra sinking speed is small in relation to the
strain of Aulacoseira (excess density ∼251 kg m−3 ) gain in size and where the loss of surface area is
were measured directly and grouped according probably insignificant.
to the number of cells in the filament. The mea- It is, of course, a feature of many pennate
surements were compared directly with the rates diatoms in the plankton to have finely cylin-
calculated for spheres of equivalent volume and drical cells. On the basis of a small number
excess density (ws calc) in Fig. 2.11; equivalent val- of measurements on a species of Synedra and
ues of ϕr (as ws calc/ws ) are also included. treating the essentially cuboidal cells as cylin-
The plot is instructive in several ways. Length- ders (hc = 128±11 µm, greatest dc ∼ 8.9 µm,
ening increases size as it does ws . Filament forma- hc /dc 13–16), Reynolds (1984a) solved ϕr (as ws
tion does not decrease sinking rate with respect calc/ws ) ∼ 4.1. For the shorter individual cuboidal
to single cells. On the other hand, the increments cells of Fragilaria crotonensis, whose length
become smaller with each cell added, with an (∼70 µm) exceeded mean width (3.4 µm) by a fac-
asymptote (in this instance) of about 15 µm s−1 , tor of over 20, ϕr was determined to be about 2.75.
reached by a filament of 11–12 cells in length In the experiments with single cells of Asterionella
sured sinking rates of killed coenobia of Fragi-

laria crotonensis are plotted against the numbers
of cells in the colonies and compared with the
curve of (ws calc) predicted for spheres of the
same volumes. Whereas, once again, Stokes’ equa-
tion predicts a continuous increase in sinking
rate against size, coenobial lengthening tends to
stability at 16–20 cells, not far from the point,
indeed, at which the lateral expanse exceeds
the lengths of the component cells and the
coenobium starts to become ribbon-like. It is
not unusual to encounter ribbons of Fragilaria
in nature exceeding 150 or 200 cells (i.e. some
500–700 µm in length) but it is perhaps initially
surprising to find that they sink no faster than
filaments one-tenth their length! A further ten-
dency that is only evident in these long chains is
that the ribbons are not always flat but are some-
times twisted, with a frequency of between 140
and 200 cells per complete spiral. The extent to
which this secondary structure influences sink-
ing behaviour is not known.
The secondary shape of Asterionella coenobia
is reminiscent of spokes in a rimless wheel, set
generally at nearly 45◦ to each other. When there
are eight such cells, they present a handsome star
Figure 2.12 Plot of sinking rates (ws ) against the number of shape (alluded to in the generic name), although
cells of killed Fragilaria crotonensis colonies compared with the the mutual attachments determine that the
sinking rates calculated for a sphere of the same volume and colony is not flat but is like a shallow spiral
density (ws calc). The ratio, φ r = (ws calc)/ws , is also shown
staircase. Cell no. 9 starts a second layer; in cul-
against the horizontal axis. Redrawn from Reynolds (1984a).
tures and in fast-growing natural populations,
coenobia of more than eight cells are frequently
formosa (length 66 µm, mean width 3.5 µm) the observed (although, in the author’s experience,
form resistance was found to fall in the range chains of over 20–24 cells are in the ‘rare’ cate-
2.3 to 2.8. Compared with the squat cylindri- gory). Following a similar approach to that used
cal shapes of the centric diatoms Cyclotella and for Aulacoseira and Fragilaria, Reynolds (1984a)
Stephanodiscus species for which measurements plotted measured sinking rates of killed coenobia
are available, the impact of attenuation on form against the numbers of cells in the colony and,
resistance is plainly evident (see Table 2.5). again, compared them with corresponding curve
(ws calc) for spheres of the same volumes (see
Colony formation Fig. 2.13). Now, although the observed sinking
Of course, both Fragilaria and Asterionella are rates, ws , suggest the same tendency to be asymp-
more familiarly recognised as coenobial algae, totic up to a point where there are 6–10 cells in
the cells in either case remaining tenuously the coenobium, it is equally clear that higher
attached on the valve surfaces, in the central numbers of cells make the colony sink faster,
region in Fragilaria (to form a sort of double-sided with no further gain in ϕr after ∼4.0. Reynolds
comb) and at the flared, distal end in Asterionella. (1984a) interpreted this result as demonstrating
These distinctive new shapes generate some the advantage to form resistance of creating a
interesting sinking properties. In Fig. 2.12, mea- new shape, from a cylinder to a spoked disc,
study of Asterionella, using different source mate-

rial, but its results fitted comfortably into the
pattern. As pointed out earlier, several separate
studies of the sinking rates of typical, eight-celled
colonies of Asterionella, using quite disparate
sources of algae, having differing dimensions
and, possibly, densities, nevertheless achieved
strikingly similar results (generally ws = 5.5–11.0
µm s−1 ), provided that the material used was
moribund, poisoned or killed prior to mea-
surement. However, the clearest demonstration
that the sinking characteristics are strongly gov-
erned by shape comes from the somewhat fortu-
itous experimentation opportunity presented to
Jaworski et al. (1988). When diatoms grow and
divide, the replication of the siliceous cell wall
is achieved by the development of a ‘new’ valve
inside each of the two mother valves. When the
daughters are eventually differentiated as new,
independent entities, one has the dimensions of
its parent cell but the other is slightly smaller,
being no larger than the smaller of the parental
valves (Volcani, 1981; Li and Volcani, 1984; Craw-
Figure 2.13 Plot of sinking rates (ws ) of killed stellate ford and Schmid, 1986). In a clone of succes-
colonies of Asterionella formosa of comprising varying numbers sive generations, one cell retains the parental
of cells, compared with the sinking rates calculated for a
dimensions but all the others are, to varying
spheres of the same volume and density (ws calc). The ratio,
extents, smaller. Average size must diminish in
φ r = (ws calc)/ws is also shown against the horizontal axis.
proportion to the number of divisions. This is
clearly not a process without limit, as the species-
specific sizes of diatoms normally remain within
which however is lost if space between cells is stable and predictable limits. Size is recovered
progressively plugged by more, dense cells that periodically through distinctive auxospore stages,
add nothing to the hydrodynamic resistance. which give rise to vegetative cells, with relatively
It is a satisfying piece of teleology that the large overall dimensions and large sap vacuoles.
maximum advantage should seem to be achieved This process is observed in culture as well as in
at the most typical coenobial size. Indeed, there nature. When it was noticed, however, that sub-
is further observational and experimental evi- cultured cells of Asterionella clone L354, one of
dence that the shape determines more of the those isolated from wild types and maintained
sinking behaviour of stellate colonies than any at the Ferry House Laboratory of the Freshwater
of the Stokesian components. First, the argu- Biological Association, were becoming unusually
ment would need to hold for Tabellaria flocculosa small, it was decided to include it in the sinking
var. asterionelloides, whose robust cells form eight- studies that were then being conducted in the
armed stellate colonies that contrast with the laboratory. These observations were commenced
more typical habits adopted by this genus (see in 1981 and were continued until 1986 (Jaworski
Knudsen, 1953). Frequently, up to 16 cells com- et al., 1988). The clone never did recover size
prise the colonies but they remain in adherent but the rate of growth became extremely slow
pairs, preserving a regular eight-radiate form. towards the end. By the time of the last measure-
It is also of significance that some of the ments, late in 1985, the cells had shrunk from
entries in Fig. 2.13 were derived from another long cuboids, 65 µm in length, to stubby ovoids,
against colony volume and compared with ws

calc of equivalent spheres having the same den-
sities. As the size of the cells diminished over
the first 2 years (during which time the mean
colony volume reduced by nearly two-thirds and
the mean density difference with the medium
almost doubled), sinking rates actually increased
slightly. The rising mean of about 7.2 to one of
8.0 µm s−1 does not violate the hypothesis that
the stellate cell arrangement dominates the sink-
ing rate. Thereafter, with the aggregate volumes
of colonies below 1200 µm3 , sinking rate dimin-
ished. The curve fitted to all sinking rate deter-
minations (curve 1 in Fig. 2.14) reveals less of
what is happening than do the two separate fit-
ted curves referring to colonies >1200 µm3 (curve
2) and those <1200 µm3 (curve 3). By taking
each of these regression lines and using them
to solve form resistance as the calculated sink-
ing rate for a sphere divided by the regression-
predicted actual sinking rate, the relationship is
further amplified. Curve (4) for all points (cor-
responding to regression 1) shows the typical
form-resistance outline, albeit rather flattened.
The right-hand side of curve 5, corresponding
to regression 2, quickly takes ϕr > 2 and > 3,
towards the ‘plateau’ level shown in Fig. 2.13.
To the left of curve 5 (corresponding to regres-
Figure 2.14 Plot of sinking rates of killed, stellate sion 3), sinking rate varies more closely with vol-
eight-celled colonies of Asterionella cells of diminuitive length ume (though not so steeply as ws calc); ϕr is in
and volume, compared with the sinking rates calculated for a the range 1.6–2.4. The change in behaviour has
sphere of the same volume and density (ws calc). Three a pivot point when the colony falls below 1200
regressions are fitted to the observed values: (1) applies to all µm3 , when individual cell volumes are <150 µm3
points, (2) to those colonies >1000 µm3 in volume and (3)
and the cells are 18–20 µm in length. The likely
to those <1000 µm3 in volume. The ratio, φ r = (ws calc)/ws
significance of this is that the hydromechanical
is shown against the horizontal axis with respect to the
estimates of ws derived from regression equation (2.1) (curve
characteristics of the star shape are eventually
4) and those combining equations (2.2) and (2.3) (curve 5). overtaken by those of a gyre of ovoids, more rem-
Redrawn with permission, from Jaworski et al. (1988). iniscent of Anabaena than Asterionella.
Some of these measurements are included
in the summaries of calculated form resistance
measuring about 5.5 µm with a basal width of owing to shape distortions.
4.2 µm and a height of 2.1 µm. Significantly, the
cells maintained mutual connections and there 2.5.4 Vital regulation of sinking rate
remained plenty of eight-armed colonies in the We may return, briefly, to the ability of live
culture. The dry mass of cells and the silica con- phytoplankton, especially diatoms, to exercise
tent of the walls diminished with size but overall this further control on their own sinking rates
density increased (from ∼1100 to 1200 kg m−3 ). below that expected from a modified Stokes equa-
The sinking rates (ws ) of killed cells measured tion, with all components properly evaluated. It
over the five year period are plotted in Fig. 2.14 has to be confirmed first that we use a correct
interpretation of facts. Smayda (1970) referred to varies interspecifically. It is not known whether
the acceleration of sinking rates in moribund the form of the valves responds to environmen-
diatoms, and the several mechanisms by which tal control, so making it possible to build longer
this might come about. These include the aggre- chains and filaments or shorter ones, according
gation of dying cells and, through their involve- to circumstances. Many papers refer to varying
ment with other planktic detritus (zooplankton numbers of cells per coenobium during growth
exuviae and faecal fragments, colloidal organ- and senescence, perhaps speculating on the role
ics, fragments of plant remains), their forma- of nutrient limitation. Observations on many nat-
tion in to larger floccular particles, collectively ural populations and of isolates treated in the
known as ‘marine snow’ (Alldredge and Silver, laboratory leads me to the view that coenobia are
1988). In these aggregates, particles may sink larger (i.e. comprise more cells) if grown rapidly
faster than might be predicted if they were sepa- in unshaken cultures but are more fragmented in
rated. However, the consistency and reproducibil- old cultures, with many moribund cells. These
ity of behaviour of killed cells having similar clearly make some impact on sinking rate but,
form resistance should prompt us to regard this as we have shown already, these are relatively
as the ‘normal’ sinking performance and to ask small compared to the sinking-rate variations
how it might be that live, healthy cells reduce attributable to changes in the physiological vital-
their sinking rates below those that Stokes’ law ity of the cells.
would predict. The physiological mechanisms regulating
The scale of these reductions is impressive, sinking rate remain stubbornly resistant to expla-
the ‘live’ rate being up to one order of magni- nation. Over a number of years, colleagues at
tude, and frequently two- to four-fold less than the Ferry House supported my efforts to develop
the ‘killed’ rate. As well as the case of Stephano- a plausible hypothesis for this behaviour. It is
discus rotula illustrated in Fig. 2.8, there is an not an entirely negative outcome to say that
abundance of data to show the sinking rates of these succeeded only in excluding several pos-
healthy, eight-celled Asterionella formosa colonies sibilities. We never found a sufficient or suffi-
to be typically 2–3 µm s−1 (about 0.2–0.3 m d−1 ) ciently responsive variation in density that would
rather than the 7–8 µm s−1 explained by the explain a two- or three-fold change in sinking
modified Stokes’ equation (Smayda, 1974; Tilman rate. While we were able to bring pressure to bear
and Kilham, 1976; Jaworski et al., 1981; Wise- on planktic Cyanobacteria to collapse gas vesi-
man and Reynolds, 1981). Similarly, variations in cles, to subject diatoms to similar treatment – up
the sinking rate of Fragilaria crotonensis may be to about 12 bars, anyway – produced no response
<0.4 m d−1 for long periods but quite quickly at all. Yet if sufficient of the specific photosyn-
increase to up to 1.1 m d−1 (≤13 µm s−1 ) when thetic inhibitor DCMU [3-(3,4-dichlorophenyl)-1,1-
cells are nutrient limited or have been exposed dimethyl urea] is added to a healthy, slow-sinking
to excessive sunlight (Reynolds, 1973a, 1983a). suspension of Asterionella cells to block their pho-
Indeed, these studies have suggested that it is tosynthesis, sinking rate rose quickly to the rates
a useful biological adaptation for an otherwise of killed or moribund examples. In time, both
non-motile organism to be able to increase sink- effects are reversible (photosynthetic capacity,
ing rate spontaneously and to ‘accelerate out of sinking-rate control are recovered). Contempor-
danger’ from excessive irradiance, especially in aneous work in our laboratory on the susceptibil-
stabilising water columns (Reynolds et al., 1986). ity of Asterionella formosa to attack by parasitic
It might appear that regulation of sinking fungi (see especially Canter and Jaworski, 1981)
rate is under the control of the diatom. Certainly, had just revealed that, under conditions of low
variability in the number of cells per colony may light or darkness, infective chytrid zoospores are
be, to an extent, self-regulated as the intercellular not attracted to Asterionella cells as they are in
links vary in structure, some being much more the light. We deduced that actively photosynthe-
amenable to separation than others. The fre- sising cells either broadcast a signal to the adja-
quency of ‘linking valves’ and ‘separation valves’ cent medium advertising their presence or that
SINKING AND ENTRAINMENT IN NATURAL TURBULENCE 67
they create a local change in the medium that is that it might be trailed in threads from cells,
exploited or avoided by the infective spores. The like a parachute or in the manner of the chitin
lateral thinking that arose from our discussions hairs of Thalassiosira. Unlike chitin, threads of
led to the hypothesis that photosynthesising cells mucilage require active maintenance by healthy
were immobilising water around their periphery. cells but would be sufficiently frail to be shed
Thus, the particle acquires a new identity and quickly, when they become a liability or too costly
the new dimensions of alga + water, in much to maintain. Traditional algal anatomists would
the similar way that an investment of mucilage concur in conversation that such mucilaginous
provides (see Section 2.5.2). threads and trails exist but there seems very little
Considering that a sufficient swathe of published to uphold a compelling case. Even the
mucilage has not been observed in these algae, use of Indian-ink irrigation, a popular technique
the candidate mechanism that we proposed was for revealing mucilaginous structures, has proved
that the surface charge on the cells was variable unhelpful to the argument. However, a recent
and that this might affect the amount of water description of mucilaginous protuberances radi-
thus immobilised. This was not an original inspi- ating from the marginal cells of Pediastrum duplex
ration but an echo of an earlier hypothesis, put colonies (Krienitz, 1990) has been confirmed in
forward by Margalef (1957). Based on his own the photomicrographs of Padisák et al. (2003a).
observations of differing polarity and electro- Back in the 1980s, we had proposed a num-
kinetic (‘zeta’) potential of Scenedesmus cells, he ber of approaches to investigate the hypothe-
developed a theory of ‘structural viscosity’, where sis, including the possibility of using WETSTEM
algae regulated the viscosity of their immedi- electron microscopy, for the observation of liv-
ate surroundings through the electrical charge ing materials at high magnification, which was
on the outer cell wall. It must be emphasised then just becoming available. However, this was
that, like any other small particles dispersed in also the time when the sponsorship of science
an electrolyte (albeit, a very weak one), algae was moving rapidly from academic, curiosity-led
carry a surface charge in any case. This is, in problems such as this. Purchase or lease of suit-
part, determined by the ionic strength of the able apparatus was less the problem than was the
medium. Moreover, several publications detail- continued support to sustain an active group of
ing direct measurements of surface charge using personnel. Resolution of the mechanism of vital
electrophoretic procedures were available (Ives, regulation of sinking rate by diatoms remains
1956; Grünberg, 1968; Hegewald, 1972; Zhurav- open to future research.
leva and Matsekevich, 1974). It became our objec-
tive to demonstrate that variable sinking rates
are related to physiologically mediated changes 2.6 Sinking and entrainment in
in surface charge. We used an electrophore-
sis microscope to determine simultaneously the natural turbulence
sinking rates and electrophoretic mobility of Aste-
rionella colonies, incubated under varying labora- 2.6.1 Sinking, floating and entrainment
tory conditions (Wiseman and Reynolds, 1981). Preceding sections of this chapter have reviewed
The outcome was quite clear, insofar as large the scales of the quantities of the two key com-
changes in sinking rate could not be correlated ponents of plankton entrainability – the veloci-
with relatively small variations in surface charge. ties of the intrinsic tendency of plankton to sink,
The experiments succeeded only in rejecting swim or float and the velocities of motion in
another hypothesis about sinking-rate regulation the medium. Both typically cover several orders
and in establishing a nice method for the direct of magnitude. The sinking rates of diatoms
measurement of sinking rates. span something like 1 µm s−1 to 6 mm s−1 .
The (as yet) unexplored alternative hypothesis The flotation rates of buoyant colonies of the
we put forward (Wiseman and Reynolds, 1981) Cyanobacteria such as Anabaena and Aphani-
referred to a quite different role for mucilage, zomenon may reach 40–60 µm s−1 , typical colonies
Figure 2.15 Ranges of sinking

(‘DOWN’) and floating (‘UP’)
velocities of freshwater
phytoplankters, or, where
appropriate of vertical swimming
rates of motile species, plotted
against unit volume. The algae are:
An flo, Anabaena flos-aquae; Aphan,
Aphanizomenon flos-aquae; Ast,
Asterionella formosa; Aul, Aulacoseira
subarctica; Cer h, Ceratium
hirundinella; Chlm, Chlorococcum; Chlo,
Chlorella; Clo ac, Closterium aciculare;
Cycl, Cyclotella meneghiniana; Fra c,
Fragilaria crotonensis; Mic; Microcystis
aeruginosa; Pla ag, Planktothrix
agardhii; Sta p, Staurastrum pingue;
Ste r, Stephanodiscus rotula; Volv,
Volvox aureus. Redrawn with
permission from Reynolds (1997a).
of Gloeotrichia and Microcystis may achieve 100–300 suspension. Larger species potentially move faster
µm s−1 , while some of the largest aggregations or farther but they need to be either flagellate
achieve 3–4 mm s−1 (Reynolds et al., 1987; Oliver, or to govern their own buoyancy to counter the
1994). Among motile organisms, reported ‘swim- tendency to sink. Indeed, there is a strong indi-
ming speeds’ range between 3–30 µm s−1 for the cation that their ability to overcome elimination
nanoplanktic flagellates to 200–500 µm s−1 for from the plankton, at least in the extant vege-
the larger dinoflagellates, such as freshwater Cer- tative stages of their life cycles, depends upon
atium and Peridinium (Talling, 1971; Pollingher, the amplification of motility that large size con-
1988) and marine Gymnodinium catenatum and Lin- fers. In essence, phytoplankton motility can be
gulodinium spp. (see Smayda, 2002). Large colonies differentiated among those that can do little to
of Volvox can attain almost 1 mm s−1 (Sommer stop themselves from sinking (mostly diatoms),
and Gliwicz, 1986) while the ciliate Mesodinium the very large, which self-regulate their move-
is reported to have a maximum swimming speed ments, and the very small for which it seems to
of 8 mm s−1 (Crawford and Lindholm, 1997). matter rather little.
At the other end of the motility spectrum, Even so, when the comparison is made, the
solitary bacteria and picoplankters probably sink range of intrinsic rates of sinking (ws ), float-
no faster than 0.01–0.02 µm s−1 (data collected ing (−ws ) and flagellar self-propulsion (us ) rep-
in Reynolds, 1987a). The data plotted in Fig. 2.15 resented in Fig. 2.15 (mostly ≤10−3 m s−1 ) are
show that, despite the compounding of several 1–6 orders of magnitude smaller than the sam-
factors in the modified Stokes equation (2.16), ple turbulent velocities cited in Table 2.2 (mostly
the intrinsic rates at which phytoplankton move ≥10−2 m s−1 ). Generally speaking, the deduction
(or potentially move) in relation to the adjacent is that ws u∗ . This does not mean that the sink-
medium are powerfully related to their sizes. ing potential (or the floating or migratory poten-
Smaller algae sink or ‘swim’ so slowly that the tial) is overcome, just that gravitating plankters
motion of the water is supposed to keep them in are constantly being redistributed. What really
matters to sinking particles is the relative magni-

tudes of ws and the upward thrusts of the turbu-
lent eddies w (see Section 2.3.1). If ws > w , noth-
ing prevents the particle from sinking. While,
however, ws < w some particles can be trans-
ported upwards faster than they gravitate down-
wards – and their sinking trajectories are reiniti-
ated at a higher point in the turbulence field. Of
course, there are, other things being equal, down-
ward eddy thrusts which add to rate of vertical
descent of the particles. Given that the upward
and downward values of w are self-cancelling, ws
is not affected. However, the greater is the mag-
nitude of w relative to ws , the more dominant
is the redistribution and the more delayed is the
descent of the particles.
Figure 2.16 The entrainment criterion, as expressed in Eq.
In this way, the ability of fluid turbulence to
(2.19). In essence, the larger is the alga and the greater its
maintain sinking particles in apparent suspen-
intrinsic settling (or flotation) velocity, then the greater is the
sion depends on the ratio of sinking speed to the turbulent intensity required to entrain it. Redrawn with
vertical turbulent velocity fluctuations. Empiri- permission from Reynolds (1997a).
cal judgement suggested that this entrainment
threshold occurs at 1–2 orders of magnitude
greater than the intrinsic motion of the par- and oceans, u∗ is a highly variable quantity (see,
ticle. In a detailed consideration of this rela- e.g., Table 2.2), with the variability being often
tionship, Humphries and Imberger (1982) intro- expressed over high temporal frequencies (from
duced a quotient (herein referred to as ) to the order of a few minutes) and, sometimes, over
represent the boundary between behaviour domi- quite short spatial scales. Whilst in near-surface
nated by turbulent diffusivity of the medium and layers even the lower values u∗ may often still be
behaviour dominated by particle buoyancy: an order of magnitude greater than the intrin-
sic particle properties, the entrainment condi-
= ws /15[(w )2 ]1/2 (2.19)
tion is not necessarily continuous in the vertical
Noting that, in open turbulence, the magnitude direction. Just taking the example of the transfer
of u∗ is not dissimilar from [(w )2 ]1/2 (see Eq. (2.4)), of the momentum of wind stress on the water
Reynolds (1994a) proposed that substitution of surface and the propagation of turbulent eddies
u∗ in Eq. (2.19) gave a useful approximation to in the water column below (Eq. 2.5), it is clear
the value of . The main line drawn into Fig. that the loss of velocity through the spectrum
2.16 ( = 1) against axes representing sinking of diminishing eddies will continue downwards
rate (ws ) and turbulent velocity (u∗ ), is proposed into the water to the point where the residual
as the boundary between effective entrainment energy is overcome by viscosity. As the penetrat-
(diffusivity dominates distribution) and effective ing turbulence decays with depth, the entraining
disentrainment (particle properties dominate dis- capacity steadily weakens towards a point where
tribution). neither u∗ nor [(w )2 ]1/2 can any longer satisfy the
The adjective ‘effective’ is important, because particle-entrainment condition. In other words,
entrainment is never complete while ws has finite the turbulence field is finite in extent and is open
value; neither is disentrainment total while there to the loss of sedimenting particles (and, equally,
is any possibility that motion in the water can to the recruitment of buoyant ones).
deflect the particle from its intrinsic vertical This leads us to a significant deduction about
trajectory. However, the main point requiring the suspension and continued entrainment of
emphasis at this point is that, in lakes, rivers phytoplankton in lakes and the sea. It is the extent
of the turbulence, rather than its quantitative number remaining in the column (Nt ) is approx-
magnitude, that most influences the persistence imately
or otherwise of various types of organism in the
N t = N 0 (1 − ws t/ h w ) (2.20)
plankton. For the same reason, the factors that
influence the depth of the turbulent mixed layer Let us suppose that the column is now instan-
and its variability are important in the survival, taneously and homogeneously mixed, such that
seasonality and succession of phytoplankton in the particles still in the column are redistributed
natural waters. These are reviewed briefly in later throughout the column but those that have
sections but it is first necessary to consider their already settled into the basal boundary layer are
behaviour within the mixed layer itself. not resuspended. This action reintroduces parti-
cles (albeit now more dilute) to the top of the
column and they recommence their downward
2.6.2 Loss of sinking particles from trajectory. Obviously, the time to complete set-
turbulent layers tling is now longer than t (though not longer
The purpose of this section is not to demonstrate than 2t ).
that sinking particles are lost from turbulent, The process could be repeated, each time leav-
surface-mixed layers but to provide the basis of ing the settled particles undisturbed but redis-
estimating the rate of loss. The converse, how tributing the unsettled particles on each occa-
slowly they are lost, is the essence of adapta- sion. If, within the original period, t , m such
tion to planktic survival. The development here mixings are accommodated at regular intervals,
is rather briefer than that in Reynolds (1984a), as separating quiescent periods each t /m in dura-
its principles are now broadly accepted by plank- tion. The general formula for the population
ton scientists. Its physical basis is rather older, remaining in suspension after the first short
owing to Dobbins (1944) and Cordoba-Molina et period is derived from Eq. (2.20):
al. (1978). Smith (1982) considered its application
N t/m = N 0 (1 − ws t /mh w ) (2.21)
to plankton. Interestingly, empirical validation of
the theory comes from using plankton algae in After the second, it will be
laboratory-scale measurements.
N 0 (1 − ws t /mh w )(1 − ws t /mh w )
Let us first take the example of a completely
static water column, of height hw (in m), open and after the mth,
at the surface with a smooth bottom, to which
N t = N 0 (1 − ws t /mh w )m (2.22)
small inert, uniform particles are added at the
top. Supposing their density exceeds that of the Because t = hw /ws , Eq. (2.22) simplifies to
water, that they satisfy the laminar-flow condi-
N t = N 0 (1 − 1/m)m (2.23)
tion of the Stokes equation and sink through
the water column at a predictable velocity, ws (in As m becomes large, the series tends to an expo-
m s−1 ), then the time they take to settle out from nential decay curve
the column is t = hw /ws (in s). If a large num-
N t = N 0 (1/e) (2.24)
ber (N0 , m−3 ) of such particles are initially dis-
tributed uniformly through the water column, where e is the base of natural logarithms (∼2.72).
after which its static condition is immediately Solving empirically,
restored, they would settle out at the same rates
N t = 0.368 N 0 (2.25)
but, depending on the distance to be travelled,
in times ranging from zero to t . The last particle This derivation is instructive in several respects.
will not settle in a time significantly less than The literal interpretation of Eq. (2.25) is that
t , which continues to represent the minimum repeated (i.e. continuous) mixing of a layer
period in which the column is cleared of parti- should be expected still to retain 36.8% of an ini-
cles. At any intermediate time, t, the proportion tial population of sinking particles at the end
of particles settled is given by N0 ws t/hw . The of a period during which particles would have
Figure 2.17 The number of

particles retained in a continuously
mixed supension compared to the
retention of the same particles in a
static water column of identical
height. Redrawn from Reynolds
(1984a).
left the same layer had it been unmixed. More- phytoplankton, because mixed layers may range
over, the time to achieve total elimination (te ) is from hundreds of metres in depth down to few
an asymptote to infinity but we may deduce that millimetres and, in any given water body, the
the time to achieve 95% or 99% elimination is variability in mixed-layer depth may occur on
(respectively) calculable from timescales of as little as minutes to hours.
te /t = log e 0.05/log 0.368 = 3.0

2.6.3 Mixed depth variability in natural
or water columns
Turbulent extent defines the vertical and hori-
te /t = log e 0.01/log 0.368 = 4.6
zontal displacement of particles that fulfil the
Using this approach, the longevity of suspension entrainment criterion (Fig. 2.16). The vertical
can be plotted (Fig. 2.17). It takes three times extent of turbulent boundary layers, uncon-
longer for 95% of particles to escape a mixed layer strained by the basin morphometry or by
than were the same depth of water left unmixed. the presence of density gradients (open turbu-
It may also be noted that the number of mixings lence), is related primarily to the kinetic energy
does not have to be vast to achieve this effect. transferred: rearranging Eq. (2.9),
Substituting in Eq. (2.23), if m = 2, Nt = 0.25 N0 ;
if m = 5, Nt = 0.33 N0 ; if m = 20, Nt = 0.36 N0 . l e ≈ (u∗ )(du/dz)−1 m (2.26)
The formulation does not predict the value
of the base time, t . However, it is abundantly where (du/dz) is the vertical gradient of horizon-
evident from the rest of this chapter that, the tal velocities (in m s−1 m−1 ) and le is the dimen-
smaller is the specific sinking rate, ws , the longer sion of the largest eddies. Entries in Table 2.2 per-
will be t for any given column of length, hm . Fur- taining to the upper layers of the open ocean and
ther, the greater is the mixed depth, then the also of a moderately large lake like the Bodensee
longer is the period of maintenance. Any ten- (Germany/Switzerland) imply an increase in mix-
dency towards truncation of the mixed depth, ing depth of about 9 m for each increment in
hm , will accelerate the loss rate of any species wind forcing of 1 m s−1 . Of course, even this rela-
that does not have, or can effect, an absolutely tionship applies only under a constant wind:
slow rate of sinking. an increase in wind speed necessarily invokes a
This confirms the adaptive significance of a restructuring, which may take many minutes to
slow intrinsic sinking rate. It has much less to organise (see below). Similarly, the contraction
do with delaying settlement directly but, rather, of the thickness of the mixed layer following a
through the extension it confers to average res- weakening of the wind stress is gradual, pending
idence time within an actively-mixed, entrain- the dissipation of inertia. The structure of turbu-
ing water layer. The relationship also provides lence under a variable, gusting wind is extremely
a major variable in the population ecology of complex!
The complexity is further magnified in small still water (such as at night), the heat trans-
basins (Imberger and Ivey, 1991; Wüest and Lorke, fer increases the density of the surface water
2003). Where the physical depth of the basin con- and causes instability. The denser water is liable
strains even this degree of dissipative order, the to tumble through the water column, acceler-
water is fully mixed by a turbulence field which ating as it does so and displacing lighter water
is, as already argued, made up by a finer grain of upwards, until it reaches a depth of approximate
eddies. Moreover, the shallower is the water body, isopycny (that is, where water has the identical
the lower is the wind speed representing the density). This process can continue so long as the
onset of full basin mixing likely to be. Thus, from heat imbalance between air and water persists,
the entries in Table 2.2, it is possible to deduce all the time depressing the depth of the density
that a wind of 3.5 m s−1 might be sufficient to gradient. The energy of this penetrative convection
mix Lough Neagh, Northern Ireland (maximum may be expressed:
depth 31 m, mean depth 8.9 m). In fact, the lake
B Q = (gγ Q ∗T )(ρw σ )−1 W m2 s−2 J−1 (= m2 s−3 )
is usually well mixed by wind and is often quite
turbid with particles entrained by direct shear (2.28)
stress on the sediment.
At all other times, the buoyancy acquired by
Density gradients, especially those due to the
the warmer water resists its downward trans-
thermal expansion of the near-surface water sub-
port through propagating eddies, whether gener-
ject to solar heating, also provide a significant
ated by internal convection or externally, such as
barrier to the vertical dissipation of the kinetic
through the work of wind. They are instead con-
energy of mixing. Although there is some out-
fined to a layer of lesser thickness. Its depth, hb ,
ward conduction of heat from the interior of
tends to a point at which the kinetic energy ( Jk )
the Earth and some heat is released in the dis-
and buoyancy ( Jb ) forces are balanced. Its instan-
sipation of mechanical energy, over 99% of the
taneous value, also known as the Monin–Obukhov
heat received by most water bodies comes directly
length, may be calculated, considering that the
from the Sun. The solar flux influences the ecol-
kinetic energy flux, in W m−2 , is given by:
ogy of phytoplankton in a number of ways but,
in the present context, our concern is solely the J k = τ (u∗ ) = ρw (u∗ )3 kg s−3 (2.29)
direct role of surface heat exchanges upon the
vertical extent of the surface boundary layer. while the buoyancy flux is the product of the
Starting with the case when there is no wind expansion due to the net heat flux to the water
and solar heating brings expansion and decreas- (Q ∗T ), also in W m−2 ,
ing water density (i.e. its temperature is >4 ◦ C), a J b = 1/2gh b γ Q ∗T · σ −1 kg s−3 (2.30)
positive heat flux is attenuated beneath the water
surface so that the heating is confined to a nar- where γ is the temperature-dependent coefficient
row near-surface layer. The heat reaching a depth, of thermal expansion of water, σ is its specific
z, is expressed: heat (4186 J kg−1 K−1 ) and g is gravitational accel-
eration (9.8081 m s−2 ).
Qz = Q∗
Te
−kz
W m−2 (2.27) Then, when J b = J k ,
where e is the base of natural logarithms and
h b = 2σρw (u∗ )3 (gγ Q ∗T )−1 m (2.31)
k is an exponential coefficient of heat absorp-
tion. Q ∗T is that fraction of the net heat flux, Q T Owing to the organisational lags and the vari-
which penetrates beyond the top millimeter or ability in the opposing energy sources, Eq. (2.29)
so. Roughly, Q ∗T averages about half the incoming should be considered more illustrative than pre-
short-wave radiation, Q S . The effect of acquired dictive. Nevertheless, simulations that recognise
buoyancy suppresses the downward transport of the complexity of the heat exchanges across the
heat, save by conduction. surface can give close approximations to actual
If the water temperature is <4 ◦ C, or if it events, both over the day (Imberger, 1985) and
is > 4 ◦ C but the heat flux is away from the over seasons (Marti and Imboden, 1986).
The effect of wind is to distribute the heat

evenly throughout its depth, hm . If the heat flux
across its lower boundary is due only to con-
duction and negligible, the rate of temperature
change of the whole mixed layer can be approxi-
mated from the net heat flux,
dθ/dt = Q ∗T (h m ρw σ )−1 K s−1 (2.32)
The extent of the turbulent mixed boundary

layer may be then viewed as the outcome of
a continuous ‘war’ between the buoyancy gen-
erating forces and the dissipative forces. The
battles favour one or other of the opponents,
depending mostly upon the heat income, Q ∗T ,
and the kinetic energy input, τ , as encapsu-
lated in the Monin–Obukhov equation (2.31). Note
that if the instantaneous calculation of hb is
less than the lagged, observable hm , buoyancy
forces are dominant and the system will become
more stable. If hb > hm , turbulence is domi-
nant and the mixed layer should be expected
to deepen. It becomes easy to appreciate how,
at least in warm climates, when the water tem-
peratures are above 20 ◦ C for sustained periods,
diurnal stratification and shrinkage of the mixed
depth occurs during the morning and net cool-
ing leads to its breakdown and extension of the
mixed depth during the afternoon or evening.
The example in Fig. 2.18 shows the outcome of
diel fluctuations in heat- and mechanical-energy
fluxes to the density structure of an Australian
reservoir.
2.6.4 Vertical structure in the pelagic

Over periods of days of strong heating and/or
weak winds, during which convective energy is
Figure 2.18 Diel variability in the mixed depth of a
insufficient to bring about complete overnight
subtropical reservoir (Wellington Reservoir, Western
mixing, there will develop a residual density dif-
Australia), reflecting the net heat exchanges with the
ference between the surface mixed layer and the atmosphere. The top panel shows the depth distribution of
water beneath it, leading to the formation of isotherms through a single day. The left-hand column of
a more enduring density gradient, or thermo- smaller graphs shows features of the evolving temperature
cline. Its resistance to mixing is acquired dur- structure (based on Imberger, 1985). The right-hand column
ing preceding buoyancy phases. This resistance proposes stages in the season-long development of enduring
is expressed by the (dimensionless) bulk Richard- stratification. Redrawn with permission from Reynolds
son number, Ri b , expressing the ratio between (1997a).
the two sets of forces:
Ri b = [ρw g h m ] [ρw (u∗ )2 ]−1 (2.33)

where ρ w is the density difference between than a few hours or days at a time (polymictic).
the surface mixed layer and the water beneath It explains the patterns of seasonal stratification
the thermocline. Imberger and Hamblin (1982) in tropical lakes wherein stable density struc-
divided Ri b by the aspect ratio (i.e. the horizon- tures are precipitated by relatively small gains
tal length of the layer, L, by, divided by its thick- in heat content but are correspondingly liable
ness, hm ) in order to test the robustness, in any to major mixing episodes for a relatively small
given system, of the density gradients detectable. drop in suface temperature (atelomictic lakes). It
This (still dimensionless) ratio, they named after can be used to test the contribution of ionic
another atmosphere scientist, Wedderburn. strength (e.g. salt content in reinforcing density
gradients).
W = Ri b [L h m ]−1 Examples of all these kinds of stratification,
= [ρw g(h m )2 ] [ρw (u∗ )2 L ]−1 (2.34) classified by Lewis (1983), may equally be viewed
from the opposite standpoint as a series describ-
Working in meters, values of W > 1 are held ing variability in the extent and duration of
to describe stable structures, resistant to further turbulent mixing. The intriguing consequence
down-mixing and incorporation of deeper water is that the depth of the turbulent mixed layer
into the surface mixed layer, without either a (hm ) may remain nearly constant, when it is
significant diminution in the value of ρw (e.g. the full depth of the water (H) in a shallow,
through convectional heat loss across the sur- wind-exposed site. In a large, deep lake, it may
face) or the sharp increase in the turbulent inten- fluctuate between <1 m and >100 m, in some
sity, (u∗2 ). Structures in which W is significantly instances, within a matter of a few hours.
<1 are liable to modification by the next phase The Wedderburn formulation equation has
of wind stress. also been used to determine whether lakes will
This relationship is especially sensitive to the stratify at all. Putting W = 1 and interpolating
onset of thermal stratification and, equally, simu- the observed summer-thermocline depths of a
lates the occurrence of mixing events. The insets series of temperate lakes, Reynolds (1992c) rear-
in Fig. 2.18 show the onset of an early-season ranged Eq. (2.34) to determine the density dif-
thermocline, when net strong daily heating and ference, ρw , between the waters separated by
the absence of sufficient wind action or night- the seasonal thermocline. In most instances, the
time convection overcome full column mixing. outcome was not less than 0.7 to 0.9 kg m−3 .
A series of days with net warming compounds At the depths of the respective thermoclines,
the stability which, in acquiring increased resis- the density difference would resist erosion by
tance, halts the downwelling mechanical energy surface-layer circulations generated by winds up
at lesser and lesser depths. The stepped gradient to ∼20 m s−1 . Winds much stronger than these
of ‘fossil’ thermoclines is typical and explicable. would cause deepening of the mixed layer and
It is only following a change (lesser heat income, depression of the thermocline. Interpolating the
greater net heat loss or the onset of storms, W corresponding values for ρw and u∗ , Eq. (2.34)
diminishing) that deeper penetration by turbu- was solved for hm against various nominated val-
lence eats into the colder water and sharpens the ues for L. The resultant slope separated almost
thermocline at the base of the mixed layer. perfectly the dataset of stratifying and non-
This is the basic mechanism for the onset and stratifying lakes assembled by Gorham and Boyce
eventual breakdown in temperate lakes and seas. (1989) (Fig. 2.19).
It also serves to track the seasonal behaviour of This outcome is a satisfying vindication of
many more kinds of system other than those of theoretical modelling. Its principal virtue in
middle- to high-latitude lakes. It applies to very plankton biology is to empiricise the relation-
deep lakes and seas, which may remain incom- ships by which familiar environmental compo-
pletely mixed (meromictic) for years on end. It nents govern the entrainment and transport of
also covers circumstances of water bodies too plankton-sized particles and how often the vari-
shallow or too wind-exposed to stratify for more ous conditions might apply.
and simultaneously lowering the depth of the

thermocline.
2.6.5 Mixing times

The intensity of turbulence required to entrain
phytoplankton covers almost 2 orders of
magnitude: non-motile algae with sinking
rates of ∼40 µm s−1 are effectively dispersed
through turbulence fields where u∗ ≤ 600 µm
s−1 (Eq. 2.19), whereas u∗ > 50 mm s−1 is suffi-
cient to disperse the least entrainable buoyant
plankters (ws ≥ 1 mm s−1 ) (see Section 2.6.1).
As has already been suggested, the intensity of
mixing is often less important to the alga than is
Figure 2.19 Depth, H, plotted against L, the length across the vertical depth through which it is mixed. The
various lakes of routinely stratifying (•) and generally depth of water through which phytoplankton
unstratified lakes (◦) considered by Gorham and Boyce is randomised can be approximated from the
(1989). The line corresponds to Reynolds’ (1992c) prediction Wedderburn equation (2.34). Putting W = 1 and
of the wind stress required to overcome a density difference u∗ ≥ 0.6 m s−1 , the numerator is equivalent to
of 0.7 kg m−3 (equivalent to u∗ = 0.025 m s−1 ). The diamond
≥0.36 for each 1000 m of horizontal distance,
symbols refer to lakes said to stratify in some years but not in
L. Dividing out gravity, the product, ρw (hm )2
others Redrawn with permission from Reynolds (1997a).
solves at ∼0.037 kg m−1 . This is equivalent to an
average density gradient of ≥0.04 kg m−3 m−1 per
This is a suitable point at which to empha- km across a lake for a 1-m mixed layer, ∼0.01 for
sise an important distinction between the thick- a 2-m layer, ∼0.0045 for a 3-m layer, and so on.
ness of the mixed layer and the depth of the The weaker is the average density gradient, then
summer thermocline. As demonstrated here, the the greater is the depth of entrainment likely to
latter really represents the transition between be. The limiting condition is the maximum pen-
the upper parts of the water column (in lakes, etration of turbulent dissipation, unimpeded by
the epilimnion) that are liable to frequent wind- density constraints. Where a density difference
mixing events and the lower part that is isolated blocks the free passage of entraining turbulence,
from the atmosphere and the effects of direct the effective floor of the layer of entrainment is
wind stress (the hypolimnion). The thickness of defined by a significant local steepening of the
the intermediate layer (in lakes, the metalimnion) vertical density gradient. Reynolds’ (1984a) con-
is defined by the steepness of the main verti- sideration of the entrainment of diatoms, mostly
cal gradient of temperature (the thermocline) or having sinking rates, ws , substantially less than
density (pycnocline) between the upper and lower 40 µm s−1 , indicated that the formation of local
layers, though neither layer is necessarily uni- density gradients of >0.02 kg m−3 m−1 probably
form itself. The top of the thermocline may repre- coincided with the extent of full entrainment,
sent the point to which wind mixing and/or that is, in substantial agreement with the above
convection last penetrated. Otherwise, the mixed averages. For the highly buoyant cyanobacterial
layer is entirely dynamic, its depth and structure colonies, however, disentrainment will occur in
always relating to the current or very recent (the much stronger levels of turbulence and from
previous hour) balance between Jb and Jk . The mixed layers bounded by much weaker gradients.
mixed layer can be well within the epilimnion Assumption of homogeneous dispersion of
or its full extent. Any tendency to exceed it, how- particles fully entrained in the actively mixed
ever, results in the simultaneous deepening of layer allows us to approximate the average veloc-
the epilimnion, the surface circulation shearing ity of their transport and, hence, the average time
off and entraining erstwhile metalimnetic water of their passage through the mixed layer. In their
elegant development of this topic, Denman and sediment traps was to have been a part of this
Gargett (1983) showed that the average time of programme and it was decided that the choice
travel (tm ) through a mixed layer unconstrained of traps and the authenticity of their catches
by physical boundaries or density gradients cor- could be tested within the special environment
responds to: of the enclosures. The plan was to add a mea-
sured quantity of alien particles to the water col-
tm = 0.2 h m (u∗ )−1 s (2.35)
umn (actually, Lycopodium spores, well steeped in
∗
Because, in this instance, both hm and u are wetting agent and preservative), then to moni-
directly scaled to the wind speed, U (Eqs. 2.5, tor the subsequent loss from suspension in the
2.11), tm is theoretically constant. Interpolation water and to compare the calculated flux with
into Eq. (2.35) of entries in Table 2.2 in respect of the sediment trap catches. Three such experi-
Bodensee permit its solution at 1416 s. The proba- ments were carried out, under differing hydro-
bility of a complete mixing cycle is thus 2 × 1416 graphic conditions. The results were published
s (≈47 minutes). (Reynolds, 1979a) but the unexpected bonus of
In the case of the wind-mixed layer of a small the experiments was the contrasting rates of loss
or shallow basin, or one bounded by a den- from suspension of ostensibly identical spores
sity gradient, the timescale through the layer is under the varying conditions.
inversely proportional to the flux of turbulent In the first experiment, carried out in win-
kinetic energy: ter, the spores were dispersed over the enclosure
surface, during windy conditions which intensi-
tm = h m (2u∗ )−1 s (2.36)
fied in the subsequent few days. A near-uniform
Following this logic, a wind of 8 m s−1 may be distribution with depth was quickly established
expected to mix a 20-m epilimnion in 2000 s (33 (see Fig. 2.20). The spores (d = 32.80 ± 3.18;
minutes) but a 2-m layer in just 200 s (3.3 min- ρc = 1049 kg m−3 ; ϕr ∼ 2.2) had a measured sink-
utes). A wind speed of 4 m s−1 would take twice ing rate (ws ) of 15.75 µm s−1 at 17 ◦ C, which,
as long in either case. adjusted for the density and viscosity of the water
These approximations are among the most at the 4–5 ◦ C obtaining in the field, predicted an
important recent derivations pertaining to the in-situ intrinsic sinking rate of 0.96 m d−1 . The
environment of phytoplankton. They have a pro- theoretical time for spores to eliminate the enclo-
found relevance to the harvest of light energy sure (at the time, H ∼ 11.8 m) was thus calcu-
and the adaptations of species to maximise the lated to be (t = ) 12.3 days. In fact, the elimi-
opportunities provided by turbulent transport nation proceeded smoothly, always from a near-
(see Chapter 3). uniformly distributed residual population at an
average exponential rate of −0.10 m d−1 , which
2.6.6 Particle settling from variable mixed value corresponds to a 95% removal in (te =) 30
layers: an experiment days. The ratio te /t is lower than predicted in Sec-
As part of an effort to improve the empirical tion 2.6.2 (2.44 against 3.0). This may be explained
description of the sedimentary losses of phyto- by probable violation of the initial assumption of
plankton from suspension, Reynolds employed full mixing of the water column throughout the
several approaches to measuring the sedimentary experiment. Although no significant density gra-
flux in the large limnetic enclosures in Blelham dient developed, continuous and complete verti-
Tarn, UK. These cylindrical vessels, 45 m in diam- cal mixing of the enclosure cannot be verified.
eter, anchored in 11–12 m of water contained Nevertheless, the outcome is sufficiently close to
sufficient water (∼18 000 m3 ) to behave like nat- the model (Fig. 2.20) solution for us not to reject
ural water columns. Their hydraulic isolation the hypothesis that entrained particles are lost
ensured all populations husbanded therein were from suspension at an exponential rate close to
captive and virtually free from external contami- −(ws /hm ).
nation (Lack and Lund, 1974; Lund and Reynolds, In the second experiment, commenced in
1982; see also Section 5.5.1). The deployment of June, spores were dispersed at the top of the
THE SPATIAL DISTRIBUTION OF PHYTOPLANKTON 77
Figure 2.20 Modelled (M) and

actual (A) depth–time distributions
of preserved Lycopodium spores (of
predetermined sinking
characteristics) introduced at the
water surface of one of the Blelham
enclosures on each of three
occasions (1, 9 January; 2, 3 June; 3,
9 September) during 1976, under
sharply differing conditions of
thermal stability. Lycopodium
concentrations plotted as cylindrical
curves; density gradients plotted as
dashed lines. ∗ – indicates no field
data are available. Redrawn from
Reynolds (1984a).
stratified enclosure during relatively calm con- mixed-layer deepening. Variability in wind forc-
ditions. Sampling within 30 minutes showed a ing was quite high and a certain degree of re-
good dispersion but still restricted to the top 1 entrainment is known to have occurred but the
m only. However, 4 days later, spores were found time taken to achieve 95% elimination from
at all depths but the bulk of the original addition the upper 9 m of the water column (te = 18.0
was accounted for in a ‘cloud’ of spores located days) at the calculated in-situ sinking rate (ws )
at a depth of 5–7 m. After a further 7 days, mea- of 1.32 m d−1 exceeded the equivalent t value
surable concentrations were detected only in the (9/1.32 = 6.82) by a factor of 2.6.
bottom 2 m of the column, meaning that, effec- The three results are held to confirm that the
tively, the addition had cleared 10 m in 11 days, depth of entrainment by mixing is the major con-
at a rate not less than 0.91 m d−1 . Adjusted for straint on elimination of non-motile plankters
the density and viscosity of the water at the top heavier than water, that the eventual elimina-
of the water column, the predicted sinking rate tion is however delayed rather than avoided, and
was 1.42 m d−1 . Thus, overall, the value of t for that prolongation of the period of suspension is
the first 10 m (= 7 days) was exceeded by the in proportion to the depth of the mixed layer,
observed te (= 11 days) by a factor of only 1.57. Part wherein u∗ ≥ 15 (ws ).
of the explanation is that sinking spores would
have sunk more slowly than 1.42 m d−1 in the
colder hypolimnion. However, the model expla- 2.7 The spatial distribution of
nation envisages a daily export of the population
from the upper mixed layer (varying between 0.5 phytoplankton
and 4 m during the course of the experiment),
calculated as N exp −(ws /hm ), whence it contin- The focus of this chapter, the conditions of
ues to settle unentrained at the rate ws m d−1 . To entrainment and embedding of phytoplankton in
judge from Fig. 2.20, this is an oversimplification the constant movement of natural water masses,
but the prediction of the elimination is reason- is now extended to the conditions where water
able. movements are either insufficiently strong or
The same model was applied to predict the insufficiently extensive to randomise the spa-
distribution and settlement of Lycopodium spores tial distribution of phytoplankton. This section
in the third experiment, conducted during the is concerned with the circumstances of plankters
autumnal period of weakening stratification and becoming disentrained and the consequences of
weakening entrainment for individuals, popula- can respond (for instance, light fluctuations are
tions and communities, as augured by spatial dif- more frequent than cell division) or so much less
ferentiation in the vertical and horizontal distri- rapidly that it is perceived as a constant (such as
butions of natural assemblages. annual temperature fluctuations having a much
Distributional variation is subject to issues of lower frequency than cell division) which will
scaling which need to be clarified. It has already be no more relevant to today’s populations than
been made plain that aquatic environments are is the onset of the next ice age to the mainte-
manifestly heterogenous, owing to spatial differ- nance of present-day forests (Reynolds, 1993b). In
ences in temperature, solute content, wind stress, between, where driver and response scales are
etc., and that each of these drivers is itself sub- more closely matched, the interactions are rather
ject to almost continuous variation. However, more profound, as in the frequency with which
while precise values are impossible to predict, new generations are recruited to a water col-
the range of variability may be forecast with umn mixed to a different extent on successive
some confidence, either on the basis of averag- days.
ing or experience, or both. We may not be able The variability in the instantaneous distribu-
to predict the intensity of wind mixing in a tion of phytoplankton may be considered in rela-
lake some three weeks or more into the future tion to an analogous spatial scale. Consider first a
but we may estimate from the knowledge base randomised suspension of unicellular flagellates,
the probability with which a given wind inten- such as Chlamydomonas or Dunaliella. Viewed at
sity will prevail. The changes in temperature, the 1–10 µm scale, distribution appears highly
insolation, hydraulic exchanges and the delivery patchy, resolvable on the basis of presence or
of essential nutrients affecting a given stretch absence. In the range 10–1000 µm, the same dis-
of water also occur on simultaneously differ- tribution is increasingly perceived to be near-
ing scales of temporal oscillation – over minutes uniform but, in the turbulence field of a wind-
to hours, night–day alternations, with changing mixed layer, variability over the 1–10 mm scale
season, interannually and over much broader may attest to the interaction of algal movements
scales of climatic change. The nesting of the with water at the viscous scale (Reynolds et al.,
smaller temporal scales within the larger scales 1993a). In the range 10–100 mm and, perhaps,
holds consequences for phytoplankters in the 10–1000 mm, the distribution may again appear
other direction, too, towards the probabilities of uniform. Beyond that, the increasing tendency
being ingested by filter-feeders, of the adequacy for there to be variations in the intensity of mix-
of light at the depths to which entrained cells ing leads to the separation of water masses in
may be circulated, even to the probability that the vertical (at the scale of tens to hundreds
the energy of the next photon hitting the photo- of metres) and in the horizontal (hundreds of
synthetic apparatus will be captured. The point metres to hundreds of kilometres), at least to
is that the reactions of individual organelles, the extent that they represent quite isolated and
cells, populations and assemblages are now gen- coexisting environments, each having quite dis-
erally predictable, but the impacts can only be tinct conditions for the survival of the flagel-
judged at the relevant temporal scales. These lates and the rate of their recruitment by growth.
responses and their outcomes are considered in This is but one example of the principle that the
later chapters in the context of the relevant pro- relative uniformity or heterogeneity within an
cesses (photosynthesis, assimilation, growth and ecological system depends mainly upon the spa-
population dynamics). However, the interrelation tial and temporal scale at which it is observed
of scales makes for fascinating study (see, for (Juhász-Nagy, 1992).
instance, Reynolds, 1999a, 2002a): in the end, Uniformity and randomisation, on the one
the distinction is determined by the reactivity of hand, and differentiation (‘patchiness’), on the
the response. This means that critical variations other, may thus be detected simultaneously
alter more rapidly than the process of interest within a single, often quite small system.
Moreover, the biological differentiation of indi- a gradient between the layers persists, albeit by
vidual patches may well increase the longer their now at a greater physical depth). Internal waves
mutual isolation persists. Thus it remains impor- may form as a consequence of differential veloci-
tant to make clear the spatial scale that is under ties, mirroring those formed at the water surface
consideration, whether in the context of vertical by the drag of high-velocity winds; and, just as
or horizontal distribution. surface waves break when the velocity differences
can no longer be contained, so internal waves
2.7.1 Vertical distribution of phytoplankton become unstable and collapse (Kelvin–Helmholtz
Against this background, it seems appropriate instabilities: see, e.g., Imberger, 1985, for details),
to emphasise that the expectation of the verti- releasing more bottom water into the upper con-
cal distributions of phytoplankton is that they vection.
should conform to the vertical differentiation of In shallow water columns, the extension of
the water column, in terms of its current (or, the convective layer is confined to the physi-
at least, very recent) kinetic structure. The lat- cal water depth, in which all energy-dissipative
ter may comprise a wind-mixed convective layer interactions and return flows must be accommo-
overlying a typically less energetic layer of turbu- dated. This necessarily results in very complex
lence that is supposed to diminish with increas- and aggressive mixing processes, often extending
ing depth, towards a benthic boundary layer in to the bottom boundary layer and, on occasions,
which turbulence is overcome by friction with penetrating it to the extent of entraining, and
the solid surface. However, in the seas as in resuspending, the unconsolidated sedimented
lakes, the horizontal drift of the convective layer material.
has to be compensated by counterflows, which Supposing wind-driven convective layers
movement promotes internal eddies and provides everywhere to be characterised by u∗ ≥ 5 ×
some turbulent kinetic energy from below. The 10−3 m s−1 (i.e. 5 mm s−1 ), they should be capa-
formation of vertical density gradients may allow ble of fulfilling the entrainment criterion for
the confinement of the horizontal circulation to plankton with sinking, floating or swimming
the upper part of the water column, leaving a dis- speeds of ≤250 × 10−6 m s−1 . Comparison with
tinct and kinetically rather inert water mass of Section 2.6.1 supports the deduction that the
the pycnocline, with very weak vertical motion. distribution of almost all phytoplankters must
Density gradients form at depths in large, deep be quickly randomised through the vertical
lakes and in the sea but do not contain basin- extent of convective layers. However, if the
wide circulations; even though the gradients may driving energy weakens, so that the convective
persist, they may be rhythmically or chaotically layer contracts (in line with the Monin–Obukhov
displaced through the interplay of the gravita- prediction [Eq. 2.31] or, without simultaneous
tional ‘sloshing’ movements of deep water masses heat gain, because u∗ diminishes), plankters that
and the convective movements of the surface were entrained towards the bottom of the layer
layer. become increasingly liable to disentrainment in
The vertical extent of the convective layer is, situ, where, ergo, their own intrinsic movements
as we have seen, highly variable and subject to begin to be expressed.
change at high frequency. It can vary from a few Four examples of distributional responses to
millimetres to tens of metres over a period of physical structure are sufficient to demonstrate
hours and between tens to hundreds of metres the basic behaviours of phytoplankters that are
over a few days. The presence of density gradients heavier than water (ρ c > ρ w ), those that are fre-
reduces the entrainability of the deeper water quently lighter than water (ρ c < ρ w ), those non-
into the surface flow. There is often great com- motile species that are more nearly isopycnic
plexity at the interface (note, even if the top layer (ρ c ∼ ρ w ) and those that are sufficiently motile for
of the deeper water is sheared off and incorpo- any density difference to be at times surmount-
rated into the convective circulation of the upper, able. The selection also employs some of the
While numerous other factors intervene in the

seasonal population dynamics and periodicity of
diatoms, the importance of the depth of the sur-
face mixed layer as another ecological thresh-
old that must be satisfied for successful main-
tenance and recruitment of diatoms has been
demonstrated many times (Reynolds and Wise-
man, 1982; Reynolds et al., 1983b; Sommer, 1988a;
Huisman et al., 1999; Padisák et al., 2003b). If
the rate of recruitment through cell growth and
replication fails to make good the aggregate rate
of all losses, including to settlement, then the
standing population must go into decline. The
extent of the mixed depth is already implicated
in the ability of a seed population to remain
Figure 2.21 Depth–time plot of the vertical distribution of in suspension. As the Lycopodium experiments
Asterionella formosa in the North Basin of Windermere demonstrate (see Section 2.6.6), the nearer this
through 1947. Isopleths in live cells mL−1 . The shaded area contracts to the surface, the faster is the rate of
represents the extent of the metalimnion. Original from Lund loss from the diminishing mixed layer itself and
et al. (1963), and redrawn from Reynolds (1984a). from the enlarging, stagnating layer beneath it.
This occurs independently of the chemical capac-
ity of the water to support growth, although it is
differing ways that distributional data may be
often influenced by spontaneous changes in the
represented.
intrinsic sinking rate (Reynolds and Wiseman,
1982; Neale et al., 1991b).
Non-motile, negatively buoyant plankters The plot of Lund et al. (1963) (Fig. 2.21) shows
(ρc > ρw ) the strong tendency in the first 3 months of the
For the first case, the classical study of Lund year towards vertical similarity in the concentra-
et al. (1963) on the season-long distribution of tion of Asterionella (with the isopleths, in cells
Asterionella cells in the North Basin of Winder- mL−1 , themselves vertically arranged), as their
mere in 1947 is illustrated (Fig. 2.21). The den- numbers slowly rise during the spring increase.
sities of diatoms mostly exceed, sometimes con- With the progressive increase in day length and
siderably, that of the surrounding water in lakes potential intensity of solar irradiance, the lake
and seas. Those bound to be negatively buoy- starts to stratify, with a pycnocline (represented
ant (they have positive sinking rates) are des- in Fig. 2.21 by the fine stippling) developing at a
tined to be lost progressively from suspension, depth of between 5 and 10 m from the surface.
at variable rates that are due to the relation- The contours reflect the segregating response of
ship between the (variable) intrinsic particle sink- the vertical distribution, with an initial near-
ing rate and the (variable) depth of penetration surface acceleration in recruitment but followed
of sufficient kinetic energy to fulfil the species- soon afterwards by rapid decline in numbers,
specific entrainment criterion. Even before the as sinking losses by dilution from the truncated
critical quantities were known, numerous stud- mixed layer overtake recruitment. The distribu-
ies had demonstrated the sensitivity of diatom tion of contours beneath the pycnocline acquire a
distribution to water movements and to the diagonal trend (reflecting algal settlement) while
onset of thermal stratification in particular, both the near-horizontal lines in the pycnocline itself
in lakes (Ruttner, 1938; Findenegg, 1943; Lund, confirm the heterogeneity of numbers in the ver-
1959; Nauwerck, 1963) and in the sea (Mar- tical direction and the strong vertical gradient
galef, 1958, 1978; Parsons and Takahashi, 1973; in algal concentration in the region of the pyc-
Smayda, 1973, 1980; Holligan and Harbour, 1977). nocline. It is not until the final breakdown of
thermal stratification (usually in December in

Windermere) that approximate homogeneity in
the vertical is recovered.
Positively buoyant plankters (ρc < ρw )

The vertical distribution of buoyant organ-
isms, which include many of the planktic, gas-
vacuolate Cyanobacteria during at least stages of
their development, is similarly responsive to vari-
ability in the diffusive strength of vertical con-
vection, save that algae float, rather than sink,
through the more stable layers. A further differ-
ence is that the population of the upper mixed
layer potentially experiences concentration by
net recruitment from upward-moving organisms
rather than dilution as downward-moving organ- Figure 2.22 Changes in the vertical distribution of
isms are shed from it. Microcystis aeruginosa colonies in a small lake (shown as
It has long been appreciated that the forma- cylindrical curves) in relation to temperature (isopleths
in ◦ C), during 28/29 July, 1971 (SS, sunset; SR, sunrise). Data
tion of surface scums of buoyant Cyanobacteria
of Reynolds (1973b) and redrawn from Reynolds (1984a).
(variously known, colloquially and in many lan-
guages, as water blooms, flowering of the waters,
etc.: Reynolds and Walsby, 1975), and involving anticyclonic weather in July 1971. The colonies
such genera as Anabaena, Anabaenopsis, Aphani- were, on average, buoyant throughout, having
zomenon, Gloeotrichia, Gomphosphaeria, Woronichinia a mean flotation rate of ∼9 µm s−1 during the
and (especially) Microcystis are prone to form in first day that increased almost twofold during
still, windless conditions (Griffiths, 1939). Buoy- the hours of darkness (Reynolds, 1973b). Note
ant Trichodesmium filaments also form locally that an established temperature gradient, extend-
dense surface patches in warm tropical seas ing downwards from a depth of about 3.5 m,
under calm conditions (Ramamurthy, 1970): the already contained the buoyant population and
little flakes of filaments also merited the sailors’ the changing vertical distribution of Microcystis
colloquialism of ‘sea sawdust’. occurred in relation to the secondary microstratifi-
The mechanisms of scum formation are not cation that developed during the course of the day
straightforward but, rather, require the coinci- (density gradient 0.1 kg m−3 m−1 ). A light breeze
dence of three preconditions: a pre-existing pop- occurred in the early evening of 28 July, before
ulation, a significant proportion of this being windless conditions resumed. Some convectional
rendered positively buoyant on the balance of cooling also occurred during the night, suffi-
its gas-vesicle content, and the hydrographic con- cient to redistribute the population to a small
ditions being such as to allow their disentrain- extent but not to dissipate the surface scum that
ment (Reynolds and Walsby, 1975). The present had formed, which, from an average concentra-
discussion assumes that the first two criteria are tion of ∼160 colonies mL−1 , increased 37-fold (to
satisfied, scum formation now depending upon 5940 mL−1 at 23.30).
the relatively short-term onset of lowered diffu- The method of plotting, greatly favoured by
sivity to the water surface, so that the magni- the early plankton ecologists, invokes the use of
tude of the flotation velocity (−ws ) is no longer ‘cylindrical curves’. These are drawn as laterally
overwhelmed by the turbulent velocity (u∗ ). The viewed solid cones or more complex ‘table-legs’,
example presented in Fig. 2.22 traces the chang- the cross-sectional diameter at any given point
ing distribution of colonies of Microcystis aerug- being proportional to the cube root of the con-
inosa, in relation to the thermal structure in a centration. These shapes capture well the discon-
small temperate lake during one 24-h period of tinuities in a given vertical distribution but may
(where, regardless of density, small size deter-

mines a low Stokesian velocity) and among rather
larger, mucilage-invested phytoplankters, which
are genuinely able to ‘dilute’ the excess mass of
cell protoplasm and structures in the relatively
large volume of water that the mucilaginous
sheath immobilises. In either case, the adapta-
tions to pelagic survival take the organisms clos-
est to the ideal condition for suspension. These
are, literally, the most readily entrainable plank-
ters according to the definition (Section 2.6.1,
Figure 2.23 The relationship between vertical patchiness Eq. 2.19).
of Microcystis in the shallow Eglwys Nynydd reservoir and
In Fig. 2.24, sequences in the vertical dis-
wind velocity (U). Data of George and Edwards (1976) and
tributions of two non-motile green algae –
redrawn from Reynolds (1984a).
the nanoplanktic chlorococcal Ankyra and the
microplanktic, coenobial palmelloid Coenochloris
be less helpful than contoured depth–time plots (formerly ascribed to Sphaerocystis) – are depicted
(such as used in Fig. 2.21) over long periods of in relation to stratification in the large limnetic
time. Neither do they necessarily convey the gen- enclosures in Blelham Tarn. The original data
eralism between vertical discontinuity and the refer to average concentrations in metre-thick lay-
physical heterogeneity. In consideration of a 2- ers (sampled by means of a 1-m long Friedinger
year series of Microcystis depth profiles in a small, trap: Irish, 1980) or in multiples thereof. The
shallow reservoir, (Eglwys Nynydd: area 1.01 km2 , data are plotted as stacks of individual cylin-
mean depth 3.5 m), George and Edwards (1976) ders, the diameters of which correspond to the
calculated a crowding statistic (x∗ , owing to Lloyd, respective cube roots of the concentrations. In
1967) and its ratio to the mean concentration both cases, the algae are dispersed approximately
over the full depth (x̄) to demonstrate the sus- uniformly through the epilimnion, while num-
ceptibility of vertical distribution to the wind- bers in the hypolimnion remain low, owing to
forced energy. Putting x∗ = [x̄ + s2 /x̄ − 1], where weak recruitment either by growth or by sedi-
s2 is the variance between the individual sam- mentation (ws < 0.1 m d−1 ). The effect of grazers
ples in each vertical series, they showed that the has not been excluded but simultaneous stud-
relative crowding in the vertical, (x∗ /x̄), occurred ies on loss rates suggest that the impact may
only at low wind speeds (U < 4 m s−1 ) and in have been small, while neither species was well-
approximate proportion to U), but wind speeds represented in simultaneous sediment-trap col-
over 4 m s−1 were always sufficient to randomise lections or samples from the surface deposits
Microcystis through the full 3.5-m depth of the (Reynolds et al., 1982a). On the other hand, deep-
reservoir (Fig. 2.23). ening of the mixed layer and depression of
the thermocline result in the immediate ran-
Neutrally buoyant plankters (ρc ∼ ρw ) domisation of approximately neutrally buoyant
In this instance, ‘neutral’ implies ‘approximately algae throughout the newly expanded layer. Such
neutral’. As already discussed above (Section 2.5), species are thus regarded as being always likely
it is not possible, nor particularly desirable, for to become freely distributed within water lay-
plankters to be continuously isopycnic with the ers subject to turbulent mixing, then to settle
medium. Nevertheless, many species of phyto- from them only very slowly and, of course, to
plankton that are non-motile and are unencum- be unable to recover a former distribution when
bered by skeletal ballast (or the gas-filled spaces mixing weakens (Happey-Wood, 1988).
to offset it) survive through maintaining a state
in which they do not travel far after disentrain- Motile plankters (us > u∗ )
ment. This state is achieved among very small Flagellates and ciliates are capable of directed
unicells of the nanoplankton and picoplankton movements that, actually as well as potentially,
Figure 2.24 Instances in the

vertical distribution of non-motile
phytoplankton in a Lund Enclosure
during the summer of 1978, shown
as cylindrical curves. Ankyra is a
unicellular nanoplankter;
Coenochloris occurs as palmelloid
colonies. Redrawn from Reynolds
(1984a).
may result in discontinuous vertical distributions vertical heterogeneity of flagellate distribution

in lakes and seas. This ability is compounded by in ponds, lakes and coastal embayments. Ref-
the capacity for self-propulsion of the alga (the erence to no more than a few investigative
word ‘swim’ is studiously avoided – see Section studies is needed (Nauwerck, 1963; Moss, 1967;
2.3.4). In terms of body-lengths per unit time, Reynolds, 1976a; Cloern, 1977; Moll and Stoer-
the rates of progression may impress the micro- mer, 1982; Donato-Rondon, 2001). Works detail-
scopist but, in reality, rarely exceed the order of ing behaviour of particular phylogenetic groups
0.1–1 mm s−1 . include Ichimura et al. (1968) and Klaveness (1988)
In general, the rates of progress that are pos- on cryptomonads; Pick et al. (1984) and Sandgren
sible in natural water columns are related to (1988b) on chrysophytes; Croome and Tyler (1984)
size and to the attendant ability to disentrain and Häder (1986) on euglenoids. Note also that
from the scale of water movements (Sommer, not all flagellate movements are directed towards
1988b). Moreover, the detectable impacts on ver- the surface. There are many instances of con-
tical distribution also depend upon some direc- spicuous surface avoidance (Heaney and Furnass,
tionality in the movements or some common set 1980; Heaney and Talling, 1980a; Gálvez et al.,
of responses being simultaneously expressed: if 1988; Kamykowski et al., 1992) and of assembling
all movements are random, fast rates of move- deep-water ‘depth maxima’ of flagellates, analo-
ment scarcely lead to any predictable pattern of gous to those of Planktothrix and photosynthetic
distribution. For instance, the impressive verti- bacteria (Vicente and Miracle, 1988; Gasol et al.,
cal migrations of populations of large dinoflag- 1992).
ellates are powerfully and self-evidently respon- The example illustrated in Fig. 2.25 shows the
sive to the movements of individual cells within contrasted distribution of Ceratium hirundinella.
environmental gradients of light and nutrient This freshwater dinoflagellate is known for its
availability. This applies even more impressively strong motility (up to 0.3 mm s−1 ) (see Section
to the colonial volvocalean migrations (Section 2.6.1) and its well-studied capacity for vertical
2.6.1) where all the flagellar beating of all the migration under suitable hydrographic condi-
cells in the colony have to be under simultaneous tions (Talling, 1971; Reynolds, 1976b; Harris et al.,
control. The point needs emphasis as the flagellar 1979; Heaney and Talling, 1980a, b; Frempong,
movements of, for instance, the colonial chrys- 1984; Pollingher, 1988; James et al., 1992). These
ophytes (including the large and superficially properties enable it quickly to take up advan-
Volvox-like Uroglena), seem less well coordinated: tageous distribution with respect to light gradi-
they neither ‘swim’ so fast, nor do their move- ents, when diffusivity permits (u∗ < 10−4 m s−1 ).
ments produce such readily interpretable distri- The right-hand profile in Fig. 2.25 shows the
butions as Volvox (Sandgren, 1988b). vertical distribution of Ceratium during windy
On the other hand, there is a large num- weather in a small, eutrophic temperate lake;
ber of published field studies attesting to the the left-hand profile shows a distribution under
Figure 2.25 Contrasted vertical

distributions of the motile
dinoflagellate Ceratium hirundinella, in
a small stratifying lake (Crose Mere,
UK), in relation to temperature
gradients (θ ) and percentage light
penetration (pecked lines). (a) was
observed under very calm
conditions; (b) under strong winds.
only weakly stratified conditions but one that is ing significant or systematic differences in con-
strongly allied to the light gradient. centration.
There have been few systematic attempts to
2.7.2 Horizontal distribution of resolve this question directly. In a rarely cited
phytoplankton study, Nasev et al. (1978) analysed the confidence
A considerable literature on the horizontal vari- interval about phytoplankton counting by parti-
ability in plankton distribution has grown up, tioning the variance attaching to each step in the
seemingly aimed, in part, towards invalidating estimation – from sampling through to count-
any preconceived assumption of homogeneity. It ing. Provided adequate steps were taken to sup-
is difficult to determine just where this assump- press the errors of subsampling and counting
tion might have arisen, as investigations readily (Javorničky, 1958; Lund et al., 1958; Willén, 1976),
demonstrate that distributions are often far from systematic differences in the numbers present in
homogeneous. It may be that a predominance the original samples could be detected at scales of
of papers in the mid-twentieth century focused a few tens of metres but, on other occasions, not
on the population dynamics and vertical distri- for hundreds. Irish and Clarke (1984) analysed the
butions of phytoplankton in small lakes paying estimates of specific algal populations of algae in
insufficient attention to simultaneous horizon- similar samples collected from within the con-
tal heterogeneity. Such assumptions, real or sup- fines of a single Blelham enclosure (area 1641 m2 ,
posed, have no place in modern plankton science. diameter, 45.7 m) at locations nominated on a
However, even now, it is important to present a stratified-random grid. They found that the coeffi-
perspective on just how much variation might be cients of variation varied among different species
expected, over what sort of horizontal distances of plankton, from about 5%, in the case of non-
and how it might reflect the contributory physi- motile, neutrally buoyant algae, to up to 22%
cal processes. for some larger, buoyancy-regulating Cyanobacte-
ria. In another, unrelated study, Stephenson et al.
Small-scale patchiness (1984), showed that spatial variability increased
Omitting the very smallest scales (see pream- with increasing enclosure size.
ble to Section 2.7), phytoplankton is generally A general conclusion is that sampling designs
well-randomised within freshly collected water underpinning in-situ studies of phytoplankton
samples (typical volumes in the range 0.5–5 population dynamics must not fail to take notice
litres, roughly corresponding to a linear scale of of the horizontal dimension. However, the size
50–200 mm). Thus, there is normally a low coeffi- of the basin under investigation is also impor-
cient of variation between the concentrations of tant. For instance, a coefficient of variation of
plankton in successive samples taken at the same even 22% is small compared with the outcome
place. The first focus of this section is the hori- of growth and cell division, where a popula-
zontal distance separating similar samples show- tion doubling represents a variation of 100% per
Box 2.1 Langmuir circulations
Langmuir circulations are elongated, wind-induced convection cells that form at the
surface of lakes and of the sea, having characters first formalised by Langmuir (1938).
They take the form of parallel rotations, that spiral approximately in the direction of
the wind, in the general manner sketched in Fig. 2.26. Their structure is more clearly
understood than is their mechanics but it is plain that the cells arise through the
interaction of the horizontal drag currents and the gravitational resistance of deeper
water to entrainment. Thus, they provide the additional means of spatially confined
energy dissipation at the upper end of the eddy spectrum (Leibovich, 1983). In
this way, they represent a fairly aggressive mixing process at the mesoscale but
the ordered structure of the convection cells does lead naturally to a surprising
level of microstructural differentiation. Adjacent spirals have interfaces where both
are either upwelling simultaneously or downwelling simultaneously. In the former
case, there is a divergence at the surface; in the latter there is a convergence. This
gives rise to the striking formation of surface windrows or streaks that comprise
bubbles and such buoyant particles as seaweed fragments, leaves and plant remains,
insect exuviae and animal products as they are disentrained at the convergences
of downwelling water.
The dynamics and dimensions of Langmuir circulation cells are now fairly well
known. The circumstances of their formation never arise at all at low wind speeds
(U < 3–4 m s−1 : Scott et al., 1969; Assaf et al., 1971). Spacing of streaks may be as
little as 3–6 m apart at these lower wind speeds, when there is an rough correlation
between the downwelling depth and the width of the cell (ratio 2.0–2.8). In the
open water of large lakes and the sea, where there is little impediment to Langmuir
circulation, the distance between the larger streaks (50–100 m) maintains this
approximate dimensional proportionality, being comparable with that of the mixed
depth (Harris and Lott, 1973; Boyce, 1974). The velocity of downwelling (w >
2.5 × 10−2 m s−1 ) is said to be proportional to the wind speed (∼0.8 × 10−2
U): Scott et al., 1969; Faller, 1971), but the average velocities of the upwellings and
cross-currents are typically less.
Consequences for microalgae have been considered (notably by Smayda, 1970,
and George and Edwards, 1973) and are reviewed in the main text.
generation time (Reynolds, 1986b). Moreover, a The mechanism concerns the Langmuir circula-
spatial difference within a closed area of water tions, which are consequent upon a strong wind
only 45.7 m across is unlikely to persist, as the acting on a shallow surface layer, when acceler-
forcing of the gradient is hardly likely to be ated dissipation from a spatially constrained vol-
stable. A change in wind intensity and direc- ume generates ordered structures. These are man-
tion is likely to redistribute the same population ifest as stripe-like ‘windrows’ of foam bubbles
within the same limited space. on the water surface. Even now, the formation
We may follow this progression of thinking of Langmuir cells is imperfectly understood but
to the wider confines of an entire small lake, their main properties are fairly well described
or to the relatively unconfined areas of the open (see Box 2.1).
sea. Before that, however, it is opportune to draw Although the characteristic current veloc-
attention to a relatively better-known horizon- ities prevalent within Langmuir circulations
tal sorting of phytoplankton at the scale of a (>10–20 mm s−1 ) would be well sufficient to
few metres and, curiously perhaps, is dependent entrain phytoplankton around the spiral trajec-
upon significant wind forcing on the lake surface. tories, the cells do have identifiable relative dead
Figure 2.26 Diagrammatic

section across wind-induced surface
flow to show Langmuir circulations.
Figure 2.27 Schematic section

through Langmuir rotations to show
the likely distributions of
non-buoyant (•), positively buoyant
() and neutrally buoyant, fully
entrained (∗ ) organisms. Based on
an original in George (1981) and
redrawn from Reynolds (1984a).
spots, towards the centre of the spiral, at the persistence (Evans and Taylor, 1980). Whereas it
base of the upwelling and, especially, at the top may take some minutes to organise and generate
of the convergent downwellings, marked by the the circulation, a wind of fluctuating speed and
foamlines (see Fig. 2.26). Smayda (1970) predicted direction will be constantly initiating new pat-
the distributions of planktic algae, categorised by terns and superimposing them on previous ones.
their intrinsic settling velocities, within a cross- This behaviour does not suppress the fact that
section adjacent to Langmuir spirals. Indepen- larger, more motile plankters remain liable to
dent observations by George and Edwards (1973) crude sorting, on the basis of their individual
and Harris and Lott (1973) on the distributions buoyant properties, into a horizontal patchiness
of real (Daphnia) and artificial (paper) markers in at the relatively small scales of a few metres to a
the field lent support for Smayda’s predictions. few tens of metres.
Although mostly well-entrained, sinking particles
(ρ c > ρ w ) take longer to clear the upwellings and Patchiness in small lake basins
accumulate selectively there, buoyant particles With or without superimposed Langmuir spirals,
(ρ c < ρ w ) will similarly take longer to clear the the horizontal drift is likely, at least in lakes, to
downwellings and those entering the foamline be interrupted by shallows, margins or islands,
will tend to be retained. A schematic, based on where the flow is subject to new constraints. Sup-
figures in Smayda (1970) and George (1981), is posing that little of the drifting water escapes
included as Fig. 2.27. the basin, most is returned upwind in subsur-
Such distributions of algae are not easy to ver- face countercurrents (see Imberger and Spigel,
ify by traditional sampling–counting methods, 1987). In small basins, there is a clear horizon-
because the behaviour depends not only on the tal circulation, which George and Edwards (1976)
match of the necessary physical conditions – the analogised to a conveyor belt. While this process
circulating velocity, the width and penetration of seems destined towards the basin-scale horizon-
the rotations are all wind-influenced – but their tal integration of populations, the movements of
Figure 2.28 Whole-lake ‘conveyor belt’ model of

non-buoyant (•) and positively buoyant (◦) phytoplankters,
proposed by George and Edwards (1976). Redrawn from
Reynolds (1984a).
plankters in the vertical plane may well super-

impose a distinct advective patchiness in the hor-
izontal plane. The mechanism is analogous to
the behavioural segregation in the Langmuir cir-
culation, though on a larger scale. Put simply,
the upward movement of buoyant organisms
is enhanced in upwind upwellings but resisted
in downwind downwellings; conversely, sinking
organisms accelerate in downwellings but accu-
mulate in the upcurrents. Positively buoyant
organisms accumulate on downwind (lee) shores;
negatively buoyant organisms are relatively more Figure 2.29 Advective horizontal patchiness of
abundant to windward (Fig. 2.28). phytoplankters in relation to wind direction: (a) positively
Such distributions of zooplankton have been buoyant Microcystis in Eglwys Nynydd reservoir (after George
observed, with concentrations of downward- and Edwards, 1976), isopleths in µg chlorophyll a L−1 ; (b)
swimming crustaceans collecting upwind (Cole- surface-avoiding, motile Ceratium in Esthwaite Water (after
brook, 1960; George and Edwards, 1976). Simi- Heaney, 1976), isopleths in cells mL−1 . Redrawn from
Reynolds (1984a).
lar patterns have been described for downward-
migrating dinoflagellates (Heaney, 1976; George
and Heaney, 1978); on the other hand, the down-
wind accumulation of buoyant Microcystis has
been verified graphically by George and Edwards
(1976). Representative maps of these contrasting
outcomes are shown in Fig. 2.29.
The representation in Fig. 2.29a is one of a
number of such ‘snapshots’ of variable patchi-
ness during a long period of Microcystis domi-
nance in the Eglwys Nynydd reservoir. The field
data allowed George and Edwards (1976) to cal-
culate a crowding index of horizontal patchi-
ness (x∗ ), analogous to that solved for the ver-
tical dimension (See Section 2.7.1), and to show
that its relationship to the mean population (x̄)
was a close correlative of the accumulated wind Figure 2.30 The relationship between horizontal
effect. Their data are redrawn here (Fig. 2.30) but patchiness of Microcystis in the shallow Eglwys Nynydd
with the horizontal axis rescaled as an equiva- reservoir and wind velocity (U). Data of George and Edwards
(1976) and redrawn from Reynolds (1984a).
lent steady wind speed. Patchiness is strongest
Large-scale patchiness
Nevertheless, the relationship does have a time
dimension, the horizontal mixing time, and this
may accommodate other sources of change. For
instance, if a constant wind of 4 m s−1 induces
a surface drift of the order of 400 m d−1 across
a 1-km basin, the probable mixing time is 5 d.
If, in the same 5 d, a patchy population is
recruited through one or more successive dou-
blings, then the same probability of its achiev-
ing uniformity requires stronger forcing or a
shorter mixing time. This relationship between
transport and recruitment becomes increasingly
prevalent in larger lake basins where the role of
the return current in establishing uniformity is
progressively diminished: maintenance of large-
scale patches (kilometers, days) needs persistent
spatial differences in recruitment rate. The latter
might be due directly to a local enhancement
Figure 2.31 Relationships between horizontal and vertical in organismic replication (because of warm or
patchiness of Microcystis (•, Figures 2.23, 2.30) and of Daphnia shallow water, or a point source of nutrient) or
populations (◦) in Eglwys Nynydd reservoir, as detected by to consistently enhanced removal rates by local
George and Edwards (1976). Redrawn from Reynolds
aggregations of herbivorous animals. However,
(1984a).
to be evident at all, the patch must give way
to the concentrations in a surrounding larger
stretch of water, through diffusion and erosion
by hydraulic exchanges at the periphery. Sev-
when winds are light but it weakens as winds eral publications have considered this relation-
start to exceed 3 m s−1 , disappearing altogether ship. Two of these, in particular (Skellam, 1951;
at U > 5 m s−1 . The work on Ceratium in Esthwaite Kierstead and Slobodkin, 1953), have given us the
Water (Heaney, 1976; George and Heaney, 1978; so-called KISS explicative model, relating the criti-
Heaney and Talling, 1980a, b) points consistently cal size of the patch to the interplay between the
to the development of horizontal patchiness only rates of reproductive recruitment and of horizon-
at wind speeds < ∼4 m s−1 . tal diffusivity. Specifically, Kierstead and Slobod-
Apart from illustrating the link between ver- kin (1953) predicted the radius of a critical patch
tical behaviour of phytoplankton and its horizon- (rc ) as:
tal distribution in small lakes, confirmed in the
rc = 2.4048(D x /kn )2 (2.37)
statistical interaction of horizontal and vertical
patchiness shown in Fig. 2.31, the information where Dx is the horizontal diffusivity and kn is
considered in this section helps to establish a the net rate of population increase or decrease.
general point about the confinement of water Interpolating values for kn appropriate to the gen-
motion to a basin of defined dimensions. It is that eration times of phytoplankton (the order of 0.1
once a critical level of forcing is applied, a certain to 1.0 doublings per day) and for typical wind-
degree of uniformity is reimposed. It is not that driven diffusivities (Dx ∼ 5 × 10−3 to 2 × 10−6
the small-scale patchiness disappears – all the cm2 s−1 : Okubo, 1971), critical radii of 60 m to
causes of its creation remain intact – so much as 32 km may be derived. This 3-order range spans
that the variance at the small scales becomes very the general cases of large-scale phytoplankton
similar at larger ones: small-scale heterogeneity patches in the open ocean considered by (for
collapses into large-scale homogeneity. instance) Steele (1976), and Okubo (1978), with
the most probable cases having a critical min- large, northern continental lakes is the vernal
imum of ∼1 km (review of Platt and Denman, patchiness of phytoplankton attributable to the
1980; see also Therriault and Platt, 1981). early-season growth in the inshore waters that
Even under the most favourable conditions are retained by horizontal temperature gradi-
of low diffusivities and localised rapid growth, ents associated with the centripetal seasonal
patches smaller than 1 km are liable to rapid dis- warming – the so-called ‘thermal bar’. This phe-
persion. Moreover, wind-driven diffusivity may be nomenon, first described in detail by Munawar
considerably enhanced by other horizontal trans- and Munawar (1975) in the context of diatom
port mechanisms, including by flow in river chan- growth in Lake Ontario, has been reported from
nels (see Smith, 1975), tidal mixing in estuaries other large lakes: Issyk-kul (Shaboonin, 1982),
(data of Lucas et al., 1999) and, in stratified small- Ladozhskoye, Onezhskoye (Petrova, 1986) and
to-medium lakes, by internal waves (Stocker and Baykal (Shimarev et al., 1993, Likhoshway et al.,
Imberger, 2003; Wüest and Lorke, 2003). In spite 1996).
of this, some instances of small patch persistence In general, it is fair to say that the KISS model
are on record. Reynolds et al. (1993a) reported is illustrative rather than deterministic, and it is
a set of observations on an intensely localised only imprecisely applicable to a majority of small
explosive growth of Dinobryon in Lake Balaton, fol- lakes subject to internal circulation and advec-
lowing a mass germination of spores disturbed by tion. Here, the predictive utility of the later gen-
dredging operations. The increase in cell concen- eral model derived by Joseph and Sendner (1958)
tration within the widening patch was overtaken is sometimes preferred. The fitted equation is
after a week or two, partly through dispersal in used to predict critical patch radius as a function
the circulation of the eastern basin of the lake of the advective velocity, us :
and into that of the western end but, ultimately,
rc = 3.67(us /kn ) (2.38)
because the rates of Dinobryon growth and recruit-
ment soon ran down. If kn is one division per day and us = 5 × 10−3
At the other end of the scale, satellite-sensed m s−1 (roughly what is generated by a wind force
distributions of phytoplankton in the ocean of 4 m s−1 ), rc ∼ 1.6 km. At five times the rate of
reveal consistent areas of relatively high biomass, horizontal advection (us = 25 × 10−3 m s−1 ), the
covering tens to hundreds of kilometres in some critical radius is increased to ∼8 km. Again, the
cases – usually shallow shelf waters, well supplied actual values probably have less relevance than
by riverine outflows, or along oceanic fronts and does the principle that patchiness in phytoplank-
at deep-water upwellings (see review of Falkowski ton developing in lake basins less than 10 km in
et al., 1998). The size and long-term stability of diameter is likely to be temporally transient and
these structures are due to the geographical per- not systematically persistent.
sistence of the favourable conditions that main-
tain production (shallow water, enriched nutri- Relevance of patchiness
ent supply) relative to the rates of horizontal Many of the mysteries of patchiness that con-
diffusivity in these unconfined locations. cerned plankton scientists in the third quarter
Such behaviour is observable in larger lakes, of the twentieth century may have been cleared
especially where there are persistent gradients up, but the issue remains an important one,
(chiefly in the supply of nutrients) that survive for two main reasons. One, self-evidently, lies in
seiching. Enduring patchiness was memorably the design of sampling strategies. If the purpose
demonstrated by Watson and Kalff (1981) along is merely to characterise the community struc-
a persistent nutrient gradient in the ribbon- ture, much information may be yielded from
like glacial Lake Memphrémagog (Canada/USA). infrequent samples collected at a single location
Persistent gradients of phytoplankton concen- (Kadiri and Reynolds, 1993) but, as soon as the
tration are evident from long-term surveys of exercise concerns the quantification of plankton
the North American Great Lakes (Munawar and populations and the dynamics of their change, it
Munawar, 1996, 2000). Of additional interest in is essential to intensify the sampling in both time
and space. Sampling design is covered in many the seasonal formation of cyanobacterial plates
methodological manuals (see, for instance, Sour- in deep alpine lakes (Bright and Walsby, 2000).
nia, 1978) but it is often useful to follow specific On the other, there is no conceptual objection
case studies where temporal and spatial variabil- to the inability of McGowan and Walker (1985) to
ity needed to be resolved statistically (Moll and demonstrate any significant variation in the rank-
Rohlf, 1981). order of species abundances in the North Atlantic
The second reason is the perspective that at spatial scales up to 800 km, despite strong
is required for the ecological interpretation of small-scale spatial and interseasonal and inter-
information on the structure and distribution of annual heterogeneity. This is attributable to the
planktic communities in nature. This is crucial large-scale coherences in the basin-scale forcing
to the ideas to be developed in subsequent chap- functions (direct measurements of the mid-depth
ters of this book. It is not just a case of defining circulation of this part of the Atlantic Ocean
the confidence intervals of quantitative deduc- have been given by Bower et al., 2002). Analogous
tions about organisms whose distribution has relationships among planktic components have
long been regarded as non-random, and over a been shown to be widespread through exten-
wide range of scales (Cassie, 1959; McAlice, 1970). sive circulation provinces of the tropical and
In order to sort out the multiple constraints on subtropical Atlantic Ocean (Finenko et al., 2003),
the selection, succession and sequencing of natu- despite significant intracompartmental variabil-
ral phytoplankton populations, it is always neces- ity in abundance and considerable intercompart-
sary to distinguish the dynamic driver from the metal structural differences. The observations
starting base. The assemblage that is observed in demonstrate the nature of the interaction of pop-
a give parcel of water at given place and at a ulation dynamics with the distinguishable move-
given time is unrelated to the present conditions ments of the water masses in shaping the species
but is the outcome of a myriad of processing con- structures of the pelagic.
straints applying to a finite inoculum of individ-
ual organisms of historic and probably inexplica-
ble provenance. 2.8 Summary
In this chapter, the focus has spanned the
dissipation of a fortuitous and localised recruit- The chapter explores the nature of the relation-
ment of algae in a relatively small, shallow lake ship that phytoplankters have with the physi-
through to the relative uniformity of the plank- cal properties of their environment. Water is a
ton composition of an oceanic basin. The one may dense, non-compressible, relatively viscous fluid,
depend upon the rapidity of growth in relation to having aberrent, non-linear tendencies to expand
diffusivity (Reynolds et al., 1993a); the other upon and contract. The water masses of lakes and seas
the extent of a single and possibly severe growth are subject to convection generated by solar heat-
constraint over an extensive area of open ocean ing and, more especially, by cooling heat losses
(Denman and Platt, 1975). Thus, it is important to from the surface. These motions are enhanced
emphasise that although the entraining motions in the surface layers interfacing with the atmo-
and horizontal diffusivity of pelagic water masses sphere, where frictional stresses impart mechan-
influence profoundly the distribution of phyto- ical energy to the water through boundary-layer
plankton, they do not confine organisms to a wave generation and frictional drag. Diel cycles
fixed position in relation to the motion. Trans- of insolation, geographical variations in heat-
port in the constrained circulation of a small lake ing and cooling, atmospheric pressure and the
or passage in an open ocean current each sets amplifying inertia caused by the rotation of the
a background for the dynamics of change and Earth represent continuous but variable drivers
variations in composition. The various outcomes of motion in aquatic environments. These can
arising from differing relative contributions of rarely be regarded as still: water is continuously
the same basic entraining processes are remark- in motion. However, the viscous resistance of the
ably disparate. On the one hand, we can explain water determines that the introduced motion
SUMMARY 91
is damped and dissipated through a spectrum exceeded by u∗ by a factor of ≥15. Thus, the best
of turbulent eddies of diminishing size, until descriptor of algal entrainability turns out to be
molecular forces overwhelm the residual kinetic its sinking rate and, the greater is the adaptive
energy. Instrumentation confirms emerging tur- ability to minimise it, the better able is the alga
bulence theory about the extent of water layers to contribute to its persistence in an adequately
subject to turbulent mixing and the sizes of the mixed water column.
smallest eddies (generally around 1 mm), which, The adaptive mechanisms for lowering the
together, most characterise the medium in which sinking velocity are reviewed in the context
all pelagic organisms, and phytoplankton in par- of the Stokes equation and its various deriva-
ticular, have to function. tives. Of the equation components, only parti-
The most striking general conclusion is that cle size, particle density and particle form resis-
most phytoplankters experience an immediate tance are considered subject to evolutionary or
environment that is characteristically viscous. Yet behavioural adaptation. Examples of each adap-
the physical scale is such that individuals of most tation are quantified. Adaptations to control or
categories of plankter (those less than 0.2 mm in offset density and the beneficial effects of distor-
size) and their adjacent media are liable to be tion from the spherical form are demonstrated.
transported wherever the characteristic motion The consequences of chain formation and cylin-
determines. From the standpoint of the plank- drical elongation (into filaments) on sinking rate
ter, the important criteria of the turbulent layer are explored and the effects of cell aggregation
are its vertical extent (hm ) and the rate at which to form the distinctive coenobia of Asterionella
it dissipates its turbulent kinetic energy (E). Both and Fragilaria are evaluated. In relation to the
are related to the intensity of the turbulence (u∗2 ) presumed vital regulatory component in sinking
and, thus, to the turbulent velocity (u∗ ). rate, some possible mechanisms are discussed.
The traditional supposition that the survival Some of the explanations offered are eliminated
strategy of phytoplankton centres on an ability but there remain others that await careful inves-
to minimise sinking is carefully updated in the tigation.
context of pelagic motion. Extended residence Some larger, motile organisms are success-
in the upper water layers remains the central ful plankters by virtue of adaptations that are
requirement at most times. This is attained, in antithetical to increasing entrainability. Large,
many instances, by maximising the entrainabil- motile species of Microcystis, Volvox, Ceratium and
ity of the plankter within the motion. Viewed Peridinium combine relatively large size, motil-
at a slightly larger scale, many phytoplankters ity and shape-streamlining to be able to escape
optimise their ‘embedding’ within the surface moderate-to-low turbulent intensities in order to
mixed layer. Criteria for plankter entrainment perform controlled migrations, at rates of several
are considered – for it can only be complete if the metres per day. Reducing sinking rate is far from
plankter has precisely the same density as (or is being a unique or universal adaptation qualify-
isopycnic with) the suspending aqueous medium. ing microorganisms for a planktic existence.
Even were this always desirable, it would be dif- In the later sections of the chapter, various
ficult to attain. Not only does the water vary in types of behaviour are illustrated through spe-
density with temperature and solute-content but cific examples of the vertical distributions of
the components of phytoplankton cells are rather planktic algae in relation to the increased differ-
more dense than water (typically amounting to entiation of density structure in the water col-
1020–1263 kg m−3 , compared to <1000; Table 2.3). umn. The impacts are extrapolated to horizon-
Following Humphries and Imberger (1982), rela- tal distribution and to the instances of small-
tive entrainment ( ) is instead suggested to be scale patchiness and advective patchiness in
governed by the relationship between particle small lakes, resolving in terms of algal migratory
buoyancy and turbulent diffusivity. Effectively, in speeds in relation to the velocity of advective cur-
order to achieve turbulent entrainment, an alga’s rents. The viability and persistence of phytoplank-
sinking rate, ws (or its flotation rate, −ws ) must be ton patches in expansive, large-scale systems,
where return currents are extremely remote, ongoing or persistent rapid recuitment of organ-
relate to the comparative rates of recruitment isms from point sources.
within the patch and of the erosion at the patch Many different distributional outcomes can
periphery. Some case studies are presented to be explained by the behaviour and dispersiveness
show some very contrasted large-scale outcomes, of particular species in given systems, although
distinguishing enduring community similarities the subjugated deployment of the same processes
over 800 km in the horizontal from sharply elsewhere may contribute to the formation of
localised patches in systems able to support quite different patterns.
Chapter 3
Photosynthesis and carbon acquisition

in phytoplankton
their energy by respiring (oxidising) the carbo-

3.1 Introduction hydrates and proteins manufactured (reduced)
by photosynthesising plants, the presently per-
The first aim of this chapter is to summarise the ceived realities of aquatic-reductant fluxes may
biochemical basis of photosynthesis in planktic seem quite counter-intuitive. The original pos-
algae and to review the physiological sensitivi- tulate is not in error: it succeeds in describ-
ties of carbon fixation and assimilation under the ing how a section of the trophic relationships
environmental conditions experienced by natu- of the pelagic is conducted. It is just that it is
ral populations of phytoplankton. These funda- far from being the whole story. For instance,
mental aspects of autotrophy are plainly rele- the photosynthetic reduction of carbon is ante-
vant to the dynamics and population ecology of dated by several hundreds of millions of years
individual algal species, functioning within the (≥0.4 Ga: Falkowski, 2002) by the chemosynthe-
constraints set by temperature and by the natu- sis by Archaeans of reduced carbon. This contin-
ral fluxes of light energy and inorganic carbon. ues to be maintained in deep-ocean hydrother-
They are also relevant to the function of entire mal vents, where there is no sunlight and only
pelagic systems as, frequently, they furnish the minimal supplies of organic nutrients (Karl et al.,
major source of energy, in the form of reduced 1980; Jannasch and Mottl, 1985). Even in the
carbon, to heterotrophic consumers. The yields of upper, illuminated waters of lakes and seas, most
fish, birds and mammals in aquatic systems are (perhaps 60–95%) of the organic carbon present
ultimately related to the harvestable and assimil- is not organismic but in solution (Sugimura and
able sources of carbon bonds. In turn, the energy Suzuki, 1988; Wetzel, 1995; Thomas, 1997). A
and resource fluxes through the entire biosphere large proportion of this is humic in character
are greatly influenced by pelagic primary produc- and, thus, supposed to be derived from terres-
ers, impinging on the gaseous composition of the trial soils and ecosystems. True, much of this car-
atmosphere and the heat balance of the whole bon would have been reduced orginally through
planet. terrestrial photosynthesis but the extent of its
Here, we shall be concerned with events at contribution to the assembly of marine biomass
the population, community and ecosystem lev- is still not fully clear. Setting this aside, the
els. However, it is necessary to emphasise at direct phagotrophic transfer of photosynthetic
the start of the chapter that recent advances primary products from phytoplankton to zoo-
in understanding of planetary carbon stores and planktic consumers is not universally achieved
fluxes assist our appreciation of the relative in the pelagic but is, in fact, commonly mediated
global importance of aquatic photosynthesis. To by the activities of free-living microbes. Thus, the
those biologists of my generation brought up dynamic relationships among phytoplankton and
with the exclusive axiom that animals derive their potential phagotrophic consumers acquire
94 PHOTOSYNTHESIS AND CARBON ACQUISITION
a new interpretative significance, which is to be been overtaken by new information. At the same
addressed in this and later chapters. time, it is possible to predict that future progress
The present chapter prepares some of the will concern the biochemical and biophysical
ground necessary to understanding the relation intricacies of control and regulation more than
of planktic photoautotrophy to the dynamics the broad principles of process and order-of-
of phytoplankton populations. After considering magnitude yields, which are generally accepted
the biochemical and physiological basis of photo- by physiological ecologists. Thus, the contempor-
synthetic production, the chapter compares the ary biochemical basis for assessing phytoplankton
various limitations on the assembly of photoau- production will continue to be valid for some
totrophic biomass in natural lakes and seas, and time to come.
it considers the implications for species selection Photosynthesis comprises a series of reac-
and assemblage composition. tions that involve the absorption of light quanta
(photons); the deployment of power to the reduc-
tion of water molecules and the release of
3.2 Essential biochemistry of oxygen; and the capture of the liberated elec-
trons in the synthesis of energy-conserving com-
photosynthesis pounds, which are used ultimately in the Calvin
cycle of carbon-dioxide carboxylation to form
It has been stated or implied several times already hexose (Falkowski and Raven, 1997; Geider and
that the paramount requirement of photoau- MacIntyre, 2002). The aggregate of these reac-
totrophic plankton to prolong residence in, or tions may be summarised:
gain frequent access to, the upper, illuminated
H2 O + CO2 + photons = 1/6[C6 H12 O6 ] + O2 (3.1)
layers of the pelagic is consequential upon the
requirement for light. The need to capture solar As with most summaries, Eq. (3.1) omits not
energy in order to drive photosynthetic carbon merely detail but several important intermedi-
fixation and anabolic growth is no different ate feedback switches, involving carbon, oxy-
from that experienced by any other chlorophyll- gen and reductant, all of which have a bear-
containing photoautotroph inhabiting the suring upon the output products and their physio-
face of the Earth. Indeed, the mechanisms and logical allocation in active phytoplankters. These
ultrastructural provisions for bringing this about are best appreciated against the background
constitutes one of the most universally conserved of the supposed ‘normal pathway’ of photosyn-
processes amongst all photoautotrophic organ- thetic electron transport. The latter was famously
isms. On the other hand, to achieve, within the proposed by Hill and Bendall (1960). Their z-
bounds of an effectively opaque and fluid envi- model of two, linked redox gradients (photo-
ronment, a net excess of energy harvested over systems) has been well substantiated, biochemi-
the energy consumed in metabolism requires cer- cally and ultrastructurally. In the first of these
tain features of photosynthetic production that (perversely, still referred to as photosystem II,
are peculiar to the plankton. Thus, our approach or PSII), electrons are stripped, ultimately from
should be to rehearse the fundamental require- water, and transported to a reductant pool.
ments and sensitivities of photosynthetic produc- In the second (photosystem I, or PSI), photon
tion and then seek to review the aspects of the energy is used to re-elevate the electrochemical
pelagic lifestyle that constrain their adaptation potential sufficiently to transfer electrons to car-
and govern their yields. bon dioxide, through the reduction of nicoti-
Enormous strides in photosynthetic chem- namide adenine dinuceotide phosphate (NADP to
istry have been made, especially over the last NADPH).
30 years or so, especially at the molecular and The (Calvin cycle) carbon reduction is based
submolecular levels (Barber and Anderson, 2002). on the carboxylation reaction. Catalysed by ribu-
This progress is not likely to stop so that, undou- lose 1,5-biphosphate carboxylase (RUBISCO), one
btedly, whatever is written here will have soon molecule each of carbon dioxide, water and
ESSENTIAL BIOCHEMISTRY OF PHOTOSYNTHESIS 95
ribulose 1,5-biphosphate (RuBP) react to yield two 46–48% of the total quantum flux. The corre-
molecules of the initial fixation product, glycer- sponding photon flux density averages 1.77 ×
ate 3-phosphate (G3P). This latter reacts with ATP 1021 m−2 s−1 . Division by the Avogadro num-
and NADPH to form the sugar precursor, glycer- ber (1 mol = 6.023 × 1023 photons) expresses
aldehyde 3-phosphate (GA3P), which now incor- the maximum flux in the more customary
porates the high energy phosphate bond. In the units, einsteins or mols, 2.94 mmol photon
remaining steps of the Calvin cycle, GA3P is fur- m−2 s−1 . The energy of a single photon, έ, varies
ther metabolised, first to triose, then to hexose, with the wavelength,
and RuBP is regenerated.
έ = h c/λ (3.2)
At the molecular level, photosynthetic reactiv-
ity is plainly sensitive to the supply of carbon and where h is Planck’s constant, having the value
water, the photon harvesting and, like all other 6.63 × 10−34 J s (e.g. Kirk, 1994). Photons at the
biochemical processes, to the ambient tempera- red end of the PAR spectrum each contain about
ture. Measurement of photosynthesis may invoke 2.84 × 10−19 J, about 57% of the content of blue-
a yield of fixed carbon, the quantum efficiency light photons (4.97 × 10−19 J).
of its synthesis (yield per photon), or the amount While a given radiation flux of light of a
of oxygen liberated. None of these is any longer single wavelength can be readily expressed in
difficult to quantify but the difficulty is still the J s−1 (and vice versa), precise conversion across
correct interpretation of the bulk results. It is a spectral band does not apply. The approximate
still necessary to consider carefully the regula- relationship proposed by Morel and Smith (1974)
tory role of the ultrastructural and biochemical for the interconversion of solar radiation in the
components that govern the photosynthesis of 400–700 nm band of 2.77 × 1018 quanta s−1 W−1
phytoplankton. Special attention is directed to (equivalent to 3.62 × 10−19 J per photon, or 218 kJ
the issues of photon harvesting, the internal elec- per mol photon) has general applicability (Kirk,
tron transfer, carbon uptake, RUBISCO activity 1994).
and the behaviour of the regulatory safeguards Photosynthesis depends upon the intercep-
that phytoplankters invoke in order to function tion and absorption of photons. Both photosys-
in highly variable environments. tems involve the photosynthetic pigment chloro-
phyll a (and, where applicable, other chloro-
3.2.1 Light harvesting, excitation and phylls), which is characteristically complexed
electron capture with particular proteins, and certain other pig-
Light is the visible part of the spectrum of ments in many instances. These are accom-
electromagnetic radiation emanating from the modated within structures known as light-
sun. Electromagnetic energy occurs in indivisi- harvesting complexes (LHC) and it is these that
ble units, called quanta, that travel along sinu- act as antennae in picking up incoming pho-
soidal trajectories, at a velocity (in air) of c ∼ 3 × tons. For instance, the light-harvesting complex
108 m s−1 . The wavelengths of the quanta define of the eukaryotic photochemical system II (LHCII)
their properties – those with wavelengths (λ) typically comprises some 200–300 chlorophyll
between 400 and 700 nm (400 – 700 × 10−9 m) molecules (mostly of chlorophyll a; up to 30%
correspond with the visible wavelengths we call may be of chlorophyll b), the specific chlorophyll-
light (and within which waveband the quanta are binding proteins and a variable number of xan-
called photons). The waveband of photosynthetically thophyll and carotene molecules, to a combined
active radiation (PAR) coincides almost exactly with molecular mass of 300–400 kDa (Dau, 1994; Gous-
that of light. The white light of the visible spec- sias et al., 2002). The prokaryotic Cyanobacte-
trum is the aggregate of the flux of photons of ria lack chlorophyll b and the light-harvesting
differing wavelengths, ranging from the shorter chlorophyll-proteins of PSII. They rely instead
(blue) to the longer (red) parts of the spectrum. on the phycobiliproteins, assembled in bodies
Relative to the solar constant (see Sec- known as phycobilisomes (Grossman et al., 1993;
tion 2.2.2), the PAR waveband represents some Rüdiger, 1994).
At the heart of the eukaryote LHCII is the chlorophyll-protein complex (known as P700 ) and
antennal chlorophyll-protein known as P680 . It is acceptor (usually denoted A). Again, photons
here that the reactions of PSII are initiated, when excite the equivalent number of P700 electrons to
the complex is exposed to light. The energy of a the point where they can be accepted by A. Next
single photon is sufficient to raise a P680 electron in the electron transfer pathway is ferredoxin,
from its ground-state to its excited-state orbital. transfer of electrons to which reoxidises A− while
Next to the P680 is the phaeophytin acceptor donation of the equivalent number from the plas-
molecule (usually referred to as ‘Phaeo’) and the toquinone pool re-reduces P700 molecules. The
two further acceptor quinones (Q A and Q B ) that electrons may be passed from ferredoxin, and
comprise the PSII reaction centre. In sequence, beyond PSI, to bring about the reduction of NADP
this acceptor chain passes the electrons to PSI. to NADPH that provides power to drive Calvin-
The reaction (P680 → P680 + ) is one of the most cycle carboxylation. The subsequent reactions of
powerful biological oxidations known to science; NADPH with carbon dioxide are not directly
the electrons are readily captured by the Phaeo dependent upon the photon flux and can con-
acceptor. In its now-reduced state, Phaeo− in turn tinue in darkness (see Section 3.2.3).
activates the Q A acceptor: its reduction to Q−A The PSI generation of the carbon-reducing
stimulates acceptance of the electron by Q B . power nevertheless also requires the photosyn-
In this way, the electrons are serially trans- thetic transfer of four electrons per atom of car-
ported towards PSI. Once it has accepted two elec- bon. Under ideal conditions, the light reactions
trons, Q B dissociates to enter a pool of reduced in photosynthesis may be summarised:
plastoquinone (‘PQ’). Molecules of PQH2 are even-
tually oxidised by the cytochrome known as b6 /f, 2NADP + 3ADP + 3P + 2H2 O + 8e
which carries the electrons to PSI. → 2NADPH + 3ATP + 3P + 2H+ + O2 (3.3)
The plastoquinone pool functions as a system
capacitor, like a sort of surge tank of reductant 3.2.2 Photosystem architecture
(D. Walker, 1992; Kolber and Falkowski, 1993), The electron transfer that this equation repre-
whose activity can be viewed in the context of sents is readily facilitated by the physical arrange-
PSII light harvesting. At quiescence, the entire ment of the two main components (PSII, PSI)
reaction centre is said to be ‘open’: P680 is in its and the intercoupling plastoquinone pool, the
reduced state, Phaeo and QA are oxidised. Then, b6 /f cytochrome complex and, in most algae
photon excitation of the P680 initiates a flow of and plants, the soluble electron carrier plasto-
electrons to the plastoquinine pool, whence they cyanin (in Cyanobacteria, cytochrome c may sub-
may be removed as rapidly as PSI can accept stitute). The basic architecture and the location
them. At the same time, the otherwise uncomple- of the biochemical functions of the photosyn-
mented positive charge of excited P+ 680 is balanced thetic units seems to be extremely well conserved
by the stripping of electrons from water (that among eukaryotic algae, plants and their ances-
is, P+
680 is reduced back to P680 ). Note that four tral cyanobacterial lines. The best-known features
photochemical reactions are necessary to gener- were revealed long ago, through light microscopy
ate one dioxygen molecule from two molecules and early transmission electron microscopy. The
of water (2H2 O → 4H+ + 4e + O2 ). It is now granule-like units, comprising LHC antennae and
understood that P+ 680 is actually reduced through the reaction centres, are strung on proteinaceous
the action of manganese ions, via a redox-active membranes, called thylakoids. In the cells of
tyrosine (Barber and Nield, 2002). However, until eukaryotes, stacks of thylakoids are contained
the P+680 molecule is re-reduced, the reaction cen- within one or more separate membrane-bound
tre is unable to accept further electrons and it envelopes, the chromophores (also called plastids
is said to be ‘closed’. It remains so until Q A is or, where they occur in chlorophyte algae and
reoxidised. all higher plants, chloroplasts) whose shape and
The light-harvesting complex and reaction arrangement is often taxon-specific. Cyanobacte-
centre of PSI are built around an analogous ria lack separate chromophores; the thylakoids
Figure 3.1 Diagram of the

configuration of the structure and
the flow of excitation energy
through the photsystems. Electrons
are extracted from water in
photosystem II and transported
through the quinone cycle and
released to photosystem I. Electrons
are accepted by ferredoxin, to bring
about the reduction of NADP to
NADPH that enables the cell to
synthesise its molecular
components. Redrawn, with
permission from Kühlbrandt (2001).
are rather loosely dispersed through the body of and, equally, its reactions to damagingly high
the cell. Apart from anchoring the various trans- light levels. The relevant ultrastructural and bio-
membrane structures (including, in the case of chemical input parameters concern how much
the Cyanobacteria, the phycobilisomes), the thy- light-harvesting capacity there is present in an
lakoid also maintains a regulatory charge gradi- alga and how much reductant it can deliver per
ent, down which the electrons are passed. unit time.
The molecular structure of the energy- The arrangement and linkage of the photo-
harvesting apparatus has become clearer as a systems are schematised in Fig. 3.1. The size of
result of the recent application of electron crys- the LHCII structures studied by Kühlbrandt et al.
tallography. Since Kühlbrandt and Wang (1991) (1994) averaged 13 nm in area and 4.8 nm in thick-
published the three-dimensional structure of ness. The PSII complexes from Synechococcus mea-
a light-harvesting complex, other investigative sured roughly 19 × 10 nm across and 12 nm thick
studies have followed, showing, at increasingly (Zouni et al., 2001). The LHCI complexes from Syne-
fine resolution, the organisation and interlink- chococcus revealed by Jordan et al. (2001) are appar-
ages of the major sub-units of PSII in plants and ently of similar size. On the basis of there being
Cyanobacteria (McDermott et al., 1995; Zouni et 200–300 chlorophyll molecules in a typical LHC,
al., 2001; Barber and Nield, 2002) and also of Reynolds (1997a) calculated that 1 g chlorophyll
PSI (Jordan et al., 2001). The recent overview and could be organised into 2.2 to 3.4 × 1018 LHCs.
model of Fromme et al. (2002) upholds that pro- Because the area that 1 g of chlorophyll subtends
posed by Kühlbrandt et al. (1994) and updates it in the light field can be as great as 20 m2 (see
in several respects. Section 3.3.3), each LHC contributes an average
The cited literature should be consulted for photon absorption of up to 10 × 10−18 m−2 (i.e.
more of the fascinating details of the struc- 10 nm2 ).
tures and organisational patterns of light har- It was also supposed that the photon absorp-
vesting and the electron-transport chain. Here, tion is in inverse proportion to the product of the
we should emphasise the generalised configura- area of the LHC and the aggregate time for the
tion and functional dynamics of the various sub- electron transport chain to accept photons and
units involved in photon absorption and electron clear electrons, ready for the next photon. Kolber
capture, for it is these which impinge upon their and Falkowski (1993) approximated the aggregate
physiological performance and their adaptability time of reactions linking initial excitation (occu-
to operation under sub-ideal conditions. As will pying less than 100 fs, or 10−13 : Knox, 1977) to the
be seen (in Section 3.3), the relevant outputs of re-oxidation of Q−A to be 0.6 ms. The principal rate-
an adequate carbon-reducing capacity relate to limiting step is the onward passage of electrons
system performance under ambient light fluxes, from the plastoquinone pool, which, depending
how it behaves in poor light (low photon fluxes) upon temperature, needed between 2 and 15 ms.
Thus, a single pathway might accommodate up to

66 reactions per second at 0 ◦ C and some 500 s−1
at 30 ◦ C, with a matching carbon-reducing power.
As a rough approximation indicates that, at
30 ◦ C, 1 g chlorophyll containing >2 × 1018 active
LHCs has the capacity to deliver >1021 elec-
trons every second and a theoretical potential to
reduce more than 1.25 × 1020 atoms of carbon
[i.e. > ∼200 µmol C (g chla)−1 s−1 ].
3.2.3 Carbon reduction and allocation

As noted above, the fixation of carbon diox-
ide occurs downstream of the energy capture,
where the reducing power inherent in NADPH
is deployed in the synthesis of carbohydrate.
The flow of reductant drives the Calvin cycle
of RuBP consumption and regeneration, during
which carbon dioxide is drawn in and glucose is
discharged. The cycle is summarised in Fig. 3.2. In Figure 3.2 The Calvin cycle. Carboxylation by RUBISCO
the algae and in many higher plants, RUBISCO- of RuBP at 1 is driven by ATP and NADPH generated by the
mediated carboxylation of RuBP yields the first light reactions of photosynthesis, and results ultimately in the
stable product of so-called C3 photosynthetic car- synthesis of sugar precursors and the renewed availability of
RuBP substrate, thus maintaining the cycle. The cycle is
bon fixation, the 3-carbon glycerate 3-phosphate
regulated at the numbered reactions, where it may be
(G3P). (Note that in this, the process differs from
short-circuited as shown. Abbreviations: DHAP,
those terrestrial C4 fixers that synthesise four- dihydroxyaceton phosphate; E4P, erythrose 4-phosphate; FBP,
carbon malate or aspartate.) fructose 1,5-biphosphate; F6P, fructose 6-phosphate; GA3P,
After the further NADPH-reduction of G3P glyceraldehyde 3-phosphate; GBP, glycerate 1,3-biphosphate;
to glyceraldehyde 3-phosphate (GA3P), the G3P, glycerate 3-phosphate; G6P, glucose 6-phosphate; Pi ,
metabolism proceeds through a series of sugar- inorganic phosphate; RuBP, ribulose 1,5-biphospharte; Ru5P,
phosphate intermediates to yield a hexose ribulose 5-phosphate; R5P, ribose 5-phospate; SBP,
(usually glucose). In this way, one molecule of sedoheptulose 1,7-biphosphate; S7P, sedoheptulose
7-phosphate; Xu5P, xylulose 5-phosphate. Redrawn with
hexose may be exported from the Calvin cycle
permission from Geider and MacIntyre (2002).
for every five of GA3P returned to the cycle of
RuBP regeneration and, ideally, one for every six
molecules of carbon dioxide imported. In this
case of steady-state photosynthesis, the following way, the overall photosynthetic Eq. (3.1) is bal-
equation summarises the mass balance through anced, at the minimal energy cost of eight pho-
the Calvin cycle: tons per atom of carbon fixed. Thus, the theoreti-
cal maximum quantum yield of photosynthesis (ϕ) is
CO2 + 2NADPH + 3ATP + 2H+ 0.125 mol C (mol photon)−1 (D. Walker, 1992).
→ 1/6C6 H12 O6 + H2 O + 2NADP However, neither the cycle nor its fixed-carbon
+ 3ADP + 3P (3.4) yield is immutable but it is subject to devia-
tion and to autoregulation, according to circum-
According to demand, the glucose may be stances. As stated at the outset, these have rever-
respired immediately to fuel the energy demands berations at successive levels of cell growth, com-
of metabolism, or it may be submitted to the munity composition, ecosystem function and the
amination reactions leading to protein synthe- geochemistry of the biosphere. The variability
sis. Excesses may be polymerised into polysac- may owe to imbalances in the light harvest and
charides (glycogen, starch, paramylon). In this carbon capture, or to difficulties in allocating
the carbon fixed. To evaluate these resourcing adequate intracellular carbon supply and upon
impacts requires us to look again at the sensitiv- the RUBISCO capacity or, at least, upon that
ity of the Calvin-cycle reactions, beginning with proportion of RUBISCO capacity that is actually
the initial carboxylation and the action of the ‘active’. To be catalytcally competent, the active
RUBISCO enzyme. site of RUBISCO has also to be carbamylated by
RUBISCO, the catalyst of the CO2 –RuBP con- the binding of a magnesium ion and a non-
junction, is a most highly conserved enzyme, substrate CO2 molecule. Under low light and/or
occurring, with little variation, throughout the low carbon availability, RUBISCO is inactivated
photosynthetic carbon-fixers (Geider and Mac- (decarbamylated), by the reversible action of an
Intyre, 2002). From the bacteria, through the enzyme (appropriately known as RUBISCO inac-
‘red line’ and the ‘green line’ of algae (see tivase), to match the slower rate of RuBP regen-
Section 1.3), to the seed-bearing angiosperms, eration. The resultant down-cycle sequestration
present-day photosynthetic organisms have to of phosphate ions and lower ATP regeneration
contend with acknowledged catalytical weak- brings about an increase in ADP : ATP ratio and,
nesses of RUBISCO. These are due, in part, to thus, a decrease in RUBISCO activity (for fur-
the fact that carbon-dioxide-based photosynthesis ther details of Calvin-cycle self-regulation, refer
evolved under different atmospheric conditions to Geider and MacIntyre, 2002).
from those that presently obtain. In particular, The action of RUBISCO inactivase is itself sen-
the progressive decline in the partial pressure of sitive to the ADP : ATP ratio and to the redox
CO2 exposes the rather weak affinity of RUBISCO state of PSI. Thus, RUBISCO activity responds pos-
for CO2 (Tortell, 2000). According to Raven (1997), itively to a cue of a light-stimulated accelera-
the maximum reported rates of carboxylation (80 tion in photosynthetic electron flow. With con-
mol CO2 (mol RUBISCO)−1 s−1 : Geider and MacIn- ditions of high-light-driven reductant fluxes and
tyre, 2002) are low compared to those mediated high CO2 availability at the sites of carboxylation,
by other carboxylases. Even these levels of activ- the limitation of photosynthetic rate switches to
ity are dependent upon a significant concentra- the rate of RuBP complexation and renewal, both
tion of carbon dioxide at the reaction site (with of which become subject to the overriding con-
reported half-saturation constants of 12–60 µM straint of the RUBISCO capacity (Tortell, 2000).
among eukaryotic algae: Badger et al., 1998). Sup- However, the kinetics of RUBISCO activity impose
posing that the cell-specific rate of carbon fixa- a heavy demand in terms of the delivery of car-
tion could be raised by elevating the amount of bon dioxide to the carboxylation sites. Although
active RUBISCO available, the investment in its many phytoplankters invoke biophysical mecha-
large molecule (∼560 kDa) is relatively expensive. nisms for concentrating carbon dioxide (see Sec-
RUBISCO may account for 1–10% of cell carbon tion 3.4), the relatively high levels needed to satu-
and 2–10% of its protein (Geider and MacIntyre, rate the carboxylation function of RUBISCO may
2002). frequently be overtaken. Circumstances that com-
Having more RUBISCO capacity is not nec- bine low CO2 with the high rates of reductant
essarily helpful either, owing to the susceptibil- and oxygen generation possible in strong light
ity of RUBISCO to oxygen inhibition: at low CO2 are liable to effect the competitive switch to the
concentrations (<10 µM) and high O2 concentra- oxygenase function of RUBISCO and the incep-
tions (>400 µM), RUBISCO functions as an oxi- tion of photorespiration.
dase, in initiating an alternative reaction that Photorespiration is a term introduced in the
leads to the formation of glycerate 3-phosphate physiology of vascular plants to refer to the
and phosphoglycolate. In the steady-state Calvin- sequence of reactions that commence with the
cycle operation, the activity of RUBISCO serves formation of phosphoglycolate from the oxygena-
to maintain the balance between NADPH gen- tion of RuBP by RUBISCO (Osmond, 1981). In the
eration and the output of carbohydrates. For a present context, the term covers the metabolism
given supply of reductant from PSI, the rate of of reductant power and controlling photosynthe-
carbon fixation may be seen to depend upon an sis at low CO2 concentrations. The manufacture
of phosphglycolate carries a significant energetic yields and the energetic efficiency of photosyn-
cost through the altered ATP balance (see Raven thetic carbon fixation. The basic equation (3.1)
et al., 2000), though this is partly recouped in indicates equimolecular exchanges between car-
the continued (albeit smaller) RUBISCO-mediated bon dioxide consumed and oxygen released (the
contribution of G3P to the Calvin cycle. Mean- photosynthetic quotient, PQ, mol O2 evolved/mol
while, the phosphoglycolate is itself dephos- CO2 assimilated, is 1). In fact, both components
phorylated (by phosphoglycolate phosphatase) to are subject to partially independent variation.
form glycolic acid. In the ‘green line’ of algae Oxygen cycling may occur within the photosyn-
(including the prasinophytes, chlorophytes and thetis electron transfer chain (the Mehler reac-
euglenophytes) and higher plants, this glycolate tion), independently of the amount of carbon
can be further oxidised, to glyoxalate and thence delivered through the system. The ‘competition’
to G3P. The full sequence of reactions has been between the carboxylation and oxidation activ-
called the ‘photosynthetic carbon oxidation cycle’ ity of RUBISCO are swayed in favour of oxy-
(PCOC) (Raven, 1997). In the Cyanobacteria and in gen production, photorespiration and glycolate
the ‘red line’ of algae, this capacity seems to be metabolism (Geider and MacIntyre, 2002). The PQ
generally lacking. When experiencing oxidative may move from close to 1.0 in normally photo-
stress at high irradiance levels, these organisms synthesising cells (actually, it is generally mea-
cells will excrete glycolate into the medium. sured to be 1.1 to 1.2: Kirk, 1994) to the range
Excreted glycolate is sufficiently conspicuous 1.2 to 1.8 under high rates of carbon-limited
outside affected cells for its production to have photosynthesis. Low photosynthetic rates under
been studied for many years as a principal ‘extra- high partial pressures of oxygen may force
cellular product’ of phytoplankton photosynthe- PQ < 1 (Burris, 1981).
sis (Fogg, 1971). It is now known that not only The effects on energy efficiency are also sensi-
glycolate but also other photosynthetic interme- tive to biochemical flexibility. Taking glucose as
diates and soluble anabolic products are released an example, the energy stored and released in the
from cells into the medium. This apparent squan- complete oxidation of its molecule is equivalent
dering of costly, autogenic products seemed to to 2.821 kJ mol−1 , or ∼470 kJ per mol carbon syn-
be an unlikely activity in which ‘healthy’ cells thesised. The electron stoichiometry of the syn-
might engage (cf. Sharp, 1977). However, it is thesis cannot be less than 8 mol photon (mol
now appreciated that, far from being a conse- C)−1 but, energetically, the photon efficiency is
quence of ill health, the venting of unusable dis- weaker. The interconversion of Morel and Smith
solved organic carbon (DOC) into the medium (1974; see above; 1 mol photon ∼218 kJ) implies
constitutes a vital aspect of the cell’s homeo- an average investment of the energy of 12.94 pho-
static maintenance (Reynolds, 1997a). It is espe- tons mol−1 . This coincides more closely to the
cially important, for example, when the pro- highest quantum yields determined experimen-
ducer cells are unable to match other growth- tally (0.07–0.09 mol C per mol photon: Bannister
sustaining materials to the synthesis of the carbo- and Weidemann, 1984; D. Walker, 1992).
hydrate base. In natural environments, the DOC Clearly, even these yields are subject to
compounds thus released – glycolate, monosac- the variability in the fate of primary photo-
charides, carboxylic acid, amino acids (Sorokin, synthate. Moreover, the alternative allocations
1999, p. 64; see also Grover and Chrzanowski, of the fixed carbon (whether polymerised and
2000; Søndergaard et al., 2000) – are readily taken stored, respired, allocated to protein synthesis or
up and metabolised by pelagic microorganisms. excreted) need to be borne in mind. It is well
The far-reaching ecological consequences of this accepted that about half the photosynthate in
behaviour are explored in later sections of this actively growing, nutrient-replete cells is invested
book (Sections 3.5.4, 8.2.1). in protein synthesis and in the replication of
So far as the biochemistry of photosynthesis cell material (Li and Platt, 1982; Reynolds et al.,
is concerned, these alternative sinks for primary 1985). However, this proportion is very suscepti-
product make it less easy to be precise about the ble to the physiological stresses experienced by
LIGHT-DEPENDENT ENVIRONMENTAL SENSITIVITY OF PHOTOSYNTHESIS 101
plankters in their natural environments as a con- adopted until it had been described in English
sequence of low light incomes, carbon deficien- (Gaarder and Gran, 1927). Although it was pos-
cies or severe nutrient depletion. These effects sible at that time to estimate carbon dioxide
are explored in subsequent sections. uptake in similar bottles, essentially through the
use of pH-sensitive indicator dyes, the measure-
ment of photosynthetic rate through changes in
3.3 Light-dependent environmental oxygen concentration in light and dark bottles
was soon adopted as a standard method in bio-
sensitivity of photosynthesis logical limnology and oceanography.
In this section, the focus moves towards the 3.3.1 Measurement of light-dependent
physiology of photosynthetic behaviour of phy- photosynthetic oxygen production
toplankton in natural lakes and seas, especially Numerous studies based on oxygen generation
its relationship with underwater light availabil- in light and darkened bottles were published
ity. According to a recently compiled history of in the 50 years between 1935 and 1985. Many
phytoplankton productivity (Barber and Hilting, of the findings were substantially covered in a
2002), quantification of pelagic photosynthesis thorough synthesis by Harris (1978). Since then,
developed through a series of sharp conceptual the method has been displaced by more direct
and (especially) methodological jumps. After a and more sensitive techniques. Nevertheless, the
rapid series of discoveries in the late eighteenth experiments based on measurements of photo-
century, establishing that plants need light and synthetic oxygen production in closed bottles sus-
carbon dioxide to produce oxygen and organic pended at selected depths in the water column
matter from carbon dioxide, there was much yielded consistent generalised results and have
slower progress in estimating the rates and mag- bequeathed to plankton science many of the con-
nitude of the exchanges. This is especially true ceptual aspects and quantitative descriptors of
for aquatic primary production, until the idea productive capacity. The set of sample results
that it could have much bearing on the tropho- illustrated in Fig. 3.3 depicts a typical depth pro-
dynamics of the sea became a matter of serious file of photosynthetic (4-h) exposures of unmodi-
debate (see Section 1.2). It was not until the begin- fied lake plankton in relation to the underwater
ning of the twentieth century, as concerns came light field in an unstratified temperate lake in
to focus increasingly on the rates of reproduction winter (temperature ∼5 ◦ C). The features of gen-
and consumption of planktic food plants, that eral interest include the following:
the pressing need for quantitative measures of
plankton production was identified (Gran, 1912). r The plot of gross photosynthetic oxygen genera-
Building on the techniques and observations of tion is shown as a function of depth in Fig. 3.3a.
Whipple (1899), who had shown a light depen- The curve is fitted by eye. However, it is plain
dence of the growth of phytoplankton in closed that photosynthesis over the 4 hours peaks a lit-
bottles suspended at various depths of water, and tle way beneath the surface, with slower rates
using the Winkler (1888) back-titration method being detected at depth. In this instance, as in
for estimating dissolved oxygen concentration, a large number of other similar experiments,
Gran and colleagues devised a method of mea- there is an apparent depression in photosyn-
suring the photosynthetic evolution of oxygen in thetic rate towards the surface.
sealed bottles of natural phytoplankton, within r The ‘gross photosynthesis’ is the measured
measured time periods. Darkened bottles were aggregate oxygen production at a given depth
set up to provide controls for respirational con- averaged over the exposure period. It is cal-
sumption. They published (in 1918) the results of culated as the observed increase in oxygen
a study of photosynthesis carried out in the Chris- concentration in the ‘light’ bottle (usually a
tiania Fjord (now Oslofjord). According to Barber mean of duplicates) plus the observed decrease
and Hilting (2002), the method was not widely in concentration in the corresponding ‘dark’
r It was neither essential nor common to calcu-

Figure 3.3 Specimen depth distributions of (a) total gross
photosynthetic rate (NP) and total gross respiration rate late the respiratory uptake, as the depletion
(NR); (b) the photosynthetic population (N), in terms of of oxygen concentration in the dark controls.
chlorophyll a; (c) chlorophyll-specific photosynthetic rate (P All that was necessary was the additional mea-
= NP/N) and respiration rate (R = NR/N); (d) underwater surement of the initial oxygen concentration
irradiance (I) in each of three spectral blocks, expressed as a at the start of the experiment. The change in
percentage of the irradiance I 0 obtaining immediately beneath
the dark bottles over the course of the exper-
the surface. P is replotted against I, either (e) as a percentage
iment, normalised to the base period, is the
of I 0 in th green spectral block (peak: 530 nm), or as the
reworked estimate of the intensity of the visible light above equivalent to the respiration of the enclosed
the water (in µmol photon m−2 s−1 ) averaged through the organisms in the dark, NR (here, expressed in
exposure period. Original data of the author, redrawn from mg O2 m−3 h−1 ); R (=NR/N) is thus supposed
Reynolds (1984a). to be the biomass-specific respiration rate (in
mg O2 consumed (mg chla)−1 h−1 ) inserted in
control, it being supposed that the respiration Fig. 3.3c. However, the extrapolation must be
in the dark applies equally to the similar mate- applied cautiously. Whereas NP, as determined,
rial in the light. The mean gross photosyn- is attributable to photosynthesis, the separate
thetic rate at the given depth (here, expressed determination of NR does not exclude respira-
in mg O2 m−3 h−1 ) represents NP, the product tion by non-photosynthetic organisms, includ-
of a biomass-specific rate of photosynthesis (P, ing bacteria and zooplankton. Moreover, it can-
here in mg O2 (mg chla)−1 h−1 ) and the biomass not be assumed that even basal respiration
present (N, shown in Fig 3.3b and expressed as rate of phytoplankton is identical in light and
the amount of chlorophyll a in the enclosed darkness: according to Geider and MacIntyre
plankton, in (mg chla) m−3 ). At this stage, it is (2002), oxygen consumption in photosynthesis-
the biomass specific behaviour that is of first ing microflagellates is 3- to 20- (mean: 7-) fold
interest: P (=NP/N) is plotted in Fig. 3.3c. The greater than dark respiration rate. Accelerated
shape of the curve of P is scarcely distorted metabolism and excretion of photosynthate in
from that of NP in this instance, owing to the starved or stressed phytoplankton may, con-
uniformity of N with depth. Discontinuities in ceivably, account for all the materials fixed in
the depth distribution of N do affect the shape photosynthesis: (P−R) is small but dark R need
of the NP plot (see below and Fig. 3.4). not be large either.
Figure 3.4 Some variations in the

basic form of depth profiles of gross
(NP) and chlorophyll-specific rates
(P), explained in the text.
r The decline in biomass-specific P below the sub- (I430 + I530 + I630 )z /3, and taking the immedi-
surface maximum (Pmax ) is supposed to be a ate subsurface irradiance (I0 ) as the average over
function of the underwater extinction. In the the experiment (in this instance, I0 = 800 µmol
present example, the underwater light attenua- photons m−2 s−1 ).
tion measured photometrically in each of three
spectral bands of visible light (blue, absorption This last plot, Fig. 3.3f, conforms to the for-
peaking at 430 nm; green, peaking at 530 nm; mat of the generic P vs. I curve. It has a steeply
and red, peaking at 630 nm) is included as Fig. rising portion, in which P increases in direct pro-
3.3d. Note that the spectral balance changes portion to Iz ; the slope of this line, generally
with depth (blue light being absorbed faster notated as α (=P/Iz ), is a measure of photosyn-
than red or green light in this case). thetic efficiency at low light intensities. Express-
r Replotting the biomass-specific rates of P ing P in mg O2 (mg chla)−1 h−1 and Iz in mol pho-
against Iz , the level of the residual light pen- tons m−2 h−1 , the slope, α, has the dimensions
etrating to each of the depths of measurement of photosynthetic rate per unit of irradiance, mg
(expressed as a percentage of the surface mea- O2 (mg chla)−1 (mol photon)−1 . As the incident
surement), gives the new curve shown in Fig. photon flux is increased, P becomes increasingly
3.3e. This emphasises the sensitivity of P to Iz , less light dependent and, so, increasingly satu-
at least at low percentage residual light pene- rated by the light available. The irradiance level
tration, and a much more plateau-like feature representing the onset of light saturation is judged
around Pmax . to occur at the point of intersection between the
r Finally, the P vs. I curve is replotted in Fig. extrapolation of the linear light-dependent part
3.3f in terms of the depth-dependent irradiance of the P − Iz curve with the back projection of
across the spectrum, approximated from Iz = Pmax , being the fastest, light-independent rate of
photosynthesis measured. This intensity, known growth and replication. In none of the experi-
as Ik , can be judged by eye or, better, calculated ments considered by Gran (1912), nor those car-
as: ried out half a century later by Lund (1949) or
Reynolds (1973a), is there any question that the
I k = P max /α mol photons m−2 h−1 (3.5)
fastest population growth rates are obtained clos-
It is usual (because that is how the photon flux is est to the water surface. In the late 1970s, some
measured) to give light intensities in µmol pho- helpful experimental evidence was gathered to
tons m−2 s−1 . In terms of being able to explain show that the surface depression was largely an
the shape of the original P vs. depth curve (Fig. artefact of the method and the duration of its
3.3a), it is also useful to be able to define the application. Taking phytoplankton from a mixed
point in the water column, z(I k ) , where the rate water column and holding them steady at super-
of photosynthesis is directly dependent on the saturating light intensities for hours on end rep-
photon flux. The left-hand, light-limited part of resents an enforced ‘shock’ or ‘stress’. In these
the P vs. Iz curve is the most useful for comparing terms, it would not be unreasonable to conclude
the interexperimental differences in algal perfor- that the algae would react and to show signs of
mances. ‘photoinhibition’ (see below). Jewson and Wood
Kirk (1994) described the several attempts that (1975) showed that continuing to circulate the
have been made to find a mathematical expres- algae through the light gradient not only spared
sion that gives a reasonable fit to the observed the algae the symptoms of photinhibition but
relationship of P to I. Jassby and Platt (1976) that the measured Pmax could be sustained. In
tested several different expressions then available an analogous experiment, Marra (1978) showed
against their own data. They found the most suit- that realistically varying the incident radiation
able expressions to be one modified from Smith received by phytoplankton avoided the apparent
(1936) and the one they themselves proposed: photoinhibition. Harris and Piccinin (1977) deter-
mined photosynthetic rates (from oxygen pro-
P = P max [1 − exp(α − I z /P max )] duction, measured with electrodes rather than
(Smith, 1936) (3.6) by titration) in bottled suspensions of Oocystis
P = P max tanh(α − I z /P max ) exposed to high light intensities (>1300 µmol
photons m−2 s−1 ) and temperatures (>20 ◦ C) for
(Jassby & Platt, 1976) (3.7)
varying lengths of time. Their results suggested
In contrast, the significance and mathematical that an elevated photosynthetic rate was main-
treatment of the right-hand part of the P vs. tained for 10 minutes or so but then declined
Iz curve, corresponding to the apparent near- steeply with prolonged exposure. Either the algae
surface depression of photosynthesis in field were photoinhibited or damaged or they had
experiments, was, for a long time, a puzzling reacted to prevent such damage (photoprotec-
feature. It was not necessarily a feature of all tion) in some way that would enable to retain
P vs. z curves: some of the possible variations their vitality to maintain growth (see Section
are exemplified in Fig. 3.4. Anomalies in the 3.3.4).
depth distribution of NP, caused by phytoplank- For these reasons, determinations of depth-
ton abundance (Fig. 3.4a), low water temperature integrated photosynthesis (NP, in mg O2 m−2
(Fig 3.4c) or by surface scumming of dominant h−1 ) need no longer need depend on the plani-
Cyanobacteria (Fig. 3.4d) fail to obscure the inci- metric measurement of the area enclosed by the
dence of subsurface biomass-specific photosyn- measurements of photosynthetic rates against
thetic rates. However, they are not seen when dull depth. They are estimable, for instance, from the
skies ensure that, even at the water surface, pho- P vs. Iz curve in Fig. 3.3f, as the area of a trapez-
tosynthetic rates are not light-saturated [z(Ik) < 0 ium equivalent to Pmax × 2 [(I0 – Ik ) + (I0 − Ic ),
m!]. where I0 is the light intensity at the water sur-
This near-surface depression in measured pho- face and Ic is the intercept of zero photosynthe-
tosynthertic rate is not, however, reflected in sis. In terms of P vs. depth, this simplifies to the
product Pmax × (the depth from the surface to

the point in the water column where the light
will half-saturate it); i.e.
NP ≈ P max × z(0.5I k) (3.8)
If the dark respiration rate, NR, is uniform with

Figure 3.5 Hypothetical P vs. I plots to contrast seasonal
depth, then the integral is simply the product of
variations in temperature on photosynthetic behaviour, with
the full depth range over which it applies (the special reference to changes in Pmax (tagged) and Ik
height of the full water column, H) depth (arrowed). In either plot, the sequence a→b→c is one of
increasing temperature. In the left-hand plot, the dependence
NR ≈ R × H (3.9) upon light is constant; in the right-hand plot, P/I varies to
keep Ik constant. Redrawn from Reynolds (1984a).
Based upon the numerous published records, sev-

eral compendia of the key indices of photosyn- lated by rearranging Eq. (3.5):
thetic oxygen production of phytoplankton in
α = Pmax /Ik = 2.28 mgO2 (mg chla)−1 h−1
closed bottles have been assembled (Harris, 1978;
Kirk, 1994; Padisák, 2003). Clearly, the gross rate /48µmol photons m−2 s−1
of photosynthesis (NPmax ) and the depth integral = 0.0475 mgO2 (mg chla)−1 h−1
(NP) respond to two variables, which are, within (µmol photons m−2 s−1 )−1
limits, highly variable. Other factors notwith- = 13.2 mg O2 (mg chla)−1 (mol photon)−1 m2
standing, observed Pmax should always be the
light-saturated maximum rate at the given tem- Comparisons among differing P vs. I plots
perature. Among the fastest reported examples often differentiate patterns of photosynthetic
are 30–32 mg O2 (mg chla)−1 h−1 (noted in warm behaviour. In the left-hand box of Fig. 3.5, the
tropical lakes in Africa, by Talling, 1965; Talling slopes (α) show similar light dependence of pho-
et al., 1973; Ganf, 1975), whereas the specific photosynthesis at low irradiances but the sequence
tosynthetic rates in temperate lakes rarely exceed of increasing Pmax values could be the response of
the 20 mg O2 (mg chla)−1 h−1 found by Bind- the same alga to increasing temperatures. In the
loss (1974). Thus, as might be expected, examples right-hand plot, data for three different algae are
of high community rates of oxygen production shown, all saturating at similar levels but with
come from warm lakes supporting large popu- differing photosynthetic efficiencies. A higher α
lations of phytoplankton algal chlorophyll and enables a faster rate of photosynthesis to be main-
when there are also ample reserves of exploitable tained when light intensity is low.
CO2 : integrals in the range 6–18 g O2 m−3 h−1
have been noted in Lac Tchad, Chad (Lévêque 3.3.2 Measurement of light-dependent
et al., 1972), Red Rock Tarn, Australia (Hammer photosynthetic carbon fixation
et al., 1973) and Lake George, Uganda (Ganf, 1975). On the basis of an equimolecular photosynthetic
The onset of light saturation, Ik , is also tempera- quotient (PQ ∼ 1), the light efficiency of photo-
ture influenced but is generally <300 µmol pho- synthesis could be calculated from the oxygen
tons m−2 s−1 . There are many citations of much evolution data to be ∼5.0 mg C (mg chla)−1 (mol
lower Ik determinations, 15–50 µmol photons photon)−1 m2 . This is actually below the average
m−2 s−1 , generally at temperatures <10 ◦ C (Kirk, of a large number of extrapolated values, which
1994). The most consistent values seem likely to fall mainly in the range 6–18 mg C (mg chla)−1
relate to the slope, α, which, at low light levels, is (mol photon)−1 m2 (Harris, 1978). However, for
less dependent on temperature and more depen- over 50 years now, it has been possible to work
dent upon light-harvesting efficiency. Taking the directly in the currency of carbon, applying the
plot in Fig. 3.3f, the slope α (i.e. P on I) is calcu- so-called 14 C method of Steemann Nielsen (1952)
Since its introduction, the method has been

Table 3.1 Temperature-sensitive characteristics of
light-dependent carbon fixation. For reported data on
improved in detail (the liquid-scintillation and
maximum photosynthetic rates ( Pmax ) and on the gas-phase counting technique is nowadays pre-
onset of light saturation ( Ik ), the majority of ferred) but, in essence, the original method
observations fall within the ranges shown in the has survived intact (Søndergaard, 2002). Provid-
brackets; extreme values are shown outside the ing proper licensing and handling protocols are
brackets. Under conditions of light limitation followed meticulously, the method is easy to
( Iz < Ik ), temperature dependence of photosynthesis is apply and yields reproducible results. Compar-
weaker and the single range of photosynthetic
isons with simultaneous measurements of oxy-
efficiencies applies to the available data
gen evolution normally give tolerable agreement,
Pmax if allowance is made for a PQ of ∼1.15 (Kirk, 1994).
in mg C (mg chla)−1 h−1 A large number of results have been published
θ 2–5 ◦ C 0.5 (1.0–2.6) 3.0 in the literature and these have been the sub-
θ 17–20 ◦ C 2.5 (3.3–8.6) 9.7 ject of a series of syntheses (including Harris,
θ 27–30 ◦ C 7–20 1978, 1986; Fogg and Thake, 1987; Kirk, 1994;
Ik Padisák, 2003). It is sufficient in the present con-
in µmol photon m−2 s−1 text simply to summarise the key characteristics
θ 2–5 ◦ C 17–30 that have been reported (maximum measured
θ 17–20 ◦ C 20 (90–150) 320 chlorophyll-specific rates of light-saturated car-
θ 27–30 ◦ C 60 (180–250) 360 bon incorporation and the chlorophyll-specific
α photosynthetic efficiencies under sub-saturating
in mg C (mg chla)−1 2 (6–18) 37 light intensities (see Table 3.1).
(mol photon)−1 Two comments are important to make, how-
m2 ever. One is the positive linkage of light satura-
tion with temperature, at least within the range
Source: Generalised values synthesised from the of the majority of observations – between 5 and
literature (Harris, 1978; Reynolds, 1984a, 1990, 25 ◦ C. As with many other cellular processes,
1994a; Padisák 2003). activity increases with increasing temperatures,
up to maximal levels varying between 25 and
40 ◦ C. There is a plain dependence for the photo-
for measuring the photosynthetic incorporation synthetic rates to accelerate with higher temper-
of carbon dioxide labelled with the radioactive atures and, so, for there to be a higher threshold
isotope. Still using darkened and undarkened bot- flux of photons necessary to saturate it. With no
tles suspended in the light field, the essential change in the strongly light-limited rates of pho-
principle is that the natural carbon source is aug- tosynthesis, α may vary little (i.e. light-limited
mented by a dose of radio-labelled NaH14 CO3 in photosynthesis is not temperature-constrained;
solution, which carbon source is exploited and left-hand plot in Fig. 3.5). Thus, Ik increases with
taken up and fixed into the photosynthetic algae. temperature, in broadly similar proportion to the
At the end of the exposure, the flask contents are increase in Pmax .
filtered and the residues are submitted to Geiger As a function of customary temperature, Pmax
counting (for a recent methodological guide, see increases non-linearly, roughly doubling with
Howarth and Michaels, 2000), and the quantity each 10 ◦ C rise in temperature. This multiple,
thus assimilated is calculated. It is supposed that formalised as the Q10 factor, is now used infre-
14
C will be fixed in photosynthesis in the same quently as a physiological index: preference is
proportion to 12 C that is available in the pool at now accorded to the slope of reactivity on
the start of the exposure. Then, the Arrhenius scale, which expresses absolute
temperature (in kelvins) as a reciprocal scale,
12
CO2 uptake/12 CO2 available 1/(temperature in K). As an example, the mea-
= 14
CO2 uptake/14 CO2 available (3.10) sured temperature sensitivity of light-saturated
ing by Steemann Nielsen himself) and, perplex-

ingly, these persist to the present day. The most
important concerns the metabolic exchanges and
cycling of carbon, in which the labelled carbon
participates relatively freely. At first, labelled car-
bon moves in only one direction, from solution to
photosynthate; it is a manifestation of gross pho-
tosynthesis. As the experiment proceeds, some of
the 14 C-labelled carbohydrate may be assimilated
but it may just as easily be used in basal respira-
tion, or it may well be subject to photorespiration
or excretion (see Section 3.2.3). This means that,
as the incubation proceeds, the method is osten-
sibly measuring something closer to net photo-
synthesis. Long incubations may determine only
Figure 3.6 Mean saturated chlorophyll-specific net photosynthetic 14 C incorporation (Steemann
photosynthetic and dark respiration rates of cultured Nielsen, 1955; Dring and Jewson, 1979). Compar-
Asterionella cells in controlled bench-top exposures. Original ing net 14 C assimilation with net oxygen pro-
data of Reynolds (1984a). duction over 24-h incubations, by which time
respired 14 CO2 is being refixed, takes PQ closer
to 1.4 (see Marra, 2002).
photosynthesis and dark respiration of labora- The switches towards this more balanced
tory isolates of Asterionella formosa are shown in state of exchanges will be approached at dif-
Fig. 3.6. The normalised, Arrhenius coefficient is ferent rates, depending upon temperature and
−18.88 × 10−3 per reciprocal kelvin. In the more the irradiance to which the incubating mate-
familiar, if incorrect, terms, Q10 is 2.18. Almost rial is exposed and on the physiological condi-
all quoted values, applying both to named species tion of the alga at the outset. The behaviour has
and to phytoplankton in general, fall in the range been expressed through various descriptive equa-
1.8–2.25 (Eppley, 1972; Harris, 1978). Some variations (notably those formulated by Hobson et al.,
tion is probable, not least because photosynthe- 1976; Dring and Jewson, 1982; Marra et al., 1988;
sis is a complex of many individual reactions. Williams and Lefevre, 1996). A probabilistic out-
However, its maximum, light-saturated rate is pri- come is that the largest proportion of the gross
marily a function of temperature (Morel, 1991), uptake of 14 C is assimilated into new protein and
even though the several descriptive equations fit- biomass in healthy cells when they are simulta-
ted to experimental data differ mutually (Eppley, neously exposed to sub-saturating light intensi-
1972; Megard, 1972) and have been shown to ties. Conversely, with approaching light satura-
be underestimates against the maximum growth tion of the growth-assimilatory demand for photo-
rates that have been observed (Brush et al., 2002). synthate (recognising that this may be constrai-
Whereas photon capture has a Q10 close to 2 ned by factors other than the supply of photo-
(see Section 3.2.1), protein assembly and inter- synthate), then more of the excess is vented
nal transport have greater temperature sensitiv- or metabolised in other ways. Thus, the ratio
ity (Tamiya et al., 1953; see also Section 5.3.2 for of net photosynthetic carbon fixation to gross
the influence of temperature on growth). The Q10 photosynthetic carbon fixation (Pn : Pg ) is in the
of steady, dark respiration rates of healthy phyto- proportion of the fraction of the gross photo-
plankters is similarly close to 2. synthate that can be assimilated, i.e. (Pg − R)/Pg
The second comment is that all these deduc- (Marra, 2002). Here, R may represent not just
tions are subject to uncertainties about the preci- the basal, autotrophic respiration (Ra ) but, in
sion of the radiocarbon method. Interpretational addition, all metabolic elimination of excess
difficulties were recognised early on (includ- photosynthate (Rh ). Even under the optimal
conditions envisaged, Ra never disappears but is where photosynthesis is entirely compensated by

always finite, being, at least, about 4% of Pmax (or respirational loss. This theme is continued in
the maximum sustainable Pg at the same temper- Section 3.3.4.
ature), and typically 7–10% (Talling, 1957b, 1971;
Reynolds, 1984a). Reynolds’ (1997a) best predictor Characterising the photosynthetic impact of
of basal respiration in a number of named fresh- the underwater light intensity
water phytoplankters at 20 ◦ C is related to the Taking first the condition of phytoplankton at a
surface-to-volume ratio [Ra20 = 0.079 (s/v)0.325 ]. On fixed or relatively stable depth in the water col-
the other hand, the sum of physiological losses umn, the light energy available to them will, in
in light-saturated and/or nutrient-deficient cells, any case, fluctuate on a diel cycle, in step with
including that vented as DOC, can be extremely daytime changes in solar radiation, and also on a
high, approaching 100% of the fixation rate: the less predictable basis brought about by changes
quotient (Pg − R − Rh )/Pg and the ratio Pn : Pg both in cloud cover and changes in surface reflectance
fall toward zero. owing to surface wind action. Besides variance
This is a plausible way of explaining diffi- in I0 , the instantaneous incident solar flux of
culties experienced in accounting for the fate PAR on the water surface, there is variance in
of the carbon fixed in photosynthesis (Talling, the irradiance in the flux passing to just beneath
1984; Tilzer, 1984; Reynolds et al., 1985) and the air–water interface, I 0 . Reflectance of incom-
the sometimes very large gaps between primary ing light is least (about 5%) at high angles of
carbon fixation and net biomass accumulation incidence but the proportion reflected increases
(Jassby and Goldman, 1974a; Forsberg, 1985). steeply at lower angles of incidence, especially
Thus, beyond gaining a broad perspective on the <30o . Wind-induced surface waves modify the
constraints acting on chlorophyll-specific photo- reflectance, both reducing and amplifying pene-
synthetic rates, it is necessary also for the physi- tration into the water at the scale of milliseconds.
ological ecologist to grasp the manner in which Near sunrise and sunset, waves are important to
the environmental conditions mould the deploy- maintaining a flux of light across the water sur-
ment of fixed carbon into the population dynam- face (Larkum and Barrett, 1983).
ics of phytoplankton. Variability in the light income to a water
body is represented in the examples in Fig. 3.7.
3.3.3 Photosynthetic production at Measured daily integrals of the solar insolation
sub-saturating light intensities input across the surface of Esthwaite Water, UK
In this section, we focus on the adaptive mech- (a temperate-region lake, experiencing a predom-
anisms which phytoplankters use to optimise inantly oceanic climate) through a period of just
their carbon fixation under irradiance fluxes that over one year are shown in the main plot. Besides
markedly undersaturate the capacity of the indi- showing a 100-fold variation between the high-
vidual light-harvesting centres. To do this, it is est (56.4 mol photons m−2 d−1 ) and lowest light
necessary to relate the light-harvesting capacity inputs (0.5 mol photons m−2 d−1 ), the plot reveals
to the cell rather than to chlorophyll per se, as we that part of this owes to the substantial annual
invoke the number of centres in existence within fluctuation in the maximum possible insola-
the cell and their intracellular distribution as tion (based on the solar constant, its latitudinal
adaptive variables, as well as the role played by correction, the deduction of non-PAR and with
accessory pigments. It is also necessary to adopt allowance for albedo scatter in the atmosphere)
a more quantitative appreciation of the diminu- due to the location of the lake (54 o N). Superim-
tion of harvestable light as a function of increas- posed upon that are the near chaotic fluctuations
ing water depth and the contribution to vertical that are due to day-to-day variations in the extent
light attenuation that the plankton makes itself. and thickness of cloud cover and which cut the
Finally, water movements entrain and transport light income by anything between 2% or as much
phytoplankton through part of this gradient, fre- as 94% of the theoretical maximum (Davis et al.,
quently mixing them to depths beyond the point 2003). The inset in Fig. 3.7 presents two day-long
Figure 3.7 Main plot: Measured

daily insolation at the surface of
Blelham Tarn (August 1999 to
October 2000), compared to the
theoretical maximum insolation, as
calculated from the latitude and
assuming a clear, dry atmosphere
(redrawn with permission from the
data of Davis et al., 2003). The insets
show the diel course of solar
radiation intensity, measured at the
same location on an overcast
January day and a cloudless day close
to the summer solstice. Based on
data presented originally by Talling
(1973) and redrawn with permission
time courses of insolation, measured at the same or

latitude by Talling (1971). Despite coming from
quite different data sets collected at quite differ- ε = (ln I 0 − ln I z )/ h(0 − z) (3.12)
ent times, these two extremes amplify the vari-
ation shown in the main plot of Fig. 3.7. The where h(0 − z) is the vertical distance from the
first corresponds to a rainy, overcast winter day surface to depth z.
on which the aggregate insolation is below 1 From Fig. 3.3d, ε(0.5 m to 1.5 m) is solved from
mol photons m−2 d−1 ; the second is a cloud- (ln I0.5 − ln I1.5 ) as ε = 1.516 m−1 . We may now
less near-solstice day when the maximum instan- apply this spectral integral of the attenuating
taneous insolation reached over 1.6 mmol pho- light to characterise the daytime changes in the
tons m−2 s−1 . underwater light field (see Fig. 3.8). The plot in
Downwelling radiation is subject to absorp- Fig 3.8a shows the reconstructed time course of
tion and scatter beneath the water surface, where the irradiance at the top of the water column
it is more or less attenuated steeply and expo- of Crose Mere on 25 February 1971. Beneath it
nentially with depth, as shown in Fig. 3.3d. This are marked the values of Ik and 0.5 Ik derived
attenuation can usefully be expressed by the coef- in the original experiment (Fig. 3.3). For clarity,
ficient of vertical light extinction, ε. On the basis the same information is plotted on to Fig. 3.8b,
of the spectral integration used to translate the now against a natural logarithmic scale. Assum-
P vs. depth to P vs. I, the coefficient may be esti- ing no change in any component save the incom-
mated on the basis of the equation ing radiation through the day, Fig. 3.8c repre-
sents the diurnal time track between sunrise and
I z = I 0 · e −εz (3.11) sunset of the depths of Ik and 0.5 Ik . Other fac-
tors being equal (including the coefficient of ver-
where e is the natural logarithmic base, and tical extinction of light, ε), the depth at which
whence chlorophyll-specific photosynthesis can be satu-
rated increases to a maximum at around the diur-
ε = −ln(I z /I 0 )/ h(0 − z) nal solar zenith, as a function of the insolation
Figure 3.9 Depth-integrated community photosynthetic

rates (NP) for selected times through the day (07.00, 09.00,
etc.) predicted interpolated values of (I 0 ) and the time track
Figure 3.8 (a) Hypothetical plot of the time-course of
of the depth of Ik developed in Fig. 3.8. Redrawn from
immediate subsurface irradiance intensity (I 0 ) on 25 February
Reynolds (1984a).
1971 (the date of the measurements presented in Fig. 3.3);
(b) the same shown on a semilogarithmic plot. In (a) and (b),
the contemporaneous determinations of Ik and 0.5 Ik are
inserted. In (c), the time courses of the water depths reached
(NP, in mg C m−2 d−1 ). A. E. Walsby and
by irradiance intensities respectively equivalent to Ik and 0.5 colleagues have been particularly successful in
Ik are plotted SR, sunrise; SS, sunset. Redrawn from Reynolds developing this approach (using Microsoft Excel 7
(1984a). software), details of which they make available
on the World Wide Web (Walsby, 2001; see also
Bright and Walsby, 2000; Davis et al., 2003). The
(I 0 ). Outside the Ik perimeter, chlorophyll-specific technique is helpful in establishing the envir-
fixation rates are light-limited, as predicted by onmental requirements and limitations on algal
the contemporaneous P vs. I curve. For instance, growth.
extrapolation of relevant data from the experi-
ment shown in Figure 3.3 allows us to fit selected The impact of optical properties of water on
reconstructions of P vs. z curves applying at vari- the underwater light spectrum
ous times of the day (see Fig. 3.9). As Kirk’s manual (1994) and several other of his
In reality, matters are more complex than contributions (see, for instance, Kirk, 2003) have
that, especially if cloud cover (and, hence, the powerfully emphasised, the attraction and use-
insolation, I 0 ), or the extinction coefficient is fulness of a single average vertical attenuation
altered during the course of the day. The greatest coefficient (ε) remains an approximation of the
depth of Ik need not be maximal in the middle complexities of the underwater dissipation of
of the day. Modern in-situ recording and teleme- light energy. First, as already established, light is
try of continuous radiation make it possible to not absorbed equally across the visible spectrum,
track minute-to-minute variability in the under- even in pure water (see Fig. 3.10). Photons travel-
water light conditions. Moreover, it is now rel- ling with wavelengths of about 400–480 nm are
atively simple to translate light measurements least likely to be captured by water molecules;
to instantaneous photosynthetic rates, from the those with wavelengths closer to 700 nm are 30
P vs. I curve, and to integrate them through times more likely to be absorbed. For this reason,
depth (NP, in mg C m−2 h−1 ) and through time water cannot be regarded as being colourless.
waters of high clarity, a second source of error

is encountered. This relates to another funda-
mental of attenuation: besides being absorbed
and scattered, according to the properties of the
water, attenuation is modified perceptibly by the
angular distribution of the light, which becomes
increasingly diffuse with increasing depth. Kirk
(2003) proposed the use of an irradiance weight-
ing of the wavelength-specific gradient, by inte-
grating light measurements over the entire water
column, so that
0 0
w
εav = εz I z dz I z dz (3.13)
∞ ∞
where ε av is the weighted average attenuation

w
coefficient and εz and Iz are the attenuation coef-

ficient and residual light values at each depth
increment.
The use of this integral improves resolution of
the subsaturating light levels in clear waters but
the simpler, less precise attenuation coefficient
derived in Eq. (3.12) remains adequate for describ-
ing the underwater light field in most lakes and
many coastal waters, where there may be more
Figure 3.10 The absorption of visible light (375–725 nm) humic material in solution and there may be
by pure water. Drawn from data in Kirk (1994).
more particulate material in suspension. More-
over, greater concentrations of algal chlorophyll
The selective absorption in the red leaves oceanic also contribute to the rapid relative attenuation
water of high clarity distinctly blue-green in of light with depth. The average attenuation coef-
colour. This effect is strongly evident at increas- ficient comprises:
ing water depths beneath the surface, where the
ε = εw + ε p + N εa (3.14)
most penetrative wavelengths come increasingly
to dominate the diminishing light field. In lakes, εw is the attenuation coefficient due to the
there is a tendency towards higher solute con- water. As already indicated, there is no unique
centrations, including, significantly, of plant or and meaningful value that applies across the vis-
humic derivatives. These absorb wavelengths in ible spectrum. The minimum absorption in pure
the blue end of the spectrum, to leave a yellow or water, equivalent to 0.0145 m−1 , occurs at a wave-
brownish tinge to the water, as the older names length of ∼440 nm. The clearest (least absorbing)
(‘Gelbstoff’, ‘gilvin’) might imply. Under these cir- natural waters are in the open ocean (Sargasso,
cumstances, the averaged extinction coefficient Gulf Stream off Bahamas), where the coefficients
(distinguished here as ε av ) becomes less steep of vertical attenuation at 440 nm (ε w 440 ) of ambi-
with depth and cannot strictly be normalised just ent visible insolation (I 0 ) are <0.01 m−1 (various
by logarithmic expression. ε av or ε as used gener- sources tabulated in Kirk, 1994). Values quoted
ally, calculated as in Eq. (3.12) from the slope of for the open Atlantic, Indian and Pacific Oceans
Iz on z, ε av really has only a local value, applying range between 0.02 and 0.05 m−1 . Among the
to relatively restricted depth bands. clearest lakes for which data are to hand, Crater
To express attenuation of light of a given Lake, Oregon has an exceptionally low coefficient
wavelength or within a narrow waveband over- of incident light attenuation (ε ∼ 0.06 m−1 : Tyler,
comes part of the difficulty but, especially in in Kirk, 1994). High-altitude lakes in the Andes
range return values of 0.12 to 0.2 m−1 (Reynolds,

1987b, and unpublished observations). Elsewhere,
the effects of solutes contribute to higher average
extinction coefficients, in coastal seas (ε w > 0.15
m−1 ) and fresh waters generally (ε w > 0.2 m−1 ). At
the other extreme, coloration due to humic stain-
ing can be intense; Kirk (1994) tabulated some
representative values from estuaries and coastal
seas receiving drainage from extensive peatland
catchments (Gulf of Bothnia, Baltic Sea; Clyde,
Scotland) of 0.4 to 0.65 m−1 , from some humic
lakes and reservoirs in Australia (1 to 3.5 m−1 )
and from peat-bog ponds and streams in Ireland
(2 to 20 m−1 ).
εp is the attenuation coefficient due to
the solid particulates in suspension in the
water. The particles may emanate from eroded
soils or resuspended silts or biogenic detri-
tus. Fine precipitates, for instance, of calcium
Figure 3.11 The chlorophyll-specific area of light
carbonate, also contribute to particulate tur-
interception by algal cells as a function of size and shape
bidity. The most persistent are clay particles (•, spherical cells; , non-spherical cells). The straight line
(typically <5 µm), eroded from unconsolidated corresponds to a diminishing efficiency of light interception
deposits, which, by absorbing and backscatter- by cells of increasing size but very small cells also lose
ing the photon flux, can impart high turbid- efficiency at sizes close to the wavelength of light. Redrawn
ity. In many lakes and seas, the contribution with permission from Reynolds (1987b).
of background turbidity to attenuation may be
trivial (<0.05 m−1 ) but Kirk (1994) cites many
examples where it is anything but trivial; canals, part, due to the size and shape of the algae (see
rivers, meltwater streams flowing into lakes or Fig. 3.11). Among microplankters, larger quasi-
reservoirs can impart attenuation coefficients of spherical shapes, having larger volumes, hold
1 to 4 m−1 , with extreme values in the range absolutely greater amounts of chlorophyll than
10–20 m−1 . Investigating the turbidity of the smaller shapes. Relative to the cross-section pre-
tidally mixed estuary of the Severn, UK, Reynolds sented to the light field, however, the chloro-
and West (unpublished data, 1988) found a cor- phyll is more compactly arranged in larger units
relation between εp and the mass of suspended and more light passes between them than in
clay and fine silt, such that an attenuation coef- smaller units carrying the same amount of
ficient of 20 m−1 corresponds approximately to 1 chlorophyll. This approximates to the expecta-
kg suspended clay m−3 , whence εp ∼ 20 m2 kg−1 . tions of the ‘packet effect’ (or ‘sieve effect’),
Nεa is the attenuation due to phytoplankton. first studied by Duysens (1956) in the context
Evolved to be able to intercept light energy, phy- of solutions and suspensions but developed for
toplankton can be, in aggregate, the main com- phytoplankton by Kirk (1975a, b, 1976). How-
ponent to vertical attenuation. The extent is pro- ever, among the free-living nanoplanktic and
portional to the mass of phytoplankton present picoplanktic cells (<100 µm3 ), the chlorophyll-
per unit volume (N, in mg chla m−3 ); however, specific area of projection is diminished, to as
the chlorophyll-specific attenuation coefficient, low as 0.004 m2 (mg chla)−1 . This may have some-
ε a , varies considerably. Most general accounts thing to do with the size of plastids being sim-
attribute attenuation coefficients of between 0.01 ilar to the wavelength of light. Alternatively, it
to 0.02 m−1 (mg chla m−3 )−1 , or, more simply, might be explained by the second anticipated
0.01 to 0.02 m2 (mg chla)−1 . Variations are, in source of variation, the chlorophyll content of
components may well dominate over the others.

In Fig. 3.13, instances of high attenuation due to
a relatively high ε w (in the humic Mt Bold Reser-
voir, Australia), εp (in the P. K. le Roux Reservoir,
South Africa) and Nε a (in an Ethiopian soda lake)
are compared with the transparency of a moun-
tain lake (2800 m a.s.l.) in the middle Andes. In
each of these relatively deep-water examples, the
gradient in Iz is exclusively related to the depth
and to the coefficient of attenuation. The same
principles apply in shallow waters but, because
light can reach the bottom and be reflected back
into the water column, the gradient of Iz may
not be so steep or so smooth as that shown in
Fig. 3.13 (discussion in Ackleson, 2003). Neverthe-
less, the present examples give a feel for the col-
umn depth in which diurnal photosynthesis can
Figure 3.12 Carbon-specific area of freshwater planktic
be light-saturated (Iz > Ik > (say) 0.1 mmol pho-
algae (ka ) plotted against the dimensionless shape index,
msv−1 . Near-spherical algae line up close to msv−1 = 6,
tons m−2 s−1 ) and, more to the point, the vertical
smaller species tending to have greater interception extent of the water column where phytoplank-
properties than larger ones. Shape distortions increase msv−1 ters need to adapt their light-harvesting potential
without sacrifice of ka . The algae are Ana, Anabaena to be able to match their capabilities for carbon
flos-aquae; Aphan, Aphanizomenon flos-aquae; Ast, Asterionella fixation.
formosa; Chla, Chlamydomonas; Chlo, Chlorella; Cry, Cryptomonas
ovata; Eud, Eudorina unicocca; Fra, Fragilaria crotonensis; Lim red,
Limnothrix redekei; Mic, Microcystis aeruginosa; Monod, Monodus;
Phytoplankton adaptations to sub-saturating
Monor, Monoraphidium contortum; Per, Peridinium cinctum; Pla
irradiances
ag, Planktothrix agardhii; Plg, Plagioselmis; Sc q, Scenedesmus
quadricauda; Tab, Tabellaria flocculosa. Redrawn with Several mechanisms exist for enhancing cell-
permission from Reynolds (1997a). specific photosynthetic potential at low levels of
irradiance. One of these is simply to increase the
cell-specific light-harvesting capacity, by adding
the cells, relative to (say) cell carbon. Plotting ka , to the number of LHCs in individual cells. This
the area projected per mol of cell carbon, against may be manifest at the anatomical level in the
an index of shape (msv−1 is the product of the synthesis of more chlorophyll and deploying it in
maximum dimension and the surface-to-volume more (or more extensive) plastids placed to inter-
ratio), shows the package effect to be upheld for cept more of the available photon flux falling on
quasi-spherical units (from Chlorella to Microcystis; the cell. Most phytoplankton are able to adjust
msv−1 = [d4π(d/2)2 /4/3π(d/2)3 ] = 6), while distor- their chlorophyll content within a range of ±50%
tion from spherical form usually enhances the of average and to do so within the timescale
area that the equivalent sphere might projec- of one or two cell generations. For example,
tion (Reynolds, 1993a) (see also Fig. 3.12). Note Reynolds (1984a) made reference to measure-
that the individual carbon-based projections are ments of the chlorophyll content of Asterionella
less liable to variation than the chlorophyll-based formosa cells taken at various stages of the sea-
derivations. sonal growth cycle in a natural lake, during
The attenuation components self-compound which the cell-specific quota fluctuated between
(Eq. 3.14) to influence the diminution of the 1.3 and 2.3 mg chla (109 cells)−1 , i.e. 1.3–2.3 pg
depth to which incident radiation of given wave- chla (cell)−1 . Supposing the capacity of the Aster-
length penetrates (Eq. 3.11). In some extreme ionella to fix carbon to have been as measured in
instances of attenuation with depth, one of the Fig. 3.3, equivalent to ∼0.8 mg C (mg chla)−1 h−1 ,
Figure 3.13 Some examples of

light penetration and (inset)
components of absorption due to
water and solutes (unshaded),
non-living particulates (hatched) and
phytoplankton (solid). Laguna Negra,
Chile (a), is a very clear mountain
lake; (b) Mount Bold Reservoir,
Australia, is a significantly coloured
water; (c) P. K. le Roux Reservoir,
South Africa is rich in suspened clay;
(d) Lake Kilotes, Ethiopia is a
shallow, fertile soda lake, supporting
dense populations of Spirulina.
Various sources; redrawn with
permission from Reynolds (1987b).
then the cells with the lower chlorophyll a com- mation for cells exposed to other than very low
plement would have fixed 1.04 mg C (109 cells)−1 light intensities.
h−1 , or 0.012 mg C (mg cell C) h−1 . Other things It appears that, over successive generations,
being equal, the cells with the higher chloro- phytoplankton vary the amount of chlorophyll,
phyll complement might have been capable of both upwards (in response to poor photon fluxes)
fixing 1.84 mg C (109 cells)−1 h−1 , or 0.022 mg C and downwards (when similar cells of the same
(mg cell C) h−1 . The main point, however, is that species are exposed to saturating light fluxes).
the cells with the higher chlorophyll content are This represents an ability to optimise the alloca-
capable of fixing the same amount of carbon as tion of cellular resources in response to the par-
those with the lower complement but at a lower ticular internal rate-limiting function, bearing in
photon flux density: in this instance, the high- mind that the synthesis and maintenance of the
chlorophyll cells achieve 1.04 mg C (109 cells)−1 light-harvesting apparatus carries a significant
h−1 not at ≥48 µmol photons m−2 s−1 but at ≥27 energetic cost and no more of it will be sponsored
µmol photons m−2 s−1 . The extra chlorophyll a than a given steady insolation state may require
increases the steepness of biomass-specific P on I (Raven, 1984). Moreover, under persistently low
(the slope α). average irradiances, there is plainly a limit to
The measurements of biomass-specific chloro- the extra light-harvesting capacity that can be
phyll a referred to in Section 1.5.4 range over installed before the returns in cell-specific pho-
an order of magnitude, between 0.0015 and ton capture diminish to zero. If all the photons
0.0197 pg µm−3 of live cell volume. This corre- falling on the cell are being intercepted, more
sponds approximately to 3 to 39 mg g−1 of dry harvesting centres will not improve the energy
weight (0.3 to 3.9%) or, relative to cell carbon, 6.5 income. On the other hand, this logic indicates
to 87 mg chla (g C)−1 . Many of the lowest val- the advantage (preadaptation?) of having a rel-
ues come from marine phytoplankters in culture atively large carbon-specific area of projection.
(Cloern et al., 1995); most of the highest come Most of the species indicated towards the top
from cultured or natural material, but grown of Fig. 3.12 are able to operate under relatively
under persistent low light intensities (Reynolds, low photon fluxes, in part, through enhance-
1992a, 1997a). The data show that the frequently ment of the chlorophyll deployment across the
adopted ratio of cell carbon to chlorophyll con- light field available. Many of the slender or fil-
tent (50:1, or 20 mg chla (gC)−1 ] must be applied amentous diatoms, as well as the solitary fila-
with caution, though it remains a good approxi- mentous Cyanobacteria (Limnothrix, Planktothrix),
which project perhaps 10–30 m2 (mol cell C)−1 , of the alga as much as the independence that
perform well in this respect. The quasi-spherical was won.
colonies of Microcystis show the opposite extreme Reynolds et al. (1983a) described an analogous
(ka : 2–3 m2 (mol cell C)−1 ). chromatic adaptation of a Cyanobacterium, now
Another physiological adaptation to persis- ascribed to Planktolyngbya, stratified in the metal-
tent light limitation is to increase the comple- imnion of a tropical forest lake (Lagoa Carioca)
ment of accessory photosynthetic pigments. In in eastern Brazil. In each of those cases where
general, this assists photon capture by widen- measurement has been made, chromatic adapta-
ing the wavebands of high absorbance, effec- tion increased the chlorophyll-specific photosyn-
tively plugging the gaps in the activity spec- thetic yield and the cell-specific photosynthetic
trum of chlorophyll a. In particular, the phy- efficiency (α). In the P. agardhii strain studied by
cobiliproteins (phycocyanins, phycoerythrins of Post et al. (1985), the efficiency (α) was ∼7 times
the Cyanobacteria and Cryptophyta) and the vari- steeper in cultures grown at 20 ◦ C on a 16 h : 8h
ous xanthophylls (of the Chrysophyta, Bacillario- light–dark cycle and a photon flux of 7 µmol pho-
phyta and Haptophyta; see Table 1.1) increase the tons m−2 s−1 than in material in similarly treated
light harvesting in the middle parts of the visi- cultures exposed to >60 µmol photons m−2
ble spectrum. The close association of accessory s−1 (0.78 vs. 0.11 mg O2 (mg dry weight)−1 (mol
pigments with the LHCs facilitates the transfer photon)−1 m2 ; or, in terms of carbon, approx-
of excitation energy to chlorophyll a. The corol- imately 0.54 vs. 0.08 mol C (mol cell C)−1 (mol
lary of widening the spectral bands of absorption photon)−1 m2 ). Supposing a basal (dark) rate of
is a colour shift, the chlorophyll green becom- respiration for Planktothrix at 20 ◦ C (derived from
ing masked with blues, browns or purples. This [Ra20 = 0.079 (s/v)0.325 ]; see Sections 3.3.2 and 5.4.1)
is acknowledged in the term ‘chromatic adap- of 0.064 mol C (mol cell C)−1 d−1 , or 0.74 × 10−6
tation’ (Tandeau de Marsac, 1977). Some of its mol C (mol cell C)−1 s−1 , it is possible to deduce
best-known instances involve Cyanobacteria for- that compensation is literally achievable at ∼1.4
merly ascribed to the genus Oscillatoria. Post et al. µmol photons m−2 s−1 . Realistically, allowing for
(1985) described the photosynthetic performance faster respiration during photosynthesis and for
of a two- to three-fold increase in the chlorophyll the dark period (8 h out of 24 h), net photosyn-
content and a three- to four-fold increase in c- thetic gain is possible over about 3 to 4 µmol
phycocyanin pigment in low-light grown cultures photons m−2 s−1 .
of Planktothrix agardhii. Photosynthetic attributes Even this performance may be consider-
of pink-coloured, deep-stratified populations of P. ably improved on by stratified bacterial pho-
agardhii were investigated by Utkilen et al. (1985b). tolithotrophs, with net growth being sustained
A remarkable case of chromatic adjustment in by as little as 4–10 nmol photons m−2 s−1 (review
a non-buoyant population of Tychonema bourrel- of Raven et al., 2000). Chromatic photoadapta-
leyi, as it slowly sank through the full light gration also sustains net photoautotrophic produc-
dient in the water column of Windermere, is tion in cryptomonad-dominated layers in karstic
given in Ganf et al. (1991). Chromatic adaptation dolines (solution hollows) at ambient photon flux
reaches an extreme claret-colour in populations densities of <2 µmol photons m−2 s−1 (Vicente
of Planktothrix rubescens, which stratify deep in the and Miracle, 1988). This seems to be accept-
metalimnetic light gradient of alpine and glacial able as a reasonable threshold for photoautotro-
ribbon lakes (Meffert, 1971; Bright and Walsby, phy in phytoplankton. Deep chlorophyll maxima
2000). The early-twentieth-century appearance of dominated by chromatically adapted cryptomon-
Planktothrix at the surface of Murtensee, Switzer- ads have also been observed in somewhat larger
land, was popularly supposed to have come from lakes and reservoirs (Moll and Stoermer, 1982),
the bodies of the army of the ruling dukes sometimes close to the oxycline, from which
of Burgundy, defeated and slain in a battle at short diel migrations, either upwards to higher
Murten in the fifteenth century. The connota- light or downwards (to more abundant nutrient
tion ‘Burgundy blood alga’ celebrates the colour resources) are possible (Knapp et al., 2003).
Photosynthetic limits in lakes and seas valid as a species-specific statement of an indi-

It is often convenient to subdivide the water col- vidual plankter’s position in the light gradient
umn on the basis of its ability to sustain net relative to its requirement to be able to compen-
photosynthesis or otherwise. The foregoing sec- sate its respirational costs.
tions demonstrate three functional subdivisions From Eq. (3.12), we can propose:
based on the criterion of light availability. In the
h p = ln(I 0 /0.5I k )ε −1 (3.15)
first (the uppermost), light is able to saturate
photosynthesis (Iz > Ik ); in the second, light is a recognising that Ik is a property of the species
constraint, being limiting to chlorophyll-specific of phytoplankton present and that its cell-carbon
photosynthesis (Iz < Ik ), but whose effects may specificity is attributable to its carbon-specific
be photoadaptively offset in order to optimise photosynthetic efficiency; i.e.
the rate of biomass-specific photosynthesis. In
I k = P /α (3.16)
the third, even biomass-specific photosynthesis is
incapable of compensating the biomass-specific where P is the carbon-specific rate of photosyn-
demands of respiration and maintenance (Iz < thesis (in mol C (mol cell C) s−1 ) and α is the effi-
IP=R ). The actual water depths for these irradi- ciency (in mol C (mol cell C)−1 (mol photon)−1 m2 ).
ance thresholds are notionally simple to calcu- This development also infers the value of the
late from the I vs. z curve but, of course, they are attenuation coefficient, ε, as a basis for intercom-
not fixed in any sense. The immediate subsurface paring aquatic environments. It has the advan-
intensity through the solar day and it is subject tage of being a property of the environment
to superimposed variability in cloud cover and (albeit a transitory one) although care is nec-
atmospheric albedo, as well in the fluctuating essary in citing the waveband being used (ε 440 ,
surface reflectance and subsurface scattering by ε530 , εav , etc.). Talling (1960, and many later pub-
particulates, induced by wind action. Irradiance lications) demonstrated a predictive robustness
thresholds translate to given depths only on an in the approximation of euphotic depth from
instantaneous basis. the minimum attenuation coefficient as the quo-
Despite the self-evident weakness of any tient, 3.7/ε min . In his examples, the least attenu-
depth–light threshold relationship, it is still valu- ation was in the green wavebands (λ ∼ 530 nm)
able to intercompare various underwater light but, as a rough guide to the depth in which pho-
environments by reference to the impacts of their tosynthesis is possible in the sea, the relationship
light attenuation properties. A commonly cited holds quite well for other wavebands. Table 3.2 is
index used in connection with the ability of a included to contrast the photosynthetic limits of
water column to support phytoplankton growth the clearest oceans and some of the most turbid
is the average depth ‘reached by 1% of surface estuarine waters, on the strength of the approx-
irradiance’; this has also been used to define the imation that, for many phytoplankters and for
depth of the so-called euphotic zone. Bearing in much of the day-light period, positive net photo-
mind that the PAR flux at the surface at mid- synthesis (Pg ≥ Ra ) is possible in the water col-
day varies within at least an order of magnitude umn defined by hp = 3.7/εmin . It is emphasised
(200–2000 µmol photons m−2 s−1 ), the 1% irradi- that almost all the net primary photoautotrophic
ance boundary is approximated no more closely production in the sea occurs within the top 100
than 2–20 µmol photons m−2 s−1 . Besides the tem- m or so and, in lakes, within the top 60 m. In
poral variability in its precise location in the both cases, it is usually much more constrained
water column, the quantity also suffers from than this.
its conceptual coarseness. Irradiances within this Another, much more convenient measure of
range could saturate the requirements of some relative transparency of natural waters is avail-
species while simultaneously failing to compen- able, the Secchi disk. A weighted circular plate,
sate the respiration of others. The depth of the painted all-white or with alternate black and
euphotic zone (hp ) is not a general property of the white quadrants, is lowered into the water and
underwater environment, although it remains the depth beneath the surface that it just
Table 3.2 Comparison of the depth of water likely to be capable of supporting net photosynthetic
production (hp ) in some representative lakes and seas, supposing hp = 3.7/εmin (cf. Talling, 1960). Values
of εmin , the minimum coefficient of attenuation across the visible spectrum are taken (1) from Kirk
(1994) or (2) from sources quoted in Reynolds (1987b)
Water εmin m−1 hp m Source

Oceans
Sargasso Sea 0.03 123 (1)
Pacific, 100 km off Mexico 0.11 34 (1)
Shelf waters 0.15–0.18 20–25 (1)
Lakes and reservoirs
Crater Lake 0.06 62 (1)
Lake Superior 0.10–0.20 18.5–37.0 (1)
Windermere (North Basin) 0.28–0.72 5.1–13.2 (2)
Crose Mere 0.32–4.20 0.9–11.6 (2)
Lake Kilotes 8.20 0.45 (2)
Mt Bold Reservoir 1.14 3.25 (2)
P. K. le Roux Reservoir 6.39 0.58 (2)
Severn Estuary 10–20 0.2–0.4 (2)
disappears from view is noted as the Secchi depth It is likely to be wholly beneficial in maintaining
or Secchi-disk depth. It is easy to use, has no work- photosynthetic vigour, just so long as the vertical
ing parts to malfunction and, in the hands of extent of entrainment is within the depth range
a single operater, it can give fairly consistent offering irradiances between I 0 to Ik .
results. However, these attractions are countered With relatively deeper mixing, however
by difficulties of quantitative interpretation of (either as a result of stronger physical forc-
its measurements (see Box 3.1). However, docu- ing or of greater underwater light attenuation),
mented records of Secchi-disk depth (zs ) span a entrained plankters are carried beyond the depth
wide range, from 0.2 m to 77 m (Berman et al., of Ik and, in many circumstances, beyond the
1985) and are adequate to separate clear waters productive compensation point. During a period
(zs ≥10 m) from the turbid (zs ≤ 3 m) and to be of time (probabilistically, the mixing time, tm :
sensitive to temporal changes in the clarity of any see Section 2.6.5), the individual plankter may be
one of them. successively exposed to light intensities that are
saturating, sub-saturating or altogether inade-
Photoadaptation to vertical mixing quate to support net photosynthetic production.
This section considers the adaptive responses These effects are represented in Fig. 3.14, where
to turbulent entrainment and vertical transport a notional ‘Lagrangian’ path of a single alga,
beyond the column compensation point. Vertical moved randomly in the vertical axis by turbulent
mixing per se is not necessarily problematic for entrainment through an equally notional light
a microplanktic photoautotroph and, in entrain- gradient (a), is exposed to a predictably fluctuant
ing plankters to and from the high solar irradi- photon flux (b). From the prediction of light-
ances that may obtain near the top of the water dependent photosynthesis (c), the instantaneous
column, may help to avoid the photoinhibition rate of photosynthesis of the phytoplankter is
response observed in phytoplankton captured in also now predictable (d). Integrating through
static bottles (Jewson and Wood, 1975). Entrain- time, it is clear that the net photosynthetic gain
ment resists the development of other restrict- is impaired below the potential of Pmax . More-
ing gradients (for instance, of diminishing dis- over, the deeper is the mixed depth (hm ) with
solved carbon dioxide or accumulating oxygen). respect to the depth of the column in which net
Box 3.1 The Secchi disk
Many variants of the Secchi disk have been employed in limnology and plankton
science but the recommended standard is made of aluminium and painted white,
is 300 mm in diameter and suspended by three cords attached to a single rope
about 30 cm above. It is lowered carefully into the water until the observer just
loses sight of it. It is often said to measure transparency or light penetration but
these are not literally accurate. The image is lost to the observer through scattering
of light rays. However, it is a sufficiently useful and serviceable instrument for there
to have been several attempts to relate measurements of Secchi-disk depth (zs ) to
more formal photometric light determinations (Vollenweider, 1974; Preisendorfer,
1976; Stewart, 1976). The quest is not aided by differences between observers
or, for a single observer, by differences between sun and cloud or between calm
and waves. On the basis of simultaneous measurements, Poole and Atkins (1929)
deduced that zs and ε are in approximate inverse proportion and, thus, that the
product zs × εmin should be roughly constant. Their evaluation of this constant
(1.44) is only just representative of the later estimates (1.4 to 3.0, with a mean of
about 2.2: Vollenweider, 1974). Then the light intensity remaining at zs is between
5% and 24% of I0 (mean ∼15%), whence the compensation depth is perhaps 1.2
to 2.7 × zs .
Figure 3.14 (a) Typical depth

profile of irradiance absorption; (b)
‘random walk’ of a phytoplankter
entrained in the mixed layer of the
same profile; (c) simultaneous plot
of the photosynthetic rate that can
be maintained at given depths
(light-saturated above Ik ); (d)
deduced instantaneous
photosyntheyic rate that is
maintained by the alga following the
trajectory depicted in (b). Redrawn
with permission from Reynolds
(1997a).
photosynthesis is possible (hp ), the more restric- be shown to be profound. Supposing I 0 is 800
tive are the mixing conditions on the prospects µmol m−2 s−1 and ε is 1.0 m−1 , then the light
of photosynthetic gain. reaching the bottom of an 8-m mixed later would
By extension of this argument, the smaller be 0.27 µmol m−2 s−1 and I∗ for the whole 8-m
is hp in relation to hm , then the more difficult layer is just under 15 µmol m−2 s−1 . Doubling the
it is to sustain any net photosynthesis at all. It mixed depth to 16 m, means that the light reach-
was long a tenet of biological oceanography that ing the bottom would be 9 × 10−5 µmol m−2 s−1
the major mechanism permitting phytoplankton and the integral for the 16-m layer would be
recruitment through growth depended upon the <0.3 µmol m−2 s−1 . A factor of two in the depth of
depth of mechanical mixing relaxing sufficiently mixing changes a light dose expected to sustain
relative to light penetration for net photosynthe- significant net photosynthetic gain to one which
sis to be sustainable. This relationship was con- will not even satisfy respiration.
sidered by Sverdrup et al. (1942), although it is Furthermore, starting on the basis of areal
the eventual mathematical formulation of what integrations of measured photosynthesis and res-
is still known as Sverdrup’s (1953) ‘critical depth piration rates versus depth ( N P and N R ),
model’ to which reference is most frequently extrapolations of net photosynthesis over 24 h
made. In particular, the idea that the spring may be approximated, as calculated as N P −
bloom in temperate waters, in lakes and rivers N R . For short enough days (temperate
as well as open seas and continental shelves, winters!), high enough attenuation coefficients
is dependent upon the onset of thermal strat- and verifiable mixing limits, it is probable
ification, at least when it is compounded by a that low or zero rates of observed phytoplank-
seasonal increase in the day length, remains a ton increase are correctly attributed to inade-
broadly plausible concept. However, it is lacking quate energy income. To illustrate this, Reynolds
in precision, is open to too literal an interpreta- (1997b) reworked some earlier data (Reynolds,
tion and is not amenable to simulation in mod- 1978a; Reynolds and Bellinger, 1992) on the year-
els. The problem is due partly to the perception round observations on the phytoplankton dynam-
of stratification (as pointed out previously, lack of ics in the turbid (ε > 0.5 m), 30-m deep, eutrophic
a pronounced temperature or density gradient is Rostherne Mere, UK. He showed that, in the
not, by itself, evidence of active vertical mixing). period January–March, net photosynthetic gain
Compounding this is the issue of short-term vari- would have been possible only when the mixed
ability and the likelihood of incomplete mixing depth was <4 m and that the development of
within a layer defined by a ‘fossil structure’ (as any significant spring diatom bloom was nor-
defined in Section 2.6.4). mally delayed until April. In a classical paper
These are important statements regarding an on the whole-column photosynthetic integrals
important paradigm, so care is needed to empha- in Windermere, Talling (1957c) showed that the
sise their essence. It is perfectly true, for instance, observed population increase of Asterionella in
that mixing to depth does not only homogenise Windermere in the first 3 to 4 months of the
probable, time-averaged integrals of insolation year would certainly account for a very high pro-
but ‘dilutes’ it as well. I used an integral, I∗ portion (around 96%) of the extrapolated inte-
(Reynolds, 1987c), to estimate light concentration gral photosynthesis (and, hence, a simultane-
in homogeneously mixed layers, based on the dif- ously very low biomass-specific respiration rate)
ference between the light availability at the sur- to be able to sustain it.
face and at the bottom, as extrapolated from the Later attempts to model the population
attenuation coefficient. growth from first principles underestimated the
oft-observed growth performance in Winder-
ln I ∗ = (ln I 0 + ln I m )/2 (3.17)
mere, unless some allowance was made for a
where Im is the extrapolated irradiance at the diminishing Monin–Obukhov length (see Section
base of the contemporary mixed layer. Solved by 2.6.3) through the spring period (Reynolds, 1990;
Eq. (3.11), as Im = I 0 · e−ε zm , deep mixing can Neale et al., 1991a). In other words, the actual
responses of each of a number (at least 20)

of plankters, being simultaneously ‘walked ran-
domly’ through a simulated light gradient, were
summated to derive an aggregate for the popu-
lation. Huisman et al. (1999) used Okubo’s (1980)
turbulent-diffusion model and a concept of resid-
ual light at the base of the turbulent layer
(devised by Huisman and Weissing, 1994) to
demonstrate the importance of a critical light
threshold and, incidentally, to show up the short-
comings of the literal ‘critical depth’ model.
Figure 3.15 Measured daily mean wind speeds at Blelham
Though this view has not passed unchallenged, it
Tarn, UK, between August 1999 and October 2000
has been supported independently in the theor-
(continuous line), and the mixed depth, as calculated from
wind speed and temperature of the water column. Redrawn
etical consideration of Szeligiwicz (1998). He also
with permission from Davis et al. (2003). verifies the point that a critical depth is not the
same for all species simultaneously, that those
with a lower critical light compensation will per-
spring growth, averaged over a number of consec- form better in deeply mixed layers and that their
utive years, is really a response to the weakening own adaptive behaviours may modify the critical
‘dilution’ of the incoming light in a diminish- light and critical depth while the environmental
ing mixed depth, even though no conventional conditions persist.
thermal stratification is yet established. In real- The nature of these adaptive responses to low
ity, this is not at all a particularly smooth tem- aggregate light doses in mixed layers is, in many
poral progression. The day-to-day variability in ways, similar to those arising from the low aggre-
cloud cover and wind forcing continues but the gate light exposure of plankters residing deep in
frequency of the sunnier, less windy days does the light gradient. However, mixed-layer entrain-
accelerate, just as the days are lengthening sig- ment offers short bursts (a few minutes) of expos-
nificantly. It follows that, as the year advances, ure to relatively high light intensities separated
there are more days on which photosynthetic by probabilistically relatively long periods (of the
gain, growth and recruitment in the upper water order of 30–40 minutes) in effective darkness.
layers is possible. It would seem important for these organisms
Another decade of improvements in moni- to undertake as much photosynthesis as pos-
toring approaches and in simulation modelling sible in the exposure ‘windows’ to non-limiting
techniques permits us to derive a greater resolu- irradiance fluxes, which requires an enhanced
tion on the variability in the underwater pho- light-harvesting capacity rather than a wide
tosynthetic environment. In Fig. 3.15, the fluc- spectrum of absorption. Phytoplankton reputed
tuations in the mixed depth of Blelham Tarn to grow relatively well under deep-mixed condi-
(between 0.5 and 12 m) through a winter–spring tions include the diatoms (especially those with
period reveal that there is neither a smooth attenuate cells or that form filaments) and cer-
nor single abrupt switch between fully mixed tain Cyanobacteria and chlorophyte genera with
and stably stratified conditions, more a chang- analogous morphological adaptations (high
ing frequency of alternation. The contemporane- msv−1 : Reynolds, 1988a); all project substantial
ous compensation depth varied between 2 and carbon-specific areas (10–100 m2 per mol cell C)
8 m (Davis et al., 2003), net growth being possible but they vary their contents of chlorophyll a
when compensation depth exceeds mixing depth. more than accessory pigments.
Two other modelling approaches anticipated Mostly these effects have been detected
similarly modified views of the Sverdrup criti- through population growth and recruitment in
cal depth model. Woods and Onken (1983) con- culture. Turning off the light for a part of the day
structed a ‘Lagrangian’ ensemble, in which the soon brings growth limitation into regulation
Experimentally imposed fluctuations in light-

exposure levels on periodicities of days to weeks
also affect the composition and diversity of phyto-
plankton assemblages (Flöder et al., 2002). These
operate through the replication, recruitment and
attrition of successive generations to populations
and will be considered in a later chapter (see
Sections 5.4.1, 5.4.2). However, to simulate in
the laboratory some of the extreme behaviours
observed in the field required observations relat-
ing to photoperiods rather shorter than an hour.
Robarts and Howard-Williams (1989) described
the response of a low-light-adapted Anabaena
species in a turbid, mixed lake (Rotongaio, New
Zealand) whose rate of photosynthesis could
Figure 3.16 Growth-rate responses of Cyanobacteria in accommodate exposure to light at the water sur-
culture at two different temperatures (10 ◦ C, 20 ◦ C) to face for 6 minutes but was slowed abruptly under
various and two photoperiods: (a) continuous light at further exposure. In this instance, the productive
27 µmol photons m−2 s−1 and (b) under a 6 h : 18 h advantages were to be gained only in the pho-
light–dark alternation. (c) shows the daily growth rate in (b)
toperiods of less than 6 minutes, to which the
extrapolated to 24 h, to show the improved efficiency of
energy use. Data of Foy et al., (1976), redrawn from Reynolds
organism had clearly adapted. These observations
(1984a). are considered in the context of photoprotection
and photoinhibition, in the next section.
by light-dependent photosynthesis more than 3.3.4 Photoprotection, photoinhibition

light-independent assimilation – exposing algae and photooxidation
to saturating light for only 12 or 6 h out of 24 h Against the background of environmental vari-
always results in a reduced daily growth rate. ability, there may be superimposed variations in
However, photoadaptative responses to shortened the contemporary ambient range of fluctuations,
photoperiod raise the rate of biomass-specific subjecting hitherto supposedly acclimated plank-
energy harvest to the extent that growth nor- ters to additional demands of accommodation.
malised per light hour is raised. This principle Among the most crucial of these is a weakening
was memorably demonstrated by Foy et al. (1976) of the mechanical forcing, either as a result of
(Fig. 3.16). a sharp reduction in the wind speed or of sharp
Litchman (2000, 2003) has taken this approach increase in the photon flux (perhaps as the cloud
further, exploring the effect of shorter exper- clears) or, as is often the case, the coincidence
imental photoperiods and their discrimination of both events. In all these instances, the abrupt
among the performances of the test algae. Fluc- shortening of the Monin–Obukhov length is, far
tuation periods were varied between 1 and 24 from being the net beneficial influence cited
h and intensities were varied between 5 and above (in Section 3.3.3), potentially highly dan-
240 µmol photons m−2 s−1 . Photoperiod evoked gerous. Part of the hitherto entrained population
little photoadaptation at the higher intensities becomes disentrained deep in the water column,
but, at lower intensities, differences in species- where the irradiance is markedly sub-saturating.
specific responses became evident. In general, Another is retained within a new, much shal-
the effect of fluctuating light tended to be lower, surface circulation, exposed to a much ele-
greater when irradiance fluctuated between lev- vated I∗ value and to a probable excess of radia-
els alternately limiting and saturating growth tion in the harmful, high-energy ultraviolet wave-
requirements. lengths. The greater the previous adaptation to
low average insolation and the more enhanced the re-reduction of P680 and, hence, the reacti-
is their light-harvesting capacity then, clearly, vation of the LHCs. The energy absorbed from
the greater is the danger of damage to the cells unused photons continuing to arrive at P680 is re-
affected. Analogous risks confront plankters near radiated as fluorescence. This is readily measurable:
the surface of lakes becalmed overnight and sub- the spectral signal of emitted fluorescence has
ject to rapid post-dawn increases in insolation. In long been used as an index of plankton biomass
a lesser way, perhaps, even short bursts of strong (an analogue of an analogue: Lorenzen, 1966).
light on a mixed layer or lulls in the wind inten- Differences in the spectral make-up of the emis-
sity acting on turbid water under bright sun- sion can also be used to separate the organis-
shine will result in potentially sharp increases in mic composition of the phytoplankton, at least
the photon flux experienced by individual algae. to the phylum level (Hilton et al., 1989). However,
Moreover, this is the fate of isolates of wild pop- because the transfer of electrons from the plas-
ulations, sampled from the water column, sub- toquinone pool to PSI is a rate-limiting step, the
sampled into glass bottles and then held captive size of the PQ pool is a measure of the photo-
at the top of the water column; it is little wonder synthetic electron transport capacity. In this way,
that their performance becomes impaired (Harris PSII fluorescence may also be exploited as a sen-
and Piccinin, 1977) (see Section 3.3.1). sitive analogue of photosynthetic activity (Kolber
In fact, photoautotrophic plankters are and Falkowski, 1993). Light-stimulated in vivo flu-
equipped with a battery of defences for coping orescence from cells exposed to a flash of weak
with and surviving exposure to excessive solar light in the dark (F0 , when all centres are open)
radiation levels. As has already been said, some is compared with the fluorescence following a
of these have the effect of cutting photosynthetic subsequent saturating flash (Fm , corresponding
rate and the response was formerly interpreted to their total closure). The presence of open cen-
as ‘photoinhibition’. Strong light certainly can tres quenches the fluorescence signal proportion-
inhibit photosynthesis and do a lot of physical ately, so the difference, (Fv = Fm − F0 ), becomes
damage to the photosynthetic apparatus. How- a direct measure of the photosynthetic electron-
ever, many of the observed responses are pho- transport capacity available and the extent of
toprotective and serve to avoid serious damage the reduction in the quantum yield of pho-
occurring to the cell. These are reviewed briefly tosynthesis caused by exposure to high light
below; the sequence is more or less the one in intensities.
which live cells, suddenly confronted by supersat- As a relatively short-term response, the
urating photon fluxes, invoke them in response. chlorophyll-a fluorescence yield alters as the
Some excellent, detailed reviews of this topic plankters are moved up and down through
include Neale (1987), Demmig-Adams and Adams the mixed layer. The measurement of fluores-
(1992) and Long et al. (1994). cence to investigate the transport and the speed
of photoadaptive and photoprotective reactions
Fluorescence of phytoplankton to variable underwater light cli-
In simply moving upwards from a sub-saturating mates is one of the exciting new areas of applied
light to a depth where the photon flux density plankton physiology (Oliver and Whittington,
supersaturates not just the demand for growth 1998).
but also the carbon-fixing ability of PSI, the
entrained cell will experience two almost concur- Avoidance reactions
rent effects, evoking two compounding reactions. Provided they are adequately disentrained and
The greater bombardment of the LHCs by pho- their intrinsic movements are adequately effec-
tons means that some of these now arrive at reactive, motile organisms migrate downwards from
tion centres that are still closed, pending reoxi- high irradiance levels. Avoidance reactions have
dation of the acceptor quinone, Q A (see Section been observed especially among the larger
3.2.1). At the same time, the accelerated accumu- motile dinoflagellates (see, especially, Heaney
lation of PQH2 in the plastoquinone pool slows and Furnass, 1980a; Heaney and Talling, 1980a)
and larger, buoyancy-regulating Cyanobacteria tion of the xanthophyll cycle in protecting PSII
(Reynolds, 1975, 1978b; Reynolds et al., 1987). For from excessive photon flux density operates by
non-motile diatoms, a rapid sinking rate may siphoning off a good part of the energy as heat.
provide an essential escape from near-surface Many details of the cycle and its fine-tuning
‘stranding’ through disentrainment, especially in are considered by Demmig-Adams and Adams
low-latitude lakes. The relatively high sinking (1992). Here, it is important to emphasise how
rates in (especially) Aulacoseira granulata may be these adaptations of phytoplankton to high light
a factor in the frequency of its role as domi- assist in maintaining photosynthetic productiv-
nant diatom in many tropical lakes where there ity. Carotenoids are especially effective in protect-
is a diel variation in mixed water depth (see ing cells against short-wave radiation and the risk
Reynolds et al., 1986). The effect may be signifi- of photooxidative stress.
cantly enhanced by spontaneous acceleration of Compounds specific to the absorption of
the sinking rate of cells coping with an abrupt ultraviolet wavelengths, previously known from
increase in insolation (Reynolds and Wiseman, the sheaths of epilithic mat-forming Cyanobac-
1982; Neale et al., 1991b), perhaps as a result of teria of hot springs (Garcia-Pichel and Casten-
the withdrawal of the alleged mechanism of vital holz, 1991), where they screen cells from dam-
regulation (see Section 2.5.4). aging wavelengths of radiation (max absorp-
tion ∼370 nm), have been found recently in the
natural phytoplankton of high mountain lakes.
Plastid orientation and contraction Laurion et al. (2002) suggested that, together
Planktic cells are generally too small for plas- with the carotenoids, these mycosporine-like
tid relocation to have the significance it does in amino acids may occur widely among limnetic
the cells of higher plants (Long et al., 1994) but, phytoplankton species, especially in response
over periods of minutes to hours of exposure to to exposure to ultraviolet wavelengths. Ibel-
high light intensities, contraction of the chro- ings et al. (1994) demonstrated just this sort
mophores of planktic diatoms lowers the cross- of acclimation of planktic species, especially
sectional areas projected by the cell chlorophyll in Microcystis, where the sustained presence
(Neale, 1987). of zeaxanthin contributes to an ongoing abil-
ity to dissipate excess excitation energy as
Protective pigmentation heat. As originally proposed by Paerl et al.
In cells exposed to frequent or continuing high (1983), the mechanism substantially protects cells
light intensities over a generation time or more, from overexposure of surface blooms to high
over-excitation of the PSII LHCs is avoided by light.
changes affecting the xanthophylls. These oxy-
genated carotenoids are subject to a series of Excretion
light-dependent reactions, which, among the In nutrient-limited cells, photosynthate is
chlorophytes (as among green higher plants), scarcely consumed in growth. Even under quite
results in the accumulation of zeaxanthin under modest light levels, simultaneous accumulation
excess light conditions and its reconversion to of fixed carbon and free oxidant in the cell risk
violoxanthin on the return of normal light con- serious photooxidative damage to the cell. This is
ditions. Among the dinoflagellates and the chrys- countered principally through the production of
ophyte orders (sensu lato, here including the antioxidants, such as ascorbate and glutathione.
diatoms), an analagous reaction involves the con- High oxygen levels may trigger the Mehler
version of diadinoxanthin to diatoxanthin, when reaction in PSI in which oxygen is reduced to
light is excessive, with oxidation back to diadi- water (Section 3.2.3). Moreover, high O2 concen-
noxanthin in darkness. The reaction is said to be trations (>400 µM) induce the oxidase reaction
about 10 times faster than the analogous reac- of RUBISCO, and the photorespiration of RuBP
tion in higher plants (Long et al., 1994, quot- to phosphoglycolic acid. Release of glycolate
ing the work of M. Olaizola). The principal func- and other photosynthetic intermediates into the
water is one of the ‘healthy’ (cf. Sharp, 1977) ways of photosynthetic production by inorganic car-
in which cells of other algal groups regulate bon has been a surprisingly contentious issue.
the internal environment by venting unusable Kuentzel (1969) considered the carbon supply to
DOC into the medium. This behaviour carries be one of several factors crucial to the develop-
important consequences for the structure and ment of algal blooms in response to lake enrich-
function of pelagic communities (see Section ment. Shapiro’s (1973) experimental demonstra-
3.5.4). tions of the ability of bloom-forming Cyanobac-
teria to grow at high pH levels (indicative of
If these fail . . . deficiencies in the reserves of CO2 in solution)
Nevertheless, prolonged exposure of phytoplank- provided very strong support for this view. On
ton cells to high light intensities over periods of the other hand, Schindler’s (1971) whole-lake
days to weeks usually results in pigment loss, manipulations in the Experimental Lakes Area of
loss of enzyme activity, photooxidation of pro- Canada pointed to the direct linkage of produc-
teins and, ultimately, death. Such dire conse- tion responses to added phosphorus and nitro-
quences to the photosystems and cell structures gen. His data were persuasive. Moreover, the intu-
certainly do enter the realm of severe photoinhi- itive supposition of an adjacent, effectively infi-
bition and photodamage. Floating scums of buoy- nite reserve of atmospheric carbon dioxide, read-
ant Cyanobacteria are especially vulnerable to ily soluble in water, was enough to allay most
photodamage; death sequences have been graph- doubts that carbon availability is a significant
ically reported by Abeliovich and Shilo (1972). constraint upon the yields of aquatic biomass.
In a more recent account, Ibelings and Maberly However, matters did not rest there and much
(1998) described the loss of photosynthetic capac- important research has ensued. Crucially, the
ity in response to excessive insolation and carbon first point that must be recognised is that these
depletion in laboratory simulations of the condi- arguments are not, in fact, directly opposed,
tions experienced in surface blooms. nor are they mutually exclusive. The aquatic
At lesser extremes, the resilience of cells and sources of inorganic carbon are, indeed, ulti-
opportunities for repair may allow recovery of mately plentiful and renewable and do not con-
physiological vigour. Thus, the many effects of stitute a biomass-limiting constraint. At the same
environmental variability that can lead to a fall time, it has to be accepted that the ambient con-
in the net planktic production of photosynthate, centration of dissolved carbon dioxide is highly
once universally labelled as ‘photoinhibition’, variable, that it is inextricably linked to the
should properly be viewed as a suite of homeo- pH-dependent bicarbonate system and that algal
static protective mechanisms. They enable phyto- production is, anyway, a principal driver in the
plankton to survive a large part of the full range transformations. The potential role of the carbon
of environmental extremes that may be encoun- supply as a rate-limiting constraint on photosyn-
tered as a consequence of pelagic embedding (see thetic behaviour is plainly indicated.
also the discussion in Long et al., 1994).
3.4.1 Sources and fluxes of available
inorganic carbon
3.4 Sensitivity of aquatic The sea-level, air-equilibrated concentration of
photosynthesis to carbon carbon dioxide in water at 0 ◦ C is ∼23 µM, falling
to ∼13 µM at 20 ◦ C (say, 0.5–1.0 mg CO2 L−1 ,
sources or ∼0.15–0.3 mg C L−1 ). Besides being sensitive
to temperature, the equilibrium concentration
Besides light, a supply of carbon dioxide is depends upon the atmospheric partial pressure
essential to normal photosynthetic production. of CO2 , which is, in turn, affected by altitude.
However, while the necessity of an instanta- In many natural waters, carbonic acid is the
neous light source to the fixation of carbon is only free acid present and, thus, the concen-
self-evidently axiomatic, the potential limitation trations of alkalinity (base ions), carbon dioxide
SENSITIVITY OF AQUATIC PHOTOSYNTHESIS TO CARBON SOURCES 125
the equilibrated mass of gas in solution (again,

generally ∼0.3 mg C L−1 ) complies with Henry’s
law and does not exceed the proportionality of
the partial pressure of the gas in contact with
the liquid.
At face value, the instantaneous carbon
capacity of natural waters to support phyto-
plankton is unlikely to exceed 0.3 mg C L−1 (or
about 0.02 mol C m−3 ). Where present, bicarbon-
ate raises the DIC reserve up to 2 orders of mag-
nitude greater. Supposing a C : Chla of 50, these
capacities are equivalent to the supportive capac-
ity for 6 to 600 µg chla l−1 . Plainly, carbon limi-
tation of the phytoplankton supportive capacity
is hardly likely to arise among the many water
bodies in the world in which biomass is severely
restricted by 1 or 2 orders of magnitude (∼0.6 to 6
µg chla L−1 ). Neither is the standing biomass of
non-calcareous waters prevented from consider-
Figure 3.17 The pH–carbon dioxide–carbonate system in
natural waters. The relative quantities of the three
ably exceeding 6 µg chla L−1 . The instantaneous
components, CO2 , HCO3 - and CO3 2− , determine the pH of carbon dioxide availability in even the soft, non-
the water, as shown in the inset. Changes in the calcareous waters may significantly exceed the
concentration of one component shifts the equilibrium. air-equilibrated concentration and some produc-
Photosynthetic withdrawal of CO2 can raise pH to the point tion may be maintained when the DIC reserve is
where CO3 2− is precipitated as a calcium salt. For a more exhausted. This inspires queries about the inter-
complete explanation, see Stumm and Morgan (1996). nal sources of carbon dioxide and the rates of
their replenishment.
The proximal sources of ‘new’ carbon include
the solution of CO2 at the air–water interface, not
and pH (acidity) are continuously interrelated. just at the surface of the water body in question
In this range, bicarbonate is generally the domi- but in the rainfall leaching the atmosphere and
nant anion (0 to ∼3.5 meq L−1 ; higher alkalinities falling directly or, indirectly, in the overland flow
may be associated with alternative solute compo- discharging into it from the surrounding water-
sition, sodium or potassium being the dominant shed. The quantities transported to lakes can be
cation, rather than calcium). Potentially, bicar- considerable but the concentrations are still sub-
bonate may dissociate to release free CO2 , accord- ject to equilibrium constraints. That fraction of
ing to the reversible reactions shown in Fig. 3.17. the inflow made up from groundwater sources
In this way, bicarbonate in solution represents can become relatively enriched with CO2 under
an exploitable store of dissolved inorganic carbon pressure (to the extent that some may vaporise
(DIC) of up to 42 mg CO2 L−1 . For much of the when normal air pressure is encountered). Direct
time it serves to buffer the water to the mildly vulcanism (through fumaroles) can provide addi-
alkaline side of neutrality (pH ∼8.3), potentially tional sources of carbon dioxide to sea water in
to the point of calcium carbonate precipitation certain locations.
(see Fig. 3.17). In base-poor lakes, this buffering Usually, the major source of DIC is derived
capacity is proportionately weaker. In extremely from chemical weathering of carbonate rocks
acidic sodium-sulphate waters (pH < 4.5), there and debris in soils, including terrestrially
is (by definition) no alkalinity at all and DIC is sequestered atmospheric CO2 , which is trans-
present only as dissolved carbon dioxide or car- ported in run-off. Anthropogenically increased
bonic acid. Here, as in the first case considered, CO2 levels and accelerated erosion have
contributed to historic sharp increases in equilibrium, were observed. At such times, the
alkalinity in some major catchments, including lake would have been losing CO2 to the atmo-
that of the Mississippi River (Raymond and Cole, sphere. In contrast, photosynthetic carbon con-
2003). Significant additional sources of carbon sumption in the summer typically depletes the
in lakes may come from deliveries of readily epilimnetic DIC to very low levels (Heaney et al.,
oxidisable organic carbon, both particulate (POC) 1986), occasionally to zero (pH ∼10.3). At these
and in solution (DOC). Anthropogenic sources times, the atmosphere becomes the main photo-
(sewage, acid deposition, mine discharge) may synthetic carbon source.
be of local importance. Over the year, this lake probably loses three
These various carbon sources are available to to four times more CO2 to the atmosphere (up to
primary producers and, thence, to assimilation 2.8 mol m−2 a−1 ) than it absorbs (Maberly, 1996).
in aquatic food webs. In lakes, some of this car- No more than 4% of the annual production of
bon may be removed as organisms, their wastes biomass was found to be attributable to CO2 solu-
or cadavers, either to the sediments, or to down- tion across the water surface. Most of the net
stream transport, eventually becoming part of resource influx arrives in the lake in solution in
the POC flux to the sea. What is usually rather the inflow streams. As elsewhere, the main part
a larger part of the biogenically assembled car- of the annual load of free CO2 is roughly propor-
bon is respired by the producers or metabolised tional to the hydraulic load and the bicarbonate
and respired by their heterotrophic consumers load is, approximately, the product of the mean
(grazers and decomposers), mostly back to car- bicarbonate alkalinity in the inflow and the num-
bon dioxide. This gas can now be vented to the ber of annual hydraulic replacements.
atmosphere by equilibration. In lakes, especially, The idea that smaller lakes and rivers are not
and at times of low biological activity, carbon necessarily net sinks for atmospheric CO2 but,
dioxide is present in solution at concentrations rather, may often be outgassing CO2 to the atmo-
considerably over those predicted by Henry’s law sphere, is relatively new (Cole et al., 1994; Cole
(Satake and Saijo, 1974). In deep, oligomictic and and Caraco, 1998). In the wet tropics, catchment
meromictic lakes, hydrostatic pressure adds to sources of CO2 can make an especially signifi-
the level of carbon-dioxide supersaturation that cant contribution to the dissolved content and
is possible. Mechanical release of such reserves, to losses back to the atmosphere (Richey et al.,
such as occurred at Lake Nyos, Cameroon, in 2002). In yet another study, Jones et al. (2001) cal-
1986, carries dire consequences for people and culated that net CO2 efflux from temperate Loch
livestock in the adjacent hinterland (see Löffler, Ness may represent around 6% of the net ecosys-
1988). tem production of the catchment.
It becomes clear that there is a wide range Proportionately, the amount of carbon diox-
of carbon availabilities among lakes and seas, ide loaded hydraulically must diminish with
as there is in the principal carbon sources. The increasing size of the water body, as (presum-
stores and supplies can often be adequate but, ably) the proportion of the carbon dioxide influx
at times, demand is capable of exhausting them contributed by net inward invasion across the
faster than they can be replenished. This can be water surface increases. Yet it is plain that, even
especially true in individual lakes having high in Esthwaite Water, there are times of high
biomass-supportive capacities and making strong carbon demand and low resource-renewal rate,
seasonal demands on the carbon flux. Maberly marked by high pH values (∼10) when the accel-
(1996) constructed a balance sheet of annual car- erated absorption of atmospheric carbon diox-
bon dioxide exchanges in Esthwaite Water (Cum- ide across the water surface must supplement
bria, UK), a small (1.0 km2 ), stratifying (15 m), soft- the truncated terrestrial and internally recycled
water (alkalinity: 0.4 meq L−1 ) but eutrophic lake. sources.
During the autumn, winter and spring, free-CO2 In water bodies much larger than Esthwaite
concentrations of up to 120 µM (1.4 mg C L−1 ), Water, the hydraulic loads are relatively very
almost seven times the expected atmospheric much smaller and the oxidation of external POC
may be equally diminished on a relative scale. pH, but they create the conditions for carbon
In this case, the exchange of carbon dioxide is limitation of their own photosynthesis, at least
mediated mainly by respiration and the dynam- until the demand falls or other sources of car-
ics tend to be dominated by metabolic turnover, bon can assuage it. In the open sea, where, it is
the loss to sedimentary depletion of particulate alleged, atmospheric dissolution represents the
carbon and the supplement of ‘new’, invading major resource of new carbon, frequent strong
atmospheric CO2 . winds may well fulfil one of the criteria of
The direction and rate of gas-exchange flux gaseous invasion. The typical low biomass rep-
(FC ) across the water surface is governed by the resented by oceanic phytoplankton assemblages
relationship rarely raises the pH far above neutrality. Even so,
the maximum rates of invasion under the con-
F C = G C ξ pCO2 (3.18)
ditions envisaged here can hardly be expected to
where ξ is the solubility coefficient (in mol m−3 supply much more than 100 g C m−2 a−1 .
atmosphere−1 ), pCO2 is the difference in partial
pressure of carbon dioxide between water and 3.4.2 Phytoplankton uptake of
air, and GC is the gas exchange coefficient, or carbon dioxide
linear migration rate (m s−1 ). In fact, the mag- Given the relatively high half-saturation con-
nitude of actual exchanges is difficult to estab- stants for RUBISCO carboxylation (12–60 µM) (see
lish. However, the work of Frankignoulle (1988), Section 3.2.3), photosynthetic carbon fixation is
Upstill-Goddard et al. (1990), Watson et al. (1991) plainly vulnerable to rate limitation by the low
and Crusius and Wanninkhof (2003), who are aquatic concentrations to which CO2 may be
among those who have attempted to determine drawn (<10 µM) (see Section 3.4.1). Even with rel-
gas-transfer rates by reference to models or to atively plentiful supplies of carbon dioxide, the
the movements of sulphur hexafluoride tracer harvesting mechanisms of aquatic plants need
(SF6 ), provides important verifications. The sea- to be well developed (Raven, 1991). In the first
sonal variability in pCO2 becomes a crucially pow- place, satisfaction of the principal requirements
erful driver in those instances when, as in Esth- of planktic cells embedded in the viscous range
waite Water, photosynthetic withdrawal from is subject to Fick’s laws of diffusion. The number
the aquatic phase takes the air–water difference of moles of a solute (n) that will diffuse across
to its maximum (up to 9 × 10−4 atmosphere, an area (a) in unit time, t, is a function of the
i.e. the fastest consumption stimulates the most gradient in solute concentration, Co , (i.e. dCo /dx)
rapid invasion of the lake). However, the trans- and the coefficient of molecular diffusion of the
fer velocity is accelerated as a function of wind substance (m):
speed and surface roughness, from ∼10−5 m s−1 ,
n = am(dC o /dx) t mol m2 s−1 (3.19)
at wind velocities beneath the critical value of
3.5–3.7 m s−1 , to an order of magnitude greater, Reynolds (1997a) used data for the single, spher-
at 15 m s−1 (Watson et al., 1991; Crusius and Wan- ical cell of Chlorella (diameter ≤4 × 10−6 m,
ninkhof, 2003). Given high values of pCO2 , the approximate surface area ≤50.3 × 10−12 m2 )
corresponding invasion fluxes are calculated to to illustrate the limits of diffusion depen-
be in the order of 3–30 × 10−8 mol m−2 s−1 , or dence. Given (i) that, for an average small-sized
between 31 and 310 mg C m−2 d−1 . Once again, solute molecule (such as carbon dioxide), m ∼
using the approximate 50 : 1 conversion, this 10−9 m2 s−1 , that (ii) the thickness of the adja-
is theoretically sufficient to sponsor a produc- cent water layer from which nutrients may be
tive increment of only 0.6 to 6 mg chlorophyll absorbed is equal to the cell radius and (iii)
m−2 d−1 . the concentration of carbon dioxide molecules
These calculations fully amplify the observa- beyond is at air equilibrium (11 µmol CO2 L−1 , or
tion that large crops of algae, especially in lakes 11 × 10−3 mol m−3 ), then Eq. (3.19) is solved to
of low bicarbonate alkalinity, do not just deplete deliver 275 × 10−18 mol s−1 . Now, let us assume
the store of CO2 available, with a sharp rise in the volume of the cell (v) is 33.5 × 10−18 m3
and contains 0.63 ×10−12 mol carbon (Tables 1.2, through the action of carbonic anhydrase. The
1.3). If we also assume every molecule of car- carbon dioxide thus available to the carboxyla-
bon dioxide so encountered is successfully taken tion of RuBP is effectively concentrated by a fac-
into the cell, then the requirement to sustain tor of 40.
the doubling of biomass without change in the The more intensively studied CCM of the
internal carbon concentration is a further 0.63 Cyanobacterium Synechococcus also transports and
× 10−12 mol C (cell C)−1 . While the concentration accumulates carbon dioxide and bicarbonate
gradient is maintained, the diffusion rate calcu- ions, achieving concentration factors in the order
lated from Eq. (3.19) is capable of delivering the of 4000-fold (Badger and Gallacher, 1987). Recent
entire carbon requirement to the cell in ∼2300 s work has revealed the mechanism and genetic
(i.e. just over 38 minutes). control of each of four separate uptake path-
For proportionately lower concentrations ways in Synechococcus PCC7942 (Omata et al., 2002:
of carbon dioxide, Eq. (3.19) delivers smaller Price et al., 2002). Two take up CO2 at a rela-
amounts per carbon per unit time and the time tively low affinity, one constitutive and the other
to accumulate the material for the next dou- inducible, involving thylakoid-based dehydroge-
bling is correspondingly extended. A concentra- nase complexes. There is a third, inducible, high-
tion of 0.3 µmol CO2 L−1 could not sustain a affinity bicarbonate transporter (known as BCT-
doubling in less than 1 day, when growth rate 1) that is activated by a cAMP receptor protein
would be considered to be carbon limited. With at times of carbon starvation (a fuller discussion
pH already close to 8.3, the (uncatalysed) dissoci- is included in Section 5.2.1). The fourth mecha-
ation of bicarbonate would support a continuing nism is a constitutive Na+ -dependent bicarbonate
supply of carbon dioxide. When that too became transport system that is selectively activated (per-
exhausted, pH drifts quickly upwards as carbon- haps by phosphorylation).
ate becomes the dominant form of inorganic CCMs represent a remarkable adaptation of
carbon. some (but not all) photosynthetic microorgan-
Many planktic algae avoid (or at least delay) isms to the onset of carbon limitation of pro-
carbon-dioxide limitation and slow bicarbon- duction rates. They are energetically expensive
ate dissociation through resort to a carbon- to operate and, not surprisingly, are invoked
concentration mechanism, or CCM. Although to assist survival and maintenance only under
the kinetic characteristics of the key RUBISCO severe conditions of DIC depletion. The photon
enzyme (especially its high half-saturation cost of fixation of CO2 concentrated by the cell as
requirement for CO2 ) cannot be modified, the opposed to the harvest at equilibration is roughly
CCM provides the means to maintain its activ- doubled (>16 mol photon per mol C fixed) and
ity by concentrating CO2 at the sites of car- the compensation point is raised to around 10
boxylation. Since their function was first recog- µmol photons m−2 s−1 (Raven et al., 2000).
nised (Badger et al., 1980, Allen and Spence, 1981;
Lucas and Berry, 1985), the mechanisms assist- 3.4.3 Species-specific sensitivities to low
ing survival of low-CO2 conditions have contin- DIC concentrations
ued to be intensively investigated. Progress has The differential ability of aquatic plants to utilise
been reported in several helpful reviews (Raven, the inorganic carbon supply in fresh waters has
1991, 1997; Badger et al., 1998; Moroney and been recognised as such for several decades.
Chen, 1998). Working with the green flagellate However, the association of particular types of
Chlamydomonas reinhardtii, Sültemeyer et al. (1991) phytoplankton with particular types of water
showed that the algal CCM involves a series of stretches back almost 100 years, to the days of
ATP-mediated cross-membrane transfers – at the the Wests and the Pearsalls (West and West, 1909;
cell wall, the plasma membrane and the chloro- Pearsall, 1924, 1932) and to the lake classifica-
plast membrane – which transport and concen- tion schemes based on biological metabolism
trate bicarbonate ions as well as carbon diox- devised by Thienemann (1918) and Naumann
ide. Breakdown of the bicarbonate is accelerated (1919). The importance of inorganic nutrients,
nitrogen and phosphorus in particular, in gov- rich in bicarbonate but usually deficient in nitro-
erning aquatic metabolism was quickly and cor- gen and, partly as a consequence of precipita-
rectly appreciated. As the broad correlations tion as hydroxyapatite (calcium phosphate), phos-
detected among indicative types of freshwater phorus. The modest phytoplanktic biomass they
phytoplankton and the metabolic state of lakes carry is, however, often dominated by the species
became developed, particular species or groups of volvocalean green algae, diatoms, dinoflag-
of species became classified as indicators of olig- ellates and bloom-forming species of Anabaena,
otrophic or of eutrophic conditions (Rodhe, 1948; Gloeotrichia or other Cyanobacteria, that Rodhe
Rawson, 1956). Many chrysophyte, desmid and (1948) had associated with nutrient-enriched sys-
certain diatom species were seen to be indicative tems. In describing the sparse phytoplankton
of oligotrophic, phosphorus-deficient conditions of Malham Tarn, situated in the carboniferous
(e.g. Findenegg, 1943). Rodhe (1948) went as far limestone formations of northern England, Lund
as suggesting that phosphorus levels >20 µg P L−1 (1961) remarked that it was ‘quantitatively typi-
may actually have been toxic to chrysophytes. On cal of an oligotrophic lake but qualitatively rep-
the other hand, Cyanobacteria, especially those resentative of a eutrophic one’. In contrast, plank-
species of Anabaena, Aphanizomenon and Microcys- tic elements most indicative of supposedly olig-
tis that became abundant as a consequence of otrophic plankton (including diatoms of the gen-
anthropogenic eutrophication, were believed to era Cyclotella and Urosolenia, such colonial green
express a preference for high-phosphorus condi- algae as Coenochloris, Paulschulzia, Pseudosphaerocys-
tions. tis and Oocystis of the O. lacustris group, desmid
Many of these differences can now be genera such as Cosmarium, Staurastrum and Stau-
explained in terms of the chemistry of carbon rodesmus and chrysophytes that might include
rather than of other nutrients. There is no ques- species of Chrysosphaerella, Dinobryon, Mallomonas
tion that the levels of biomass of phytoplankton and Uroglena) were conspicuously lacking.
that may be sustained in a pelagic system are Moss (1972, 1973a, b, c) conducted an impor-
related to the resources available and that the tant series of experimental investigations of the
amounts of accessible phosphorus or nitrogen or factors influencing the distributions of algae
(in the oceans) iron may well be the biomass- associated with the eutrophication of erstwhile
limiting resource (see Chapter 4). Because car- oligotrophic lakes. He found systematic differ-
bon is unlikely ever to be a capacity-limiting ences in the dynamic responses of algae to vari-
resource and because a large body of litera- able pH and/or variable carbon sources to be
ture projects a weight of experimental evidence more striking than those due to variation in
for species-specific differentials in the uptake the amounts of nutrient supplied (see especially
capabilities and requirements in respect of Moss, 1973a), with clear separation between the
(especially) phosphorus and nitrogen, it is under- characteristically eutrophic species that could
standable that interpretations of species selec- maintain growth at relatively high pH and low
tion in terms of available nutrients should per- concentrations of free carbon dioxide, and the
sist (Reynolds, 1998a, 2000a). In fact, the experi- oligotrophic, ‘soft-water’ species, which could
mental evidence for interspecific differentiation not. Moss’s (1973c) conclusion that the response
among the dynamics of planktic algae on the of natural phytoplankton assemblages to nutri-
basis of their differential abilities to exploit the ent enrichment (‘eutrophication’) is not depen-
supplies of carbon has been to hand for many dent on the principal variant (more or less nutri-
years. Now, detailed biochemical and physiologi- ent) but on the productivity demands on the
cal explanations are available to support the criti- totality of resources.
cal role of carbon in distinguishing ‘oligotrophic’ Talling’s (1976) more detailed experiments
and ‘eutrophic’ assemblages. on the capacity of freshwater phytoplankton
An indicative anomaly is the example of cal- to remove dissolved inorganic carbon from the
careous (marl) lakes set in karstic, limestone water established a series (Aulacoseira subarctica →
upland areas. Their waters are, by definition, Asterionella formosa → Fragilaria crotonensis →
Ceratium hirudinella/Microcystis aeruginosa) of otrophic species have no, or only modest, abilities
increasing tolerance of CO2 depletion and an in this direction. The species of Aulacoseira and
increasing capability of staging large population Anabaena studied by Talling (1976) are intermedi-
maxima under alkaline, CO2 -depleted condi- ate on this scale. Talling’s deduction that the CO2
tions. The work of Shapiro (1990) confirmed system in natural waters ‘plays a large part in
the apparent high carbon affinities of several determining the qualitative composition as well
Cyanobacteria, especially of Anabaena and Micro- as the photosynthetic activity of the freshwater
cystis, which could maintain slow net growth at phytoplankton’ was prophetic.
pH > 10. That the supply of carbon, rather Evidence is accumulating to suggest that the
than any other factor, is limiting under such carbon dioxide system may be similarly selective
high-pH conditions is supported by the fact that in the sea. Normally, upward pH drift in the sea
bubbling with CO2 will restore the growth rate used to be considered unusual. With the excep-
of Microcystis (Qiu and Gao, 2002). tion of Emiliana huxleyi, the formation of whose
On the other hand, Saxby-Rouen et al. (1998; coccoliths was investigated by Paasche (1964), evi-
see also Saxby, 1990; Saxby-Rouen et al., 1996) dence for the ability of marine phytoplankters to
showed convincingly that the chrysophyte Synura use bicarbonate was still lacking as recently as
petersenii is unable to use bicarbonate at all and the mid-1980s (Riebesell and Wolf-Gladrow, 2002).
gave strong indications that species of Dinobryon The investigations of Riebesell et al. (1993) con-
and Mallomonas probably also lack the capability. firmed that certain species of marine diatom
Ball (in Moroney, 2001) has presented evidence (Ditylum brightwellii, Thalassiosira punctigera, Rhi-
that a number of chrysophyte species, includ- zosolenia alata) appear to depend exclusively on
ing Synura petersenii and Mallomonas caudata, lack the diffusive flux of dissolved CO2 . Growth rates
any known kind of carbon-concentrating mech- became carbon limited at DIC concentrations
anism. Lehman (1976) had already shown that below 10–20 µM (0.012–0.024 mg C L−1 ; pH > 8.1)
high phosphorus concentration was no bar to the and stalled completely at <5 µM. The dependence
growth of Dinobryon. Reynolds’ (1986b) manipula- upon diffusive transport and non-catalysed con-
tions of phytoplankton composition in the large version of bicarbonate becomes more problem-
limnetic enclosures in Blelham Tarn (see Section atic among those larger phytoplankters that have
5.5.1), showed that phosphorus was as stimula- a relatively low ratio of surface area to volume,
tory to the growth of chrysophytes (Dinobryon, for the flux to the boundary layer and the nat-
Mallomonas, Uroglena) as to any other kind of phy- ural dissociation of bicarbonate is just too slow
toplankter, provided that the pH did not exceed to compensate the CO2 deficit at the cell surface
8.5. To emphasise the point: neither phosphate in the wake of a high photosynthetic demand. In
nor bicarbonate interferes with the growth of contrast, however, some shelf-water species, such
these chrysophytes, so long as they have access as Skeletonema costatum and Thalassiosira weissflogii,
to free CO2 . show no growth-rate dependence on free-CO2 con-
It is now easy to interpret these various find- centrations, even at pH levels (>8.5) requiring use
ings in the light of understanding about differ- of bicarbonate and/or some method of carbon
ential abilities to exploit the various available concentration (Burkhardt et al., 1999). Whether
sources of DIC. Eutrophic phytoplankters, includ- carbon dioxide or bicarbonate predominates as
ing colonial volvocaleans, many Cyanobacteria the proximal carbon source for the alga is not
and several dinoflagellates, are those that toler- altogether clear. Bicarbonate may be taken up
ate the low free-CO2 conditions of naturally high- and converted to carbon dioxide by the action of
alkalinity lakes. The species found in soft waters the enzyme carbonic anhydrase, hydroxyl being
in which enrichment with nitrogen and phos- excreted to balance the charge, so adding to the
phorus stimulates greater demands on the DIC prevalence of bicarbonate. Carbonic-anhydrase
reserves may well be selected by their ability to activity is also detectable on the outer surface
exploit bicarbonate directly and/or to focus car- of these phytoplankters where the use of bicar-
bon supplies on the sites of carboxylation. Olig- bonate is accelerated, especially in response to
CAPACITY, ACHIEVEMENT AND FATE OF PRIMARY PRODUCTION AT THE ECOSYSTEM SCALE 131
reducing concentrations of free CO2 (Nimer et al., to the point of abandoning chlorophyll pigmen-
1997; Sültemeyer, 1998). However, even the bene- tation. Among the Chromulinales, the resort to
fits that this ability brings are finite, amount- phagotrophy seems very strongly associated with
ing, in effect, to an acceleration of the re- a shortage of nutrients; though normally pig-
establishment of the carbonate system (Riebe- mented, however, cells resorting to bacterivory
sell and Wolf-Gladrow, 2002). Carbonic anhydrase are much paler and show reduced photosyn-
activity is said to reach its peak at CO2 concen- thetic capacity above a threshold prey density
trations of ∼1 µM, (Elzenga et al., 2000) when (references in Geider and MacIntyre, 2002). On
dissociation of bicarbonate is likely to yield not the other hand, some of these species (e.g.
more than 10–20% of the carbon flux occurring Ochromonas) are prominent nanoplanktic bacteri-
at the air equilibrium in sea water. Carbon limi- vores and fulfil a key stage in the microbial loop
tation of photosynthetic assimilation and poten- (see Section 3.5).
tial growth rate of marine phytoplankton is cer- An ability of Chlorococcales to take up glu-
tainly possible and may occur more frequently in cose and other soluble sugars derivatives has
high-production waters than has previously been been inferred or demonstrated on several occa-
acknowledged. sions (Algéus, 1950: Berman et al., 1977; Lewi-
tus and Kana, 1994). Their habitats, which fre-
3.4.4 Other carbon sources quently include organically rich ponds, provide
To be able to fix carbon in photosynthesis is the opportunities for assimilating organic solutes
the abiding property separating the majority but the relationship appears not to be obligate.
of photoautotrophic organisms (‘plants’) from Though sometimes representing a major step in
the majority of phagotrophic heterotrophs (‘ani- the carbon dynamics of ponds, the trait is far
mals’). This division is by no means so obvious from being obligate; algal heterotrophy neverthe-
among the protists where nominally photosyn- less can play an important role in the pelagic car-
thetic algae show capacities to ingest particu- bon cycle of large lakes.
late matter and bacteria as a facultative or a
quite typical feature of their lifestyles. This kind
of mixotrophy is seen among the dinoflagellates 3.5 Capacity, achievement and fate
and certain types of chrysophyte (mostly chromu- of primary production at the
lines). Alternatively, the bacterium-like ability
to absorb selected dissolved organic compounds
ecosystem scale
across the cell surface (‘osmotrophy’) is possessed
among some chlorophyte algae (of the Chlorococ- This section is concerned with the estimation
cales) and in certain Euglenophyta, and among of the capacity of phytoplankton-based systems
cryptomonads (Lewitus and Kana, 1994). to fix carbon, how much of that capacity is
Whereas osmotrophy clearly represents a realised in terms of primary product assembled
means of sourcing assimilable carbon, without and how much of that is, in turn, processed
the requirement for its prior photosynthetic into the biomass of its consumers, including
reduction, mixotrophy is generally regarded as a at higher trophic levels. This is no new chal-
facultative ability to supplement nutrients other lenge, the questions having been implicit in the
than carbon (chiefly N or P) under conditions of earliest investigations of plankton biology. The
nutrient limitation of production (Riemann et al., stimulus to pursue them has varied perceptibly
1995; Li et al., 2000). However, there is a valuable over this period, beginning with the objective
energetic subsidy to be derived too, although the of understanding the dynamics of biological sys-
exploitative opportunity varies (Geider and Mac- tems thitherto appreciated only as steady states.
Intyre, 2002); some marine dinoflagellates are From the 1970s, advances in satellite observa-
said to be ‘voraciously heterotrophic’, ingesting tion have greatly enhanced the means of detect-
other protists. In some lakes, Gymnodinium helvet- ing planetary behaviour and function, while,
icum seems to be predominantly phagotrophic, at the other end of the telescope (as it were),
revolutionary changes in understanding how from a small number of short, representative

fixed carbon is transferred among ecosystem field measurements (i.e. to extrapolate the values
components have greatly enhanced the interpre- NP , NR ). With alternative techniques for
tation of the sophisticated techniques available proxy estimates of biomass and production rates
for their remote sensing. In the 1990s, study of (remote variable chlorophyll fluorometry; Kolber
the fluxes of materials between atmosphere and and Falkowski (1993) and see Section 3.3.4), the
oceanic systems and the means by which they are usefulness of models to relate instantaneous esti-
regulated has acquired a fresh urgency, born of mates to water-column integrals over periods of
the need to understand the nature and potential hours to days is self-evident.
of the ocean as a geochemical sink for anthro- Integral solutions for calculating primary pro-
pogenically enhanced carbon dioxide levels. duction over 24 h were devised some 50 years
These researches help to substantiate the fol- ago. Several, generically similar formulations are
lowing account, even though it is approached available, differing in the detail of the manner in
through a hierarchical sequence of integrals of which they overcome the difficult diurnal inte-
capacity, from those of organelles to metapopula- gration of the light field, especially the diel cycle
tions of plankters stretching across oceans. More- of underwater irradiance in reponse to the day
over, although the productive potential is shown time variation in I 0 and its relationship to Ik .
to be vast, plankton biomass remains generally For instance, Vollenweider (1965) chose an empir-
very dilute and broadly stable, and has remained ical integral to the diel shift in I 0 , employing
so despite recent increases in atmospheric CO2 the quotient, [(0.70 ± 0.07) I 0 max ], where is
partial pressures, and it is also argued that the length of the daylight period from sunrise
the current flux of carbon through this rarefied to sunset. Extrapolation of N P , in respect of
catabolic system is probably as rapid as it can be. the measured N P max (or, rather, an empirically
fitted proportion, [(0.75 ± 0.08) N P max ], that com-
3.5.1 Primary production at the local scale pensates for the proportion of the day when I 0
Subject only to methodological shortcomings < Ik ) invokes the coefficient of underwater light
and the free availability of exploitable inorganic extinction (ε). The Vollenweider solution is:
carbon, the most useful indicator of the poten-
NP = (0.75 ± 0.08)NP max ×
tial primary production, Pg (or Pn , sensu Pg − Ra )
(see Section 3.3.2) comes from the areal integra- × ln([0.70 ± 0.07]I 0 max /0.5I k ) × 1/ε
tion of the instantaneous measurements of pho- (3.20)
tosynthetic rate ( N P , in mg C fixed m−2 h−1 ) Talling (1957c) had earlier tackled the integration
(see Section 3.3.1). The productivity, sensu produc- problem by treating the daily light income as
tion per unit biomass, measured in mg C fixed a derivative of the daytime mean intensity, I 0 ,
(mg biomass C)−1 h−1 , is a valuable comparator. applying over the whole day (). He expressed
However, for periods relating to the recruitment I 0 in units of light divisions (LD), where LD =
of new generations, it is helpful to measure (or to ln(I 0 max /0.5Ik ) / ln2), having the dimensions of
extrapolate) carbon uptake over longer periods, time. The daily integral (LDH) is the product of
comparable, at least, with the generation time LD and , approximating to:
required for cell replication to occur. This gen-
LDH = [ln(I 0 max /0.5I k )/ln 2] (3.21)
erally means designing experimental exposures
of 12 or 24 h. Such designs increase the risk of The completed Talling solution, equivalent to Eq.
measurement error (through depletion of unre- (3.20) is:
plenished carbon, possible oxygen poisoning and
NP =ln 2 × NPmax × × L D H × 1/ε
the increasing recycling of ‘old’ carbon; see Sec-
tion 3.3.2). These problems may be overcome by (3.22)
mounting contiguous shorter experiments (see Numerically, the two solutions are similar.
the notable example of Stadelmann et al., 1974) Applied to the data shown in Fig. 3.3, for exam-
but it is generally desirable to integrate results ple, where N = 47.6 mg chla m−3 , Pmax = 2.28 mg
O2 (mg chla)−1 h−1 , = 10.8 h, I 0 max = 800 and about 0.564 g C m−2 d−1 . Relative to the producer
0.5 Ik = 24 µmol photons m−2 s−1 ; and suppos- population (4.8 m × 47.6 mg chla m−3 ) this repre-
ing ε = 1.33 (ε min ) = 1.33 (ε w + εp + Nε a ), where sents some 2.47 mg C (mg chla)−1 d−1 , or around
(εw + εp ) = 0.422 m−1 and ε a = 0.0158 m2 (mg 0.05 mg C (mg cell C)−1 d−1 .
chla)−1 , NP is solved by Eq. (3.20) to between Such estimates of local production are the
1531 and 2020 mg O2 m−2 d−1 . For the Talling basis of determining the production of given
solution, the daily mean integral irradiance (406 habitats (Pn ), the potential biomass- (B-)specific
µmol photons m−2 s−1 ) is used to predict N P productive yields (Pn /B) and the organic carbon
= 2119 mg O2 m−2 d−1 . For comparison, interpo- made available to aquatic food webs. The pro-
lation of the profiles represented in Fig. 3.9 sum- ductive yield may sometimes seem relatively triv-
mates to approximately 2057 mg O2 m−2 d−1 . ial in some instances. Marra (2002), reviewing
Most of the variations in the other integra- experimental production measurements, showed
tive approaches relate to the description of the daily assimilation rates in the subtropical gyre of
light field. Steel’s (1972, 1973) models used a pro- the North Pacific in the order of 6 mg C m−3 d−1 .
portional factor to separate light-saturated and However, this rate was light saturated to a
sub-saturated sections of the P vs. I curve. Tak- depth of 70 m. Positive light-limited photosyn-
ing advantage of advances in automated serial thetic rates were detected as deep as 120 m.
measurements of the underwater light field, A. E. Thus, area-integrated day rates of photosynthesis
Walsby and his colleagues have devised a means (∼570 mg C m−2 d−1 ) could be approximated that
of estimating column photosynthesis at each iter- are comparable to those of a eutrophic lake. How-
ation and then summating these to gain a direct ever, the concentration of phytoplankton chloro-
estimate of NP (Walsby, 2001) (see Section phyll (∼0.08 mg chla m−3 through much of the
3.3.3). upper water column reaching a ‘maximum’ of
We may note, at this point, that all these 0.25 mg chla m−3 near the depth of 0.5 Ik ) indi-
estimates of gross production (Pg ; on a daily cates chlorophyll-specific fixation rates in the
basis, the equivalent of NP ) need correc- order of only 60 mg C (mg chla)−1 d−1 . Carbon-
tion to yield a useful estimate of potential specific rates of ∼1.2 mg C (mg cell C)−1 d−1 are
net production (Pn = Pg − Ra , as defined in indicated. This is more than enough to sustain a
Section 3.3.2). Again, taking the example from doubling of the cell carbon and, in theory, of the
Fig. 3.3, we may approximate NR as the population of cells in the photic layer. It is curi-
product, 24 h × H × NR , where mean H ous, not to say confusing, that eutrophic lakes
= 4.8 m, R = 0.101 mg O2 (mg chla)−1 h−1 and are often referred to as being ‘productive’ when
N = 47.6 mg chla m−3 to be NR ∼ 554 mg ultraoligotrophic oceans and lakes are described
O2 m−2 d−1 . The difference with Pg gives the daily as ‘unproductive’. This may be justified in terms
estimate of Pn (strictly, we should distinguish it of biomass supported but, taking (Pn /B) as the
as NP n ): index of productivity, then the usage is diamet-
rically opposite to what is actually the case.
NPn = NP − NR In areal terms, reports of directly mea-
= 2057 − 554 mg O2 m−2 d−1 sured productive yields in lakes generally range
= 1.503 g O2 m−2 d−1 . (3.23) between ∼50 mg and 2.5 g C m−2 d−1 (review of
Jónasson, 1978). These would seem to embrace
Since many of these approaches were developed, directly measured rates in the sea, according
it has become more common, and scarcely less to the tabulations in Raymont (1980). Estimates
convenient, to measure, or to estimate by anal- of annual primary production run from some
ogy, photosynthetic production in terms of car- 30–90 g C fixed m−2 a−1 (in very oligotrophic,
bon. Supposing a photosynthetic quotient close high-latitude lakes and the open oceans that
to 1.0 (reasonable in view of the governing condi- support producer biomass in the order of 1–5
tions – see Section 3.2.1), the net carbon fixation mg C m−3 through a depth of 50–100 m; or,
in the example considered might well have been say, ≤500 mg C m−2 ), to some 100–200 g C fixed
m−2 a−1 (in more mesotrophic systems and shelf absorption and fluorescence attributable to the
waters able to support some 50–500 mg producer phytoplankton (Geider et al., 2001). To do this
C m−3 through depths between 15 and 30 m; i.e. with confidence requires methodological calibra-
∼15 g C m−2 ), and to 200–500 g C m−2 a−1 among tion and the application of interpolative produc-
those relatively eutrophic systems sustaining the tion models (discussed in detail in Behrenfeld
fixation of enriched lakes and upwellings, capa- et al., 2002). Most of the latter employ traditional
ble of supporting perhaps 1500–5000 mg C m−3 functions (such as those reviewed in the previ-
through depths of only 3–10 m i.e. ∼15 g C m−2 . ous section); calibration is painstaking and pro-
This series is rather imprecise but it serves tracted but the remakable progress in interpret-
to illustrate the fact that although the supporting photosynthetic properties of phytoplankton
ive capacity of pelagic habitats probably varies has led to synoptic mapping both of the distri-
over 3 or 4 orders of magnitude, the annual pro- bution of phytoplankton at the basin mesoscale
duction that is achieved varies over rather less and of analogues of the rate of its carbon fixa-
than 2 (maximum annual fixation rates of up tion (Behrenfield and Falkowski, 1997; Joint and
to 800–900 g C m−2 are possible in some shallow Groom, 2000; Behrenfield et al., 2002).
lakes and estuaries). This is mainly because the The beauty, the global generality and the
exploitation of high supportive capacity results simultaneous detail of such imagery are awe-
in a diminished euphotic depth. There is thus an some. However, its scientific application has been
asymptotic area-specific maximum capacity and, first to confirm and to consolidate the previous
on many occasions, a diminishing productivity, generalised findings of biological oceanographers
sensu production rate over biomass (see also Mar- (e.g., Ryther, 1956; Raymont, 1980; Platt and
galef, 1997). Sathyendranath, 1988; Kyewalyanga et al., 1992;
see also Barber and Hilting, 2002, for review).
In essence, the main oceans (Pacific, Atlantic,
3.5.2 Primary production at the Indian) are deserts in terms of producer biomass
global scale (<50 mg chla m−2 ) while net primary produc-
The rapid advances in, on the one hand, air- tion is assessed to be generally <200 g C fixed
borne and, especially, satellite-based remote- m−2 a−1 . In the high latitudes, towards either
sensing techniques and, on the other, the tech- pole, biomass and production tend to be more
niques and resolution for analysing the signals seasonal, with maximum production in the six
thus detected, have verified and greatly ampli- summer months and least in winter. The great-
fied our appreciation of the scale of global net est annual aggregates (200–500 g C fixed m−2 a−1 )
primary production (NPP). In barely 20 years, the are detected mainly on the continental shelves.
capability has moved from qualitative observa- Production ‘hotspots’ (500–800 g C fixed m−2 a−1 )
tion, to remote quantification of biomass from are located in particularly shallow areas (e.g.,
the air (Hoge and Swift, 1983; Dekker et al., 1995) the Baltic Sea, the Sea of Okhotsk), in shelf
and on to the detection of analogues of the rate waters receiving nutrient-rich river outfalls (the
of its assembly and dissembly (Behrenfeld et al., Yellow Sea, the Gulf of St Lawrence) and in the
2002). The newest satellite techniques can pro- upwellings of major cold currents (e.g., the Peru,
vide the means to gather this information in a around Galápagos; and the Benguela, Gulf of
single overpass. For terrestrial systems, NPP is Guinea).
gauged from the light absorption by the plant Satellite remote sensing has also helped to
canopy (APAR, absorbed photosynthetically active improve the resolution of global NPP aggregates
radiation) and an average efficiency of its util- and their relative contribution to the global car-
isation (Field et al., 1998). For aquatic systems, bon cycle. The estimates of total oceanic NPP,
sensors are needed to derive the rate of under- based on imagery, converge on values of around
water light attenuation (from which the magni- 45–50 Pg C a−1 . It is interesting that previous esti-
tude of the photosynthetically active flux den- mates, all based on summations and extrapola-
sity at depth is estimable) and the rates of light tions of various in-situ measurements, are mostly
Table 3.3 Annual net primary production (NPP) of various parts of the sea and
of other major units of the biosphere
Energy
invested
NPP (Pg C) ( J × 10−18 )
Marine domains
Tropical/subtropical trades 13.0 509
Temperate westerlies 16.3 638
Polar 6.4 250
Coastal shelf 10.7 419
Salt marshes, estuarine 1.2 47
Coral reef 0.7 27
Total 48.3 1890
Terrestrial domains
Tropical rainforests 17.8 697
Evergreen needleleaf forest 3.1 121
Deciduous broadleaf forest 1.5 58
Deciduous needleleaf forest 1.4 54
Mixed broad- and needleleaf forest 3.1 121
Savannah 16.8 658
Perennial grassland 2.4 94
Broadleaf scrub 1.0 39
Tundra 0.8 31
Desert 0.5 19
Cultivation 8.0 313
Total 56.4 2205
Source: Based on Geider et al. (2001), using data of Longhurst et al. (1995)
and Field et al. (1998).
within about 50% of this (20–60 Pg C a−1 : see per million by volume, or 0.2 g C m−3 ). Signifi-
Barber and Hilting, 2002). Only Riley’s (1944) esti- cant natural abiotic exchanges of carbon with
mate of 126 Pg C a−1 , based on oxygen exchanges, the atmosphere include the removal due to car-
now seems exaggerated. bonate solution and silcate weathering (∼0.3 Pg
It is interesting to compare the estimates for C a−1 ) but this is probably balanced by releases
various parts of the ocean and with other major of CO2 through calcite precipitation, carbon-
biospheric units (see Table 3.3). Shelf waters con- ate metamorphism and vulcanism (Falkowski,
tribute nearly a quarter of the total oceanic 2002). As is well known, however, ambient atmo-
exchanges despite occupying less than about 1/20 spheric carbon dioxide concentration is currently
of the area of the seas. Nevertheless, marine pho- increasing. This is generally attributed to the
tosynthesis is responsible for just under half the combustion of fossil fuels (presently around 5.5 ±
global NPP of about 110 Pg (or 1.1 ×1017 g) C a−1 . 0.5 Pg C a−1 and rising) but the oxidation of ter-
Much of this carbon is recycled in respiration restrial organic carbon as a consequence of land
and metabolism and reused within the year (see drainage and deforestation also makes significant
also Section 8.2.1). Net replenishment of atmo- contributions (some 1.6 ± 1.0 Pg C a−1 : data from
spheric carbon dioxide would contribute a steady- Sarmiento and Wofsy, 1999, as quoted by Behren-
state concentration (currently around 370 parts feld et al., 2002). In spite of this anthropogenic
annual addition to the atmosphere of ∼7 Pg C, tion by cloud and backscattering by dust, it is

the present annual increment is said to be ‘only’ unrealistic to expect much more than 30% of
3.3 (±0.2) Pg C a−1 ; say ∼0.6% a−1 relative to an this to be available to photosynthetic organisms
atmospheric pool of ∼500 Pg). The ‘deficit’ (∼3.8 (say 3.6 GJ m−2 PAR). Using the relationships con-
Pg C a−1 ) is explained, in part, by a verified dis- sidered in Section 3.2.3, the potential maximum
solution of atmospheric CO2 into the sea and, in yield of fixed carbon is 3.6 GJ m−2 /470 kJ (mol C
part, by transfers of organic and biotic compo- fixed)−1 , i.e. in the order of 7 kmol m−2 a−1 , or
nents of unverified scale. about 85 kg C m−2 a−1 . This is about 100 times
Considering that it seems more painful (in the greater than measured or inferred optima. Even
short term) to cut anthropogenic carbon emis- if we start with something a little less optimistic,
sions, there is currently a great deal of interest such as the PAR radiation measured at the tem-
in augmenting net annual flux of carbon to the perate latitude of the North Sea (annual PAR
oceanic store of dissolved carbon of around 40 × flux ∼1.7 GJ m−2 ), we still find that no more than
1018 g C (Margalef, 1997). about 1.5% of the energy available is trapped into
the carbon harvest of primary production.
3.5.3 Capacity Looking at the issue in the opposite direction,
The achievement of even 48.3 Pg across some 360 the minimum energy investment in the carbon
× 106 km2 of ocean (note, areal average ∼133 g C that is harvested (Table 3.3) by planktic primary
fixed m−2 a−1 ) should be viewed against the typ- producers is usually a small proportion of that
ically small active photosynthetic standing-crop which is available. To produce 50–800 g C m−2 a−1
biomass (on average, perhaps just 1–2 g C m−2 , requires the capture of a minimum of 9–144 mol
Pn /B ∼ 100). The supportive potential of this photons m−2 a−1 , equivalent to 2–32 MJ m−2 a−1 ,
conversion is certainly impressive but it repre- or less than 2% of the available light energy.
sents poor productive yield against the potential There are many reasons for this apparent inef-
yield were every photon of the solar flux cap- ficiency. The first consideration is the areal den-
tured and its energy successfully transferred to sity of the LHCs needed to intercept every photon.
carbon reduction. It is possible that the rate is This can be approximated from the area of the
constrained; or that the rate is rapid but the individual centre, the (temperature-dependent)
producer biomass is constrained; or, again, that length of time that the centres are closed follow-
other processes impinge and the productive yield ing the preceding photon capture and, of course,
is constrained. the photon flux density. It was argued, in Sec-
These are problems of capacity – how rapidly tion 3.2.1, that the area of a PSII LHC (σ X ) cov-
can production proceed before the metabolic ers about 10 nm2 . Thus, in theory, it requires
losses might balance the productive gains or a minimum of 1017 LHCs to cover 1 m2 . It was
before the assembly of biomass might become, also shown that 1 g chlorophyll could support
subject only to the supply of the material com- between 2.2 and 3.4 ×1018 LHCs and, so, might
ponents. The problem will not be addressed fully project an area of ∼22 m2 and harvest photons
before the physiology of resource gathering and over its entire area. At very low flux densities, the
growth have been addressed (Chapters 4 and 5). intact LHC network (0.045 g chla m−2 ) has some
For the present, we will examine the constraints prospect – again, in theory – of intercepting all
of insolation and carbon flux on the biomass of the bombarding photons but, as the flux density
phtoplankton that may be present. is increased, it is increasingly likely that photons
Commencing with the productive capacity will fall on closed centres and be lost. Saturation
of the solar quantum flux to the top of the of the network (Ik ) is, in fact predicted by:
atmosphere (see Fig. 2.2), the income of 25–45
I k = (σX tc )−1 photons m−2 s−1 (3.24)
MJ m−2 d−1 provides the basis for a theoretical
annual income of 12–13 GJ m−2 annually. Tak- where tc is the limiting reaction time (in s
ing only the PAR (46%) and subjecting the flux photon−1 ). At 20 ◦ C, when tc ∼ 4 ×10−3 s (see Sec-
to absorption in the air by water vapour, reflection 3.2.1), Ik is solved from Eq. (3.24) as (10 ×
10−18 m2 × 4 × 10−3 s photon−1 )−1 , i.e. about 2.5

× 10−19 photons m−2 s−1 , or ∼42 µmol photon
m−2 s−1 .
With supersaturation, more of the photons
pass the network or bounce back. Increasing the
areal density of LHCs increases the proportion
of the total flux intercepted but, in the oppo-
site way, the increasing overlap of LHC projection
means that individual LHCs are activated less
frequently.
There is no light constraint on the upper limit
of light-intercepting biomass surface per se, but
the cost of maintaining underemployed photo-
synthetic apparatus and associated organelles is
significant. Ultimately, the productive capacity is
set not by the maximum rate of photosynthe-
sis but by the excess over respiration. Reynolds
(1997b) developed a simple model of the capac-
ity of the maximum photon flux to support
Figure 3.18 Comparison of the light-harvesting potential
the freshwater unicellular chlorophyte Chlorella.
of phytoplankton as a function of biomass, compared with the
The alga was chosen for its well-characterised
energetic costs of respiration and maintenance. The
growth and photosynthetic properties. The ‘max- maximum supportable biomass is that which continues to fix
imum’ set was 12.6 MJ m−2 d−1 , based upon the just enough carbon in photosynthesis to offset its
flux at the summer solstice at a latitude of 52 maintenance requirements. Redrawn with permission from
◦
N and supposing a dry, cloudless atmosphere Reynolds (2002b).
throughout (equivalent daily photon flux, 57.6
mol). Higher daily incomes occur in lower lati-
tudes but the small shortfall in the adopted value an areal concentration of 1/(20.7 m2 (mol cell
need not concern us here). C)−1 ) = 0.0483 mol cell C m−2 (or ∼0.58 g C, or
The available light energy is represented by about 11 mg chla m−2 ). Above this threshold,
the vertical axis in Fig. 3.18. The temperature is the cell-specific carbon yield decreases while the
assumed to be invariable at 20 ◦ C. A relatively cell-specific maintenance remains constant. Thus,
small numbers of cells of Chlorella (diameter 4 for increasingly large populations, total main-
µm, cross-sectional area 12.6 µm2 , carbon con- tenance costs increase absolutely and directly
tent 7.3 pg C or 0.61 × 10−12 mol C cell−1 , carbon with crop size; total C fixation also increases but
specific projection 20.7 m2 (mol cell C)−1 ) are now with exponentially decreasing efficiency. While
introduced into the light field. They begin to har- the daily photon flux to the lake surface remains
vest a small fraction of the flux of photons. At unaltered, the asymptote is the standing crop
low concentrations, the cells have no difficulty which dissipates the entire 12.6 MJ m−2 d−1 with-
in intercepting the light. At higher concentra- out any increase in standing biomass (somewhere
tions, cells near the surface will partly shade out to the right in Fig. 3.18).
those beneath them. Even if the whole popula- Taking the basal respiration rate of Chlorella
tion is gently mixed, the probability is that cells at 20 ◦ C to be ∼1.1 × 10−6 mol C (mol cell C)−1
will experience increasingly suboptimal illumi- s−1 (Reynolds, 1990), the energy that would be
nation and approach a condition where a larger consumed in the maintenance of the biomass
aggregate portion of each day is passed by each present is supposed to be not to be less than
cell in effective darkness. The onset of subopti- 0.095 mol C (mol cell C)−1 d−1 . To calculate the
mal absorption begins when there is a proba- maximum biomass sustainable on a photon flux
bilistic occlusion by the Chlorella canopy, i.e. at of 57.6 mol photon m−2 d−1 , allowance must be
made for the quantum requirement of at least 8 retical phytoplankton concentration that would
mol photons to yield 1 mol C and for the frac- generate these light conditions is demonstrably
tion of the visible light of wavelengths appropri- equivalent to 650 mg chla m−3 and that it would
ate to chlorophyll excitation (∼0.137: Reynolds, absorb some 97% of the incoming light. However,
1990, 1997b). Then the daily photon flux required if the mixing extends to 10 m, Nεa = 4.7, the
to replace the daily maintenance loss is equiva- chlorophyll a (470 mg m−2 ) can now have a max-
lent to 0.095 × 8/0.137 = 5.55 mol photons (mol imum mean concentration of only 47 mg m−3 ,
cell C)−1 d−1 . This is the slope of the straight line accounting for 70% of the light absorbed. If mix-
inserted in Fig. 3.18. It follows that the maximum ing extends 30 m, Nε a = 0.7 and the chlorophyll
sustainable population is that which can har- a capacity (70 mg m−2 ) can have a concentration
vest just enough energy to compensate its respi- of no more than 2.3 mg chla m−3 , which absorbs
ration losses under the proposed conditions, i.e. no more than 10% of the incoming light.
57.6/5.55 = 10.4 mol cell C m−2 (∼125 g C m−2 ). Finally, capacity may also be approximated
Not surprisingly, there is little evidence that from the integral equations for primary produc-
quite such high levels of live biomass are tion and respiration. Taking the Vollenweider Eq.
achieved, much less maintained, in aquatic envi- (3.20) for NP , for instance, capacity is deemed
ronments, although it is not unrepresentative to be filled when it is precisely compensated
of tropical rainforest. There the true producer by respiration, i.e. when NP = NR , and
biomass of up to 100 g C m−2 is massively aug- N P / N R = 1. Given that N R is equiv-
mented by 10–20 kg C m−2 of biogenic necromass alent to the product, 24 h × H × N R , we may
(wood, sclerenchyma, etc.: Margalef, 1997). In approximate that:
enriched shallow lakes, phytoplankton biomass is
[0.75 NPmax × × ln (0.70 I 0 max /0.5I k )
frequently found to attain areal concentrations
equivalent to 600–700 mg chla m−2 (Reynolds, ×1/ε]/[24 h × H × NR] = 1
1986b, 2001a), more rarely 800–1000 mg chla m−2 Whence, supposing ε = (εw + εp ) + N εa ,
(Talling et al., 1973; i.e., 30–50 g C m−2 ). This rel-
ative poverty is due, in part, to the markedly N = (1/εa )[0.75 (P max /R ) × (/24)
sub-ideal and fluctuating energy incomes that × ln(0.70I 0 max /0.5I k )
water bodies actualy experience. Moreover, fre-
× (1/H ) − (εw + εp )] (3.25)
quent, weather-driven extensions of the mixed
layer determine a lower carrying capacity for the The general utility of this equation is realised in
entrained population (Section 3.3.3). two ways. First, it can be used to gauge the annual
Light absorption by water is a powerful detrac- variation in photosynthetic carrying capacity of
tion from the potential areal carrying capacity a water body of known intrinsic light-absorbing
that increases with the depth of entrainment. properties (ε w + εp ) and known seasonal varia-
If we solve Eq. (3.17), for instance, against nom- tion in mixed-layer depth (hm in substitution of
inated values for I0 = 1000 and for Im = 1.225 H). In the application and spreadsheet solution
µmol photons m−2 s−1 , the mixed-layer integral of Reynolds and Maberly (2002), (Pmax /R) was set
is I∗ = 35 µmol photons m−2 s−1 , which is suf- at 15 and εa at 0.01 m2 (mg chla)−1 .
ficient to impose a significant energy constraint Second, the equation has been used by
on further increase in biomass. From Eq. (3.11), Reynolds (1997a, 1998a) to illustrate the impact of
we can work out that the coefficient of attenua- the depth of convective mixing on phytoplankton
tion equivalent to diminish 1000 down to 1.225 carrying capacity (reproduced here as Fig. 3.19).
µmol photons m−2 s−1 is equivalent to εz = –6.7. Against axes of mixed depth (hm ) and background
Now, supposing that this extinction occurs in a light extinction (ε w + ε p , with a minimum set,
mixed layer extending through just the top metre arbitrarily, for fresh waters at 0.2 m−1 ) and sup-
(z = 1), we may deduce from Eq. (3.14) that ε = posing = 12 h, Fig. 3.19 shows graphically
6.7 m−1 = (ε w + ε p ) + Nε a . Putting (εw + ε p ) = how the maximum chlorophyll-carrying capacity
0.2 m−1 and ε a = 0.01 m2 (mg chla)−1 , the theo- is diluted by mixing, from ∼150 mg chla m−3 in
low (Talling, 1966; Pollingher and Berman, 1977;

Peterson, 1978; Hecky and Fee, 1981). Zero net
gains in biomass relative to photosynthesis (i.e.
carbon fixation is balanced or exceeded by
net losses: Reynolds et al., 1985) are observable
when plankter growth is resisted by the total
exhaustion of one or other of the essential
nutrients. The relatively stable biomass of low-
latitude oceanic phytoplankton, in spite of pos-
itive carbon-fixation rates (Karl et al., 2002), con-
forms to the latter diagnosis. The general pattern
is that the coupling of net growth to photosyn-
thesis is closest in well mixed, light-limited and
nutrient-replete (but possibly carbon-deficient)
water columns and weakest under conditions
characterised either by stratification, or light sat-
uration to a substantial depth, or extreme nutri-
Figure 3.19 Chlorophyll-carrying capacity (as µg chla L−1 )
ent limitations, or any combination of these.
of water columns as a function of the depth to which they are
That there should be a gap between the
mixed (hm ) and the background coefficient of light extinction
due to colour and suspended tripton (εw + εp ). Solutions amounts of carbon fixed and those eventually
assume that Pmax /R = 15 and day length is 12 h. Redrawn constituting new biomass is not in itself sur-
with permission from Reynolds (1997a). prising, neither is the relative magnitude of
the difference. It was once supposed that the
shortfall was explicable in terms of mortalities
of producer biomass, chiefly to settlement and
to consumers and pathogens (Jassby and Gold-
a 10-m layer to ∼20 mg chla m−3 in a 40-m mixed
man, 1974a). Estimating loss rate of biomass was
layer and to <1 mg chla m−3 in a layer mixed
not then well-advanced but the magnitude of
to 80 m. Steel and Duncan (1999) developed a
biomass losses necessary to explain productive
similar model to emphasise the advantages of
shortfalls of this order is, on purely intuitive
destratifying the eutrophic Thames Valley Reser-
grounds, unrealistically large. It was another per-
voirs supplying London in order to lower their
ceptive analysis (Forsberg, 1985) that pointed out
plankton carrying capacity below that of the
that, in study after study, the alleged loss of
nutrients.
biomass was so close to the measured photosyn-
thetic gain that perhaps the ‘lost biomass’ had
3.5.4 Photosynthetic yield to the planktic never been formed in the first place. Instead, the
food web losses of fixed carbon are predominantly phys-
The purpose of this section is to comment on iological (for instance, through enhanced respi-
some aspects of the fate of photosynthetic prod- ration), as had indeed been suggested by both
ucts at the local and global level. The profound Talling (1984) and Tilzer (1984).
impact that they exert on carbon cycling in The fate and allocation of what, to the pho-
plankton-based aquatic systems is also addressed. tosynthetic microorganism, is excess, unassimil-
The preceding sections show a wide variation in able and mainly unstorable photosynthate, has
the eventual allocation of the carbon fixed in taken a little longer to diagnose. Certainly, a
photosynthesis. Well corroborated in the liter- proportion is respired directly, or is fully pho-
ature, the range extends from some 92% to torespired to carbon dioxide and water. It had
95% investment in new biomass (Talling, 1957c; already been clear for over a decade, however,
de Amezaga et al., 1973; Knoechel and Kalff, that a proportion is excreted as DOC, espe-
1978; Geider and Osborne, 1992) to disparately cially when cells are stressed by high insolation
or depleted nutrititive resource fluxes (Fogg, referred to as the ‘microbial food web’ (Sherr and
1971; Sharp, 1977). The circumstances of gly- Sherr, 1988) or, perhaps, as the ‘oligotrophic food
colate excretion, in particular, had been diag- web’. Its short-cutting by direct herbivory is then
nosed (Fogg, 1977), well before the circumstances to be seen to be the luxurious exception, possible
of its regulated production through the acceler- only when a threshold of relative abundance of
ated oxygenase activity of the RUBISCO enzyme nutrient resources is surpassed, sufficient to sus-
(see Section 3.2.3) and its beneficial role in pro- tain a ‘eutrophic food web’ (Reynolds, 2001a) (see
tecting against photooxidative stress had been also Section 8.2.4).
elucidated (Geider and MacIntyre, 2002). Many For the present discussion, it is the mecha-
other organic compounds are now known to be nisms and pathways of phototrophically gener-
released by algae into the water, often in solu- ated carbon that are of first interest. Progress has
tion but not necessarily all to do with metabolic been hampered somewhat by an insufficiency of
homeostasis. They include monosaccharides, car- information about the identities of the main bac-
bohydrate polymers, carboxylic acid and amino terial players and their main organic substrates.
acids (Sorokin, 1999; Grover and Chrzanowski, Knowing which ‘bacteria’ use what sources of
2000; Søndergaard et al., 2000). ‘DOC’ sources is essential to the ecological inter-
Though they may be unusable and unwanted pretation of the behaviour of pelagic systems.
by the primary producers, at least in the imme- Until quite recently, the identification of bac-
diate short term, these organic solutes provide teria relied upon shape recognition, stain reac-
a ready and exploitable resource to pelagic bac- tion and substrate assay. Now, microbiology is
teria (Larsson and Hagström, 1979; Cole, 1982; adopting powerful new methods for the isolation
Cole et al., 1982; Sell and Overbeck, 1992). The of nucleic acids (DNA and especially the 16S or
existence of bacteria in the plankton, both free- 23S ribosomal RNA), their amplification through
living and attached to small mineral and detri- polymerase chain reaction (PCR) and their match-
tal particles in suspension, has for long been ing to primers specific to particular bacterial
appreciated but, for many years, their role in taxa. The approach is similar for both marine
doing much more than recycling organic detritus and freshwater bacterioplankton (good examples
and liberating inorganic nutrients was scarcely of each are given by Riemann and Winding, 2001;
appreciated. Interest in the ability of bacteria to Gattuso et al., 2002). Methods of enumeration
assimilate the organic excretory products of pho- have advanced to routine automated counts by
toautotrophs increased rapidly with the realisa- flow cytometry. The use of highly fluorescent
tion that a large part of the flux of photosyn- nucleic acid stains makes for rapid and easy cyto-
thetically fixed carbon is passed to the food web metric determination of microbe abundance and
through a reservoir of DOC rather than through size distribution in simple bench-top apparatus
the direct phagotrophic activity of metazoans (Gasol and del Giorgio, 2000). In consequence,
feeding on intact algal cells (Williams, 1970, the knowledge of the composition, abundance
1981; Pomeroy, 1974). It soon emerged that the and dynamics of planktic microorganisms is now
chain of consumption of bacterial carbon – nor- developing rapidly.
mally by phagotrophic nanoflagellates, then suc- In both lakes and the seas, the free-living
cessively ciliates, crustacea and plantivorous fish heterotrophic bacteria occur in the picoplanktic
– resulted in the transfer of carbon to the higher size range (0.2–2 µm; <4 µm3 ), which they share
trophic levels. This alternative to the more tra- with the photoautotrophic synechococcoids and
ditional view of the pelagic alga → zooplank- eukaryote picophytoplankton (in lakes) and coc-
ton → fish food chain soon became known as coid prochlorophytes (in the open ocean). Other
the ‘microbial loop’ (Azam et al., 1983). Indeed, it key participants in the oceanic microbial food
now seems that the ‘loop’ is often the only viable webs (viruses, protists) are also prominently rep-
means by which diffusely produced organic car- resented in those of lakes. These structural sim-
bon can be exploited efficiently by the fauna of ilarities between marine and freshwater micro-
resource-constrained pelagic systems. It is better bial communuties suggest that they both have
Table 3.4 Typical (upper) densities of bacterioplankton in lakes and seas and daily production rates.
Standing population Standing biomass Production

(× 10−6 mL−1 ) (mg C m−3 ) (mg C m−3 d−1 )
Deep (>800 m) tropical 0.01–0.02 <0.4 <0.02
oceanic waters
Surface tropical oceans 0.1–0.4 2–6 2–10
Oligotrophic lakes 0.5–0.8 8–15 4–12
Antarctic waters (summer) 1–2 20–60 ?
Temperate ocean 1–2 20–70 10–50
Mesotrophic lakes 1–3 20–90 10–70
Inshore waters, estuaries 1.5–3 40–90 20–130
Oceanic coastal upwellings 2–5 90–500 40–150
Eutrophic lakes 3–8 150–500 70–150
Hypertrophic lakes, polluted 5–40 500–2500 120–700
lagoons
Source: Based on Sorokin, 1998.
ancient and, possibly, common origins. The typi- primary production (GPP) rate of ∼230 mg C m−2
cal cell concentrations present in either broadly d−1 was considerably exceeded by bacterial respi-
fit within 2 orders of magnitude (105 –107 mL−1 ), ration rate (1740 mg C m−2 d−1 ) but with little
although pronounced seasonal variation is often change either to the bacterial abundance or to
detectable, depending upon temperature, the the DOC pool (indicating almost no biomass accu-
abundance of organic substrate and the inten- mulation and complete pelagic cycling of CO2 ).
sity of bacterivorous grazing. It is clear that num- Significantly, an upwelling event, with a conse-
bers generally reflect the trophic state and they quent pulse of nitrogen, stimulated GPP (to over
are responsive to enhanced primary-producer 900 mg C m−2 d−1 ) and to the net recruitment
activity (Nakano et al., 1998), especially in the through growth of phytoplankton, leading to an
wake of phytoplankton ‘bloom’ periods (Coveney enhanced biomass of larger species ‘exportable’
and Wetzel, 1995; Sorokin, 1999; Ducklow et al., as food or in the sedimentary flux. The bacter-
2002) but there is no constant proportionality. ial biomass decreased, relatively and absolutely,
Indeed, relative to algal biomass, bacterial num- presumably in response to the rerouting of photo-
bers (104 –107.6 mL−1 : Vadstein et al., 1993; Sorokin, synthetic carbon. Where conditions are typically
1999) (see also Table 3.4) diminish with higher more eutrophic, supporting biomasses of >2.5
nutrient availability (Weisse and MacIsaac, 2000). mg algal C L−1 , bacterial mass may scarcely
Using relationships resolved by Lee and Fuhrman exceed 0.5 mg C L−1 .
(1987) for marine bacterioplankton, heterotrophs The species structure of the bacterioplank-
and photoautotrophs may each account for ton is highly varied and, collectively, includes
(roundly) up to 0.1 mg C L−1 of the phytoplank- species capable of oxidising substrates as varied
ton biomass of substantially oligotrophic sys- as carbohydrates, various hydrocarbons, proteins
tems. Here, bacterial activity may exceed that and lipids (Perry, 2002). Presumably, the numbers
of algae (Biddanda et al., 2001) and, thus, make of the particular types fluctuate in response to
relatively the greatest contribution to organic- substrate supply and to grazing: distinct species
matter cycling. In another recent study of the ‘successions’, in time and in space, have been
carbon flux through the oligotrophic microbial demonstrated as the community composition
community of the Bay of Biscay ([chla] < 0.7 mg responds to the dissipation of substrate pulses
m−3 ), González et al. (2003) found that a gross moving through linear mainstem reservoirs (see
especially Šimek et al., 1999). Interestingly, how- the producers retain proportionately more photo-
ever, the clearest trends in species composition synthate and invest it in the production of their
seem to respond – like the phytoplankton itself – own biomass.
to the availability of inorganic nutrients. Olig- In reality, planktic systems are rather more
otrophic and mesotrophic assemblages in lakes complex than this simple model might indicate.
(e.g. Šimek et al., 1999; Lindström, 2000; Riemann One major distorting factor is the complicat-
and Winding, 2001; Gattuso et al., 2002) and in ing and paradoxical role played by other, usu-
the sea (Fuhrman et al., 2002; Cavicchioli et al., ally much more abundant, sources of dissolved
2003; Kuuppo et al., 2003; Massana and Jürgens, organic matter in pelagic environments. In par-
2003) are commonly dominated by species of ticular, dissolved humic matter (DHM) is often,
the Cytophaga–Flavobacterium group and/or vari- by far, the major component of the DOC con-
ous genera of α- and γ-proteobacteria. It also tent of natural waters. Indeed, in the open ocean,
seems likely that these are the main groups of where there is a fairly invariable base concentra-
bacteria colonising particulate organic detritus tion of ∼1 mg L−1 of DOC (Williams, 1975; Sug-
(Riemann and Winding, 2001), some of which imura and Suzuki, 1988), in lakes where concen-
anyway decomposes and disintegrates rapidly trations typically fall in the range, 1–10 mg C L−1
(Legendre and Rivkin, 2002a). The detailed stud- (Thomas, 1997), and in brown, humic waters
ies of Cavicchioli et al. (2003) on the dynam- draining swamps and peatlands and in which
ics of the proteobacterium Sphingopyxis reveal humic matter accounts for 100–500 mg C L−1
the relevant properties of an oligotrophic het- (Gjessing, 1970; Freeman et al., 2001), DHM may
erotroph. Besides its small size and high surface- represent some 50–90% of all the organic car-
to-volume ratio, this obligately aerobic bacterium bon (including organisms) in the pelagic (Wetzel,
has a high affinity for nutrient uptake. Pop- 1995; Thomas, 1997). The supposed origin of this
ulation growth rates are sensitive to availabil- varied material – decomposing terrestrial plant
ity of substrates (which include malate, acetate matter – is plainly self-evident in lake catch-
and amino acids) and to the supply of inorganic ments, although neither the flux to the sea
ions. nor its persistence in the ocean has been fully
It is clear from this that, although plank- verified.
tic photoautotrophs and heterotrophs have quite DHM has the reputation of resistance,
independent carbon sources, they nevertheless or recalcitrance, to degradation by bacteria.
have to compete for common sources of limit- Humic material appears in water as substances,
ing inorganic nutrients. Moreover, it is likely that mainly phytogenic polymers, of relatively high
the bacteria are superior in this respect (Gurung molecular weight and complexed with various
and Urabe, 1999). Potentially, a mutualism devel- organic groups, which include acetates, formates,
ops between nutrient-deficient autotrophs and oxalates and labile amino acids. By the time
carbon-deficient heterotrophs. The elegant exper- they leach into water some decomposition has
iments of Gurung et al. (1999) on the plank- already taken place. The diversity of humic mate-
ton of the oligotrophic Biwa-Ko, Japan, illustrate rials, already large, is increased further (Wer-
how this balance might be maintained. Under shaw, 2000): to make any kind of general assess-
low light, photosynthesis is low and bacterial ment of the availability of DHM to pelagic bac-
growth is constrained by low organic carbon teria is still difficult, awaiting more research.
release. Increasing the light to nutrient-limited However, Tranvik’s (1998) thorough evaluation
phytoplankton stimulates the supply of extra- of the bacterial degradation of DOM in humic
cellular organic carbon and the growth of het- waters presents some well-considered analysis.
erotrophs (and of their phagotrophic consumers). Many humic compounds are amenable to bac-
Raising the resources available to the photoau- terial decomposition but, generally, the yield of
totrophs, however, interferes with the organic energy to bacteria is rather poorer than non-
carbon release to the increasing limitation of het- humic DOM. Most is relatively refractory but the
erotroph production. With increased nutrients, resultant pools are not unimportant as bacterial
SUMMARY 143
substrates, even though they turn over slowly. in pelagic bacterial biomass of 4.5 to 7.1 (Kirch-
The rate of oxidation is influenced by the avail- man, 1990) are interpreted as being indicative
ability of other nutrients, their tendency to floc- more of carbon than of nitrogen limitation of
culate and their exposure to sunlight and photo- bacterial growth (Goldman and Dennett, 2000).
chemical cleavage. This last turns out to be cru- Clearly, the relatively abundant forms of DOC
cial, as the photodegradation of organic macro- in the oceanic pools often fail to satisfy the
molecules to more labile and more assimilable requirements of the most abundant planktic het-
products is now known to occur under strong erotrophs, which must therefore rely predomi-
visible and ultraviolet irradiance (Bertilsson and nantly on the excretion of phototrophs, much as
Tranvik, 1998, 2000; Obernosterer et al., 1999; the Gurung et al. (1999) model suggests. Equally
Ziegler and Benner, 2000). As a result, many rela- clearly, it is a relationship of high resilience
tively simple, low-molecular-weight organic radi- (Laws, 2003).
cals may become available to microbes and may
not necessarily be readily distinguishable from
the DOM released by photoautotrophs (Tranvik 3.6 Summary
and Bertilsson, 2001).
The emphasis may still be on the restricted Pelagic primary production is the outcome of
nature of the photodegradation and its confine- complex interplay among biochemical, physiolog-
ment to surface layers, for the general impression ical and ecological processes that include pho-
of slow decomposition of humic matter endures. tosynthesis and the large-scale dynamics of vari-
It is likely that it is only in shallow-water sys- ous forms of carbon. Photosynthesis is the photo-
tems where allochthonous inputs of DOM might chemical reduction of carbon dioxide to carbohy-
sustain the predominantly hetertrophic activity drate, drawing upon radiant energy to synthesise
that the relative abundance of organic carbon a store of potential chemical energy, pending its
would lead us to expect. Elsewhere, it is mainly discharge when the carbohydrate (or its deriva-
the non-humic, autochthonously produced DOC tives) is oxidised (respiration). As in other pho-
that seems likely to underpin heterotroph toautotrophs, algae and photosynthetic bacteria
activity. employ two sequenced, chlorophyll-based photo-
This last deduction fits most comfortably with systems. In the first, electrons are stripped from
the previously noted general coupling between water and transported to a reductant pool. In
bacterial mass and primary production: the sup- the second, photon power re-elevates the electro-
position that, on average, around half the pri- chemical potential sufficiently to transfer elec-
mary production of the oligotrophic pelagic trons to carbon dioxide, through the reduction
passes through the DOC reservoir requires that of nicotinamide adenine dinuceotide phosphate
this must also be the more dynamic source of (NADP to NADPH). The carbon reduction process
carbon and this supports the more active part of is built around the cyclical regeneration of ribu-
bacterial respiration (Cole et al., 1988; Ducklow, lose 1,5-biphosphate (RuBP). RuBP is first com-
2000). Of course, the relationship is approximate, bined with (carboxylated) carbon dioxide and
it is difficult to predict precisely and is plainly water to form sugar precursors, under the con-
subject to breakage. High rates of bacterivory, for trol of the enzyme RUBISCO, and from which
instance, would cause one such mechanism. How- hexose is generated and RuBP is liberated (the
ever, bacterial growth can become nutrient lim- Calvin cycle). The hexose may be polymerised (e.g.
ited in very oligotrophic waters, to the extent to starch or glycogen) or stored.
of positive DOC accumulation (Williams, 1995; The theoretical photosynthetic quantum yield
Obernosterer et al., 2003), just as easily as it can is 1 mol carbon for 8 mol photon, or 0.125 mol
be substrate limited in (say) nutrient-rich estu- C (mol photon captured)−1 . Actual efficiency is
aries (Murrel, 2003). In this context, it is espe- closer to 0.08 mol C (mol photon)−1 , equivalent
cially interesting to note the reports of oceanic to 2.821 kJ (mol C fixed)−1 , or ∼470 kJ (g C)−1 . The
microbiologists referring to atomic C : N ratios maximum rates of photosynthesis are related to
the rate of electron clearance from the reductant recycles. Indeed, most smaller lakes probably
pool (and which responds to the photon flux), release more CO2 to the atmosphere than they
as well as to an adequate supply of CO2 to the take from it. They are considered to be net het-
RUBISCO reaction (if a concentration of >0.01 erotrophic. Only in very large, oligotrophic sys-
mM is not maintained, the enzyme acts as an tems does the sedimentary export of carbon bal-
oxygenase). ance the atmospheric inorganic uptake flux (at
Physiologically, photosynthetic rate is sensi- some 50–90 g C m−2 a−1 ).
tive to temperature, to light and carbon dioxide Globally, pelagic photosynthesis accounts for
availability. Even at 30 ◦ C, given saturating light around 45% of the planetary carbon fixation.
and an adequate carbon supply, photosynthesis In some circumstances, when photosynthesis is
achieves <20 mg C (mg chla)−1 h−1 . Maximum constrained (especially by light dilution) the
photosynthetic rates are generally halved for each carbon is invested in the growth of the pho-
10 ◦ C drop in temperature. Below saturation (usu- toautotroph. These organisms become potential
ally <150 mol photons m−2 s−1 ), photosynthetic food to pelagic grazers. In many other cases,
rates fall in a light-dependent manner, in the pro- light saturation or nutrient depletion result in
portion 6–18 mg C (mg chla)−1 (mol photon)−1 carbon fixation in excess of contemporaneous
m2 . Carbon dioxide concentrations below air growth requirements and photosynthate is either
saturation may also limit photosynthetic rates. reoxidised or excreted as dissolved organic car-
Some algae are extremely efficient in adapting bon (DOC). This augments an already relatively
to photon harvesting under very low light fluxes large pool of dissolved humic matter (DHM) but
or in the fluctuating light experienced by phy- presents a much more amenable substrate for
toplankton entrained in mixed water columns. pelagic bacteria. Like those of photoautotrophs,
Some algae are restricted to carbon dioxide as concentrations of heterotrophic bacteria reflect
a carbon source and are sensitive to the very the availability of inorganic nutrients and there
low concentrations experienced at pH >8. Others is mutual competition. However, bacterial growth
can use bicarbonate or employ energy-consuming is often more carbon limited while the main
carbon-concentrating mechanisms to focus the producers are usually nutrient limited. Besides
limited fluxes at the sites of synthesis. In this the mutualism that this situation engenders,
way, low light and low carbon availability select the acquisition by bacteria of organic carbon
strongly for well-adapted species. products of the phytoplankton and the con-
On a local basis, it is possible to calculate sumption of bacteria by microzooplankton rep-
the carrying capacity of the environment and resents the main route of pelagic photosynthate
the rates of biomass assembly that might be sus- to the pelagic food web. This ‘microbial loop’
tainable. Down-mixing and light dilution place commonly dominates the first steps in the food
important limits on both. The carbon flux from chain, is certainly of great antiquity, and should
the atmosphere is potentially – and, at times, no longer be regarded as a special exception to
is – a constraint on area-specific photosynthe- alga–herbivore–fish linkages. It is the latter that
sis but is avoided in most lakes and at most are the exception, being sustainable only in rela-
times by inflowing CO2 -saturated and internal tively resource-rich conditions.
Chapter 4
Nutrient uptake and assimilation

in phytoplankton
There is a huge literature on this topic. The

4.1 Introduction purpose here is not to review the findings in
detail or to give more than the sketchiest outline
This chapter addresses the resource requirements of the historical development of the understand-
for the assembly of photoautotrophic biomass. In ing of nutrient limitation. Even the elements
addition to light and carbon, growth of phyto- most often implicated in the constraint of phyto-
plankton consumes ‘nutrients’ and, equally, may plankton growth (nitrogen, phosphorus, iron and
often be constrained by their availability and one or two other trace elements, together with
fluxes. Put at the most basic level, every repli- the well-known constraint on diatom growth set
cation of a phytoplankton cell roundly demands by its skeletal requirement for considerable quan-
the uptake and assimilation of a quota of (usu- tities of silicon) are sufficiently well and clearly
ally) inorganic nutrients similar to that in the known not to require any long and detailed
mother cell, if her daughters are to have the account of how this recognition came about. The
similar composition. Ignoring skeletal biominer- approach that I have adopted is first to consider
als for the moment, we may recall from Sec- the mechanisms of nutrient uptake and the gen-
tion 1.5.3 that, in addition to carbon, the liv- eral constraints that govern the successful assimi-
ing protoplast comprises at least 19 other ele- lation and anabolism of resources by phytoplank-
ments. Some are needed in considerable abun- ton. Then, mainly by reference to the key limiting
dance (hydrogen, oxygen, nitrogen), others in elements (N, P, Fe, etc.), I seek to show how the
rather smaller amounts (phosphorus, sulphur, abilities of phytoplankton to gather the resources
potassium, sodium, calcium, magnesium and necessary to support cell growth and replication
chlorine), for the assembly and production of might impinge upon the dynamics and ecology
the organic matter of protoplasm. Others occur of populations.
as vital traces in support of cellular metabolism Uptake and assimilation of these nutrients do
(silicon, iron, manganese, molybdenum, copper, need to be considered in rather more detail, as
cobalt, zinc, boron, vanadium). However, it is impairment to these processes, mostly through
less the amounts in which these elements are resource deficiencies, is frequently implicated in
required that constrains growth than does the the comparative dynamics and relative abun-
ease or otherwise with which they are obtained. dance of phytoplankters. It is important to
It is the demand (D) relative to the supply (S) recognise that the first impediment to be over-
that is ultimately critical, bearing in mind that a come is that just about all of the nutrients
measurable presence is not a measure of avail- to be drawn from the water occur in concen-
ability if the element in question is not both trations that, relative to their effective concen-
soluble and diffusible and, so, assimilable by trations within the cell, are extremely dilute,
cells. or rarefied. How these steep gradients are
146 NUTRIENT UPTAKE AND ASSIMILATION IN PHYTOPLANKTON
overcome is an appropriate starting point for our sites of their anabolism into proteins and, even-
consideration. tually, into organelles. The cell is ordered, with
relative compositional homeostasis based on bal-
anced resource deployment and controlled com-
4.2 Cell uptake and intracellular position. Outside the cell, the external medium
transport of nutrients is chaotic: besides signalling irregular and rapid
fluctuations in the photon flux, the solutes to
which the cell is exposed are often patchily dis-
To describe adequately the main structures of
tributed, even at the scale of a few millimetres.
a eukaryotic unicellular phytoplankter that are
Some initial calculations illustrate the mag-
involved in the uptake, transport and assembly
nitude of the uptake requirement. Starting from
of inorganic components, it is helpful to refer
the premise that the ash-free dry mass of the
to the simplified and stylised diagram in Fig. 4.1.
cytoplasm accounts for between 0.41 and 0.47
Inside the multiple-layered plasmalemma (shown
pg µm−3 of live volume and that between 46%
as a single line), there is a nucleus contain-
and 56% of the ash-free dry mass is carbon, then
ing the genomic proteins (marked ‘DNA’); the
it follows that the carbon concentration in the
ribosomal centres of protein synthesis are rep-
replete, healthy, live cell is in the range 0.19 to
resented by ‘RNA’ and part of the structure of
0.26 pg C µm−3 , or 225 ± 35 g C L−1 . This is
the chloroplast and the thylakoid membranes
equivalent to 18.8 mols C L−1 . Against the air-
are also sketched. Superimposed upon the cell is
equilibrium concentration of carbon dioxide in
a series of arrows that provides a fragmentary
water (0.5–1 mg L−1 , or between 11 and 23 µmol
indication of the key pathways located within
L−1 ), the growing cell is literally accumulating
the protoplast. The arrows refer, in part, to the
carbon atoms against a concentration gradient
dynamics of photosynthetic reduction of inor-
in the order of 1 000 000 to 1. Moreover, in order
ganic carbon dioxide and, in part, to the uptake
to accomplish a doubling of cell material, it has
and intracellular delivery of key nutrients to the
to acquire another 1 mol carbon for every 1 mol
of carbon in the newly isolated daughter tissue.
The corresponding calculations for the average
cell concentrations of nitrogen (∼2.8 mols N L−1 )
and phosphorus (∼0.18 mols P L−1 ) are of a simi-
lar magnitude greater than they might typically
occur in natural waters (2–20 µmol N L−1 ; 0.1–5
µmol P L−1 ). In relation to the carbon require-
ment, each cell has to draw on the equivalent of
∼151 mmol N and 9.4 mmol P for each mol of C
required to replicate the cell mass.
4.2.1 Supply of nutrients

Based on the example of carbon, the well-
developed nutrient harvesting capabilities of
algae have already been indicated (see, especially,
Section 3.4.2). However, it is not simply a mat-
ter of engineering a high affinity for the carbon
dioxide (or, indeed, other nutrient in the adja-
Figure 4.1 Diagram of a phytoplankton cell to show the cent medium) as the mechanisms can only be
essential pathways for the gathering and deployment of the effective over a short distance beyond the cell.
key resources. Based on an illustration of Harris (1986) and The operational benefits are really restricted to
reproduced with permission from Reynolds (1997a).
within the boundary layer adjacent to the cell.
CELL UPTAKE AND INTRACELLULAR TRANSPORT OF NUTRIENTS 147
Here, the movement of solutes are subject to Fick- where t is time, m is the coefficient of molecular
ian laws of diffusion (cf. Eq. 3.19). The renewal, diffusion of the solute (as in Eq. 3.19) and N =
or replenishment, of nutrients in this immedi- (δ/δx, δ/δy, δ/δz) is an integral of the gradients in
ate microenvironment of the cell can also be the x, y and z planes. Supposing steady state in a
critical and, hence, so is any attribute of the symmetrical sphere, this will reduce to:
organism that enhances the rate of entry of
d/drb rb2 dC /drb = 0 (4.2)
essential solutes into that boundary layer. Such
adaptations in this direction may raise directly where rb is the radial distance from the centre
the effectiveness of nutrient gathering by the of the sphere. It may be solved for the space to
cell. the edge of the boundary, Csurface = C(rb = a), and
The importance of the movement of water beyond, Cbulk = C(rb → ∞), so that:
relative to the phytoplankter (or, as we now
recognise, to the phytoplankter plus its bound- C (rb ) = C bulk − (C bulk − C surface )a/rb (4.3)
ary layer) was famously considered by Munk and The flux (Fa ) of the solute to the cell is calculable
Riley (1952). They were among the first to point as:
out that the effect of motion – either active
‘swimming’ or passive sinking or flotation – is F a = 4πa 2 mdC /drb |rb=a
to increase the solute fluxes to the cell above = 4πam(C bulk − C surface ) (4.4)
those that would be experienced by one that is
non-motile with respect to the adjacent medium. If the live cell now retains the inwardly diffusing
This seemingly axiomatic statement was verified solute molecules, Csurface diminishes to zero and
through the experiments of Pasciak and Gavis the flux increases towards a maximum:
(1974, 1975) and the interpolation of the results F a max = 4πamC bulk (4.5)
to the benefits to nutrient uptake kinetics of
a diatom of sinking through nutrient-depleted The effect of the motion of the cell, sinking, float-
water. In consideration of these data, Walsby ing or ‘swimming’ through water is to increase
and Reynolds (1980) determined the trade-offs the flux to the diffusive boundary layer at the
between sinking and uptake rates in sinking same time as compressing its thickness (Lazier
diatoms: there was always a positive benefit in and Mann, 1989). The distribution of a nutrient
material delivery but at the ambient external solute next to the cell is modified with respect to
concentrations critical to sufficiency, the com- Eq. (4.1) by the advection owing to the hydrody-
pensatory sinking rates become unrealistically namic flow velocity, u:
large. In other words, motion relative to the δC /δt + uN C = m(N )2 C (4.6)
medium undoubtedly assists the renewal and
delivery of nutrients to the immediate vicinity The advection–diffusion equation is not easily
of the plankter but, ultimately, is no guarantee soluble. The approach of Riebesell and Wolf-
of satisfaction of the plankter’s requirements at Gladrow (2002) was to rewrite the problem in
low concentrations. dimensionless Navier–Stokes terms, using parti-
A modern, empirical perspective on this topic cle Reynolds (Section 2.3.4 and Eq. 2.13), Péclet
has been pursued in the work of Wolf-Gladrow and Sherwood numbers. The Péclet number (Pe)
and Riebesell (1997; see also the review of Riebe- compares the momentum of a moving particle
sell and Wolf-Gladrow, 2002). Starting from the to diffusive transport. For a phytoplankton cell
perspective of the single spherical algal cell with whose movement satisfies the condition of non-
an adjacent boundary layer of thickness a, the turbulent, laminar flow (Re < 0.1; Section 2.4.1),
concentration (C) of a given nutrient in the imme-
Pe = (us d) m−1 (4.7)
diate microenvironment is subject only to diffu-
sive change, in conformity with the equation: where us is the intrinsic velocity of a spherical
cell of diameter d. In the present context, where
δC /δt = m(N )2 C (4.1) the particle is introduced into the flow field u
influence of the turbulent shear rate (Karp-Boss

et al., 1996). They showed:
Sh = 1.014 + 0.015Pe2 (4.9)
when the Péclet number was derived from:
Pe = (d/2)2 (E /ν)2 m−1 (4.10)
in which equation, d is the diameter of a spher-
ical cell, E is the turbulent dissipation rate, in
m2 s−3 , ν is the kinematic viscosity of the water
(in m2 s−1 ) and m is the coefficient of molecular
diffusion of the solute (for further details, see
Section 2.3.3).
A further deduction and reinterpretation of
the comment of Walsby and Reynolds (1980; see
Figure 4.2 The Sherwood number as a function of the above) is that relative motion does not in itself
Péclet numbers for steady, uniform flow past an algal cell overcome rarefied nutrient resources. However,
(solid line) and turbulent shear (dashed line). Figure redrawn chronic and extensive resource deficiencies must
with permission from an original in Riebesell and exact a greater dependence of larger algae on tur-
Wolf-Gladrow (2002). bulence to fulfil their absolute resources require-
ments to sustain growth requirements than they
do of smaller ones. Conversely, smaller cells
(x, y, z, Re), the Péclet number also expresses the are less dependent upon turbulent diffusivity
ratio of the scales of advective (u N C) and dif- to deliver their nutrient requirements than are
fusive solute transport (m(N )2 C). The Sherwood larger ones.
number (Sh) is the ratio between the total flux of
a nutrient solute arriving at the surface of a cell 4.2.2 Moving nutrients into the cell
in motion and the wholly diffusive flux. Riebesell The transfer of nutrients from the enveloping
and Wolf-Gladrow (2002) showed that, for parti- boundary layer into and within the cells is
cles in motion with very low Reynolds numbers, biologically mediated, being effected principally
Sherwood numbers are non-linearly related to through a series of substance-specific membrane
Péclet numbers but, in the range 0.01 ≤ Pe ≤ 10 transport systems. Modern molecular biology is
(which embraces the sinking motions of algae providing the means to investigate both the
from Chlorella to Stephanodiscus; see Section 2.4.1), mechanisms by which cells marshal and assem-
the relationship is adequately described by: ble components in cellular synthesis and how
Sh = 2 + 2(1 + 2Pe)a (4.8) their operations are regulated. In the case of
membrane transport systems, working against a
The relationship (sketched in Fig. 4.2) shows concentration gradient, structure and function
that, for small cells embedded deeply in the tur- conform to a generalised arrangement common
bulence spectrum (Pe < 1), the benefit in terms to most living cells. In essence, these accept spe-
of nutrient supply is marginal (Sh ∼1). For larger cific target molecules and transfer them to the
units and motile forms generating Re > 0.001 and sites of deployment. These movements are gener-
Pe > 1, the dependence on turbulence for the ally not spontaneous and, so, require the expen-
delivery of nutrients becomes increasingly signif- diture of energy. Power, fuelled by ATP phosphor-
icant (Sh > 1). ylation, is used to generate and maintain ion
The conclusion fits comfortably with the gradients and proton motive forces, through the
demonstration of a direct relationship of algal coupling of energy-yielding reactions to energy-
size to Sherwood scaling, mediated through the consuming steps (Simon, 1995).
CELL UPTAKE AND INTRACELLULAR TRANSPORT OF NUTRIENTS 149
tor by the reaction with the target becoming the

excitation of the next. The sequenced reactions of
the transporter proteins provide a redox-gradient
‘channel’ along which the target molecule is
passed. The whole functions rather like a line of
people helping to douse a fire. The first lifts the
filled bucket and passes it to the next, who, in
turn passes it to a third. Only after the second
has accepted a bucket from the first can the first
turn to pick up another bucket. The second can-
not accept another bucket until he has passed
on the last and is once more receptive to the
next.
In much the same way as the fire-fighters
might be supplied with filled buckets more
rapidly than they can be dispatched down the
line, so the molecular sequence can become sat-
urated and the fastest rate of uptake then fails
to deplete the supply of target nutrients. At
the other extreme, exhaustion of the immedi-
Figure 4.3 Basic structure of a receptor–excitation
ate source of buckets or targets leaves the entire
assembly, used to capture, bind and transport specific target
molecules into and within the phytoplankton cell. Based on a sequence idle but the full transport capacity
figure in Simon (1995) and reproduced with permission from remains ‘open’ and primed to react to the stimu-
Reynolds (1997a). lation of the next arriving molecule.
The activity state of the transport system is
communicated to the controlling genes. It is
In the specific case of nutrient uptake, the extremely important that the cell can react to
linkages involve sequences of protein–protein the symptoms of shortages of supply (of, say,
interactions in which the binding of a spe- phosphorus) by regulating and closing down the
cific target ligand at a peripheral receptor stim- assembly processes before the supply of compo-
ulates an excitation of the transfer response. The nents (or a particular component) is exhausted.
basic structure of the transmembrane assembly What happens is that a second group of regu-
is sketched in Fig. 4.3. The receptor region is latory proteins associated with the transporters
periplasmic and constitutes the ligand-specific activate the transcription of particular genes
protein. Reaction with the target molecule stimu- called operons. While the uptake and transport
lates a molecular transformation, which, in turn, mechanism is functioning normally, the operons
becomes the excitant substrate to the proteins repress the expression of further genes which
of the transmembrane region. The central reac- regulate the reactions to nutrient starvation
tion within this complex is to catalyse the phos- (Mann, 1995). For instance, an external shortage
phorylation of the substrate. This is, of course, of a given nutrient (say, orthophosphate ions)
the principal reaction through which cells regu- will result in a diminishing frequency of recep-
late the transfer of redox power. The high-energy tor reactions and a weakening suppression of the
pyrophosphate bond between the second and genes that will activate the appropriate cellular
third radicals of ATP is broken by a kinase, the response. The response may be to produce more
conversion to the diphosphate releasing some phosphatases or to promote the metabolic close-
33 kJ (mol)−1 of chemical energy. down of the cell, including entry into a resting
The further proteins in the series react analo- stage, before the cell starves to death (see also
gously and sequentially, the excitation of a recep- Sections 4.3.3, 5.2.1).
4.2.3 Empirical models of nutrient uptake

Many of the well-established paradigms relat-
ing to the uptake and deployment of nutrients
and the ways in which these impinge upon
the growth and dynamics of phytoplankton are
founded on a welter of experimental observa-
tions. However, it is a small number of classic
studies that have provided much of the insight
and understanding, into which the majority of
observations can be interpolated. For instance,
the model of Dugdale (1967) recognised the Figure 4.4 Uptake rates of molybdate-reactive phosphorus
saturable transport capacity (corresponding to by cells of Chlorella sp., pre-starved of phosphorus, as a
the fully ‘open’ condition of the receptors) and function of concentration, according to data of Nyholm
(1977). Redrawn with permission from Reynolds (1997a).
the simultaneous dependence of the rate of
uptake upon resource availability in the adja-
cent medium. Dugdale showed that the rela-
tionship between the rate of uptake of nutri- found to be less satisfactory for the descrip-
ents in starved cells and the concentrations in tion of nutrient-limited growth rates. This may
which the nutrient is proffered conforms to be attributed, in part, to inappropriate applica-
Michaelis–Menten enzyme kinetics, as expressed tion and a misplaced assumption – analogous to
by the Monod equation. Dugdale’s (1967) general the one about photosynthesis and growth – that
equation stated: growth rates are as rapid as the relevant materi-
als can be assembled. In fact, growing cells may
VU = VUmax S/(K U + S) (4.11) take up nutrients when they are abundant much
more rapidly than they can deploy them, just as
The equation recognises that the rate of uptake they can sustain growth at the expense of inter-
of a nutrient by fully receptive cells is a func- nal stores at times when the rate of uptake may
tion of the resource concentration, S, up to a be constrained by low external concentrations.
saturable limit of VUmax . KU is the constant of Droop (1973, 1974) cleverly adapted the Dugdale
half saturation (i.e. the concentration of nutri- model to include a variable internal store in
ent satisfies half the maximum uptake capacity). order to represent the impact of the cell quota
There is no way to predict these values accu- on the rate of growth. The impacts of nutrient
rately save by experimental determination. Many deficiencies on cell replication are considered in
measurements show close conformity to the pre- Chapter 5 (see Sections 5.4.4, 5.4.5). However, it is
dicted behaviour and, hence, to the generalised useful to introduce here the concept of an inter-
plot in Fig. 4.4, showing the uptake of phospho- nal store in the context of its influence on uptake
rus by Chlorella, as described by Nyholm (1977). and, indeed, its relevance to how we judge ‘lim-
Because the uptake rates and affinities are, how- itation’ and its role in interspecific competition
ever, very variable among individual phytoplank- for resources.
ton species, they are most conveniently intercom- It is, firstly, quite plain that the intracellular
pared by reference to the magnitudes of alga- content of the cell starved of a given particular
and nutrient-specific values of VUmax and KU . For resource will not just be low (leaving the cell very
instance, a relatively high VUmax capacity com- responsive to new resource) but it will probably
bined with a low KU is indicative of high uptake be close to the absolute minimum for the cell to
affinity for a given nutrient. stay alive. This is Droop’s ‘minimum cell quota’
The Monod model has been widely applied (q0 ) and it is too small to be able to sustain any
and found to describe adequately the uptake growth. Secondly, raising the actual internal con-
by algae of essential micronutrients, under the tent (q) above the minimal threshold (essentially
starvation conditions described. It has been through uptake) makes resource available to
PHOSPHORUS 151
deployment and growth. At low but steady rates the literature. In the following considerations,
of supply, some proportionality between the rates the usage is necessarily precise, adhering to the
of growth (r ) and resource is expected to be evi- definitions shown in Box 4.1.
dent. Interpolating (q − q0 ) for S, r for VU and rmax

for VUmax in Eq. (4.11), the Droop equation states:

4.3 Phosphorus: requirements,
r = rmax

(q − q0 )/(K r + q − q0 ) (4.12) uptake, deployment in
As it cannot be assumed that growth and uptake phytoplankton
are half-saturated at the same concentration,
we must also substitute a half-saturation con- The phosphorus relations of phytoplankton cells
stant of growth, (Kr ). By analogy, we might also provide a good example of the ways in which
assume that, again at a low but steady rate of the adaptations for gathering of an essential
supply, r and VU and rmax

and VUmax are mutu- but frequently scarce resource impinge upon
ally interchangeable. Such equivalence is demon- the dynamics of populations and the species
strable, provided the steady-state condition holds structure of natural assemblages. As a compo-
(Goldman, 1977; Burmaster, 1979). However, nent of nucleic acids governing protein synthesis
when supply rates exceed deployment and the and of the adenosine phosphate transformations
internal store increases, the uptake rate must that power intracellular transport, phosphorus
slow down, even when a high external concen- is an essential requirement of living, functional
tration obtains. It might then be proposed that plankters. As observed earlier (Section 1.5.3), the
nutrient uptake is better described by: phosphorus content of healthy, resource-replete,
VU = [(qmax − q)/(qmax − q0 )] actively growing phytoplankton cells is generally
close to 1–1.2% of ash-free dry mass (Round, 1965;
[VU max S/(K U + S)] (4.13)
Lund, 1965), with a molecular ratio to carbon
where (qmax ) is the replete cell quota. The larger of around 0.0094 (106 C : P). The minimum cell
is the instantaneous cell quota, q, the smaller quota (q0 ) may vary intespecifically, most proba-
will be the effective rate of uptake. The mul- bly, between 0.2% and 0.4% of ash-free dry mass
tiple allows the cell to accumulate even scarce (some 320–640 mol C : mol P) but, reportedly,
resources (of, say, phosphorus) from low concen- almost an order lower in some species (Asteri-
trations so long as the uptake of another (say, onella ∼0.03% of ash-free dry mass: Rodhe, 1948;
nitrogen) is controlling (‘limiting’) the rate of the Mackereth, 1953), equivalent to molecular C : P
deployment of both in the structure of new cell ratios of ∼4000). Conversely, intracellular stor-
material. An independent increase in the supply age capacity of phosphorus may allow q to rise
of the second resource (nitrogen), however, with in some species to ≥3% of dry mass (≤40 C : P).
no simultaneous alteration in the availability of The interesting deduction is that, as a result of
the first (phosphorus), might very quickly leave this so-called ‘luxury uptake’, the cell may con-
the rate of supply of the first as the limiting contain 8–16 times the minimum quota and that,
straint, when the internal quota is likely to be as a consequence, it is theoretically able to sus-
drawn down. This principle underpins the use of tain three or possibly four cell doublings without
intracellular nutrient ratios to indicate the nutri- taking up any more phosphorus.
ent status of cells and, thus, the identity of the
instantaneously ‘limiting factor’. Interspecific dif- 4.3.1 The sources and biological availability
ferences in the competitive abilities of algae to of phosphorus in natural waters
function at low resource availability are also held The natural sources of phosphorus in water are
to influence the structure of communities. the small amounts that occur in rainfall (gen-
The terms ‘limitation’ and ‘competition’ (in erally 0.2 to 0.3 µM), augmented by phosphates
the context of satisfying resource requirements) derived from the weathering of phosphatic min-
have been used variously and inconsistently in erals, especially the crystalline apatites, such
Box 4.1 Limitation and competition in the nutrient rela-

tions of phytoplankton
The word ‘limitation’ has been used, variously, to explain the control of phyto-
plankton growth dynamics, the poverty of plankton biomass and the dearth of the
supportive nutrients. All non-toxic environments have a finite supportive capacity,
which is generally based upon the notion that available resources are deployed in
the assembly of biomass, ideally, in quasi-fixed quotas, up to a maximum, Bmax =
Ki /q0 , where = Ki is the steady-state concentration of the ith resource and q0 is
the minimum cell quota in the biomass, supposing a uniform, Redfield-type com-
position, or the minimum cell quota in the biomass of the jth species. In this usage,
the limiting capacity is the lowest of the individual supportive capacities, Ki . By impli-
cation, the ratios among the components of the cell will show supra-ideal values
to the one that is sub-ideal and, thus, biomass limiting (Reynolds, 1992a; Reynolds
and Maberly, 2002). The capacity limitation by the factor least available relative to
demand is the expression of von Liebig’s ‘Law of the Minimum’ (von Liebig, 1840).
In practical terms, the identity of the capacity-limiting factor is revealed by the
magnitude of the response to its augmentation. Gibson (1971) usefully deduced
that a substance is not capacity-limiting if an increase in that factor produces no
stimulation to the biomass that can be supported.
Growth dynamics may also limited, in the sense that the rate of biomass elab-
oration is determined by the rate of resource supply. Moreover, the rate-limiting
factor is the one upon by whose rate of supply determines the rate of elaboration.
Analysing data on the growth of the diatom Asterionella in Windermere over a
period of 50 years, Reynolds and Irish (2000) were able to confirm Lund’s (1950)
view that the biomass capacity was set by the winter concentration of soluble
reactive silicon. However, they also showed that the rate of its attainment had
been phosphorus limited and that the timing of the silicon-limited maximum had
advanced over the period of the documented enrichment of available phosphorus
in this lake.
‘Competition’ is used inconsistently by biologists. However, the term ‘competi-
tor’ is applied by aquatic ecologists, with great consistency, to refer to species that
eventually rise, tortoise-like, to a steady-state dominance. Unfortunately, in the
parlance of terrestrial plant ecologists, a good competitor is dynamic, fast-growing
and applicable to Aesop’s fabled hare. Mindful of the place that competition the-
ory occupies in ecological and evolutionary theories, it seems important to have
robust definitions. In this book, I use ‘competition’ in the sense of Keddy’s (2001)
definition as ‘The negative effects that one organism has upon another by con-
suming, or controlling access to, a resource that is limiting in its availability.’ Thus, a
competitive outcome has only transpired if the activities of species 1 denies access
to the resources required to nourish the activities of species 2. Being able to grow
faster when fully resourced does not, by itself, make species 1 ‘more competitive’
than species 2. It is merely more efficient in converting adequate resources into
biomass. On the other hand, the behavioral or physiological flexibility of species
2 to better exploit a critically limiting resource affords a significant competitive
advantage over species 1, at such times when that resource limitation is operative.
PHOSPHORUS 153
as fluorapatite and hydroxylapatite, and the ganese), although there is the further complica-
amorphous phosphorite. All are forms of calcium tion of their redox sensitivities. At redox poten-
phosphate, which has a low solubility in water tials below +200 mV, the higher-oxidised ion,
at neutrality, and the bioavailability of phospho- Fe3+ , is reduced to the divalent Fe2+ . Whereas the
rus in drainage waters tends to be low (Emsley, hydrolysis of the trivalent ion leads to the precip-
1980). Terrestrial plants and the ecosystems of itation of insoluble ferric hydroxide, divalent fer-
which they are part share analogous problems rous ions remain in solution. Raising the redox
of phosphorus sequestration. Not surprisingly, potential favours the opposite reaction (Fe2+ −
forested catchments, especially, remove and accu- e → Fe3+ , although it is usually enhanced by
mulate much of the modest quantities of inor- microbial oxidation): the floccular ferric hydrox-
ganic phosphorus with which they are supplied, ide precipitate scavenges orthophosphate ions,
leaving little in the export to receiving waters again in exchange for hydroxyls. At close to neu-
save as organic derivatives of biogenic products. trality, the orthophosphate ions are substantially
The losses of inorganic phosphorus to water immobilised (‘occluded’) to the extent that they
can be greatly enhanced through anthropogenic are scarcely any longer available to algal or micro-
activities (quarrying, agriculture and tillage and, bial uptake. Only a further change in redox or
especially, the treatment of sewage) but the gen- an increase in the ambient alkalinity of the
eral condition of natural waters draining from medium alters this position. The phosphate ions
any but desert catchments and/or ones with an that are released into solution are, potentially,
abundant occurrence of evaporite minerals is to fully bioavailable (Golterman et al., 1969).
be moderately or severely deficient in inorganic Redox-mediated changes in phosphate solubil-
phosphorus (Reynolds and Davies, 2001). ity in sediment water and in limnetic hypolimnia
In all its biologically available (or ‘bioavail- were described over 60 years ago (Einsele, 1936;
able’) forms, phosphorus occurs in combination Mortimer, 1941, 1942). Since then, many of the
with oxygen in the ions of orthophosphoric fears about the impacts of phosphorus enrich-
acid, OP(OH)3 (Emsley, 1980). Orthophosphoric ment on aquatic ecosystems have continued to
acid itself is a weak tribasic acid and is freely be dominated by the renewed bioavailability
water soluble. The relative proportions of the to phytoplankton of sediment phosphorus. Of
various anions (PO3− 4 , HPO4
2−
and H2 PO− 4 ) vary course, there still needs to be phytoplankton
with pH. The hydrogen radicals are all replace- access to these phosphorus sources. Though the
able by metals. The orthophosphates of the ‘release’ of orthophosphate to the water seems
alkali metals (except lithium) are also solu- just as likely an occurrence, most of this should
ble but those of the alkaline earth metals be re-precipitated with ferric iron, once the
and the transition elements are quite insolu- redox is raised sufficiently (≥+200 mV). Under
ble. Three of these – calcium, aluminium and severe reducing conditions (redox potential ≤
iron – are especially relevant to the consider- −200 mV), however, sulphate ions are reduced to
ation of phosphorus availability and plankton sulphide ions. These readily precipitate with fer-
behaviour. The precipitation of calcium phos- rous iron, thus scavenging the water of Fe2+ ions.
phate effectively removes orthophosphate ions The consequence then is that, on re-oxidation,
from solution, in stoichiometric proportions. The the residual iron content will be diminished, less
bioavailability of orthophosphate ions can be sig- ferric hydroxide will precipitate and less phos-
nificantly affected through exchange with the phate may be scavenged. High phosphate levels in
hydroxyl ions that are otherwise immobilised, eutrophic systems may be more influenced by the
in large, non-stoichiometric numbers, on the redox transformations of sulphur than by those
surfaces of aluminium oxides; this sorption of of iron.
the orthophosphate ions effectively renders them Certainly, the solubility transformations at
biologically unavailable. A similar behaviour high pH and the behaviours of other elements
characterises the reactions of phosphate with at low redox can have profound effects on the
the precipitated hydroxides of iron (and man- bioavailability of phosphate in natural waters
Table 4.1 Phosphorus-containing fractions in water: nomenclature and availability
Phase Abbreviation Chemical sensitivity and bioavailability

Dissolved P DP Free orthophosphate ions, some in
combination with organic derivatives.
Assumed to be freely bioavailable
Soluble, MRP (or SRP) DP + fine colloidal organic material.
molybdate-reactive P Demonstrably bioexhaustible and supposed
to be freely bioavailable
Particulate P PP Phosphorus not in solution or in fine colloids
but bound to suspended solids; fraction
subdivisble as:
Water-extractable PP IMRP Phosphorus moves into solution in irrigating
water; most frequently encountered in intact
sediments, where it is mainly from the
interstitial and is conditionally bioavailable
NH4 Cl-extractable PP NH4 Cl-P Particle-bound phosphorus, ion exchangeable
and conditionally bioavailable
Citrate-dithionate- Na2 S2 O4 -P Iron-bound phosphorus. Scarcely bioavailable,
extractable PP dependent upon low redox or high pH
NaOH-reactive P NaOH-rP Iron- and aluminium-bound phosphorus,
sensitive to high pH. Otherwise scarcely
bioavailable
Non-alkali-reactive PP NaOH-nrP PP that is not soluble in strong alkali; fraction
includes:
HCl-reactive P HCl-P Phosphorus in compound with alkaline metals,
esp. apatite. Scarcely bioavailable
non-HCl-reactive resP (Organic) PP soluble only in powerful oxidant
residue (e.g. perchloric acid). Not bioavailable
HClO4 -digestible P TP Perchloric acid digestion releases all known
combinations of phosphorus. Phosphorus
quoted as ‘TP’ is only partially bioavailable as
roughly determined by serial analysis of the
above sequence
Source: Based on Table 1 of Reynolds and Davies (2001), compounded from various sources.
(Lijklema, 1977). On the other hand, the mecha- eral and biogenic particulate phosphorus in the
nisms favouring increased bioavailability of phos- water) generally under 1–2 µM. Moreover, only
phorus are more active in environments that are a small proportion of this TP may be in solu-
already relatively enriched with respect to this tion or be so readily soluble to be measurable by
and other nutrients. These are not exceptional or the standard molybdenum-blue method of Mur-
uncommon conditions among shallow, enriched phy and Riley (1962). Most of the balance will
lakes. However, the wide acceptance that a major- already be constituent in pelagic biomass or in
ity of lakes and many seas conform to a model of non-bioavailable colloids and fine particles.
pristine conditions characterised by low phospho- These various fractions are potentially sep-
rus availability is well justified. Such habitats fre- arable by serial assays, each step using a pro-
quently carry a total-phosphorus concentration gressively more aggressive chemical cleavage
(TP, being the aggregate of all dissolved, min- (see Table 4.1). Prior to these methods being
PHOSPHORUS 155
developed, the understanding of phosphorus phosphorus. One of these, a 32-kDa polypep-

dynamics was poised between the detection of tide, was localised in the cell wall, linked to
small amounts of molybdate-reactive phosphorus an intracellular 100-kDa polypeptide. Together,
(MRP), with poor sensitivity, and the MRP con- these conform to the typical structure of a recep-
tent of companion samples after digestion with tor transport system (cf. Fig. 4.3). These polypep-
powerful oxidants, supposedly corresponding to tides showed 35% identity and 52% similarity
the TP concentration. Neither necessarily affords with those of E. coli. They also showed that the
a clear notion of the supportive capacity of the encoding genes were almost identical to those
bioavailable forms (BAP): the MRP content, if reli- isolated from other Synechococcus strains, which
ably measurable at all, is an underestimate of had already been linked to the induction of phos-
BAP, with some or most of what is available hav- phatase activity in Synechococcus PCC7942 (Ray et
ing already been biologically assimilated. The TP al., 1991). Alkaline phosphatases are a well-known
determination is always likely to include frac- group of zinc-based enzymes which break phos-
tions that are chemically immobilised and, in phate ions from organic polymers in the external
the short term, biologically inert. The whole issue medium close to the cell. These are also said to
of what is or is not bioavailable is complex and be produced only under conditions of declining
requires a different approach (see Section 4.3.2). external orthophosphate concentrations. Ihlen-
Yet it is perfectly clear, from studies of systems as feldt and Gibson (1975) noted phosphatase pro-
far apart as Windermere and Lake Michigan, that duction in a freshwater Synechococcus at external
vernal ‘blooms’ of planktic diatoms, featuring sig- concentrations of <4 µM P.
nificant increments in chlorophyll concentration Eukaryotic phytoplankton has not been inves-
and a ≥30-fold increase in the concentrations of tigated to this level of biochemical detail. How-
cells in suspension take place against only small ever, it seems likely that analogous mecha-
changes in MRP concentration. As revealed by nisms and similar sensitivities apply among the
conventional chemical analyses, these scarcely many phytoplankters that inhabit aquatic envi-
exceed 0.1 µM (i.e. ∼3 mg P m−3 : Reynolds, ronments in which phosphate concentrations are
1992a). Some planktic algae and bacteria, at least, frequently <1 µM P and, often, an order of mag-
are sufficiently well adapted to gather phospho- nitude less again (<0.1 µM P, i.e. <10−7 mol L−1 ,
rus to fund several cell doublings despite chron- <3 µg P L−1 ).
ically low ambient MRP concentrations. Most of our present knowledge of the
phosphorus-uptake kinetics of phytoplankton
4.3.2 MRP-uptake kinetics comes from the numerous laboratory studies on
Phosphorus uptake and transport in microorgan- named species, carried out mainly in the mid-
isms are thought to depend on two separate dle years of the last century. Some of these
uptake mechanisms. In the Enterobacteria, such have been used in the compilation of compen-
as Escherichia coli, which can normally experi- dia and reviews (Reynolds, 1988a, 1993a; Padisák,
ence a much higher external concentration than 2003). In order to make valid interspecific com-
a free-living phytoplankter, a low-affinity mem- parisons (such as those in Fig. 4.5), it is nec-
brane transport system normally operates (Rao essary to convert often disparate measurements
and Torriani, 1990). If external phosphorus con- to appropriate common scales. Only volume- or
centration falls, however, to ≤20 µM (∼0.6 mg P carbon-specific uptake rates of planktic cells lend
l−1 ), the second, high-affinity system is activated. themselves to some generalisations. One of these
This one is ATP-driven and is linked directly is that the maximum phosphorus-uptake rates
to periplasmic phosphate-binding sites. Working (VUmax ) of a range of freshwater phytoplankton
with a cultured strain of the marine cyanobac- are comparable, at least within 2 orders of mag-
terium, Synechococcus sp. WH7803, Scanlan et al. nitude, when expressed per unit area of algal
(1993) demonstrated the accelerated synthesis of unit surface (Fig. 4.5b: 0.5 to 35 × 10−19 mol P
several intracellular polypeptides as their cul- µm−2 s−1 ). When normalised to cell carbon, the
tures became increasingly depleted of soluble same data translate to maximum uptake rates
Figure 4.5 (a) Absolute maximal phosphorus uptake rates too concerned about phosphorus-limited uptake
of phytoplankton cells and colonies as reported in the rates that remain capable of saturating growth
literature reviewed by Reynolds (1988a) and expressed on a rates down to external concentrations (reading
common scale. (b) The same data normalised to the surface from Fig. 4.4) of the order of 0.1 × 10−6 mol P L−1
areas of the cells or colonies as appropriate. (c) The same (see also Section 5.4.4).
data normalised to cell carbon (the shaded part of the Nevertheless, ambient concentrations of MRP
histograms correspond to the fastest carbon-specific rate of
in some natural waters may be chronically con-
P assimilation in growth; the balance represents spare
strained to this order and, in many others, be fre-
capacity (note the logarithmic scales). The algae are Ana,
Anabaena flos-aquae; Ast, Asterionella formosa; Chla, quently drawn down to such levels. We may read-
Chlamydomonas sp.; Chlo, Chlorella; Din, Dinobryon divergens; ily accept that phytoplankton tolerant of such
Eud, Eudorina unicocca; Mic, Microcystis aeruginosa; Per, conditions must invoke high-affinity mechanisms
Peridinium cinctum; Plaa, Planktothrix agardhii; Scq, Scenedesmus for phosphorus uptake. Our interest should be
quadricauda: Vol, Volvox aureus. Data presented in Reynolds sharply focused on the shape of the uptake curve
(1993a) and redrawn from Reynolds (1997a) with permission. at the extreme left-hand side of Fig. 4.4 and the
beneficial distortion represented by a relatively
low half-saturation concentration (KU ). Indeed,
of between 0.1 and 21 × 10−6 mol P (mol cell the lower is the concentration required to half-
C)−1 s−1 . Against the theoretical requirement for saturate the uptake of phosphorus, then the
phosphorus to sustain a doubling of the cell car- greater is the likely ability of the alga to fulfil
bon (9.4 × 10−3 mol P (mol cell C)−1 ), these max- its requirements at chronically low external con-
imal P-uptake rates (VUmax ) would be sufficient centrations. The faster is the uptake capacity at
to meet the growth demand in from 440 to 94 the low, markedly sub-saturating resource levels,
000 s (∼7 minutes to 26 h), supposing a saturat- then the greater is the alga’s affinity for phospho-
ing concentration and a constant rate of uptake. rus and the greater is its ability to compete for
The steady-state phosphorus requirements of the scarce resources.
same planktic species growing at their respec- Pursuing this reasoning further, Sommer’s
tive maximal cellular growth rates (from Chap- (1984) experiments distinguished several differ-
ter 5) are inserted in Fig. 4.5c to emphasise a ing adaptive strategies among freshwater phyto-
second generalisation. It is that we need not be plankton for contending with variable supplies of
PHOSPHORUS 157
Table 4.2 Some species-specific values of maximum phosphorus uptake rate (VU max ) at ∼20 ◦ C and the
external concentration of MRP required to half-saturate the uptake rate (KU , being the concentration required
to sustain 0.5 VU max )
Species VUmax µmol P KU µmol P L−1 Referencesa

(mol cell C)−1 s−1
Chlamydomonas reinhardtii 7.35 0.59 Kennedy and Sandgren
(unpublished, quoted by
Reynolds, 1988a, with
permission)
Chlorella pyrenoidosa 25.12 0.68 Nyholm (1977)
Asterionella formosa 0.51 1.9–2.8 Tilman and Kilham (1976)
Dinobryon sociale 0.21 0.39 Lehman (1976)
Scenedesmus quadricauda 3.4 1.2–4.0 Nalewajko and Lean (1978)
Anabaena flos-aquae 10.5 1.8–2.5 Nalewajko and Lean (1978)
Peridinium sp. 0.11 6.3 Lehman (1976)
Eudorina elegans 0.34 0.53 Kennedy and Sandgren
permission)
Planktothrix agardhii 5.41 0.2–0.3 van Liere (1979),
Ahlgren (1977, 1978)
Volvox aureus 5.24 1.62 Kennedy and Sandgren
permission)
Microcystis aeruginosa 1.95 0.3 Holm and Armstrong (1981)
a
The original data come from the works cited, as recalculated to a common scale of cell-carbon specificity
by Reynolds (1988a).
phosphorus. Species might be relatively velocity- interexperimental variability, even for the same
adapted, in which high rates of cellular growth species. The values noted in Table 4.2 are those
and replication (r ) are matched by suitably rapid used in the construction of Fig. 4.5. To an extent,
rates of nutrient uptake (VUmax ), or else, they may VUmax is necessarily greater than the rate of
be more storage-adapted, in which rapid, oppor- deployment of phosphorus in new cell mate-

tunistic uptake rates exceed relatively slow rates rial, supposing that this corresponds to rmax (at
◦
of deployment in growth, thereby permitting a 20 C) and that the cell quota remains con-
net accumulation of an intracellular reserve of stant. From this, it may also be deduced that
phosphorus. These adaptations are said to be dis- the uptake rate, VU , has to be markedly under-
tinguished by differences in the species-specific saturated for a considerable time before r can

ratio (VUmax /rmax ). Species may also show a ten- be said to be P-limited. The most helpful adap-
dency to be more or less affinity-adapted accord- tation to enable algae to deal with chronically
ing to the species-specific ratio, VUmax /KU ; as sug- low external MRP concentrations is a very low
gested, high affinity is imparted by a low KU half-saturation coefficient. However, it is clearly
requirement. relevant for such algae to be able still to func-
Sommer’s terminology is helpful but the tion on a relatively low internal phosphorus
derived measures are themselves subject to quota. Davies’ (1997) recent investigations of the
cell-phosphorus-related growth kinetics of nat- In another line of investigation, it has been

ural Asterionella populations during the spring- shown that some strains of Cyanobacteria are
bloom period in the English Lakes are illus- able to maintain full growth down to exter-
trative. Plotting cell-increase rates against the nal concentrations of 100 nmol P L−1 (∼3 µg
corresponding cell P quotas at the time of their L−1 ), without producing any of the regulator pro-
sampling, Davies (1997) was able to fit a sin- teins that signal the activity state of the trans-
gle, statistically significant Michalis–Menten-type port system to the controlling operons (Mann,
curve, that suggested growth rate was fully sat- 1995; Scanlan and Wilson, 1999). As suggested
urated by cell quotas of 5–10 pg P (cell)−1 (or, in Section 4.2.2, the presence of regulator pro-
roughly, 0.023–0.045 mol cell P (mol cell C)−1 ) teins is indicative of incipient cell starvation,
and half-saturated at about 0.7 pg P (cell)−1 (i.e. triggering the appropriate intracellular defensive
∼0.003 mol cell P (mol cell C)−1 ). Plotting cell reactions. Work on the bacterium Vibrio (Kjelle-
phosphorus as a function of the MRP concentra- berg et al., 1993) showed that symptoms include
tion in the lake water at the time of collection, a sharp slowdown in cell growth, following an
she showed that maintenance of a quota of this abrupt deceleration in the rates of protein syn-
magnitude was possible at an external concen- thesis. Assembly of macromolecules is halted by
tration of round 0.75 µg P L−1 (0.024 µmol P the action of synthesis inhibitors, followed by the
L−1 ). Plainly then, good growth is still possible reorganisation of the cell components and the
in the face of external depletion so long as the adjustment of the fatty-acid content of the mem-
cell quota is maintained. In contrast, at the very branes to resist lysis. In turn, these actions are
minimum cell quota (e.g. of Mackereth, 1953) cor- followed by a decline in the rate of respiration
responding to ∼0.0003 mol cell P (mol cell C)−1 , and other metabolic activity.
growth is quite impossible. Central to these reactions are the transduc-
ing signals. Certain nucleotides are known to
increase in response to falling nitrogen concen-
4.3.3 Metabolic-rate limitation by tration and amino-acid synthesis. One of these,
phosphorus guanosine 3 ,5 -bipyrophosphate (ppGpp), is gen-
How cells function in the face of low internal P erated in nitrogen-starved E. coli (Gentry et al.,
resources and very low external P supplies has 1993) and Vibrio (Kjelleberg et al., 1993). Homo-
been investigated in recent years, using a vari- logues to these are found in cyanobacterial cells
ety of alternative techniques that overcome the experiencing a sharp reduction in photon flux
problem of how to quantify chemically the small (Mann, 1995). Incipient starvation and ribosomal
amounts of determinand present. For instance, stalling are thought to lead to ppGpp synthe-
Falkner et al. (1989) applied force-flow functions, sis and, thence, to the communication of star-
derived by Thellier (1970), to demonstrate that vation. Mann’s group was able to grow plank-
the typical external concentrations of phosphate tic cyanobacteria in media in which phosphorus
below which cells of Cyanobacteria fail to balance concentrations fell to <0.1 µM (i.e. less than 3 µg
their minimal maintenance requirements indeed P L−1 ) before compounds like ppGpp began to
fall within the range 1–50 nmol L−1 (0.03–1.5 µg appear in the cells. This is strongly suggestive
P L−1 ). In the application of Aubriot et al. (2000), of the probability that cells do not experience
the importance of the affinity of uptake mecha- phosphorus shortages in media containing MRP
nisms and of the opportunism to invoke them in concentrations greater than this.
the face of erratic supplies was especially empha- Finally, in this context, the emerging tech-
sised. Hudson et al. (2000) applied a radiobioas- nique of using fluorimetric labelling to detect
say technique which was also able to demonstrate the intracellular transients induced by incipient
that the amount of phosphorus in the medium nutrient starvation (the so-called NIFT, nutrient-
supporting active phytoplankton populations can induced fluorescent transients) has been applied
fall less than 10 nmol L−1 (i.e. <10−8 M), without to microalgae grown under P-replete and P-
necessarily impairing productivity. deficient conditions to identify the reactivity
PHOSPHORUS 159
of the cells. According to the experiments of those photosynthetic organisms capable of

Beardall et al. (2001), NIFT responses were wholly supplementing or, perhaps, fulfilling their
lacking in each of four species of freshwa- requirements for nutrients and carbon by ingest-
ter microalgae in media containing ≥0.13 µM ing organic particulates are called mixotrophs.
(4 µg P L−1 ). The best-known examples come from among the
These various threads lead to a strong consen- dinoflagellates (marine and freshwater Gymno-
sus that phosphorus availability does not limit diniales and Gonyaulacales) and from among
phytoplankton activity and growth before the the Chromulinales, including Ochromonas (Rie-
MRP concentration in the medium falls almost mann et al., 1995; Geider and MacIntyre, 2002).
to the limits of conventional analytical detec- Certain pigmented cryptomonads are reputedly
tion. At this point, phytoplankton may draw on mixotrophic (Porter et al., 1985): this need not be
internal reserves such that activity is not imme- surprising insofar as the phagotrophic abilities
diately suppressed by lower external concentra- of the typically colourless cryptomonad genera
tions. Even at <0.1 µM, it is not the concentra- (such as Katablepharis and Cyathomonas) have long
tion of phosphorus that is critical so much as the been recognised (Klaveness, 1988). As a source of
capacity of the intracellular storage and the affin- phosphorus, bacterivory and phagotrophy offer
ity of the biological uptake mechanism for the a rich alternative to scarce dissolved inorganic
small amounts of bioavailable phosphorus being sources and, unlike phosphatase secretion,
turned over in the system (Hudson et al., 2000). the available resource would seem to be less
Two other mechanisms for contending with restricted. However, a low-phosphorus environ-
MRP limitation of metabolic activity are avail- ment pervades all its trophic levels: bacterivory
able to certain species of phytoplankton. The first is not a sustainable alternative to deficient MRP
involves the production of extracellular phos- if the bacteria are themselves simultaneously P-
phatases. Many freshwater species, in fact, pro- limited. Mixotrophy is particularly beneficial as a
duce alkaline phosphatases which liberate phos- supplementary source of nutrient in those (gen-
phate from organic solutes that can then be erally smaller) water bodies that receive inputs
absorbed by the alga (Cembella et al., 1984). They of terrestrial organic matter) but are otherwise
are produced in response to external MRP defi- quite oligotrophic (Riemann et al., 1995).
ciency, almost as soon as it develops (Healey,
1973). In the past, phosphatase activity has been 4.3.4 Capacity limitation and potential
considered to be indicative of phosphorus limita- phosphorus yield
tion (Rhee, 1973). There is little doubting the fact While it is clear that (probably all) phytoplank-
that phosphorus thus sequestered increases the ton can take up and assimilate the entire mea-
resource availability to the cell. For phosphatase surable MRP resource base, without first experi-
production to be able to offer any survival advan- encing rate limitation, and that, thereafter, some
tage, however, the phosphatase must be retained species at least are extremely effective in main-
at or close to the cell surface (Turpin, 1988). Phos- taining their biomass, it is no less clear that
phatase activity might then raise significantly the the maximum supportable biomass, Bmax , cannot
ability of algae to tolerate chronically P-deficient exceed the capacity of the most scarce resource
conditions. There is little evidence to suggest that relative to demand (Box 4.1). It remains generally
phosphatase production does much to enhance true that, in a large number of larger, deeper
the growth dynamics of assemblages, or any lakes in the higher latitudes, phosphorus is the
component species, when inorganic phosphorus nutrient that is exhausted first and, thus, the one
sources are effectively exhausted (Reynolds, that imposes the upper limit on the supportive
1992a). capacity of the location. The generality is sup-
The second mechanism involves the ported by the well-known ‘Vollenweider model’
phagotrophic ingestion of organic particles, and the impressive fit of the average phytoplank-
including especially other organisms such as ton biomass present in a selection lakes to the
bacteria. As indicated earlier (Section 3.4.4), corresponding average phosphorus availability in
the same lakes (Vollenweider, 1968; 1976; Organ- known resource availability. This same regression
isation for Economic Co-operation and Develop- equation (4.13) has been shown to be applicable
ment, 1982; see also Section 8.3.1). Several generi- to the prediction of the maximum algal concen-
cally similar regression models from the same era tration in other British lakes in which the MRP
(Sakamoto, 1966; Dillon and Rigler, 1974; Oglesby falls to analytically undetectable levels (Reynolds,
and Schaffner, 1975) provide analogous findings. 1992a; Reynolds and Davies, 2001), enhancing the
It has to be recognised that the datasets are supposition that the resource-limited yield is pre-
dominated by information from just the well- dictable from the available resource. It has also
studied, northern-hemisphere oligotrophic lakes been used to estimate the chlorophyll-carrying
in which later understanding confirms the max- capacity of the MRP resource in lakes where the
imum phytoplankton carrying capacity is deter- maximum crop is susceptible to other limita-
mined by the availability of phosphorus. It has tions (Reynolds and Bellinger, 1992) and is now
equally to be recognised that the condition does incorporated into the capacity-solving model of
not apply everywhere – it is less likely to apply Reynolds and Maberly (2002).
to large, continental lakes at low latitudes, espe- Whereas, it was originally estimated that:
cially to lakes in arid regions, and to smaller,
log[chla]max = 0.585 log[MRP]max + 0.801
shallower lakes at all latitudes. It certainly does
not apply to the open oceans, although its rele- (4.14)
vance to coastal waters should not be dismissed. where [MRP]max is the highest observed concen-
Nevertheless, there remains a danger in suppos- tration of MRP and [chla]max is the predicted
ing that the Vollenweider-type equations can be maximum chlorophyll (both units in µg l−1 or
used to predict phytoplankton biomass in a given mg m−3 ), the later applications are used to pre-
individual lake. Plainly, the criterion of capacity dict an instantaneous yield against the supposed
limitation by phosphorus must be demonstrable. bioavailability of P. Thus,
Moreover, the equations are statistical and not
[chla]max = 6.32[BAP]0.585 (4.15)
predictive, indicating no more than an order-of-
magnitude probability of average biomass that Estimating exactly what is bioavailable, without
may be supported. The trite circumsciption of enormous analytical effort, remains problematic.
lakes or seas as being ‘phosphorus-limited’ (or However, on the assumption that phosphatase
‘nitrogen’, or ‘anything else’ limited’) is to be activity will raise the supportive capacity only
avoided completely. negligibly and that mixotrophic enhancement
Several authors have tried to express the max- rarely applies outside the habitats in which it is
imum yield of biomass as a function of nutrient recognised (see Section 4.3.3), then the resource
availability (Lund, 1978; Reynolds, 1978c). Lund currently available to the phytoplankton is rep-
regressed maximum summer chlorophyll against resented by the unused MRP in solution plus
total phosphorus in a small lake (Blelham Tarn, the intracellular phosphorus already in the algae.
UK) in each of 23 consecutive years during which The minimum estimate of the resource in intra-
the lake underwent considerable eutrophication. cellular store can be gauged simply by reversing
Reynolds (1978c) chose to regress the chloro- Eq. (4.15) to solve the BAP invested in the standing
phyll concentrations measured in several con- crop.
trasted lakes in north-west England at the times
[cell P]min = (0.158[chla])1.709 (4.16)
of their vernal maxima against the correspond-
ing MRP concentrations at the start of the spring BAP can be estimated by first solving Eq. (4.16),
growth. Despite certain obvious drawbacks to based on the existing chlorophyll concentration,
this approach (no allowance was made for inter- then adding the equivalent intracellular cell P
mediate hydraulic exchange and nutrient supply, content thus predicted to the existing MRP con-
neither were any other loss processes computed), centration. Substituting this solution in Eq. (4.15)
the regression comes close to expressing the gives an instantaneous carrying capacity and
notion of a direct yield of algal chlorophyll for a potential chlorophyll yield.
NITROGEN 161
L−1 (i.e. 6.32 µg chla (µg BAP)−1 ); 10 µg BAP L−1

will support ≤24 µg chla L−1 (i.e. 2.4 µg chla
(µg BAP)−1 ), whereas the return on 100 µg BAP
L−1 is ≤91.4 µg chla L−1 (i.e. 0.91 µg chla (µg
BAP)−1 ). A small, biomass-limiting BAP has to be
used very efficiently but a larger base, one that
perhaps challenges the next potential capacity
of the biomass, is used with more ‘luxury’, at
least before the external resource is exhausted.
In terms of biomass, the effect may be even more
striking, bearing in mind the tendency towards
relatively lower biomass-specific chlorophyll con-
tents of phytoplankters in sparse, light-saturated,
nutrient-limited populations. Supposing a con-
stant quota of 0.02 µg chla (µg cell C)−1 (see
Section 1.5.4), the phytoplankton carbon yields
available from 1–100 µg BAP L−1 may be cal-
culated to be from 316 down to 45.7 µg C (µg
BAP)−1 . The corresponding range of phosphorus
quotas, 0.0012–0.0085 mol P (mol cell C)−1 , neatly
spans the condition of cells close to their min-
imum (q0 ) to being close to the Redfield ideal
(C : P ratio >800 to 118). This outcome is pos-
sibly more realistic than the direct solution of
Bmax = Ki /q0 (as proposed in Box 4.1), which may
greatly exaggerate outcomes extrapolated from
abundant resource bases.
Figure 4.6 (a) Observed maximum chlorophyll 4.4 Nitrogen: requirements,

concentrations in lakes in north-west England as a function of
bioavailable phosphorus, as detected by Reynolds (1992a) and
sources, uptake and
the regression originally proposed by Reynolds (1978c) on metabolism in phytoplankton
the basis of a study of just three lakes (see Eqs. 4.14, 4.15).
(b) A later, larger dataset of observed chlorophyll maxima
from UK lakes plotted against the predtion of the Reynolds Nitrogen is the second element whose relative
regression. As expected, a majority of points lie below the scarcity impinges upon the ecology of phyto-
predicted maximum but a number are above it, sometimes plankton. As a constituent of amino acids and,
substantially so. Graphs redrawn from Reynolds and Davies thus, all the proteins from which they are syn-
(2001) and Reynolds and Maberly (2002). thesised, nitrogen accounts for not less than 3%
of the ash-free dry mass of living cells (about 0.05
mol N (mol C)−1 ). This rises to around 7–8.5% in
Besides being broadly verifiable from observa- replete cells, capable of attaining rapid growth
tions (see Fig. 4.6b), Eq. (4.15) is consistent with (0.12–0.15 mol N (mol C)−1 , i.e. 6.6–8.2 C : N) (see
the underpinning physiology. The slope of the also Section 1.5.3), and to 10–12% in cells stor-
equation as plotted in Fig. 4.6a predicts a higher ing condensed proteins. However, molecular C : N
return in chlorophyll for the BAP invested at low ratios of <6 in vegetative cells are usually con-
availabilities. Thus, 1 µg BAP L−1 is predicted strued to be symptomatic of carbon deprivation
to be capable of supporting up to 6.32 µg chla (see Section 3.5.4). Relative to cell phosphorus,
the nitrogen content of replete cells is generally extreme, temperate shelf waters, especially those
in the range 13–19 mol N (mol P)−1 ; higher molec- influenced by large fluvial outfalls, may have
ular ratios (>30 N : P) are indicative of intracellu- nitrate levels of 60–70 µM. In lakes and rivers,
lar phosphorus deficiency; lower ratios (<10 N : P) especially in the temperate regions, the nitrate
are consistent with nitrogen shortages. availability may reach 50–65 µM in late win-
ter (generally the time of minimum biological
demand, slowest terrestrial denitrification and
4.4.1 The sources and availability of maximum leaching: George, 2002). In regions
nitrogen to phytoplankton subject to intensive modern agriculture and rel-
Despite the abundance of the element in the atively heavy applications of nitrogen fertiliser,
atmosphere, relative inertness of nitrogen gas leachate may raise the dissolved inorganic nitro-
rather restricts most photautotrophic exploita- gen concentration in receiving river waters to up
tion to nitrogen compounds. The element is also to 1 mM (14 g N m−3 ). However, on the ancient
poorly represented in the Earth’s crust: its occur- continents at lower latitudes and, especially, in
rence is largely restricted to biogenic layers in arid regions, the amount of nitrate lost from
sedimentary rocks. The principal forms of com- catchment topsoils is usually small and subject
bined nitrogen available to photoautotrophs are to further microbial denitrification. Thus, receiv-
the ions nitrate, nitrite and ammonium (NO− 3, ing waters tend to be relatively more deficient in
NO− 2 and NH +
4 ), although this may not be exclu- nitrate (1–10 µM, 15–150 mg N m−3 ) than in phos-
sively true for all phytoplankton (see below). Very phorus (Reynolds, 1997a). Even at temperate lat-
little of the available resource, either in lakes itudes, however, barren upland catchments may
or in the sea, is due to direct atmosphere-to- be capable of delivering only low concentrations
water linkages: most of the sources of combined of nitrate (≤15 µM). There is considerable evi-
nitrogen in water are imported from terrestrial dence (Soto et al., 1994; Diaz and Pedrozo, 1996)
systems or are recycled within the aquatic sys- of a nitrogen-regulated carrying capacity in the
tem. The ready solubility of most inorganic nitro- oligotrophic lakes of Patagonia and the southern
gen compounds, the rarity of their occurrence Andes (total N <300 µg N L−1 , some <100 µg
in secondary polymers and the redox sensitivity N L−1 ; equivalent to 7–20 µM as nitrate supplied).
of their ionic configurations assist the frequency These observations prompt questions about com-
of transformations and relocation. The biogeo- parative current deliveries of nitrate in northern-
chemical cycling of nitrogen is mediated mainly hemisphere rainfall, which may have been rela-
by organisms. Accordingly, its turnover is regu- tively more augmented by industrial airfill than
lated predominantly at the physiological level, in the southern hemisphere.
and so is extremely rapid when compared to the Nitrate ions are sensitive to the low-redox
cycling of other elements such as phosphorus or conditions (<+300 mV) in sediments, the deep
silicon. water of stratified, eutrophic lakes and seas and
Of the three main sources of inorganic com- in other (usually polluted) waters experiencing
bined nitrogen, it is the highest-oxide form that high biochemical oxygen demand. Reduction to
occurs most widely in solution in lakes and seas. lower oxides (nitrite), to nitrogen gas and ammo-
In the deep oceans, nitrate concentrations are nia is accelerated through microbial oxidation of
generally in the range 20–40 µM (280–560 mg organic carbon and its requirement for alterna-
N m−3 ) but, towards the surface (the upper tive electron acceptors to the diminishing quan-
50–100 m or so), they may be drawn down tities of oxygen. Specifically, the activites of the
severely as a result of algal and microbial uptake, denitrifying nitrate reducers like Thiobacillus deni-
to levels close to the limits of conventional anal- trificans and various pseudomonad bacteria result
ysis (∼1 µM). Among the most nitrate-deficient in the venting of nitrogen gas to the atmo-
waters are those of the North Pacific Gyre, the sphere. Nitrate ammonification occurs through
subtropical Atlantic (including the Sargasso) and the agency of facultatively anaerobic bacteria,
Indian Oceans (McCarthy, 1980). At the other such as Aeromonas, Bacillus, Flavobacterium and
NITROGEN 163
Vibrio, first reducing nitrate to nitrite. This may DON pool as well as benefit from the microbial
be excreted or, under appropriate conditions, liberation of DIN.
some of these organisms reduce the nitrite fur-
ther, to hydroxylamine (NH2 OH) and ammonium 4.4.2 Uptake of DIN by phytoplankton
(Atlas and Bartha, 1993). Phytoplankton are generally capable of active
The ammonium ions are more soluble (so less uptake of DIN from external concentrations as
volatile) than nitrogen, hence the reduction is low as 3–4 mg N m−3 (0.2–0.3 µM). Although
more of a transformation within the pool of inor- nitrate is usually the most abundant of the DIN
ganic nitrogen, denoted by DIN (dissolved inor- sources in the surface waters of lakes and seas,
ganic N), rather than a loss therefrom. Whereas ammonium is taken up preferentially if concen-
nitrate may dominate the DIN fraction in the trations exceed some 0.15–0.5 µM N (2–7 mg
open water of seas and lakes, in-situ biologically N m−3 ). This is because the initial intracellu-
mediated redox transformations may lead to the lar of assimilation of nitrogen proceeds via a
accumulation of comparable quantities of nitrite reductive amination, forming glutamate, and a
and, especially, ammonium (to >1 g N m−3 , subsequent transamination to form other amino
70 µM) in microaerophilous or anoxic environ- acids. The substrate is apparently always ammo-
ments. Ammonium is typically also present in nium (Owens and Esaias, 1976). Thus, it is both
oxic, unpolluted surface waters, though rarely in probable and energetically preferable that the
excess of ∼150 mg N m−3 (or ≤10 µM) (Reynolds, alga should use ammonium directly; nitrate and
1984a). nitrite have to be reduced prior to assimilation in
The sources of nitrogen available to phyto- reactions catalysed by (respectively) nitrate reduc-
plankton may be supplemented by certain dis- tase and nitrite reductase, so adding to the ener-
solved and bioavailable organic nitrogen com- getic cost of nitrogen metabolism. This differ-
pounds (DON). These include urea (McCarthy, ence in the energy requirement for the assim-
1972), which is produced mainly as an excretory ilation of nitrate and ammonium is reflected in
metabolite of animal protein metabolism, as well the photosynthetic quotient, being about 1.1 mol
as through the bacterial degradation of purines O2 (mol CO2 )−1 when ammonium is assimilated
and pyrimidines. McCarthy’s (1980) compilation and 1.4 when nitrate is the substrate (Geider and
of urea concentrations recorded in the litera- McIntyre, 2002, quoting Laws, 1991). Eppley et
ture reveals concentrations under 3 µg-atoms al. (1969a) devised an assay for nitrate-reductase
N L−1 (≤3 µM N) in the sea and up to ∼9 µM activity in natural populations and which shows,
in some North American rivers. Other sources consistently, that it is suppressed by ammonia
of organic nitrogen directly available to phyto- concentrations exceeding 0.5–1.0 µg-atoms N L−1
plankton include the small amounts (generally (0.5–1.0 µM N) (see also McCarthy et al., 1975).
<1 µM N) of free amino acid present in lakes More recently, the genes encoding the kinases
and seas (McCarthy, 1980). The relevant deam- for bacterial nitrogen transport have been recog-
inases are said to be produced by microalgae nised (Stock et al., 1989) and the action of ammo-
only under conditions of DIN deficiency (Saubert, nium in suppressing them has been similarly
1957; Turpin, 1988). demonstrated (Vega-Palas et al., 1992).
Of course, the size and dynamics of the DON The kinetics of DIN uptake by marine phy-
pool is of additional indirect relevance to the toplankton have been studied extensively; those
pelagic function. Far from being refractory, DON of freshwater species having received relatively
is frequently the major source of nitrogen avail- less attention. Half-saturation concentrations for
able to planktic microbes (>80% of the nitrogen uptake (KU ) by named, small-celled oceanic
available in oceanic surface waters is organic: species in culture (Eppley et al., 1969b; Caperon
Antia et al., 1991) and some of it is evidently and Meyer, 1972; Parsons and Takahashi, 1973)
metabolised rapidly (in days rather than weeks; fall within the range 0.1–0.7 µM N (when
see the review of Berman and Bronk, 2003). Plank- nitrate is the substrate) and 0.1–0.5 µM N
tic algae and cyanobacteria may contribute to the (with ammonium). Among neritic diatoms, the
corresponding ranges are 0.4–5.1 µM NO3 .N and associated with prokaryotes. This ability to ‘fix’
0.5–9.3 µM NH4 .N. Some half-saturation con- (reduce) elemental dinitrogen to ammonia is a
centrations for nitrate uptake among freshwa- widespread trait among obligate heterotrophic
ter plankters are available (Lehman et al., 1975; chemolithotrophic bacteria, the photosynthetic
Reynolds, 1987a; Sommer, 1994), typically falling bacteria and the Cyanobacteria. Certain of the
in the range 0.3–3.0 µM N. The maximum rates latter, most especially, some of the nostocalean
of DIN uptake at 20 ◦ C (calculated to be generally genera, are the only members of phytoplankton
equivalent to 0.6 to 35 µmol N (mol cell C)−1 s− 1 ) to have this capacity. Nitrogen fixation may have
are competent to saturate growth demand (D/S been a crucial step in the evolution of autotrophy
< 0.1 to 0.2: Riebesell and Wolf-Gladrow, 2002). in an increasingly oxygenic atmosphere, because
As in the general case (see Section 4.2.3), uptake of the relative volatility and extreme sparseness
and consumption achieve parity at steady rates of of nitrogen in the lithosphere. Ammonia was also
growth, at external DIN concentrations generally rare owing to its photolysis in an atmosphere
≤7 µmol N L−1 . relatively undefended against ultraviolet radia-
Conversely, nitrogen availability is unlikely tion. The early emergence of biological fixation,
to constrain phytoplankton activity and growth through the production of the dinitrogen reduc-
before the DIN concentration in the medium falls tase enzyme, provided the first means of entry
to below 7 µmol N L−1 (∼100 mg N m−3 ) in into ecosystems of large quantities of combined
the case of large, low-affinity species or below nitrogen (Falkowski, 2002). The enzyme catalyses
∼0.7 µmol N L−1 (∼10 mg N m−3 ) in the case of the reduction of dinitrogen to ammonium using
oceanic picoplankton). Activities become severely reductant produced via carbohydrate oxidation.
constrained once the cell nitrogen content falls Nitrogen fixation is a respiratory reaction:
below ∼0.07 mol N (mol C)−1 , when the cell reacts
2N2 + 4H+ + 3[CH2 O] + 3H2 O6 4NH+
4 + 3CO2
to its internal N deficiency by closing down non-
essential processes. The minimum cell quota (q0 ) (4.17)
of nitrogen in phytoplankton cells is said to be Interestingly, dinitrogen reductases are based on
0.02–0.05 mol N (mol C)−1 (Sommer, 1994). iron–sulphur prosthetic groups that are redox
Applying the statistic (K/q0 ), the ultimate sensitive: the enzymes operate only under strictly
yield or carrying capacity of the available inor- anaerobic conditions (as they did when they
ganic combined nitrogen is around 20 mol C evolved). Nitrogen fixation is rapidly inactivated
(mol N)−1 , with a possible extreme of ∼50 (17– in the presence of oxygen (Yates, 1977). In order
42 g C : g N). Before the internal nitrogen to fix nitrogen in an oxic ocean, the enzyme must
becomes yield limiting, however, the equivalence be protected from poisoning by oxygen. As Paerl
is not likely to much exceed 10 and perhaps as (1988) remarked, for compatibility between oxy-
little as 5 mol C (mol N)−1 (say, 8.5 to 4.2 g C : g genic photosynthesis and anoxic nitrogen fixa-
N). In terms of chlorophyll yield, the supportive tion to have developed represents a remarkable
capacity of 5–20 mol C (mol N)−1 is equivalent evolutionary achievement for the Cyanobacteria.
to some 0.08–0.34 g chla : g N. The factor used Until the 1960s, the nitrogen-fixing capability
in the capacity-solving model of Reynolds and of Cyanobacteria had only been suspected from
Maberly (2002), which is biassed by data from sys- nitrogen budgets (e.g. Dugdale et al., 1959). The
tems that are more likely to be P-deficient, is 0.11 introduction of the acetylene-reduction assay for
g chla : g N. nitrogenase activity (Stewart et al., 1967) made it
possible to investigate which species fixed nitro-
4.4.3 Nitrogen fixation gen, under what conditions and at which loca-
The ability to exploit the atmospheric reservoir of tions. Among the freshwater Nostocales, fixation
nitrogen gas (or, at least, that fraction dissolved is confined to the heterocysts (sometimes called
in water: at sea-level air-equilibrium, ∼20 mg heterocytes). These are specialised cells differenti-
nitrogen L−1 at 0 ◦ C, falling to ∼11 mg L−1 ated at intervals along the vegetative filaments
at 20 ◦ C) as a source of nutrient is exclusively (Fay et al., 1968). Their thick walls defend the
NITROGEN 165
intracellular anaerobic conditions necessary for trations has been demonstrated in the laboratory
the enzyme function. They are differentiated in (Ohmori and Hattori, 1974; see also Kerby et al.,
life from normal vegetative cells responding to 1987), albeit at higher levels. On the other hand,
nitrogen deficiency. Besides the thickening of the isolates of non-nitrogen-fixing Cyanobacteria
the wall, the cells lose their blue-coloured phy- from nitrogen-deficient lakes (Merismopedia, Micro-
cocyanin. However, they retain a chlorophyll- cystis, Synechococcus) used in the experiments of
based light-harvesting capacity, attached to a Blomqvist et al. (1994), all responded much more
functional PS I and ferredoxin transfer pathway positively to ammonium enrichment than they
to NADP reduction (cf. Section 3.2.1) but the did to nitrate additions. They also responded less
oxygen-evolving PS II is, of course, defunct (Wolk, positively to nitrate additions than did plank-
1982; Paerl, 1988). tic eukaryotes, including a Peridinium. Bearing in
Heterocysts are not permanent features. Natu- mind that the group evolved in an ammonium-
ral populations of Anabaena, Aphanizomemon, etc. scarce, nitrate-free, anoxic environment, a high
can increase to significant levels of biomass with- affinity for ammonium nitrogen and a low-redox
out producing heterocysts. Differentiation is a mechanism for its intracellular enhancement
facultative response to falling external DIN con- would appear to be useful adaptations. Both
centrations, to the extent that their relative fre- retain a relevance to survival and relative suc-
quency (heterocysts : vegetative cells) has been cess of distinctive members of the group in mod-
taken by ecologists to be a sign of incipient ern environments that are extremely nitrogen-
nitrogen limitation (Horne and Commins, 1987; deficient or where relatively high phosphorus
Reynolds, 1987a). Their induction and distribu- levels drive biomass accumulation of produc-
tion are regulated genetically by DNA promoters ers able to exploit extraneous sources of
(see Mann, 1995, for details). Most of the rele- nitrogen.
vant observations on their production (reviewed Besides low external DIN concentrations and
in Reynolds, 1987a) report the incidence of a low-redox, microaerophilous proximal environ-
increased heterocyst frequency, from <1 : 10 000 ment, adequate nitrogen fixation remains depen-
vegetative cells to as high as 1 : 10, in popula- dent upon high electron transport energy as
tions of Anabaena, Aphanizomenon and Nodularia, well as high rates of endogenous respiration
coincident with DIN concentrations falling below (Paerl, 1988), driven (in this instance) by photo-
300–350 mg m−3 (19–25 µM). The higher ratios synthesis and good insolation. The low reactiv-
are noted particularly at DIN concentrations ≤80 ity of N2 requires that large amounts of ATP and
mg m−3 (≤6 µM N). Given that these concentra- reducing power are invested in the nitrogenase
tions will, ostensibly, at least half-saturate the reaction (Postgate, 1987). Nitrogen fixation also
maximum rates of uptake of combined nitrogen requires phosphorus: Stewart and Alexander
and completely saturate nitrogen demand (see (1971) showed that nitrogenase activity was
above), the sensitivity of heterocyst production steadily lost in cultures of heterocystous Anabaena
to rather higher DIN concentrations is puzzling. and other nostocalean species transferred to P-
One possible explanation is that the heterocyst free medium and was not restored without the
production and, indeed, the nitrogenase activ- addition of phosphate to the medium to a con-
ity that they accommodate are actually sensitive centration equivalent to 5 mg P m−3 (∼0.16 µM
to the external concentrations of ammonium, P). The availability of molybdenum and/or vana-
which may represent much the smaller fraction dium/iron for the core of the nitrogenase enzyme
of the total DIN pool and is also the one that is is biochemically essential to nitrogen fixation
the more rapidly drawn down. This would also (Postgate, 1987; Rueter and Petersen, 1987). It
have to imply that the nitrogen-fixation response cannot be assumed that low ambient nitrogen
is a preferential reaction to low external levels of levels are automatically compensated by nitro-
NH4 .N (<0.5 µM N, or <7 mg N m−3 ) and not of gen fixation and the successful exploitation of
nitrate. Direct sensitivity of nitrogenase produc- such conditions by N2 -fixing Cyanobacteria, with-
tion in Anabaena flos-aquae to ammonium concen- out demonstrable satisfaction of the constraints
imposed by light, phosphorus and micronutrient dant than it is and contributing a larger part
deficiencies. to the oceanic turnover of nitrogen. Zehr’s (1995)
Nitrogen fixation can occur, or has been careful exploration of these questions confirmed
induced, in other non-heterocystous genera of the widespread occurrence among the Cyanobac-
freshwater Cyanobacteria (Plectonema: Stewart teria of the nitrogenase-encoding DNA but that
and Lex, 1970; Gloeocapsa: Rippka et al., 1971). its expression in nitrogenase activity was as
The maintenance of an oxygen-free microenviron- limited as previously circumscribed. It is not
ment remains a paramount precondition. One known to be expressed among the picoplanktic
way in which this can be achieved is through Cyanobacteria but fixers were sometimes incor-
the dense adpositioning of trichomes into dense porated in the microaerophilic zones of sinking
bundles or rafts (Carpenter and Price, 1976; Paerl, particulate clusters of marine snow. Expression of
1988). The effect is further enhanced by bathing nitrogenase activity among other members of the
filaments in mucilage containing reducing sul- Oscillatoriales is confined to microaerophilous
phydryl groups (Sirenko et al., 1968). Nitrogen conditions, which Trichodesmium is uniquely able
fixation also occurs among mat-forming littoral to contrive through its own growth habit. It is
species of Oscillatoria but only during darkness not relatively abundant where more nutrients
when there is no photosynthetic oxygen genera- (including DIN) deny to Trichodesmium its dynamic
tion (Bautista and Paerl, 1985). For many other advantage. As to why it is not more abundant in
common freshwater genera (Microcystis, Woroni- the low-DIN oceans, it was supposed, for a time,
chinia, Gomphosphaeria), no such facility has been that micronutrient deficiencies (in particular, of
demonstrated. iron and molybdenum) so interfere with nitro-
In the marine phytoplankton, nitrogen fix- gen fixation that the otherwise obvious potential
ers are represented by the non-heterocystous advantage that nitrogen fixation might confer is
marine species of oscillatorialean genus of Tri- suppressed. In fact, as is now well known, rel-
chodesmium. Each of the three recognised species atively high-nitrogen, low-chlorophyll regions of
has adopted the life habit of forming large macro- the great oceans augur that the biomass capacity
scopic rafts, or flakes, of uniseriate filaments. is constrained by micronutrient availability per
These bundles were sufficiently prominent in the se (see Section 4.5), which access to alternative
very clear tropical and subtropical seas where exploitable sources of nitrogen fails to alleviate.
they mainly occur for early mariners to have Not even the diatom Hemiaulus, with its nitrogen-
named them ‘sea sawdust’ (given the reddish- fixing endosymbiont, Richiella (Heinbokel, 1986),
brown accessory pigmentation of the flakes, the is able to gain much advantage over other species
name is apposite, elegantly conveying a good of the tropical gyres. The problem is still to sat-
description of their appearance). Living in envir- isfy the simultaneous requirements of nitrogen
onments of the Atlantic Ocean maintaining fixation. The ability to fix nitrogen really provides
very low levels of inorganic combined nitro- an advantage only in those parts of the sea where
gen (often <1 µM DIN), Trichodesmium thiebau- DIN is truly limiting and where energy, phospho-
tii nevertheless fixes nitrogen, aerobically and rus and adequate sources of iron, molybdenum
whilst photosynthesising, sufficient to satisfy the and vanadium are simultaneously sufficient to
bulk of its nitrogen requirements (Carpenter and support it.
McCarthy, 1975). This metabolism is energeti-
cally expensive and relatively slow but is ade-
quate to support Trichodesmium dominance over 4.5 The role of micronutrients
almost all other, nitrogen-starved phytoplankters
(Carpenter and Romans, 1991; Zehr, 1995). After the six elements that each contribute ≥1%
This being so, it has long been puzzling to of the ash-free dry mass of phytoplankton cells
ecologists why there are not more genera of (in descending order of contribution by mass,
nitrogen fixers in the nitrogen-deficient oceans C, O, H, N, P, S), all the others figure in rela-
or even why Trichodesmium is not more abun- tively small fractions of the cytological structure
THE ROLE OF MICRONUTRIENTS 167
or participate in its function. Some of these are zinc, copper, molybdenum and cobalt are nec-
used in small quantities, despite being normally essary additions to culture media (Huntsman
relatively abundant in the solute content of lake and Sunda, 1980), even if the information about
and sea water, where they are constitute some of their deployment and effects is difficult to inter-
the major ions (Na, K, Ca, Mg, Cl). Most of the pret. For instance, the cellular content of man-
remainder used by plankton cells in small quan- ganese (Mn) ranks next to that of iron. The finite
tities also generally occur naturally at low con- requirement for its central role in re-reducing
+
centrations. These used to be known as the ‘trace P680 in photosynthesis (see Section 3.2.1) is usu-
elements’ but are now more commonly referred ally fulfilled by the amounts present in lakes,
to as micronutrients. which may be sufficiently abundant to bring
Much of the early knowledge of the impor- about external deposition on the cell walls. In
tant part played by micronutrients in the growth (unspecified) excesses of manganese ions are sup-
and physiological well-being of phytoplankton posed to inhibit algal growth. Although there
came not from analysis of lake or sea water are occasional references to growth being stim-
but from the attempts to grow algae in pre- ulated by the addition of manganese (Goldman,
pared artificial media. The use of carefully formu- 1964), there is little evidence to suggest that the
lated solutions, contrived and refined through metal is ever a significant growth-regulating fac-
the experimental pragmatism of such pioneers as tor. Similar conclusions apply to zinc, copper and
Chu (e.g. 1942, 1943), Pringsheim (1946), Gerloff cobalt, insofar as each participates vitally in one
and co-workers (1952), Provasoli (see especially or more enzymic or cytochrome reactions. In
Provasoli et al., 1957) and Gorham et al. (1964; solution at concentrations >10 nmol L−1 each is
see also Stein, 1973) progressively identified the seriously toxic to a majority of algae. Copper sul-
additional ‘ingredients’ necessary to keep labora- phate is still widely used as an algicide (although,
tory clones in a healthy, active, vegetative state. in many countries, its use in waters eventually
More recently, of course, chromatographic appli- supplied for drinking is banned) and is effec-
cations, atomic-absorption spectroscopy and X- tive at concentrations of 0.3–1.0 mg CuSO4 L−1
ray microanalysis have helped to confirm and (2–6 µmol Cu L−1 ). Toxicity varies interspecifically
greatly amplify the elemental composition of among algae and in relation to the organic con-
planktic cells, in field samples as well as in lab- tent of the water (Huntsman and Sunda, 1980).
oratory cultures, even to the specific intracellu- Possible toxic effects of redox-sensitive metal
lar locations (Booth et al., 1987; Krivtsov et al., species may be magnified in relevant habitats,
1999). Yet more recently, a method for measur- including lakes subject to seasonal deep-water
ing the chemiluminescence emitted in reactions anoxia, where bioavailable species may be recy-
between metals and luminol appear to be both cled (Achterberg et al., 1997).
precise and sensitive at very low sample concen- Clear incidences of regulation (or ‘limitation’)
trations (Bowie et al., 2002). of algal activity through deficiency of these ele-
ments are scarce. In contrast, both molybdenum
4.5.1 The toxic metals and (especially) iron are known to fulfil, on occa-
The known micronutrients, as they are now sions, this key limiting role in pelagic systems.
understood, include several metals whose avail- The case for molybdenum has been made on a
ability in natural waters may vary between defi- number of occasions (Goldman, 1960; Dumont,
ciency and toxic concentrations. Some (barium, 1972). In the best-known case, additions of a
vanadium) are required in such trivial amounts few micrograms of molybdenum per litre to
that their specific inclusion in artificial culture water from Castle Lake (in an arid part of Cal-
media is considered unnecessary; their presence ifornia) were sufficient to promote, quite strik-
as impurities in other laboratory-grade chemi- ingly, the growth and the attainment of a higher
cals or among the solutes that leach from the standing biomass of phytoplankton, where pre-
containing glassware suffices for most practical viously, despite the presence of adequate lev-
purposes. On the other hand, iron, manganese, els of bioavailable P and DIN, activities had
been severely constrained (Goldman, 1960). In a than is needed by cells supplied with assimilable
later investigation of the same lake, molybde- DIN to sustain the equivalent yield of cell carbon
num addition was shown to stimulate carbon (Rueter and Peterson, 1987).
fixation and nitrogen uptake rates, especially The iron content of phytoplankton cells is
when nitrate dominated the nitrogen sources reckoned to be between 0.03% and 0.1% of ash-
(Axler et al., 1980). Molybdenum is specifically free dry mass, or about 0.1–0.4 mmol Fe : mol C.
involved in the nitrogen metabolism of the cell, The problem that cells have in meeting even this
participating as a co-factor in the action of nitrate modest requirement lies principally in the low
reductase and (in Cyanobacteria) nitrogenase, solubility of the hydrous ferric oxides that pre-
and in the intracellular transport of nitrogen cipitate in aerated, neutral waters. Thus, despite
(Rueter and Peterson, 1987). The cell requirement a relative abundance of total iron in fresh waters
is estimated to be about 1/50 000 of that for P generally (10−7 –10−5 M Fe), most of it is present in
(∼0.2 µmol Mo (mol cell C)−1 ), which can be flocs and particulates, extremely little (≤10−15 M)
accumulated from external concentrations of the being in true solution (other than at very low
order of 10−11 M. According to Steeg et al. (1986), pH: Sigg and Xue, 1994). In the open oceans, the
Mo deficiency nevertheless results in symptoms concentration of total iron is generally weaker
of nitrogen limitation, including heterocyst for- (10−8 –10−7 M Fe). Particular interest has been
mation among members of the Nostocales, even directed towards the eastern equatorial Pacific,
though rates of nitrogen fixation are themselves the sub-Arctic Pacific and the Southern Ocean
seriously impaired. (north of the Antarctic front) where concen-
trations are considerably lower still (perhaps
4.5.2 Iron <10−11 M) and where iron limitation of photosyn-
Iron, being, by weight, the most important of thesis and growth is demonstrable (Martin, 1992;
the trace components in algal cells, has long Martin et al., 1994; and see below).
been implicated in the ecology of phytoplank- It has not always been clear just how phy-
ton. The two most energy-demanding processes toplankton cells satisfy their iron needs and at
in the cell – photosynthetic carbon reduction and what point these become compromised by an
nitrogen reduction – involve the participation inadequacy of availability. Supplying inorganic
of iron-containing compounds deployed in elec- iron to cultures and maintaining it in solution
tron transport (such as ferredoxin, nitrogenase) requires the inclusion of chelating ligands, such
and pigment biosynthesis. Among the recognised as citrate (Gerloff et al., 1952). It was supposed
direct symptoms of iron deficiency are reduced that this role was fulfilled naturally by the humic
levels of cytochrome f (Glover, 1977) and the or fulvic acids, and that it was suitably imi-
blockage of chlorophyll synthesis (and, where rel- tated by including in media formulations aque-
evant, phycobilins: Spiller et al., 1982). Shortage ous extracts of soil or by substitution of repro-
of iron also impairs the structural assembly of ducible solutions of nitrilotriacetic acid (NTA)
thylakoid membranes (Guikema and Sherman, or trishydroxymethyl-aminomethane (tris). Proce-
1984). Thus, iron-deficient cells are able to har- dures soon standardised on the use of ethylene
vest relatively fewer photons than iron-replete diamine tetra-acetic acid (EDTA, usually as its
ones and photon energy is utilised less efficiently. sodium salt) in media in which cultures could
Iron limitation results in poor photosynthetic be maintained over many generations. EDTA has
yields of fixed carbon, lower reductive power and, proved very successful in this context.
hence, an impaired growth potential. Iron defi- It seems that the large molecules of EDTA are
ciencies also restrict directly the synthesis of too large to be absorbed directly by algae. The
nitrite reductase. For active dinitrogen fixers, the role of the chelate is to maintain the source of
requirement for iron is relatively greater. The iron in the medium, stably and accessibly, and
synthesis of nitrogenase and the electron power which algae are then able to exploit directly,
demand for the reduction of nitrogen draws through a process of ligand exchange. The
upon the availability of up to 10 times more iron importance of providing an iron source was an
THE ROLE OF MICRONUTRIENTS 169
incidental confirmation of the procedures for the dilute to support any growth of similarly-starved
bioassaying of natural-water samples that became algae. Within this three-order span, results were
popular in the 1960s and 1970s. The essence of erratic, either showing some or no growth but
the technique is to grow a test alga, under as which could be stimulated by the addition of
near-reproducible laboratory conditions as possi- Fe-EDTA or, on occasions, by EDTA alone. These
ble, in water sampled from a given lake or sea performances could not be explained satisfacto-
under investigation, and in samples of the same rily. Some of the variability is attributable to the
water selectively enriched (‘spiked’) with the sus- difficulties of manipulating such low concentra-
pected regulatory nutrients, separately and in tions, when even the impurities present interfere
various combinations. Thus, the chemical compo- with the nominal interpretation. It would appear,
nent that most enhanced the yield of test organ- however, that concentrations of residual TFe in
ism relative to that in the unspiked water was the range 10−11 –10−10 M may well be exploitable
deemed to be the capacity-regulating (‘limiting’) by some algae, provided chelators continue to
factor in the original sample (Skulberg, 1964; mediate their availability.
Maloney et al., 1973). The method would readily In all these cases, the maintenance of iron in
confirm previous suspicions about P or N defi- solution by organic chelates is properly empha-
ciencies but frequently, the tests would point sised. However, equal emphasis is due to the exis-
to a direct and previously unsuspected limita- tence of a mechanism for transferring chelated
tion by iron. Alternatively, the effects of N or iron in the medium into the cell. It seems most
P spikes were substantially enhanced when iron- likely that the uptake and assimilation of iron
EDTA and the relevant spike were added to the in the cell relies on reduction of the Fe-chelate
medium (Lund et al., 1975). This was true even for at or near the cell surface. In turn, this presup-
water samples from particular lakes previously poses that a redox enzyme is produced close to
and deliberately enriched with iron (Reynolds the cell membrane and whose action is to cleave
and Butterwick, 1979). The explanation for this iron from the organic chelates adjacent to the
behaviour lay almost wholly in the method and cell surface.
its requirement that lake water submitted to The minimum iron requirements of active
bioassay be first fine-filtered of all algal inocula nitrogen fixers must, fairly obviously, be rela-
and as many bacteria as possible, prior to the tively higher than those of facultative or obli-
introduction of the test organism. Reynolds et gate users of DIN. It has been suggested that
al. (1981) used serial filtrations and intermedi- dinitrogen-fixing Cyanobacteria need up to 10
ate analyses of total iron (TFe) to identify where times more iron than algae of the same species
the loss of iron fertility occurred. Even coarse growing on DIN at the same rate (Rueter and
filtration (50 µm) removed up to one-third of Peterson, 1987). However, Kustka et al. (2003) have
the TFe (as floccular material or finer precipi- explored the complexities of iron-use efficiency
tates on the algae) and glass-filter filtration (pore in the diazotrophic Trichodesmium species and cal-
size 0.45 µm) removed over half of the remain- culated the fixed-carbon and fixed-nitrogen quo-
der. From initial TFe concentrations close to 10−5 tas required to sustain daily growth rates of
M (560 µg Fe L−1 ), the passage of ∼10−6 M TFe 0.1 d−1 . The iron use efficiency was such that
iron in fine, near-colloidal suspension would nev- 1 mol Fe sustained the elaboration of between
ertheless sustain the subsequent growth of test 2900 and 7700 mol C d−1 (0.13–0.34 mmol
algae, at least to the point of exhaustion of the Fe : mol C incorporated), thus requiring the sup-
conventional N or P additions, without any fur- ply of 27–48 µmol Fe (mol cell C) d−1 . To sup-
ther enhancement of the iron or the EDTA. On ply the iron demand of a population equivalent
the basis of further experiments, reviewed in to 0.4–4 × 10−6 mol C L−1 (∼0.1–1 µg chla L−1 )
Reynolds (1997a), similar results were obtained requires an iron source of 1–2 ×10−10 M TFe.
with iron-starved algae reintroduced into arti- It is relevant to point out that many Cyanobac-
ficial media containing ≥10−8 M TFe, whereas teria (though not just the dinitrogen fixers) are
media containing <10−11 M were consistently too able to acquire and transport iron through the
production of extracellular iron-binding com- to iron deficiency. A part of the enhanced

pounds (called siderophores) that comprise part iron pool (15–40%) supported the production of
of their own high-affinity iron-transport sys- autotrophic diatoms and flagellates while the bal-
tems (Simpson and Neilands, 1976). Production ance persisted (for over 40 days after fertilisation)
is induced under iron stress and repressed by within the tight cyclic linkages involving pelagic
its relative availability. This ability is said to con- bacteria and microzooplanktic grazers.
fer some competitive advantage to Cyanobacteria We may deduce that, at least for these oceanic
over other algae (Murphy et al., 1976), though this locations, the natural iron levels are simply too
would seem to apply only to nitrogen fixers under low (∼10−10 M) to support any more autotrophic
conditions of simultaneous nitrogen stress. biomass than they do (i.e. iron availability is abso-
Piecing together the (mainly circumstantial) lutely yield limiting and it is not just nitrogen
evidence, it seems scarcely likely that iron exerts fixation that is constrained). Moreover, the struc-
any serious regulation on the activities of fresh- ture of typical iron-limited communities ensures
water phytoplankton. Algae are exposed to rel- that bioavailable iron is retained, as far as possi-
atively high concentrations of TFe supplied by ble, in surface waters.
terrestrial soils and that sufficient of the iron
is normally maintained in dissolved or colloidal 4.5.3 Organic micronutrients and vitamins
complexes with organic carbon (DOFe) for the Besides the provision of a full spectrum of inor-
carrying capacities of the nitrogen, phosphorus ganic nutrients, successive generations of sub-
and light to be simultaneously iron-replete. In cultured axenic (bacteria-free) clones of many
the sea, however, iron is much more dilute. species of microalgae are known to benefit
Apart from fluvial inputs, concentrations are, from organic supplements at low concentrations.
in part, augmented directly by wet deposition In particular, thiamine, biotin and cyanocobal-
of dust, derived from arid terrestrial (aeolian) amine (vitamin B12 ) have been shown to be essen-
sources (Karl, 2002), as well as from deep ocean tial nutrients to some species of algae at least.
vents. In much of the sea, organic ligands prob- The reviews of Provasoli and Carlucci (1974) and
ably complex sufficient of this iron to ensure its Swift (1980) highlight the widespread depen-
availability to phytoplankton (again, as DOFe) but dence upon organic micronutrients among algae
there remain areas of the ocean where there is of the ‘red’ evolutionary line. The centric diatoms
just too little iron to avoid a deficiency of sup- and several pennate species have been shown to
ply to autotrophs. It had been suggested that pro- require a supply of vitamin B12 . Most dinoflag-
duction in the oligotrophic ocean, long supposed ellates studied also require vitamin B12 , either
to be regulated by nitrogen, would be stimu- alone or in combination with thiamine or biotin
lated by iron additions permitting more nitrogen or both. Among the Haptophyceae, a majority of
(and thus carbon) to be fixed (Falkowski, 1997). species tested have a requirement for thiamine,
In the relatively high-nutrient but iron-deficient sometimes with B12 as well. Among the Chryso-
low-chlorophyll areas of the southern Pacific and phyceae, most species require the supply of two
the circumpolar Southern Ocean, subject to the or three vitamins.
IRONEX fertilisation (Martin et al., 1994), it was The biochemical requirement for these sub-
photosynthesis that was first stimulated by iron stances is general: thiamine is a co-factor in
addition (Kolber et al., 1994). In a later fertilisa- the decarboxylation of pyruvic acid; biotin is
tion experiment in the Southern Ocean (SOIREE; a co-factor in the carboxylation and transcarb-
see Bowie et al., 2001), a pre-infusion concentra- oxylation reactions of photosynthesis. Vitamin
tion of 0.4 nM was raised to give a dissolved iron B12 mediates reactions involving intramolecu-
concentration of 2.7 nM. This was very rapidly lar recombinations involving C---C bond cleav-
depleted within the fertilised patch, to under 0.3 ages (Swift, 1980). The point is that many
nM. Part of this was due to patch dilution but bacteria (including the Cyanobacteria), most
a distinct biological response confirmed that the green algae and higher plants are capable of
biomass limitation was exclusively attributable synthesising these products themselves and are
MAJOR IONS 171
independent of an external supply. As a conse-

quence of their metabolism, however, thiamine,
biotin and cyanocobalamine are generally mea-
surable components of the labile DOC fraction
in the sea. Moreover, each is present in concen-
trations (equivalent to 0.5 to 5 ng C L−1 : Williams,
1975) sufficient to saturate the demands of
individual plankters (said to be in the order
10−12 –10−10 mol L−1 : Swift, 1980). The demand
and supply of organic trace substances seem not
to exert any strong ecological outcome on the
competitive potential of plankters in the wild.
4.6 Major ions

Figure 4.7 The major ionic constituents dissolved in sea
water. Redrawn with permission from Harvey (1976).
Although the major ions in lake and sea water
(including Ca, Mg, Na, K, Cl) are no less important
to planktic cells than are P, N or the micronutri- a weak acid, they allow the strongly alkaline
ents, they are treated in less detail here because ions to press pH above neutrality, although this
their ecological role in regulating species compo- is resisted by the presence of free carbon diox-
sition and abundance is either trivial or unclear. ide in solution. Dissociation of the bicarbonate
to release free CO2 serves to buffer the water
4.6.1 Cations at a mildly alkaline level, as well providing an
Calcium belongs in the second of these cate- additional resource of photosynthetic DIC (Sec-
gories. It does have an immediate and direct rel- tion 3.4.1; see Fig. 3.17). This crucial participation
evance to the species that deploy calcium salts in the carbon-dioxide–bicarbonate system means
(usually carbonate in the form of calcite) in that calcium can have a strong selective influence
skeletal structures. Among the most notable of among phytoplankton that are variously sensi-
these are the marine coccolith-producing hap- tive to pH and carbon sources and, ultimately,
tophytes. For such organisms, there is a spe- those with an acknowledged capacity for car-
cific and finite requirement for calcium but the bon concentration (CCM; see Section 3.4.2). In
amounts dissolved in sea water are, globally, gen- this way, many chrysophyte species seem to be
erally uniform. Of the mean solute content (35 ± confined to soft-water (low-Ca) systems (unless
3 g kg−1 : Section 2.2.1), calcium accounts for there are local sources of CO2 from organic fer-
just over 1% by mass (typical calcium content, mentation), while many Cyanobacteria are sup-
∼0.4 g L−1 , i.e. ∼10−2 M, or 20 meq L−1 ) (see posed to have an affinity for calcareous waters.
Fig. 4.7). Any preference of particular species for Indeed, a large number of cyanobacterial gen-
particular water masses is unlikely to be gov- era is reputed to be relatively intolerant of acid
erned by differences in ambient calcium concen- conditions (pH <6.0; Paerl, 1988; Shapiro, 1990).
trations. Among fresh waters, however, calcium The mechanism underpinning the observation
is frequently the dominant cation (≤120 mg L−1 , has not been satisfactorily explained. The gen-
≤6 meq L−1 ), although concentrations (and spe- erality about pH sensitivity in these organisms
cific ‘hardnesses’) in individual water bodies vary is doubtless confounded by the extreme toler-
throughout the available range. The importance ance of high pH by many planktic Cyanobacteria,
of calcium hardness resides in its relation to the yet some species (e.g. of Merismopedia, Chroo-
anions; electrochemical balance in fresh waters coccus) are plainly able to function adequately in
is often contributed, substantially or in part, by notably acidic waters (pH ≥5.5; author’s observa-
bicarbonate ions. Being derived from the salt of tions). The green volvocalean alga Phacotus, whose
single cells are enclosed in a calcified envelope, is dant of the four in cytoplasm. A recently pub-
a true calcicol and its presence in sediment cores lished experimental study (Jaworski et al., 2003)
is taken to indicate highly calcareous phases in has reappriased the situation. The authors were
the development of the lake whence the cores unable to show any dependence of the growth
were extracted (Lund, 1965). rate of the freshwater diatom Asterionella on
Magnesium is the second most abundant (to potassium concentrations above 0.7 µM (27.5 µg
sodium) cation in sea water (Fig. 4.7) and is the K L−1 ), while yields were only diminished in
other common divalent cation (with calcium) in cultures in which the initial concentration was
freshwater. Although an essential component of below this level. Another diatom (Diatoma elonga-
the chlorophylls, magnesium – a single atom is tum) and a cryptomonad (Plagioselmis nannoplank-
chelated at the centre of a tetrapyrole ring – is tica) also showed no dependence on potassium
not known to limit phytoplankton production in initially supplied at >0.8 µM (31 µg K L−1 ). We
nature. Supposing Mg to be under 3% of the mass may conclude that the regulation of phytoplank-
of a chlorophyll molecule of 894 da, then even ton growth by potassium or sodium is unlikely
a very large phytoplankton population (1000 mg to occur naturally.
chla m−3 is sustained by 30 mg Mg m−3 , or ∼1.25 It is relevant to remark briefly on another sug-
µM Mg) is unlikely to challenge the availability in gestion in the early ecological literature (Pearsall,
the majority of natural waters. 1922) that the species composition of the fresh-
Similarly, sodium and potassium are rarely water phytoplankton might be sensitive to the
considered to have much influence on algal com- variable ratio of monovalent to divalent cations
position, save through their impact on ionic (M : D). Pearsall (1922) deduced that diatoms are
strength and on the effects of ionic osmosis more abundant in relatively calcareous waters
across the cell membrane. Marine algae equipped (M : D < 1.5) whereas many desmids and some
to deal with a medium containing some 10.7 g chrysophyceans occur in softer water. Talling and
Na L−1 (i.e. ∼0.47 M, or 470 meq L−1 ) would sim- Talling (1965) showed desmids increased in major
ply burst if immersed in a soft, oligotrophic lake African lakes of high alkalinity (>2.5 meq L−1 )
water containing between (say) 0.1 and 0.5 meq but where sodium rather than calcium is the
(or 2–10 mg Na) L−1 . Conversely, phytoplankters dominant cation. The observations are, broadly,
from dilute fresh waters shrivel and lyse in sea upheld by later work but the underpinning mech-
water. Neither would tolerate the extreme salin- anisms can be explained by sensitivity to the car-
ities of certain endorheic inland waters; the dis- bon sources in the media concerned.
solved sodium content of the Dead Sea is ∼1090
meq (25 g, or 1.1 mol) L−1 . The cationic strength is 4.6.2 Anions
compounded by some 140 meq (5.5 g) L−1 potas- Apart from the crucial behaviour of bicarbon-
sium and 2540 meq (58 g) L−1 calcium. At the ate, the major anions (chloride, sulphate) do not
other extreme, minimum sodium concentrations appear to limit algal production. Chloride is, by
required by algae and Cyanobacteria are to be mass, the most abundant of the dissolved ions
found in the literature, the highest of these being in sea water (∼19.3 g L−1 , i.e. ∼0.54 M, or 540
4–5 mg Na L−1 (say 0.2 meq L−1 ) for a cultivated meq L−1 ) (Fig. 4.7) and the principal agent in its
Anabaena (Kratz and Myers, 1955). However, many salinity and halinity. Among softer fresh waters,
studies on other algae and Cyanobacteria report it is generally the dominant anion (0.1–1 meq
much lower thresholds than this (Reynolds and L−1 , 4–40 mg Cl L−1 ) but, elsewhere, it may be less
Walsby, 1975). abundant than either bicarbonate or sulphate
The requirements of phytoplankton for potas- or both. The anions influence the medium and,
sium are probably similarly non-controversial. in extreme, distinguish the properties of high-
This is despite its well-known importance as a chloride, high-sulphate and carbonate-hydroxyl
constituent of agricultural fertilisers (acknowl- (soda) lakes. Sulphate also normally saturates the
edging soil deficiencies) and, in both the sea and sulphur uptake requirements of algae down to
in many fresh waters, it being the least abundant concentrations of 0.1 meq L−1 (4.8 mg SO2− 4 ,
of the four major cations, yet the most abun- 0.05 mM).
SILICON 173
In the context of sulphur biogeochemistry, In the context of the present chapter, however,
this is an appropriate point to mention the bio- the focus of interest is the extensive deploy-
logical production of dimethyl sulphide (DMS). ment of silicon polymers in the scales of the
This volatile compound evaporates from the sea Synurophyceae and, especially, in the skeletal
to the air, where it constitutes the main natural, reinforcement of the pectinaceous cell walls of
biogenic source of atmospheric sulphur. As the the Bacillariophyceae (diatoms). As the diatoms
molecules in the air act as condensation nuclei, are among the most conspicuous and abundant
DMS production has consequences on the radia- groups represented in the phytoplankton of the
tive flux to the ocean surface. In the 1980s, when sea and of fresh waters, the biological interven-
the DMS fluxes were first recognised, excitement tion in the movements of silicon are of profound
was engendered by the idea that the release by ecological and biogeochemical significance. Sev-
marine microalgae of such a substance might eral useful reviews of this topic (e.g. Werner,
contribute to the regulation of global climate. It 1977; Paasche, 1980; Sullivan and Volcani, 1981;
was cited as a practical demonstration of Love- Reynolds, 1986a) appeared 20 to 30 years ago but
lock’s (1979) Gaia principle, with living systems cre- more modern treatments are scarce. However,
ating and regulating the planetary conditions for many aspects of the pelagic availability, uptake,
their own survival. Since then, it has been recog- deployment and dynamics of silicon have been
nised that the source of the DMS is a precursor well established for some time. In lots of ways, it
osmolyte, dimethyl-sulphonioproponiate (DMSP), is an ideal nutrient to study. Later work has con-
which is synthesised by marine microalgae and siderably amplified, rather than revolutionised,
bacteria as a counter to excessive water loss. Mea- the earlier findings.
surements noted by Malin et al. (1993) showed cor- Despite its chemical similarities to carbon,
relations between DMS concentrations ranging the second most common element in the Earth’s
between 1 and 94 nmol L−1 (mean 12) and those crust is somewhat less reactive. It occurs almost
of the DMSP precursor and between the concen- invariably in combination with oxygen (as in the
trations of DMSP and chlorophyll a during a sum- minerals quartz and cristobalite) and often also
mer bloom of coccolithophorids in the north- with aluminium, potassium or hydrogen (kaolin-
east Atlantic Ocean. The volatile DMS metabo- ite, feldspars and micas – the so-called clay miner-
lite is released mainly as a consequence of the als). These are only sparingly soluble, but hydrol-
operation of the marine food web (Simo, 2001). ysis of aluminium silicates, aided by mechani-
Using 35 S-labelled DMSP, Kiene et al. (2000) have cal weathering, allows silicon into aqueous solu-
shown that DMSP supports a significant part of tion. Below pH ∼9, the dissolved reactive sili-
the carbon metabolism of the marine bacterio- con available that is exploitable by diatoms and
plankton and that it impinges upon the availabil- other algae is the weak monosilicic acid (H4 SiO4 ).
ity of chelated metals (see Section 4.5.2). The argu- Its upper concentration (at neutrality and 20 ◦ C:
ments for and against the tenancy of the Gaian ∼10−2.7 M, or ∼56 mg Si L−1 ) is regulated by the
hypothesis of supra-organismic regulation of the precipitation of amorphous silica (Siever, 1962;
planetary biogeochemistry notwithstanding, it is Stumm and Morgan, 1996). The concentrations of
plain that the DMSP–DMS metabolism plays a sig- soluble reactive silicon (SRSi) that can be encoun-
nificant role in the ecological structuring of the tered in most open fresh waters are 1 or 2 orders
oceans. lower (0.7–7 mg Si L−1 , 25–250 µM); the maxi-
mal levels found in oceanic upwellings (∼3 mg
Si L−1 ) also fall within this range. In both habi-
4.7 Silicon: requirements, uptake, tats, the concentration may be drawn down sub-
stantially, as a consequence of uptake and growth
deployment in phytoplankton by diatoms, by other algae, by radiolarian rhi-
zopods and sponges.
All phytoplankton have a requirement for the Uptake and intracellular transport of H4 SiO4
small amounts of silicon involved in protein syn- proceed by way of a membrane-bound carrier
thesis (<0.1% of dry mass; see Section 1.5.2). system that conforms to Michaelis–Menten
kinetics (Azam and Chisholm, 1976; Paasche, rect measure of the rate of recruitment of new
1980 Raven, 1983). Reported species-specific half- cells.
saturation constants (KU ) for the uptake of Nevertheless, interspecific differences in size,
monosilicic acid by marine planktic diatoms are shape and vacuole size determine that the
generally within the range 0.3–5 µmol L−1 , while amount of silicon needed to complete the new
those for freshwater species are slightly higher cell varies in relation to the mass of cytoplasm
(Paasche, 1980). The next steps, leading to the (and, hence, its content of C, P and N). Among
precipitation of the opal-like cryptocrystalline sil- the selection of diatoms included in Tables 1.4
ica polymer used in the skeletal elements and and 1.5, Si : C varies between 0.76 and 1.42 by
the species-specific morphogenesis and organi- mass (supposing the non-silica dry mass to equal
sation into the punctuate plates, ribs, bracing ash-free dry mass and 50% of this to be carbon.).
spars, spines and other diagnostic features that This fact is not to be confused with the differing
both characterise the group and facilitate their concentrations at which various diatoms experi-
identification, are closely regulated and coordi- ence growth-rate limitation by silicon availability
nated by the genome of the living cell (Li and (see Section 5.4.4).
Volcani, 1984; Crawford and Schmid, 1986). The Diatoms make a major impact upon the abi-
resultant structures, that may survive long after otic geochemical silicon cycle. This is due partly
death and which can be isolated and purified to the global abundance of diatoms but it is
by chemical treatment, inspire the deep interest compounded by their behaviour and propensity
of diatomists and taxonomists alike. Their forms to redissolve. Owing to the high density of opa-
are celebrated in compendia of scanning elec- line silica (∼2600 kg m−3 ), most diatoms are sig-
tromicrographs (for instance, Round et al., 1990), nificantly heavier than the water they displace
published essentially as aids to diatom identifi- and, hence, they sink continuously. Populations
cation, although they generally project a power- are correspondingly dependent upon turbulent
ful artistic appeal too. It is worth emphasising entrainment for continued residence in surface
that the active uptake of silicic acid is neces- waters and the loss rates through sinking remain
sary not simply to sustain the amounts of sil- sensitive to fluctuations in mixed depth (see Sec-
ica used in the skeleton but also to generate the tions 2.7.1, 6.3.2). Much of the production is even-
saturated intracellular environment essential to tually destined to sediment out or to be eaten
aid its deposition in the wall (Raven, 1983). The by pelagic consumers. Either way, there is a flux
mechanism is known to be extremely effective: of diatomaceous silicon towards the abyss. Dur-
external concentrations of SRSi can be lowered to ing its passage through the water column, sili-
barely detectable levels (≤0.1 µmol L−1 in some cic acid is leached from the particulate material,
instances: Hughes and Lund, 1962). at variable rates that relate to the size of parti-
On the other hand, the cellular silicon cles, their degree of aggregation, pH and water
requirement for a cell of a given species and size temperature, though probably not much exceed-
is quite predictable (see Section 1.5.2 and Fig. 1.9). ing 0.1 × 10−9 mol m−2 s−1 (∼3 pg Si m−2 s−1 )
The skeletal structure in diatoms comprises two (Wollast, 1974; Werner, 1977; Raven, 1983). Match-
interlocking frustules, or valves. At cell division (see ing these to the sinking rates of dead diatoms
Section 5.2.1), each of the daughter cells takes (many of which sink at <1 m d−1 ), the time
away one of the separating maternal frustules taken to dissolve completely (200–800 d) will have
and elaborates a new one to just fit inside it. The elapsed before they have settled through 1000 m
amount of silicon taken up does not much differ (calculations of Reynolds, 1986a). Larger, faster-
from the amount required to form the new frus- sinking centric diatoms may reach rather greater
tules and it is absorbed at the time of demand. depths but the point is that very little silica
This carries some negative implications for the reaches the deep ocean floor. Most will have been
new cell (see Section 5.2.2) but, because no more returned to solution in the water above, and its
silicon is withdrawn than is required, the rate of availability to future diatom growth is restored. It
its removal from solution provides a useful indi- is easy to concur with Wollast’s (1974) view that
SUMMARY 175
45% of the mass of siliceous debris in the sea tude more dilute in the medium than in the
is redissolved between the surface and a depth living cell. They have to be drawn from the water
of 1000 m. This uptake, incorporation, transport against very steep concentration gradients. Plank-
and resolution of silicon from diatoms plainly ters have sophisticated, ligand-specific membrane
shortcuts the abiotic movements of silicon; Wol- transport systems, comprising receptors and exci-
last (1974) estimated the annual consumption of tation responses, for the capture and internal
silicon by diatoms (∼12 Pg Si) reduced the resi- assimilation of target nutrient molecules from
dence time of oceanic silicon from ∼13 000 to within the vicinity of the cell. These work like
just a couple of hundred years. pumps and they consume power supplied by ATP
In contrast, however, in the relatively trun- phosphorylation. However, cells are still reliant
cated water columns of most lakes (between, say, on external diffusivities to renew the water in
5 and 40 m in depth), a significant proportion their vicinities; quantitative expressions are avail-
of the spring ‘bloom’ of settling pelagic diatoms able to demonstrate the importance of relative
does reach the sediments intact (Reynolds and motion. The work of Wolf-Gladrow and Riebe-
Wiseman, 1982; Reynolds et al., 1982b). This sell (1997) suggests that small cells may experi-
does not prevent re-solution from continuing ence an advantage over larger cells in rarefied,
but, once part of the superficial lake sediment, nutrient-poor water, as they are less reliant upon
the rate of solution quickly becomes subject turbulent motion to replenish their immediate
to regulation by the (usually high) external environments.
concentration of silicic acid in the interstitial The uptake of nutrients supplied to starved
water and by re-precipitation on the frustule planktic cells conforms to the well-tested mod-
surfaces. Other substances, including organic els based on Michaelis–Menten enzyme kinetics.
remains, may interfere (Lewin, 1962; Berner, Performances are characterised by reference to
1980). Thus, the pelagic diatoms become pre- the maximum capacity to take up nutrient (VUmax )
served in the accumulating sediment, providing and the external concentration (KU ) that will
a superb record of unfolding sedimentary events half-saturate this maximum rate of uptake (i.e.
over a long period of time and environmental that that will sustain 0.5 VUmax ). Clearly, a high
change. In this instance, the limnetic silicon flux biomass-specific uptake rate and/or an ability to
enters a highly retarded phase of its potential half-saturate it at low concentrations represent
biogeochemical cycle. advantageous adaptations. Actual performances
are conditioned by what is already in the cell
and its ‘blocking’ of the assimilation pathways
(according to the Droop ‘cell quota’ concept). A
4.8 Summary new expression (Eq. 4.12) is ventured to show how
the uptake rate is conditoned by the contempo-
The chapter deals with the components required raneous quota. These formulations are used to
to build algal cells and the means by which distinguish among uptake mechanisms that are
they are obtained, given the background of some- variously ‘velocity adapted’, ‘storage adapted’ or
times exceedingly dilute supplies. The materi- ‘affinity adapted’. The usage of the terms ‘limi-
als are needed in differing quantities, some of tation’ and ‘competition’ (in the context of satis-
which are in relatively plentiful supply (H, O, fying resource requirements) is also rationalised
S), some are abundant in relation to relatively (see Box 4.1).
modest requirements (Ca, Mg, Na, K, Cl), while Phosphorus generally accounts for 1–1.2% of
some occur as traces and are used as such (Mn, the ash-free dry mass of healthy, active cells, in
Zn, Cu, Co, Mo, Ba, Va). Four elements for which the approximate molecular ratio to carbon of
failure of supply to satiate demand has impor- 0.009. Minimum cell quota (q0 ) may vary inte-
tant ecological consequences are treated in some specifically, generally to 0.2–0.4% of ash-free dry
detail (P, N, Fe, Si). However, even the more abun- mass but some species may survive on as little
dant of these nutrients are orders of magni- as 0.03%. Natural concentrations of bioavailable
P (usually less than the total concentration in late winter (concentrations 50–70 µM). Temper-
the water but often rather more than the sol- ate lakes and rivers may offer similar levels
uble, molybdate-reactive fraction (MRP, or SRP), of resource but, again, anthropogenic activities
are frequently around 0.2 to 0.3 µM. These (especially the agricultural application of nitroge-
are variously augmented by the weathering of nous fertiliser) may augment these, up to 1 mM.
phosphatic minerals, especially in desert catch- On the continental masses at lower latitudes
ments. Forested catchments may restrict even and, especially, in arid regions, DIN losses from
this supply but anthropogenic activities (quarry- catchment topsoils are small and subject to fur-
ing, agriculture and tillage and, of course, the ther microbial denitrification, so that receiving
treatment of sewage) may significantly augment waters are often DIN-deficient in consequence
them. As phosphorus is often considered to be (1–10 µM).
the biomass-limiting constraint in pelagic ecosys- Nitrate is redox-sensitive. Ammonification is
tems, P enrichment can provide a significant mediated by facultatively anaerobic bacteria.
stimulus to the sustainable biomass of phyto- Ammonium is less volatile than elemental nitro-
plankton. Many species can take up freely avail- gen, so DIN levels are not necessarily depleted as
able phosphorus at very rapid rates, sufficient a consequence of anoxia.
to sustain a doubling of cell mass in a matter Phytoplankton generally takes up DIN from
of a few (≤7) minutes. The external concentra- external concentrations as low as 0.2–0.3 µM.
tions required to saturate the rates of growth are Although nitrate is usually the most abundant
generally under 0.13 µM P and the most affinity- of the DIN sources in surface waters, ammo-
adapted species can function at concentrations nium is taken up preferentially while concen-
of between 10−8 and 10−7 M. In the presence trations exceed 0.15–0.5 µM N. Half-saturation
of MRP concentrations >0.1 µM P, phytoplank- of DIN uptake by small oceanic phytoplankters
ton is scarcely ‘phosphorus limited’. The con- occurs at concentrations of 0.1–0.7 µM N with
clusion is supported by each of four quite dis- nitrate as substrate and 0.1–0.5 µM N with ammo-
tinct approaches to determining whether cells nium. Nitrogen availability is unlikely to con-
are experiencing P regulation. strain phytoplankton activity and growth before
Persistent P deficiency can be countered in the DIN concentration in the medium falls to
appropriately adapted species by the production below 7 µmol N L−1 (∼100 mg N m−3 ), in the case
of phosphatase (which cleaves P from certain of large, low-affinity species, or below ∼0.7 µmol
organic binders) or by phagotrophy (consump- N L−1 , in the case of oceanic picoplankton.
tion of P-containing organic particles or bacteria). In the effective absence of DIN, phytoplank-
Both rely on the sustained availability of these ton exploits the pool of dissolved organic nitro-
alternative sources of P. gen, including urea. Certain groups of bacte-
Nitrogen accounts for 7–8.5% of the ash-free ria, including the Cyanobacteria, are addition-
dry mass of healthy, active cells, in the approxi- ally able to reduce (‘fix’) dissolved nitrogen gas.
mate molecular ratio to carbon of 0.12–0.15. Min- The relevant enzymes operate only under anaer-
imum cell quota (q0 ) may not be less than 3% ash- obic conditions. In the Nostocales, usually the
free dry mass of living cells. Nitrogen is relatively most effective dinitrogen fixers in the freshwa-
unreactive, organisms having to rely on sources ter plankton, fixation is confined to specialised
of the element in inorganic combination (nitrate, cells called heterocysts. They are produced fac-
ammonium) but which are extremely soluble. ultatively under conditions of depleted DIN. Dif-
Aggregate concentrations of dissolved inorganic ferentiation commences against a background of
nitrogen (DIN) in the open sea are generally falling DIN, below 19–25 µM. It is likely that the
in the range 20–40 µM but are often depleted reaction actually responds to depletion of ammo-
towards the surface. The most N-deficient waters nium nitrogen (to <0.5 µM NH4 .N). Successful
are those of the North Pacific, the Sargasso and fixation also depends upon threshold levels of
the Indian Ocean. Shelf waters may be relatively light, phosphorus, iron and molybdenum being
more replete, especially in temperate waters in satisfied.
SUMMARY 177
In parts of the Atlantic and Indian Oceans means that the nitrogen fixers are not free to
characterised by low DIN levels (often <1 µM exploit the situation.
DIN), nitrogen is fixed by Trichodesmium spp. The Silicon plays a regulatory role in the plank-
plankters succeed over non-fixers but dinitro- ton, not as a conventional nutrient but as a
gen fixers are not more widespread in DIN- vital skeletal requirement of diatoms. The crypto-
deficient seas because the energy and micronu- crystalline, opal-like silica polymer that makes
trient requirements are not simultaneouly up the structure of the diatom frustules is pre-
satisfied. cipitated and organised within the cell from
Iron is a micronutrient, the availability of the dissolved reactive monosilicic acid that the
which is rarely problematic except in the large cells take up from solution. The transformations
oceans. However, the amounts that occur in true between external solution, internal deposition
solution are extremely small (∼10−15 M) and the and re-solution of the frustule after death are
availability to algae depends upon its chelation regulated, in part, by the solubilities of the sili-
by fine organic colloids. Some, at least, of this cic acid and of the silica polymer after death of
fraction is accessible to phytoplankton. A total- the cell.
iron (TFe) content of 10−8 M seems adequate The amounts of silicon that are deposited in
to support the needs of most species of phyto- cell walls vary interspecifically and, intraspecifi-
plankton, in which iron constitutes some 0.03% cally, with size. Cell-specific silicon requirements
and 0.1% of ash-free dry mass (about 0.1–0.4 range between 0.5% (in marine Phaeodactylum)
mmol Fe : mol C). On the other hand, media and 37% of dry mass (in freshwater Aulacoseira);
containing <10−11 M are too dilute to support among well-studied diatoms, Si : C varies between
sustained growth of algae. It seems most likely 0.76 and 1.42 by mass. Populations draw down the
that minimal productivity of the relatively high- concentrations present in natural waters, from a
nitrogen, low-chlorophyll areas of the Southern typical range, 25–250 µM, until depleted to half-
Ocean are absolutely iron deficient. The mini- saturation levels (KU ) of about 0.3–5 µM. How-
mum iron requirements of active nitrogen fix- ever, uptake of Si scarcely exceeds the amounts
ers are suggested to be relatively higher, at about deposited; silicon consumption provides an
1–2 × 10−10 M TFe. Lack of nitrogen may pre- accurate guide to the numbers of diatoms
clude most other algae but lack of sufficient iron produced.
Chapter 5
Growth and replication of phytoplankton
of intracellular processes leading to the comple-

5.1 Introduction: characterising tion of the cell cycle, net or otherwise of respi-
growth ratory costs as specified. ‘Increase rate’ (presented
as rn ) will be used in the context of the accu-
Whereas the previous two chapters have been mulation of species-specific biomass, though fre-
directed towards the acquisition of resources quently as detected by change in cell concentra-
(reduced carbon and the raw materials of tion. The potential increase in the biomass pro-
biomass), the concern of the present one is the vided by the frequency of cell division and the
assembly of biomass and the dynamics of cell intermediate net growth of the cells of each gen-
recruitment. Because most of the genera of phyto- eration will be referred to as the biomass-specific
plankton either are unicellular or comprise rel- population replication rate, signified by r . In each
atively few-celled coenobia, the cell cycle occu- instance, the dimensions relate the increment to
pies a central position in their ecology. Division the existing mass and are thus expressible, in spe-
of the cell, resulting in the replication of simi- cific growth units, using natural logarithms. In
lar daughter cells, defines the generation. On the this way, the net rate of increase of an enlarging
same basis, the completion of one full replica- population, N, in a unit time, t, is equivalent to:
tion cycle, from the point of separation of one t
δ N /δt = N t /N 0 = N 0 ern (5.1)
daughter from its parent to the time that it too
divides into daughters, provides a fundamental where N0 is a population at t = 0 and Nt is the
time period, the generation time. population at time t; whence, the specific rate of
Moreover, provided that the daughters are, increase is:
ultimately, sufficiently similar to the parent, the
rn = ln(N t /N 0 )/t (5.2)
increase in numbers is a convenient analogue
of the rate of growth in biomass. The rate of Then, the replication rate is the rate of cell pro-
increase that is thus observed, in the field as duction before any rate of loss of finished cells to
in the laboratory, is very much the average of all mortalities (rL ), so
what is happening to all the cells present and
r = rn + rL (5.3)
is net of simultaneous failures and mortalities
that may be occurring. The rate of increase in Replication rate is net of all metabolic losses, not
the natural population may well fall short of least those due to respiration rate (R). Where the
what most students understand to be its growth computation of cytological anabolism (r) is before
rate. It is, therefore, quite common for plank- or after respiration and other metabolic losses
ton biologists to emphasise ‘true growth rates’ will be stated in the text.
and ‘net growth rates’. In this work, ‘growth rate’ On the above basis (Eq. 5.2), one biomass
(represented by r) will be used to refer to the rates doubling is expressed by the natural logarithm
THE MECHANICS AND CONTROL OF GROWTH 179
of 2 (ln 2 = 0.693) and its relation to time pro- Results accumulated over a period of algal
vides the rate. Growth rate may be expressed per change can be processed to determine the mean
second but growth of populations is sometimes rate of population change over that period. Often
more conveniently expressed in days. Thus, a dou- it is convenient to find the least-squares regres-
bling per day corresponds to a cell replication sion of the individual counts for each species
rate of 0.693 d−1 , which is sustained by an aver- ln(N1 ), ln(N2 ) . . . ln(Ni ), on the corresponding occa-
age specific growth rate of not less than 8.0225 × sions (t1 , t2 . . . ti ). The slope of ln(N) on t is, mani-
10−6 s−1 , net of all metabolic and respirational festly, equivalent to rn . Much of the information
losses. to be presented, in this and subsequent chapters,
The present chapter is essentially concerned on the net rates of change in algal populations
with the growth of populations of planktic algae, and the rates of growth and replication that may
the generation times they occupy and the factors be inferred, is based on this approach. Modifi-
that determine them. It establishes the fastest cations to this technique have been devised in
replication rates that may be sustained under respect of colony-forming algae, such as Micro-
ideal conditions and it rehearses their suscepti- cystis and Volvox (Reynolds and Jaworski, 1978;
bility to alteration by external constraints in sub- Reynolds, 1983b).
ideal environments. In most instances, informa-
tion is based upon the observable rates of change
in plankton populations husbanded in experi- 5.2 The mechanics and control
mental laboratory systems, or in some sort of
field enclosures or, most frequently, on natural, of growth
multi-species assemblages in lakes or seas. They
are expressed in the same natural logarithmic 5.2.1 The cell growth cycle
units. Their derivation necessarily relies upon the The ‘cell cycle’ refers to the progression from
application of rigorous, representative and (so far the newly separated daughter to the point where
as possible) non-destructive sampling techniques, it itself separates into daughters, allocating its
to ensure that all errors due to selectivity or accumulated mass and structure between them.
patchiness of distribution are minimised. Of course, the process depends upon the ade-
Estimating the numbers or the biomass of quate functioning of the cell’s resource gathering
each species in each of the samples provides a and organisation (Chapters 3 and 4) but it also
further problem. To use such analogues as chloro- requires the accomplishment of several other key
phyll, fluorescence, light absorption or scatter is assimilatory steps. The correct proteins and lipids
convenient but each compounds the errors of must be formed; they have to be arranged in the
sampling through misplaced assumptions about relevant cytological structures and, eventually,
the biomass equivalence. They also lose a lot allocated between the pro-daughter cells. Mean-
of species-specific information. There really is while, the entire nucleic acid complement has to
no substitute for direct counting, using a good have been copied, the chromosomes segregated
microscope and based on a pre-validated subsam- and (at least among eukaryotes) the nuclear sep-
pling method, subject to known statistical confi- aration (karyokinesis) initiated. All these processes
dence (Lund et al., 1958). However, the original take finite and significant periods of time to com-
iodine-sedimentation/inverted microscope tech- plete. Throughout, their organisation and control
nique (of Utermöhl, 1931) has largely given way to are orchestrated by the genome, including espe-
the use of flat, haemocytometer-type cells (Young- cially the RNA of the ribosomes.
man, 1971). With the advent and improvement The regulation of the life-cycle events among
of computer-assisted image analysis and recogni- eukaryotes is remarkably conserved, many of
tion, algal counting is now much less of a chore. their features having analogues in the bacteria
Properly used, the computerised aids yield results (Vaulot, 1995). This much is well known, the
that can be as accurate as those of any human process of nuclear division having been described
operator. by nineteenth-century microscopists studying
180 GROWTH AND REPLICATION OF PHYTOPLANKTON
Figure 5.1 The cell cycle. The

sketch shows the vegetative growth
(G) of the new cell and, at maturity,
the break into mitotic division (M)
and separation of the daughter cells.
Completion depends crucially upon
(S) the replication of the DNA
which occurs only after the
regulatory proteins have been
‘satisfied’ that the cell has all the
resources necessary to sustain the
daughters. Sketch based on figures
in Murray and Kirschner (1991) and
Vaulot (1995), and redrawn with
organisms as mutually disparate as yeasts and tal plates between the separating daughters: each
frogs. As is also now well known, the primary has to produce and assemble the replacement
alternation is, of course, between nuclear inter- parts (for details, see Pfiester and Anderson,
phase, during which the cell increases in mass but 1987). The cytokinetic bequest of one parental
the nucleus remains intact, and mitosis, during frustule requires each daughter diatom to pro-
which the nucleus is replicated. Mitosis follows a duce anew the relevant complementary (inter-
strict sequence of steps, starting with the break- nal) valve (see Crawford and Schmid, 1986, for
down of the nuclear membrane (prophase); the more details). Similar issues of scale or coc-
duplicated chromosomes first align (metaphase) colith replacement respectively confront divid-
and then separate to the poles of the nuclear ing synurophyceaens (Leadbeater, 1990) and
spindle (anaphase), before the propagated chromo- coccolithophorids (de Vrind-de Jong et al.,
somes become re-encapsulated in separate nuclei 1994).
(telophase). The rest of the cell contents divide After separation, the next generation of
around the two daughter nuclei, the original daughters resumes growth during the next
maternal cytoplasm thus becoming divided to period of nuclear interphase. Following the dis-
contribute the substance of each of the two new covery of Howard and Pelc (1951) that DNA syn-
daughter cells, which are eventually excised from thesis is discontinuous and confined to a distinct
each other to complete the cell division. It is not segment of the cell development, the interphase
yet wholly clear how elaborate organelles (such as is also subdivided into corresponding the peri-
flagella, vacuoles, eye-spots) are reallocated but ods. These are denoted by S, signifying synthesis,
the daughters soon copy what they are miss- G1 for the preceding gap and G2 for the succeed-
ing. Cell division in the markedly asymmetric ing gap, terminated by the inception of mitosis
dinoflagellates allocates the armoured exoskele- prophase (see Fig. 5.1).
THE MECHANICS AND CONTROL OF GROWTH 181
Assembling the growing cell has been analo- transport and protein-synthesis pathways (see
gised to a factory production line, with a highly Sections 3.4.2, 4.2.2). Transcription of the relevant
sensitive quality supervision over each process operon genes is stimulated by the group of cAMP
(Reynolds, 1997a). Close coordination is required receptor proteins (or CRP) associated with the
to sequence the events of interphase in the cor- active pathways. Thus, marked resource under-
rect order, to check stocks and marshal the com- saturation or slow delivery are reflected in weak-
ponents and, if something is missing, to deter- ened CRP flow and weakened operon activity.
mine that production should be suspended. It is Much like the air-brakes on a train, active trans-
also needed to initiate mitosis in a way that keeps port and synthesis act to suppress the cell’s pro-
the size of the daughter cells so evidently similar tective features but, as soon as normal functions
to that of the parent cell prior to its own division. begin to fail, the mechanisms for closing non-
The ‘supervision’ is, in fact, achieved through the essential processes and conserving cell materials
activity of a series of regulatory proteins, knowl- are immediately expressed. Incipient starvation
edge of which has developed relatively recently and the stalling of the relevant ribosomes lead to
(Murray and Hunt, 1993; good summaries appear the activation of the inhibitory nucleotides, such
in Murray and Kirschner, 1991; Vaulot, 1995). A as ppGpp (see Section 4.3.3). These do not just
maturation-promoting factor (MPF) occurs at a arrest maturation but may induce the onset of a
low concentration in newly separated daughter resting condition, with a substantial reduction
cells but it increases steadily throughout inter- in all metabolic activities, including RNA and
phase. Purification of MPF showed it to be made protein synthesis and respiration. The machin-
of two kinds of protein. One of these, cyclin ery is said to remain in place for at least 200 h
B, is one of several cyclins produced and peri- after such shutdowns (Mann, 1995). If renewed
odically destroyed through the cell cycle, each resources permit, however, renewed protein syn-
being specific to a particular part of the cycle. thesis may be induced within minutes of their
Cyclin B is the one that reaches its maximum availability.
concentration at the start of mitosis. The second So long as the conditions remain benign, opti-
part of the MPF is a kinase. It had been first mal functioning is supported and CRP generation
discovered in a strain of mutant yeast, codified is upheld. The cell recognises that it has suffi-
cdc. The mutation concerned the presence of a cient in reserve to be able to fulfil the mitosis and
kinase-encoding gene, called cdc2, and the 34- so allow the DNA replication to proceed. Then,
kDa kinase protein was referred to as p34cdc2. there really is no escape from the commitment.
Like all kinases, it is active only when phospho- A further group of substances, called licensing
rylated (cf. Section 4.2.2). In this state, it triggers factors, are bound to the chromosomes but are
prophase spindle formation. After completion of destroyed during DNA replication. Licensing fac-
the mitosis, the kinase is dephosphorylated and tors cannot be re-formed while the nuclear mem-
the cyclin B is rapidly degraded (by a cyclin pro- brane is intact, so that the DNA synthesis can
tease), thus leading to the destruction of MPF. occur only once per generation.
In the daughter cells, cyclin B is synthesised Though they lack a membrane-bound nucleus
again and MPF begins to accumulate for the next and any of the physical structure of a spindle
division. upon which to sort the replicated chromosomes,
The cues are critical. In this instance, it the prokaryotes have no lesser need than eukary-
is clearly the instruction to phosphorylate the otes for a closely controlled, phased cell cycle.
kinase that triggers the mitosis. The commit- At slow rates of growth, the DNA replication
ment to nuclear division is, however, made ear- of E. coli occupies only a fraction of the gen-
lier, when, following a sequence of signals pro- eration time, giving a delineation analogous to
cessed through the preceding G1 period of inter- the G1–S–G2 phasing represented in Fig. 5.1. In
phase, the DNA is finally replicated (the S phase the fast-growing Synechococcus, DNA replication
in Fig. 5.1). Activation depends upon satisfaction may occupy relatively more of the cell genera-
of resource adequacy, which is communicated tion time. Completion of the DNA duplication
by the operons informed by the intracellularis the cue for chromosome separation and cell
division (Armbrust et al., 1989). Cycle regulation transport and the intricacies of frustular mor-
by analogues of MPF is probable. phogenesis in forming the two new frustules
required to complete each cell division (see Sec-
5.2.2 Cell division and population growth tion 4.8), the actual construction of the new silica
All the eukaryotic species of phytoplankton that structures is confined to a relatively short period.
have been investigated conform to G1–S–G2 phas- The latter extends from just after nuclear divi-
ing (Vaulot, 1995). Curiously, the best-studied sion to the point of eventual cell separation. Hav-
freshwater species belong to the Chlorococcales ing passed G1 of interphase in its vegetative con-
or to the Volvocales, which undergo a relatively dition, the cell commences to form the new valve
prolonged growth period that is followed by a in a silica-deposition vesicle just beneath the
fairly rapid series of cell divisions, resulting in plasmalemma (Drum and Pankratz, 1964). The
the formation of between four (as in genera of origins of the silicon deposition vesicle and of
Chlorococcales such as Chlorella and Scenedesmus) the control of the highly species-specific pattern-
to 16 or 32 (in Eudorina) or, perhaps, as many ing of the new valves are described in detail in
as 1000 (Volvox itself ) daughter cells. Suppressed Pickett-Heaps et al. (1990). The trigger for the pro-
though the smooth transition between genera- cess is DNA replication. Thus, no new wall forms
tions may be, it is also quite evident that, over without the initiation of the nuclear division.
two or three generations, the increase in specific Equally, the commitment to division is taken
biomass adheres closely to a smooth exponen- before the parent cell has taken up sufficient sol-
tial rate (Reynolds and Rodgers, 1983). Reynolds’ uble reactive silicon either to fulfil the skeletal
(1983b) deductions on the increase in biomass of demand of the cell division, or to be able to main-
a field population of Volvox aureus are also contain the necessary internal concentration (Raven,
sistent with a smoothing of both cell growth 1983). This carries ecological consequences: if the
and population growth with respect to the cell- requirement is pitched against low or falling
division sequence. To treat the growth rate and external concentrations of monosilicic acid, it
the times of consecutive generations (tG ) in the is possible that, in a growing population, cells
same terms as simple binary fission times may begin division in a silica-replete medium but
be cautiously justified. Based on Eq. (5.2), we encounter deficiency before it is completed. Many
deduce: cells may fail to complete the replication and die
(Moed, 1973).
r = ln 2(tG )−1 (5.4)
While external concentrations continue to
On the other hand, every confidence can be satisfy uptake requirements (KU : 0.3–5 µmol L−1 ;
accorded to deductions about the observable rate Section 4.7), the demand is assuaged and the
of population growth over consecutive genera- completion of the next generation can reason-
tions of diatoms. There is a manifest similarity ably be anticipated. For all phytoplankton, the
of size between parent and daughter cells, owing processes of gathering of raw materials equiva-
to the shared bequest of the parental frustules. lent to its current mass, of assembling them into
The cellular requirements of each species are also species-specific proteins and lipids and, then, into
remarkably constant, at least when cell size is the correct cytological structures and organelles,
taken into account (Lund, 1965; see Tables 1.4, each occupy a finite period of time. Once accom-
1.5). Cells take up little more monosilicic acid plished, further time is required for the com-
than they require to sustain the skeletal demands pletion of the S, G2 and mitosis phases, prior
of the imminent division (Paasche, 1980), includ- to the occurrence of the final cytokinetic sepa-
ing that needed to maintain the necessary inter- ration. As stated above, this event or, at least,
nal concentration (see Section 4.7). These char- the frequency with which it occurs, is of fun-
acters lend themselves to accurate computation damental significance to the plankton ecologist.
of biomass increase from the division of cells To predict, measure accurately or model the
and its direct analogy to silicon uptake (Reynolds, rate of growth of cells has long been an ambi-
1986a). Despite the complicated kinetics of sil- tion of students of the phytoplankton. Most of
icon uptake, the constraints on intracellular the convenient, traditional determinations of the
THE DYNAMICS GROWTH AND REPLICATION 183
measured rates of change in numbers are, of be increasing rapidly and, generally, will be dou-
course, surrogates of cell growth and they are bling its mass at approximately regular intervals.
always net of metabolic losses and, often, also net It is early in this exponential phase that the max-
of mortalities of replicated cells. The rates of cell imal rate of replication is achieved, when r is
replication are rarely predictable from separate supposed to be equal to the observed net rate
determinations of the capacities deduced from of increase, rn , solved by Eq. (5.2). Later, as the
experimental measurements of photosynthetic resources in the medium become depleted or the
rate or nutrient uptake rate (although these rep- density of cells in suspension begins to start self-
resent upper limits). It is easy to share the frustra- shading, the rate of increase will slow down con-
tions of all workers who have struggled with the siderably (eventually, the stationary phase). Care is
problem of the determination of in-situ growth taken to discount the biomass increase in this
rates. phase from the computation of the exponent of
the maximum specific growth rate, r .
The fastest published rate of replication for
5.3 The dynamics of phytoplankton any planktic photoautroph is still that claimed
growth and replication in for a species of Synechococcus (at that time named
controlled conditions Anacystis nidulans) by Kratz and Myers (1955).
At a temperature of 41 ◦ C, the Cyanobacterium
increased its mass 2896-fold in a single day,
There is another way to address the potential through the equivalent of 11.5 doublings (tG =
rates of cell replication of phytoplankton and 2.09 h), and sustaining a specific rate of expo-
that is through the dynamics of isolates under nential increase (r ) of 7.97 d−1 (or 92.3 × 10−6
and carefully controlled conditions in the labora- s−1 ). Numerous other algal growth rates are
tory. We might begin by assessing how well the recorded in the literature cover wide ranges
cells of a given species perform under the most of species, culture conditions and temperatures.
idealised conditions it is possible to devise. Once Several notable attempts to rationalise and com-
an optimum performance is established, further pile data for interspecific comparison include
experiments may be devised to quantify the influ- those by Hoogenhout and Amesz (1965), Reynolds
ences of each of the suspected controlling fac- (1984a, 1988a) and Padisák (2003). The selection of
tors. Finally, the various impositions of sub-ideal entries in Table 5.1 is hardly intended to be com-
growth conditions can be evaluated. The follow- prehensive but it does refer to standardised meas-
ing sections apply this approach to a selection of urements, made at or extrapolated to 20 ◦ C, on a
freshwater species of phytoplankton. diversity of organisms of contrasting sizes, shapes
and habits. The temperature is critical only inso-
5.3.1 Maximum replication rates as a far as it is uniform and that it has been a popular
function of algal morphology standard among culturists of microorganisms. It
The collective experience of culturing isolates of is probably lower than that at which most indi-
natural phytoplankton in the laboratory has been vidual species (though not all) achieve their best
summarised by Fogg and Thake (1987). The fastest performances (see Section 5.3.2 below).
rates of species-specific increase are attained in The entries in Table 5.1 show a significant
prepared standard media, designed to saturate range of variation (from ∼0.2 to nearly 2.0 d−1 ).
resource requirements, when exposed to con- There is no immediately obvious pattern to the
stant, continuous light of an intensity sufficient distribution of the quantities – certainly not
to saturate photosynthesis, and at a steady, opti- one pertaining to the respective phylogenetic
mal temperature. Even then, maximal growth affinities of the organisms, nor to whether they
rates are not established instantaneously. There are colonial or unicellular. The variations are
is usually a significant ‘lag phase’ during which not random either, for in instances where more
the inoculated cells acclimatise to the ideal world than one authority has offered growth-rate deter-
into which they have been introduced. Within a minations for the same species of alga, the
day or two, however, the isolated population will mutual agreements between the studies has been
d−1 ) reported for some freshwater species of phytoplankton

Table 5.1 Maximum specific growth rates (r20
in laboratory cultures, under continuous energy and resource saturation at 20 ◦ C

Phylum Species r20 d−1 References
Cyanobacteria Synechococcus sp. 1.72a Kratz and Myers (1955)
Planktothrix agardhii 0.86 Van Liere (1979)
Anabaena flos-aquae 0.78 Foy et al. (1976)
Aphanizomenon flos-aquae 0.98 Foy et al. (1976)
Microcystis aeruginosab 1.11 Kappers (1984)
Microcystis aeruginosac 0.48 Reynolds et al. (1981)
Chlorophyta Chlorella strain 221 1.84 Reynolds (1990)
Ankistrodesmus braunii 1.59a Hoogenhout and Amesz (1965)
Eudorina unicocca 0.62 Reynolds and Rodgers (1983)
Volvox aureus 0.46 Reynolds (1983b)
Cryptophyta Cryptomonas ovata 0.81a Cloern (1977)
Eustigmatophyta Monodus subterraneus 0.64a Hoogenhout and Amesz (1965)
Chrysophyta Dinobryon divergens 1.00 Saxby (1990), Saxby-Rouen et al.
(1997)
Bacillariophyta Stephanodiscus hantzschii 1.18 Hoogenhout and Amesz (1965)
Asterionella formosa 1.78 Lund (1949)
Fragilaria crotonensis 1.37 Jaworski, in Reynolds (1983a)
Tabellaria flocculosa var.
asterionelloides 0.66 G. H. M. Jaworski (unpublished data)
Dinophyta Ceratium hirundinella 0.21 G. H. M. Jaworski (unpublished data)
a
Rate extrapolated to 20 ◦ C.
b
Unicellular culture.
c
Colonial culture.
generally excellent (Reynolds, 1988a, 1997a). replication rates shown in Table 5.1 (save the
Where there has been a significant departure, as entry for Dinobryon that was not available to the
there is in the case of the two entries for Micro- 1989 compilation) and the corresponding species-
cystis, it is attributable to the difference between specific surface-to-volume ratios noted in Table
working with a colonial strain and, as is com- 1.2, is reproduced in Fig. 5.2. The regression line
mon among laboratory cultures, one in which fitted to points plotted in Fig. 5.2 is:
the colonial habit had been lost. This turns out

to be an important observation, for it provides r20 = 1.142(s/v)0.325 d−1 (5.5)
the clue to the robust pattern that accounts
for a large part of the variability in organis- where s is the approximate area of the algal sur-
mic replication rate, which relates to organismic face (in µm2 ) and v is the corresponding vol-
morphology. The apparent dependence of growth ume (in µm3 ). Both dimensions were estimated
rate on algal size and shape, suggested by ear- from microscope measurements and the com-
lier analyses (Reynolds, 1984a) was convincingly pounding of relevant geometric shapes (Reynolds,

confirmed when the replication rates (r20 ) were 1984a). The coefficient of correlation is 0.72; thus,
plotted against the surface-to-volume ratio (sv−1 ) 52% of the variability in the original dataset is
ratio of the life-form, regardless of whether it was explained.
a unicell, coenobium or mucilage-bound colony The outcome is instructive in several ways.
(Reynolds, 1989b). The relationship, between the First, it is satisfying that surface-to-volume
volume. It was pointed out first by Lewis (1976)

that the morphologies of marine phytoplankton,
despite embracing a range of sizes covering 6 or
7 orders of magnitude, are such that many of the
larger ones are sufficiently non-spherical for the
corresponding surface-to-volume values to vary
only within 2. He deduced that this conservatism
of (sv−1 ) was not incidental but a product of nat-
ural selection. When Reynolds (1984a) attempted
the analogous treatment for a selection of fresh-
water phytoplankton, nearly 3 orders of magni-
tude of variation in (sv−1 ) were found, against
over 9 orders of variation in the corresponding
unit volumes (see 1.7). The cytological relation-
Figure 5.2 Maximum growth and replication rates of ship of cell surface to cell mass is clearly a rele-
phytoplankters in continuously light- and nutrient-saturated
◦ vant factor in cell physiology.
media at constant temperatures of 20 C, plotted against their
The second interesting feature of Fig. 5.2 is the
respective surface-to-volume ratios. The algae are: An flo,
Anabaena flos-aquae; Aphan, Aphanizomenon flos-aquae; Ast,
slope of r20 on (s/v): the exponent, 0.325 agrees
Asterionella formosa; Cer h, Ceratium hirundinella; Chlo, Chlorella closely with Raven’s (1982) theoretical argument
sp.; Cry ov, Cryptomonas ovata; Eud, Eudorina unicocca; Fra c, for growth conforming to a model relationship
Fragilaria crotonensis; Mic, Microcystis aeruginosa; Monod, of the type, r = a constant × (cell C content)−0.33 .
Monodus sp.; Monor, Monoraphidium sp.; Pla ag, Planktothrix Raven’s supposition invoked the slower increase
agardhii; Ste h, Stephanodiscus hantzschii; Syn, Synechococcus sp.; of surface (as the square of the diameter) than
T fl, Tabellaria flocculosa; Volv, Volvox aureus. The fitted the volume (as its cube) of larger-celled organ-
least-squares regression is r 20 = 1.142 (s/v) 0.325 . Redrawn
isms. Assuming, for a moment, carbon content
with permission from Reynolds (1997a).
to be a direct correlative of protoplasmic vol-
ume (as is argued in Section 4.2), and the sur-
should provide such a strong allometric state- face area to be a 2/3 function of increasing vol-
ment about the capacity of the alga to fulfil its ume, then the arithmetic of the indices to (s/v)
ultimate purpose. The empirical relation between clearly sum to 1/3. Raven’s (1982) derivation dif-
these two attributes is, of course, not constant, fers from the more frequently quoted deduction
even among individual spherical cells. Rather, it of Banse (1976), namely r = a constant × (cell C
diminishes with increasing diameter (d), with a content)−0.25 , which had been based upon marine
slope of exactly 2/3. The unit in which (sv−1 ) is diatoms. Reynolds’ (1989b) equation (i.e. that in
expressed, (µm2 /µm3 =) µm−1 , has the dimen- Fig. 5.2) is not necessarily at odds with Banse’s
sion of reciprocal length and conveys the idea findings, as the assumption of a constant rela-
that the decline in assimilation and growth effi- tionship of C to external cell volume does not
ciency might be primarily a function of the hold for the (larger) diatom cells characterised
intracellular distance that metabolites must be by large internal vacuolar spaces.
conducted within the cell. The implication is A further satisfaction about the relationship
that small spherical cells are metabolically more comes from the fact that the scatter of points
active than large ones. If there is to be an advan- shown in Fig. 5.2 becomes a cluster at the end
tage in being larger, it must either be at the of the regression line fitted by Nielsen and Sand-
expense of the potential for rapid growth, or Jensen (1990) to the growth rates of higher plants
it should invoke a simultaneous distortion in as a function of their surface-to-volume ratios.
shape. As the sphere is bounded by the least pos- The slopes of the two regressions are almost iden-
sible surface enclosing a given three-dimensional tical. Presumably, small size, the consequent rel-
space, any distortion from the spherical form atively high surface-to-volume ratio, structural
increases the surface area relative to the enclosed simplicity and the exemption from having to
allocate resources to the production of mechan- plotting (Fig. 5.3), data were normalised by relat-
ical and conducting tissues provide the main ing the logarithm of daily specific replication rate
reasons for the high rates of specific biomass at the given temperature, log(rθ ), to the given
increase among planktic microalgae relative to temperature (θ ◦ C) rendered on an Arrhenius
those of littoral bryophytes and angiosperms. scale. The latter invokes the reciprocals of abso-
The ability of the plankter to exchange materials lute temperature (in kelvins). Thus, 0 ◦ C (or 273 K)
across and within its boundaries is a key determi- is shown as its reciprocal, 0.003663. For manipu-
nant of its potential physiological performance. lative convenience, the units shown in Fig. 5.3a
That ability is strongly conditioned by the ratio are calculated in terms of A = 1000 [1/(θ K)]. The
of its surface to its volume. thousand multiple simply brings the coefficient
into the range of manageable, standard-form
5.3.2 The effect of temperature numbers.
Sourcing from the various literature compi- In this format, the temperature-response plots
lations mentioned in the previous section, reveal several interesting features. These include
Reynolds (1984a) deduced that, with the excep- interspecific differences in the temperature of
tion of acknowledged cold-water stenothermic maximum performance, ranging from appearing
and thermophilic species, most laboratory strains at a little over 20 ◦ C (a little under 3.42 A) in Apha-
of planktic algae and cyanobacteria then tested nizomenon flos-aquae and Planktothrix agardhii (orig-
achieved their maximal specific rates of replica- inal data of Foy et al., 1976) but somewhere >41 ◦ C
tion in the temperature range 25–35 ◦ C. A few (<3.19 A) in the Synechococcus of Kratz and Myers
(like the Synechococcus of Kratz and Myers, 1955) (1955). The differences in the normalised parts
maintain an accelerated function beyond 40 ◦ C of the species-specific slopes reflect interspecfic
but, exposed to their respective supra-optimal differences in sensitivity to variation in tempera-
temperatures, the replication rates of most of ture. Thus, the slope fitted to the data for Synecho-
the species considered here first stabilise and, coccus, for example, has the value β = −3.50 A−1 .
sooner or later, fall away abruptly. From 0 ◦ C to Cast in the more traditional terms of the Q10
just below the temperature of the species-specific expression for the factor of rate acceleration over
optimum, the replication rates of most plankters a 10- ◦ C step in customary temperature (usually
in culture appear, as expected, to increase expo- that from 10 to 20 ◦ C), the normalised response
nentially as a function of temperature. However, has a Q10 of ∼2.6. In contrast, the slope for colo-
the degree of temperature sensitivity of the div- nial Microcystis, β = −8.15 A−1 (Q10 ∼ 9.6), reveals
ision rate is evidently dissimilar among plank- a relatively greater temperature sensitivity. The
ters. In some, growth rates vary by a factor of ∼2 slopes, of course, reflect interspecific differences
for each 10 ◦ C step in temperature, as Lund (1965) in the sum of metabolic responses to tempera-
recognised; for others, the temperature depen- ture fluctuation, some of which are themselves
dence of growth rate is more sensitive and the differentially responsive to thermal influence.
slope of r on temperature is steeper. Whereas, for instance, photosynthetic electron
Seeking some general expression to describe transfer has a Q10 of approximately 2 over a 30 ◦ C
the sensitivity of algal replication rates to temper- Range (see Section 3.2.1) and both light-saturated
ature, Reynolds (1989b) used the same data com- photosynthesis and dark respiration carry Q10 val-
pilations to identify sources of relevant informa- ues in the range 1.8–2.25 (Section 3.3.2), protein
tion on growth performances. What was needed assembly has a Q10 said to exceed 2.5 (Tamiya
were the maximum rates of replication of named et al., 1953). Whereas the rates of growth of the
algal species maintained in culture under con- plankters whose (relativly gentler) slopes appear
stant, photosynthesis-saturating light conditions towards the top of Fig. 5.3a might reflect tem-
and initially growth-saturating levels of nutrients perature constraints on individual assembly pro-
but at two or (ideally) more constant tempera- cesses, the (steeper) slopes towards the bottom
tures. Data satisfying this criterion were found of the figure refer to larger, low (s/v) forms
for 11 species. For the purpose of comparative and coenobial Cyanobacteria whose responses
of temperature sensitivity of replication (β, from

Figure 5.3 Temperature dependence of light-saturated Fig. 5.2a) against the corresponding organismic
growth in a selection of freshwater phytoplanktic species. (a) (sv−1 ) value. β is a significant (p < 0.05) correlative
Datasets normalised against a temperature on an Arrhenius of the surface-to-volume attributes of the eleven
scale (103 /K); (b) the slopes of the regressions plotted algae tested. The regression, β = 3.378 − 2.505
against the surface-to-volume ratios of the algae. The algae log (sv−1 ), has a coefficient of correlation of 0.84
are: Ana, Anabaena flos-aquae; Ana cyl, Anabaena cylindrica; Ank, and explains 70% of the variability in the data.
Ankyra judayi; Aph, Aphanizomenon; Ast, Asterionella formosa:
The open circles entered in Fig 5.3a are not part
Coel, Coelastrum microporum; Cry er, Cryptomonas erosa; Cry ov,
Cryptomonas ovata; Dict, Dictyosphaerium pulchellum; Fra c,
of the generative dataset but come from a previ-
Fragilaria crotonensis; Lim r, Limnothrix redekei; Mic, Microcystis ously untraced paper of Dauta (1982) describing
aeruginosa; Monor, Monoraphidium sp.; Ped b, Pediastrum the growth responses of eight microalgae. They
boryanum; Pla ag, Planktothrix agardhii; Sc q, Scenedesmus have been left as a verification of the predictive
quadricauda; Syn, Synechococcus. The least-squares regression value of the regression.
inserted in (b) is fitted only to the solid points; its equation is The hypothesis that algal morphology also
β = 3.378 –2.505 log (sv−1 ). Redrawn with permission from regulates the temperature sensitivity of the
Reynolds (1997a).
growth rate is not disproved. For the present, β
can be invoked to predict the growth rate of an
are determined by the temperature sensitivity of alga of known shape and size at a give tempera-
the slowest processes of intracellular assimilation ture, θ, following Eq. (5.6):
and relative rates of surface exchange (Foy et al.,
1976; Konopka and Brock, 1978).
log(rθ ) = log(r20

) + β[1000/(273 + 20)
This pattern of behaviour is emphasised in
the plot (in Fig. 5.2b) of the species-specific slopes − 1000/(273 + θ )]d−1 (5.6)
5.3.3 Resourcing maximal replication (1990) calculated a possible carbon delivery rate
That the slowest anabolic process should set of 35.7 × 10−6 g C (g cell C)−1 s−1 . This is suffi-
the fastest rate of growth leads to an impor- cient to meet the full growth requirement in 19
tant corollary of concern to the ecologist. It is 416 s, or about 5.4 h. Interestingly, it is also pos-
simply that it cannot be obtained without each sible to deduce, from the number of photosyn-
of its resource requirements being supplied at thetic reaction centres represented by 1 g chla
demand-saturating levels. At steady state, the rate (between 2.2 and 3.4 × 1018 ) (see Section 3.2.1)
of photosynthesis and the rates of uptake of each and their operational frequency at 20 ◦ C (∼250
of its nutrients match the growth demand. More- s−1 ), that photosynthetic electron capture might
over, as pointed out in Chapters 3 and 4, the pho- proceed at between 0.55 and 0.85×1021 (g chla)−1
tosynthetic and uptake systems carry such excess s−1 . The potential fixation yield is thus between
capacity that they can sustain higher demands 0.07 and 0.11 × 1021 atoms of carbon per second,
than those set by maximum growth rate. This or between 0.11 and 0.18 ×10−3 mols carbon per
is not to deny that the demands of cell replica- (g chla)−1 s−1 , or again, between 4.9 and 7.6 mg C
tion might not at times exceed the capacity of (mg chla)−1 h−1 . Putting C : chla = 50, a carbon
the environment to supply them or that growth delivery rate of between 27 and 42 × 10−6 g C (g
rate might not, indeed, fall under the control cell C)−1 s−1 may be proposed, which, again, is
of (become limited by) the supply of a given well up to supplying the doubling requirement
resource. Following this logic, we can nominate in something between 4.6 and 7.1 h (Reynolds,
the demand for resources (D) set by the sustain- 1997a). The calculations suggest that photosyn-
able growth rate and compare this with the abil- thesis can supply the fixed-carbon requirements
ities of the harvesting mechanisms to perform of the dividing cell in a little over half of the gen-
against diminishing supplies (S). eration time. However, they make no allowance
As a case in point, the Chlorella strain used for respirational or other energetic deficits (see
in the analysis of growth rate (Table 5.1 and Section 5.4.1 below).
Reynolds, 1990) achieved consistently a maximal For comparison, the well-resourced Chlorella
rate of biomass increase of 1.84 d−1 at 20 ◦ C. cell has no problem in gathering the carbon diox-
Given the size of its spherical cells (d ∼4 µm; ide to meet the photosynthetic requirement. The
s ∼50 µm2 ; v ∼33 µm3 ), this is faster than is pre- diffusion rate calculated from Eq. (3.19) in sec-
dicted by the Eq. (5.6) of the regression (Fig. 5.2), tion 3.4.2 indicated that delivery of the entire

which gives r20 = 1.31. Staying with the real data, doubling requirement of carbon in ∼2300 s (i.e.
one biomass doubling (taken as the equivalent just over 38 minutes). In order to maintain
of an orthodox cell cycle culminating in a sin- steady internal Redfield stoichiometry, the grow-
gle division) defines the generation time; by re- ing cell must absorb 9.43 × 10−3 mol P (mol
arrangement of Eq. (5.3), tG = 9.05 h. During this C incorporated)−1 per generation (i.e. 1 mol/106
period of time, the alga will have taken up, assim- mol C). The phosphorus requirement for the dou-
ilated and deployed 1 g of new carbon for every 1 bling of the cell-carbon content of 0.63 × 10−12
g of cell carbon existing at the start of the cycle. mol is 5.9 × 10−15 mol P cell−1 , which, at its maxi-
It is not assumed that the increase in biomass mal rate of phosphorus uptake (13.5 × 10−18 mol
is continuously smooth but the average exponen- P cell−1 s−1 (Fig. 4.5), the cell could, in theory,
tial specific net growth rate over the generation take up in only 0.44 × 103 s, that is, in just 7.3
time is (1.84/86 400 s =) 21.3 × 10−6 s−1 . This, in minutes! Note, too, that an external concentra-
turn requires the assimilation of carbon fixed in tion of 6.3 × 10−9 mol L−1 is sufficient to supply
photosynthesis at an instantaneous rate of 21.3 the entire phosphorus requirement over the full
× 10−6 g C (g cell C)−1 s−1 . From the maximum generation time of 9.05 h (see Fig. 4.6). Solving
measured photosynthetic rate at ∼20 ◦ C [17.15 mg Eq. (4.12) for the supply of nitrogen to the same
O2 (mg chla)−1 h−1 ] and, assuming a photosyn- cell of Chlorella, a concentration of 7 µmol DIN
thetic quotient of 1 mol C : 1 mol O2 (12 g C : 32 L−1 should deliver ∼175 ×10−18 mol N s−1 , suf-
g O2 ) and a C : chla of 50 by weight, Reynolds ficient to fulfil the doubling requirement of
REPLICATION RATES UNDER SUB-IDEAL CONDITIONS 189
(0.151 × 0.63 × 10−12 =) 0.095 × 10−12 mol N calculations in Section 5.3.3, that growth perfor-
in 540 s (9 minutes). mance is most vulnerable to interruptions to net
These calculations help us to judge that, for photosynthetic output and the supply of reduced-
the rate of cell growth to qualify for the descrip- carbon skeletons to cell assembly. It is self-evident
tion ‘limited’, whether by light or by the avail- that photoperiod truncation by phases of real or
ability of carbon, phosphorus or any other ele- effective darkness must eventually detract from
ment, then it has to be demonstrable that the the ability to sustain rapid growth. For small
generation time between cell replications is pro- algae, even maintaining any adequate reserve of
longed. Moreover, it has to be shown that the condensed photosynthate (such as starch, glyco-
additional time corresponds to that taken by the gen, paramylum, etc.) soon becomes problematic
cells of the present generation to acquire suffi- during extended periods of darkness, whereas all
cient of the ‘rate-limiting’ resource to complete light-independent anabolism will become rapidly
the G1 stage and thus sustain the next division. starved of newly fixed carbon.
Of the responses to light/dark alternation
that might be anticipated, the most plausible is
5.4 Replication rates under that the rate of cell replication becomes a direct
sub-ideal conditions function of the aggregate of the light periods.
If the cell required a 24-h period of continuous
light exposure in which to complete one genera-
The corollary of the previous section is that tion, then, other things being equal, a day/night
the achievement of maximal growth rates is alternation of 12 h should determine that at least
not dependent upon maximal photosynthetic two photoperiods must be passed (not less than
rates and nutrient uptake rates being achieved: 36 h real time) before the cell can complete its
the resource-gathering provisions have capacity replication. By extension of this logic, day/night
in excess over the heaviest resource demands alternations giving 6 h of light to 18 h of dark-
of growth. This luxury can be enjoyed only ness, or 3 h to 21 h will each double the real
under ideal conditions of an abundant supply of time of replication. We may note that, on the
resources, which scarcely obtain in the natural basis of this logic, alternations of 6 h light to
environments of phytoplankton: darkness alter- 6 h dark (or, for that matter, 1-minute alterna-
nates with periods of daylight, depth is equated tions from light to dark) should not extend the
with (at best) sub-saturating light levels and a bal- generation time beyond the 12-h cycle of alter-
anced, abundant supply of nutrients is extremely nations.
rare. The question that arises relates to the lev- Surprisingly, there is not a lot of experimen-
els at which resources start to impose restric- tal evidence to confirm or dismiss these conjec-
tive ‘limitations’ on the dynamics of growth. The tures. Although there are indications that the
question is not wholly academic, as it impinges logic is neither ill-founded nor especially unre-
on the prevalent theories about competition for alistic, it does exclude two influential effects.
light and nutrients. Some, indeed, may need revi- One is the ongoing maintenance requirement
sion, while growth-simulation models founded of the cell – all those phosphorylations have
upon resource harvesting are likely to be erro- to be sustained! The power demand, which per-
neous, except at very low nutrient levels. sists through the hours of darkness and light,
is met through the respirational reoxidation of
5.4.1 The effect of truncated photoperiod carbohydrate. Unfortunately, it is still difficult
Perhaps the most obvious shortcoming that real to be certain about precisely how great that
habitats experience in comparison with idealised demand might be, at least partly because of his-
cultures is the alternation between light and toric difficulties in making good measurements
dark: this is avoidable in nature but only at polar and, in part, because sound testable hypotheses
latitudes and, even then, for just a short period about maintenance have been lacking (see Sec-
of the year. It also seems likely, following the tions 3.3.1, 3.3.2 and 3.5.1). Most of the available
information about the respiration rates of plank- excretion, or photorespiration (see Section 3.3.4),
tic algae and Cyanobacteria comes from the ‘dark the carbon losses from cells must be expected
controls’ to experimental measurements of pho- often to exceed basal metabolism. The experi-
tosynthetic rate (of, for instance, Talling, 1957b; ments of Peterson (1978) show how easily the
Steel, 1972; Jewson, 1976; Robarts and Zohary, coupling between respiration and growth is bro-
1987). Rates, appropriately expressed as a propor- ken and why the fractions of photosynthetic pro-
tion of chlorophyll-specific Pmax , typically range duction lost to respiration, reported in the liter-
between 0.04 and 0.10, over a reasonable range ature (reviewed by Tang and Peters, 1995), are so
of customary temperatures. Translating from the variable.
chlorophyll base to one of cell carbon, Reynolds The second reservation about extrapolating
(1990) deduced specific basal metabolic rates for growth over summated photoperiods concerns
Chlorella, Asterionella and Microcystis at 20 ◦ C of the photoadaptation of cells. Culturing cells
(respectively) 1.3, 1.1 and 0.3 × 10−6 mol C (mol under ideal irradiances actually tends to lead to a
cell C)−1 s−1 . These fit sufficiently well to a regres- reduction in the cell-specific chlorophyll content,
sion parallel to the one describing maximal repli- often to as little as 4 mg (g ash-free dry mass)−1 ;
cation rate (Eq. 5.4, Fig. 5.2), that it is possible to Reynolds, 1987a; C : Chl ∼100). On the other
hypothesise that basal respiration at 20 ◦ C (R20 ) hand, populations exposed to low light intensi-
conforms to something close to: ties are able to adapt by increasing their pigment
content (see Section 3.3.4). It is almost as if the
R20 = 0.079(sv−1 )0.325 (5.7)
cell’s photosynthetic potential varies to match
In this case, R is roughly equivalent to 0.055 r . the growth requirement, rather than the oppo-
Moreover, there is no a priori reason to suppose site, as is generally presumed. Turning off the
that basal respiration rate necessarily carries a light for a part of the day provokes photoadap-
temperature sensitivity different from other bio- tative responses in respect of the shortened pho-
chemical processes in the cell (that is, Q10 ∼ 2). It toperiod to the extent that the energy available
is suggested, again provisionally, that the propor- to invest in growth, normalised per light hour, is
tionality of Eq. (5.6) holds at other temperatures, compensated, as shown by the data of Foy et al.
i.e. R θ ∼ 0.055rθ . (1976) (see Fig. 3.16).
In the context of growth, however, the value There are limits to this argument, of course.
is academic rather than deterministic, for it is On the other hand, the abilities of certain
by no means proven that the basal respiration diatoms (Talling, 1957b; Reynolds, 1984a) and,
applies equally in the light and in the dark. The especially, of some of the filamentous Cyanobac-
implication of photosynthetic quotients (PQ) of teria, such as Planktothrix (formerly Oscillatoria)
1.1–1.15 (Section 3.3.2) is that 10–15% more oxy- agardhii ( Jones, 1978; Foy and Gibson, 1982; Post
gen is generated in photosynthetic carbon fixa- et al., 1985) to function on very low light doses
tion than is predicted by the stoichiometric fixed- has been well authenticated. The curves plotted
carbon yield and that the ‘missing’ balance repre- in Fig. 5.4 represent a selection of experimen-
sents respirational losses in the light. Ganf (1980) tally derived fits of specific growth rates of named
observed that when Microcystis colonies were phytoplankters at 20 ◦ C in cultures fully acclima-
transferred from zero to saturating-light inten- tised to the daily photon fluxes noted. Far from
sities, their respiration rate accelerated rapidly, being a linear function of light dose, except at
from ∼25 to ∼50 µmol O2 (mg chla)−1 h−1 , and very low average photon fluxes, the more adapt-
did not fall back until after the colonies had able species are able to increase biomass-specific
been transferred back to the darkness. The time photosynthetic efficiency so that the growth
taken to return to base rate was proportional demand can continue to be saturated at signif-
to the time spent at light saturation. Insofar icantly lowered light intensities.
as the same inability to store excess photosyn- How far the photosynthetic apparatus can
thate requires some homeostatic defence reac- be pushed to turn photons into new biomass
tion, such as accelerated respiration, glycolate is ultimately dependent upon the integrated
Figure 5.5 The initial slopes, α r , of growth rate on light

intensity from Fig. 5.4, plotted against the corresponding
values of msv−1 , the product of maximum dimension and the
surface-to-volume ratio, of each of the same selection of
species of algae: Ana, Anabaena flos-aquae; Aphan,
Aphanizomenon flos-aquae; Coel, Coelastrum microporum; Dict,
Dictyosphaerium pulchellum; Fra b, Fragilaria bidens; Lim red,
Limnothrix redekei; Mic, Microcystis aeruginosa; Monor,
Figure 5.4 Light dependence of growth rate at 20 ◦ C, as a Monoraphidium sp.; Ped b, Pediastrum boryanum; Pla ag,
function of intensity, in a selection of freshwater Planktothrix agardhii; Scen q, Scenedesmus quadricauda. The
phytoplanktic species. The algae are: Ana, Anabaena flos-aquae; least-squares regression fitted to the points is α r = 0.257
Aphan, Aphanizomenon flos-aquae; Coel, Coelastrum microporum; (msv−1 )0.236 . Redrawn with permission from Reynolds
Dict, Dictyosphaerium pulchellum; Fra b, Fragilaria bidens; Lim (1997a).
red, Limnothrix redekei; Mic, Microcystis aeruginosa; Monor,
Monoraphidium sp.; Ped b, Pediastrum boryanum; Pla ag,
Planktothrix agardhii; Scen q, Scenedesmus quadricauda. light antennae, at least when oriented correctly
Redrawn with permission from Reynolds (1997a). in the photon-flux field. The filamentous arrange-
ment cells in the Oscillatoriales seems to be
supremely efficient in this context, as both the
flux density, which does, eventually, force a numerous studies referred to above and the affin-
dependence of r on I. The steeper is the slope of ity the Planktothrix and Limnothrix species for tur-
light-dependent growth (α r ), the more efficient bid, well-mixed lakes (see Section 7.2.3) would
is the dedication of harvested light energy. Val- indicate. Recalculating from the data of Post et
ues of α r (expressed in units of specific replica- al. (1985), acclimated P. agardhii may maintain a
tion rate (r) d−1 (mol photon)−1 m−2 per d, which maximum growth rate of 9.8 × 10−6 mol C (mol
simplifies to (mol photon)−1 m2 ), are derived cell C)−1 s−1 at 20 ◦ C, to as low as 18 µmol pho-
from the initial, light-dependent slopes in Fig. tons m−2 s−1 ; α r ≈ 0.54 mol C (mol cell C)−1 (mol
5.4. Following Reynolds (1989b), they are plot- photon)−1 m2 . The analogous experimental data
ted in Fig. 5.5 against the dimensionless prod- for the diatom Asterionella have not been located
uct of the surface-to-volume ratio (sv−1 ) and the but, piecing together information from field pop-
maximum dimension of the alga (m). This had ulations, Reynolds (1994a) showed it to rival the
been found to provide the most satisfactory mor- reputation of Planktothrix as a low-light adapting

phological descriptor of the interspecific variabil- organism. From a specific growth rate, r20 = 20.6
−6 −1 −1
ity in α r . It also corresponded with the inter- × 10 mol C (mol cell C) s and a chlorophyll
pretation that the greatest flexibility of algae to content of 2.3 pg cell−1 , that is ∼0.324 g chla (mol
enhance their light-dependent growth efficiency cell C)−1 , the sustaining chlorophyll-specific yield
evidently resided with those having the great- is 63.6 × 10−6 mol C (g chla)−1 s−1 . In turn, this
est morphological attenuation of form. Ostensi- requires a photon flux of, theoretically, not less
bly, slender and flattened shapes make the best than 509 × 10−6 mol photons (g chla)−1 s−1 and
∼700 × 10−6 mol photons (g chla)−1 s−1 , on the but these are not followed here. In a turbid envi-
best performances measured by Bannister and ronment, much of the light available for intercep-
Weidemann (1984). The maximum area projected tion is already scattered and at these low aver-
by a single Asterionella cell is approximately 200 age intensities, changes of orientation prove to
× 10−12 m2 or, with this chlorophyll content, 87 be of little consequence. Well-distributed light-
m2 (g chla)−1 . The requisite active photon flux is harvesting complexes are everything.
calculable as 700 × 10−6 mol photons per 87 m2 , These considerations emphasise the influen-
or just 8 µmol photons m−2 s−1 . This assumes tial nature of the relationship between algal
all wavelengths of visible light are utilised but, if shape in the interception of light energy and the
only half were usable, the growth-saturating light impact of algal size in governing its metabolism.
intensity would be similar to the level measured Morel and Bricaud (1981) recognised this rela-
for Planktothrix agardhii. tionship some years ago, referring to the ‘pack-
It is not, at first sight, at all obvious that aging’ effect on pigment deployment (cf. Duy-
low-light adaptation should be related to algal sens, 1956), where the area projected by the
morphology when it is functionally dependent pigments assumed the same relevance as the
upon appropriate enhancements in pigmenta- concentration of LHC receptors. It is thus inex-
tion. However, it is easily demonstrated that cell tricably linked to the contestable size allome-
geometry and orientation raise the efficiency of try of growth rate (with its −1/3 slope instead
light interception by the pigment complement of the expected −1/4; see Section 5.3.1 above
(Kirk, 1975a, b, 1976). A spherical cell, d µm in and Finkel, 2001). A general relationship between
diameter, has a volume, v = 4/3π (d/2)3 , while the projection and morphometry is shown in Fig.
area that it projects is that of the equivalent disc, 3.12. The independent variable is, again, the
a = π(d/2)2 . Because the carbon content is, pri- index msv−1 , the product of the maximum
marily, a function of volume, the carbon-specific cell dimension (m) and the surface-to-volume
projection (ka ) of spherical algae diminishes with ratio. Note that it is a dimensionless prop-
increasing diameter. For example, we may calcu- erty, length always cancelling out. For spheres,
late that, for a single cell of Chlorella (a = 12.6 × m = d, and msv−1 is a constant [d × 4π (d/2)2
10−12 m2 ; C content = 0.61 × 10−12 mol C), ka = ÷ 4/3π(d/2)3 = 6]. For any shape representing
20.7 m2 (mol cell C)−1 ; for a spherical Microcystis distortion from the spherical form, msv−1 > 6.
colony (d = 200 µm), comprising 12 000 cells, Figure 3.12 also shows that the smaller algae
each containing 14 pg C (Reynolds and Jaworski, generally project large carbon-specific areas but
1978), in which a = 31.4 × 10−9 m2 for a content larger ones have to be significantly subspherical
of 14 × 10−9 mol C, ka = 2.24 m2 (mol cell C)−1 . to match the ka values of the smaller ones. It is
In the case of non-spherical algae that are flat- especially interesting to observe that the algae
tened in one, like those of Pediastrum, or in two that already project the greatest area in relation
planes, like those of Closterium, Synedra or Asteri- to their cell-carbon content are also mostly those
onella, the area projected depends upon orienta- with the maximum photoadaptive potential.
tion. The maximum area projected is when the
two longest axes are perpendicular to a unidirec- 5.4.2 The effect of persistent low
tional photon source. The typical cell of Asteri- light intensities
onella in a colony lying flat on a microscope Rather than experiencing alternations of dark
slide is ∼65 µm in length and shows a tapering interludes with windows of saturating light, phy-
valve with an average width of ∼3.3 µm. In rela- toplankton of so-called crepuscular habitats are
tion to its approximate carbon content of 85 pg exposed to variability that offers only windows of
(7.08 × 10−12 mol C; Reynolds, 1984a), the maxi- gloom. The algae forming metalimnetic swarms
mum value of ka is ∼ 28.2 m2 (mol cell C)−1 . In and deep chlorophyll maxima (DCM) in stable lay-
other orientations relative to a single source of ers in seas and lakes experience the same circa-
light, the area projected may greatly diminish. dian alternations of night and day perceived by
Kirk’s (1976) calculations compensated for this terrestrial and littoral plants but, because they
are located so relatively deep in the light gradi- (Ganf et al., 1991). Taking as the most extreme
ent, the daytime irradiances they experience are case of photosynthetic efficiency from Post et al.
low. (1985) for P. agardhii as a proven example of the
The circumstances of a cell placed at a con- low-light adaptation that is possible at 20 ◦ C (viz.
stant depth and receiving a dielly fluctuating 0.54 mol C (mol cell C)−1 (mol photon)−1 m2 ; Sec-
but low-intensity insolation differ from those tion 5.4.1) and comparing it with the supposed
of one receiving short bursts of high illumina- basal rate of respiration of an alga of its dimen-
tion, even though the daily photon flux might sions, 0.079 (s/v)0.325 , i.e. ∼0.064 mol C (mol cell
be similar. However, the ultimate objective – to C) d−1 , or 0.74 ×10−6 mol C (mol cell C)−1 s−1
maximise absorption of the photons available at the same temperature, it is possible to deduce
– does invoke certain similarities of response. that compensation is achieved at ∼1.4 µmol pho-
All the organisms that successfully exploit sta- tons m−2 s−1 , certainly in the order of 3–4 µmol
ble layers have to be capable of maintaining ver- photons m−2 s−1 , if allowance for the dark period
tical position with respect to the light gradient. is accommodated.
They are either motile (e.g. the flagellated chrys-
ophytes that form layers in oligotrophic, soft- 5.4.3 The effect of fluctuating light
water lakes and certain species of cryptophyte Applying the results of laboratory experiments
of slightly more enriched lakes), or they regu- to the extrapolation of field conditions or to
late buoyancy (as do some of the solitary filamen- the interpetation of field data requires caution.
tous Cyanobacteria, including, most familiarly, In the context of the growth responses of phy-
Planktothrix rubescens and other members of the toplankton entrained in mixed layer, insolation
prolifica group of species, and Planktolyngbya lim- may change rapidly, either increasing or decreas-
netica). Buoyancy-regulating sulphur bacteria of ing at random (Fig. 3.14). Depending on the light
the Chromatiaceae and Chlorobiaceae may strat- gradient and on the depth of turbulent entrain-
ify in the oxygen gradient, provided this lies ment, phytoplankters might experience anything
simultaneously within a stable density gradient from a probable period of 30–40 minutes in effec-
and is also reached by a few downwelling pho- tive darkness with a few minutes of exposure
tons (usually ≤5 µmol m−2 s−1 ). Here, photoau- to high light (deep mixing, steep light gradient),
totrophs function on inputs of light energy that to a similar time period of fluctuating light lev-
are invariably low. The extent of photoadaptation els that are nevertheless adequate to support net
demanded of them depends essentially upon the photosynthesis throughout (mixing within the
depth in the light gradient at which the organ- photic zone). The generic nature of the adaptive
ism is poised. This determines the quantity of responses available, discussed here and in Section
penetrating irradiance and the wavelengths of 3.3.3, is clearly aimed towards optimising growth
the residual light least absorbed at lesser depths against highly erratic drivers. However, the prob-
(the quality of the irradiance). Beneath relatively abilistic, Eulerian aggregation of the responses
clear layers, with absorption predominantly in of the whole population does not take account
the blue wavelengths, the pigmentation may be of the photoprotective and recovery behaviour
expected to intensify but without obvious chro- on the growth rates of individual cells to what
matic adaptation. The less is the residual light in are sometimes sharp, sudden and possibly cru-
the red wavelengths, however, the more advan- cial changes in insolation.
tageous is the facultative production of acces- There are experimental data that lend per-
sory pigments, such as phycoerythrin and phy- spective to this issue. Litchman (2000) designed
cocyanin, to the organism’s ability to function experiments that brought irradiance fluctuations
phototrophically at depth. to each of four cultured microalgae, in each
As discussed in Section 3.3.3, the adaptation of three ranges (15–35, 15–85 and 65–135 µmol
is most plainly observed in depth-zonated popula- m−2 s−1 ) and over three wavelengths of fluctua-
tions of Cyanobacteria (Reynolds et al., 1983a) or tion (1, 8 and 24 h). Variations in the low-light
in populations slowly sinking to greater depths range (15–35 µmol m−2 s−1 , wherein growth rate
is expected to be proportional) were fairly neu- circumstances, able to take up nutrients far faster
tral. The growth rate of the diatom Nitzschia was than they can deploy them. Moreover, they can
slightly increased, that of the green alga Sphaero- continue to do so until very low resource con-
cystis schroeteri was slightly depressed, compared centrations are encountered. Even then, the lux-
to growth at a constant 20 µmol m−2 s−1 . In the ury uptake in generations experiencing resource
saturating range (65–135 µmol m−2 s−1 ), little plenitude may support two or three generations
effect was experienced, except that Anabaena flos- born to resource deficiency.
aquae was slightly increased over its performance At first sight, there is certainly a rapid tran-
at a steady 100 µmol m−2 s−1 . Over the wide sition from there being no competition for
range of fluctuations (15–85 µmol m−2 s−1 , span- resources to there being little left over which
ning limitation to saturation), growth responses to compete. There is a counter to this deduc-
differed significantly from the behaviour under tion, which refers back to the distinctions in the
steady exposure. Growth of all species was main- strategies of velocity, storage and affinity adapta-
tained on the short (1-h) cycle; in each case, the tion (Section 4.3.2). Naturally, fast-growing algae
relatively short exposure to high light remained must be able to garner resources with equal
within the capacity of the species-specific phys- velocity, from diminishing external concentra-
iologies. On the longer cycles, however, growth tions. To be able to store resources in excess
rate was impaired in all species, though not all promises an advantage when external concentra-
to the same extent. The reaction would once tions have been diminished, although it might be
have been described as ‘photoinhibition’ but more beneficial to species that have the ability to
would now be better referred to photoprotection migrate between the relative resource-richness of
and the first steps toward photoadaptation (see deep-water layers and resource-depleted but inso-
Section 3.3.4). The experiments of Flöder et al. lated surface waters. To survive, even to thrive, in
(2002) investigated the influence of fluctuating waters chronically deficient in one (rarely more)
light intensities (range 20–100 µmol m−2 s−1 ) major resource might well call for a superior
on growth rates of natural phytoplankton assem- competitive ability to win the scarce supplies and
blages collected from Biwa-Ko, imposed on cycles deny them to individuals of other species.
of 1, 3, 6 or 12 days. These ably illustrate the Evidence for the existence and implementa-
population responses consequential upon differ- tion of these strategies is to be discussed later in
ential growth rates altering the composition of this chapter. For the moment, the first task is to
the assemblage. discern the resource levels that separate famine
from bounty. It is convenient to consider these
5.4.4 The effects of nutrient deficiency nutrient by nutrient.
A cornerstone role has long been accorded to
nutrients in the regulation of productive capac- Phosphorus deficiency
ity in the plankton and in shaping the species Satisfaction of the alga-specific P requirements
composition. A very large literature on the nutri- for growth has been suggested to rest upon the
ent limitation of phytoplankton and the inter- ability to maintain a stoichiometric balance of
specific competition for resources reflects the assimilates approximating to 1 P atom to every
importance of their availability in pelagic ecol- 106 of carbon. This determines the generalised
ogy, even if these key processes seem, at times, requirement that 9.4 × 10−3 mol P (mol C)−1 is
to have been misrepresented. The point has been incorporated during each single replication time.
made earlier that the least-available resource, rel- As already indicated, the alga’s uptake capacity is
ative to the minimum requirements of organ- likely to be such that, under resource-rich condi-
isms, sets a finite ‘carrying capacity’ for the habi- tions, it may achieve this in minutes rather than
tat. To establish the limiting role of nutrients on hours. It is not likely that any phytoplankter takes
growth rate is more difficult, for two reasons. As longer than its achievable generation time while
has been shown in Chapter 4 and again in Sec- external MRP concentrations exceed 0.13 × 10−6
tion 5.3.3, most phytoplankters are, under ideal M (4 µg P L−1 ). For many species, indeed, this
could be true at MRP concentrations as low as (Section 4.3.2), found that growth rate at 20 ◦ C is
10−8 M (0.3 µg P L−1 ; see Sections 4.3.3, 5.3.3). half-saturated when the cell contains about 0.7
Even then, the reserves accumulated during pre- pg P (or about 0.003 mol cell P (mol cell C)−1 ).
vious resource-replete generations may sustain The external MRP concentration required to bal-
one or two generations before the cell recog- ance this quota is about 0.75 µg P L−1 (Kr = 0.024
nises impending shortages and perhaps three, µM). The extensive work of Tilman and Kilham
or even four, before the cell quotas approach (1976) using semi-continuous cultures of diatoms
q0 and exhaustion. Fast-growing species, having pointed to the half saturation of growth in Asteri-
high sv−1 (>1.3 µm−1 , which, at about 30 ◦ C, onella formosa falling between 0.02 and 0.04 µmol
might permit specific growth rates of up to ∼4 P L−1 (0.6–1.2 µg P L−1 ). However, the correspond-
d−1 , or ∼50 × 10−6 s−1 , to be attained), would ing value for another diatom, Cyclotella menegh-
be expected to reach exhaustion proportinately iniana, was substantially higher than this (Kr =
sooner. Yet averaging and normalising the total 0.25 µmol P L−1 , or nearly 8 µg P L−1 ). Finally, in
P requirements to the starting biomass carbon, this context, Nalewajko and Lean (1978) used 32 P-
the P demand of two generations (∼0.5 × 10−6 labelled phosphorus to track phosphorus uptake
mol P (mol initial cell C)−1 s−1 ) is still likely to and turnover in cultures of Anabaena flos-aquae
be deliverable from a starting concentration of and Scenedesmus quadricauda, and remarked on
0.03–0.13 × 10−6 M (1–4 µg P L−1 ). the ‘low’ concentrations (∼6 µg P L−1 ; ∼0.2 µmol
Data compilations that compare growth rates, P L−1 ) at which full exchange activity is main-
nutrient-uptake rates and their half-saturation tained.
coefficients are available (see Padisák, 2003), from Certainly, most of these data uphold the view
which it ought to be a straightforward exer- that the onset of phosphorus ‘famine’, below
cise to verify, on the basis of hard, experimental which growth rate may be regulated by the sup-
results, some of the above conjectures. The trou- ply of P, generally occurs at <0.1 µM P (∼3
ble is that many of the half-saturation concen- µg P L−1 ). This may not apply to all species:
trations refer to uptake by starved cells, rather Scenedesmus and the Cyclotella of Tilman and Kil-
than to that needed to half-saturate the require- ham are possible exceptions. For most species,
ment to maintain growth (Kr ). These quantities allusions to growth-rate limitation of phytoplank-
require the analysis of a lot of batch cultures, ton by phosphorus when external MRP concentra-
across a range of concentrations, or the appli- tions exceed 0.1 µM P may be doubted. It is pos-
cation of a semi-continuous technique in which sible that the higher species-specific thresholds
the algal culture is diluted with test medium are a consequence of adaptation to the relatively
at a rate adjusted to keep the algal concen- P-rich habitats in which they occur. Conversely,
tration constant (that is, to balance the rate differential species-specific affinities uptake (Kr
of growth). As a consequence, there are few values spanning 0.01–0.2 µM P) can be said to
good data to confirm the half-saturation con- influence where species may live and how com-
stants of P-limited growth. Rhee (1973), using a petitive they might be for truly limiting supplies
Scenedesmus, Ahlgren (1985), with Microcystis wesen- of phosphorus.
bergii, and Spijkerman and Coesel (1996a), using
two desmids (Cosmarium abbreviatum and Stauras- Nitrogen deficiency
trum pingue), each found that the algae would Analogous calculations concerning algae and
grow at about one half the nutrient-saturated their nitrogen requirements are also available.
rate at 20 ◦ C, in media supplying <6.0 µg P L−1 Supposing the satisfaction of the alga-specific N
(0.19 µM). All but the Scenedesmus did so at requirements for growth to be similarly based
<1.2 µg P L−1 (<0.04 µM). Cosmarium growth upon the stoichiometric balancing against car-
rate was half-saturated at 0.35 µg P L−1 (0.011 bon in the atomic ratio 6.6 : 1, the generalised
µM). Davies (1997), whose work with Asterionella demand approximates to 151 × 10−3 mol N (mol
has been highlighted earlier in the context of C)−1 . Again the DIN-uptake capacity is well up
the interaction between P-uptake and cell quota to this under nitrogen-replete conditions. From
the information on DIN uptake in Chlorella (Sec- imprecise. In many of the nitrogen-deficient but
tion 5.3.3), it may be deduced that the nitrogen simultaneously phosphorus-poor lakes of the
complement needed to sustain a doubling of Andean–Patagonian lakes investigated by Diaz
mass could proceed 60 times more slowly and and Pedrozo (1996), the phytoplankton biomass
still fulfil the maximum growth rate. Moreover, is demonstrably (by assays and by mathematical
the external concentration needed to supply DIN regession) constrained by nitrogen availability.
at this rate would be about 0.12 µmol N L−1 , The significant incidence of dinitrogen-fixing
or 2 µg N L−1 ). The half-saturation constants Cyanobacteria is relatively restricted to lakes,
for nitrate and ammonium uptake (KU ) among such as Bayley-Willis and Fonck, that have higher
small-celled oceanic phytoplankton are of simi- TP contents (and where, incidentally, the SRP is
lar order or only slightly higher (see Section 4.4.2) drawn down in summer to growth-limiting levels
and are, thus, unlikely to experience symptoms as identied above, viz. to <0.1 µM P). Elsewhere,
of nitrogen-limited growth at external DIN con- the unsupplemented combined levels of nitrate-,
centrations >1–2 µg N L−1 . On the other hand, nitrite- and ammonium- nitrogen levels stand at
the half-saturation constants for nitrate and ≤0.8–1.0 µM (≤11–14 µg N L−1 ), perhaps falling
ammonium uptake (KU ) among larger diatoms of in summer to <0.2 µM (see Diaz et al., 2000).
inshore waters may be up to an order greater The revised verdict on the DIN levels repre-
again (0.5–5 µM N), so that problems of obtaining senting the onset of nitrogen limitation among
sufficient nitrogen to even half-saturate growth non-fixers is 0.1 µM (1.5 µg N L−1 ) for oceanic
might be experienced at external DIN concentra- picoplankton and 1–2 µM (15–30 µg N L−1 ) for
tions in the range 0.2–2 µM N (say 3–30 µg N many microplankters in lakes and seas. If other
L−1 ). conditions are satisfied (see Section 4.4.3), nitro-
The lower limits of the range of DIN concen- gen fixers may avoid altogether the constraint of
trations able to support phytoplankton growth low DIN concentrations.
in inland waters are less well researched. It was
once deduced (Reynolds, 1972) that, based on Silicon deficiency
the assumption that the appearance of nitrogen- Silicon limitation of growth in diatoms is abso-
fixing organelles (nostocalean heterocysts) at DIN lute and well understood. Lund’s (1949, 1950) clas-
levels of 20–25 µM provided a simultaneous sical studies on Asterionella in Windermere and
advantage to the fixers over non-fixers, the non- its relationship with the availability of silicon
fixers were experiencing supply difficulties. As helped to establish the importance of nutrients
a result of later observations but without vary- in planktic ecology. The clarity of the effects
ing the essential logic, this critical range was and the precision of the chemical thresholds
revised downwards to ∼6–7 µM N (80–100 µg N were, and are still, impressive. That they were
L−1 ) (Reynolds, 1986b). As it now seems probable never emulated in the studies of other elemen-
that heterocyst production responds to ammonia tal requirements is attributable to the nature of
concentration and not DIN per se (discussion, sec- the nutrients and the margin around the empiri-
tion 4.4.3), it no longer follows that their appear- cism of the requirements that the organisms
ance necessarily coincides with nitrogen shortage introduce through storage and luxury uptake. As
among the nitrate users. Current evidence indi- stated (Section 4.6), the diatoms are the biggest
cates that they are competent to draw down DIN planktic consumers of silicon; they take up no
to concentrations to levels of 0.3–3 µM without more than is required to build the silica valves
growth-rate limitation. of the frustules of the current generation, and
Eventually, the ability of heterocystous the silicon polymer is laid down under close
cyanobacteria to fix nitrogen when DIN is simul- genetic supervision. All this leads to the sili-
taneously depleted influences the recruitment of con requirement of each new cell of a given
to natural communities (Riddolls, 1985) but the species and size being readily predictable. The
concentration threshold favouring Cyanobacte- concentrations of silicic acid capable of deliv-
rial dominance, between 2 and 6 µM N, remains ering the silica requirement have been found
through observation and experiment. The conse- established an important conceptual theory of
quence of silicon ‘limitation’ is easy to detect as resource-based competition. Simply, if two species, A
its consequence is that cells cannot complete the and B, having similar resource-saturated growth
growth cycle. Moreover, failure of the putative rates, are cultured together in a gradient of
cell to build its new frustular valve is fatal. growth-limiting concentrations of a resource S1 ,
In the case of Lund’s Asterionella, the average the one with the higher uptake rate (VU ) is clearly
complement of 140 pg (SiO2 )n cell−1 could be sat- able to sustain a faster rate of growth at low con-
isfied from a concentration equivalent to ≥0.5 centrations of [S1 ] than the other. The theoretical
mg (SiO2 ) L−1 . Translation of the units notwith- growth performances of A and B are shown in
standing, both quantities have been abundantly Fig. 5.6a. Against a second resource, S2 , however,
verified. Solving the regressions of Jaworski et al. it is species B that performs better at low concen-
(1988), an Asterionella cell 65 µm in length has trations (Fig. 5.6b). Placed together in a medium
a probable Si content of 65 pg (2.3 pmol cell−1 ). deficient in both S1 and S2 , it is possible, depend-
Using the experimental data of Tilman and Kil- ing on the relative concentrations of either nutri-
ham (1976), the silicic acid concentration that ent, for the species to be simultaneously lim-
will half-saturate the Si requirement is 3.9 µM ited by different nutrients. Plotting against the
(equivalent to 109 µg Si L−1 , or 0.23 mg (SiO2 )n concentrations of both resources (Fig. 5.6c), the
L−1 ). respective limitations can be used to predict com-
Equivalent data from many other experi- petitive outcomes of variable resource combi-
ments, reviewed in Tilman et al. (1982) and nations (Fig. 5.6d). Thus, at low concentrations
in Sommer (1988a), show that growth rates of of (S1 ), an outcome will always be favoured in
freshwater diatoms at 20 ◦ C tend to be half- which A dominates; at low concentrations of
saturated at between 0.9 (in Stephanodiscus min- ions of (S2 ), it will be B that is favoured. At
utus) and 20 µM Si (in Synedra filiformis). That slightly higher concentrations of both resources,
of Cyclotella meneghiniana is half-saturated at 1.4 A and B may coexist successfully, while the
µM Si (Tilman and Kilham, 1976). Among the one (B) remains S1 -limited and the other is
investigated clones of marine Thalassiosira pseudo- still limited by S2 . This prediction was pre-
nana and T. nordenskioeldii, half-saturation of the cisely the outcome of their investigations (Tilman
growth rate at 20 ◦ C tends to occur in the range and Kilham, 1976) of phosphorus- and silicon-
0.2 to 1.5 µM (data of Paasche, 1973a, b). limited growth between Asterionella formosa
The 100-fold range in uptake thresholds has (Kr (P), ∼0.03 µmol P L−1 , Kr (Si) 3.9 µmol Si
ecological consequences, to be considered in the L−1 ) and Cyclotella meneghiniana (Kr (P), ∼0.25 µmol
next section. For the moment, the deduction P L−1 , Kr (Si) 1.4 µmol Si L−1 ); Cyclotella domi-
is that silicon concentrations begin to interfere nated over Asterionella in mixed cultures at low
with the growth of diatoms at concentrations Si : P ratios. The opposite was true at high Si : P
below ∼0.5 mg Si L−1 (say 1 mg L−1 as equivalent ratios. On the basis of later investigations of
silica, or ∼20 µM). In most lacustrine instances, other species, Tilman et al. (1982) emphasised the
growth-limiting concentrations are encountered differential competitive abilities of diatoms to
mainly below about 0.1 mg Si L−1 (say <4 µM) Si : P by plotting the experimentally solved Kr (P)
and, in the sea, below 0.03 mg Si L−1 (<2 µM). against Kr (Si) for each (see Fig. 5.6e). The plot ably
arranges species on the basis of Si : P preferences.
5.4.5 The effect of resource interactions: For most other plankton, silicon is a minor
nutrients and light nutrient and, not surprisingly, they tolerate Si:P
Resource-based competition ratios very much lower than Cyclotella’s limit
The able demonstrations by Tilman and his co- of 5.6 (Holm and Armstrong, 1981; Sommer,
workers of the interspecific differences in the 1989). However, it is the relationships between
capability of diatoms to take up the silicon the nitrogen and phosphorus requirements that
and phosphorus required to sustain their growth have aroused enormous interest, especially in the
from relatively low external concentrations also context of a widely held belief that low
Figure 5.6 Resource competition and species interactions. Parts (a) and (b) compare the nutrient-limited growth rates two
species of phytoplankter, Sp. A and Sp. B, against low, steady-state concentrations of resources S1 and S2 . Growth of either may
be limited (c) by the availability of either resource. Tilman’s theory of resource-based interspecific competition acknowledges that
the uptake constraints acting on Sp. A and Sp. B differ sufficiently for (d) Sp. A to dominate over Sp. B when [S1 ] is low and Sp. B
to do so when [S1 ] is low, but Spp. A and B do not compete when limited by different resources. The relative competitive abilities
of named diatoms for silicate and phosphate, as determined by Tilman et al. (1982), are shown in (e): A.f., Asterionella formosa; C.m.,
Cyclotella meneghiniana; D.e., Diatoma elongatum; F.c., Fragilaria crotonensis; S.f., Synedra filiformis; S.m., Stephanodiscus minutulus; T.f.,
Tabellaria flocculosa. In (f), the effects on phytoplankton assemblages in a selection of natural lakes of differing N and P availabilities
are represented: C, Crose Mere, and W; Windermere, are in UK; E is Esrum, Denmark; K, Kasumiga-Ura, and S, Sagami-Ko,
Japan; Me, Mendota, T, Tahoe, and Wa, Washington, in USA; Mg is Maggiore, Italy/Swizerland and Ml, Mälaren, in Sweden. Area 1
applies to low-P lakes, dominated by diatoms and chrysophytes; area 2 covers lakes in which nitrogen-fixing Cyanobacteria are
abundant through substantial parts of the year; area 3 lakes are are dominated by Microcystis for long periods. The composite
combines figures redrawn from Reynolds (1984a) and Reynolds (1987b).
nitrogen-to-phosphorus ratios favour the (usually ing normal growth (when both storage effects
unwelcome) dominance of Cyanobacteria (Smith, and deficiencies should have been minimal). Out-
1983). Rhee (1978) suggested that the evident comes ranged between 7, for the diatom Stephano-
mutual competition along N : P gradients is influ- discus binderanus, and 20–30, for three species of
enced by differing N : P optima in the cells Chlorococcales. For the Cyanobacterium Microcys-
of various species. In a major programme of tis aeruginosa, it was 9. These optima differ from
laboratory experimentation, Rhee and Gotham the ideal (‘Redfield’) N : P stoichiometry centred
(1980) showed systematic differences in the ratios at 16 molecular and from the suggested (Section
of species-specific optimal N and P quotas dur- 4.4) range of normality of 13–19. Given the range
in physiological variability in N : C and P : C con- Stoichiometry

tent of individual cells, extremes of N : P of 4 and These plausible deductions invoked parallel
108 are theoretically tenable but values within developments in the appreciation of organismic
the range 10 to 30 are scarcely indicative of nutri- stoichiometry, prompted, in part, by the work of
ent stress. Hessen and Lyche (1991) on the differential ele-
Nevertheless, the findings of Rhee and mental make up (chiefly C:N:P) of zooplankton.
Gotham (1980) do not violate any supposition Stoichiometric differences between the trophic
about the importance of N : P ratios in favour- components imply consequences for the system
ing Cyanobacteria or otherwise. However, there as a whole. For instance, animals with a relatively
are difficulties (see below) in applying resource low N : P content feeding on algal foods having
ratios to either the interpretation or the pre- a high N:P make-up will retain P preferentially
diction of the composition of natural communi- and, so, recycle wastes with a yet higher N:P con-
ties. Besides, ratios of available resources change tent. This approach has been developed further
quite rapidly through time, without necessarily by J. Elser and co-workers (usefully reviewed in
precipitating immediate changes in species com- Elser et al., 2000). They were able to demonstrate
position. If the total nitrogen and phosphorus striking connectivities among the molecular sto-
resources delivered to and present within a water ichiometries of growing cells of a wide range
body are considered, some broad compositional of algal species, their rates of growth and the
trends are discernible. The distribution of lakes evolutionary pressures underpinning their life-
shown in Fig. 5.6f against axes of total N and history traits. That is to say, the evolution of fast
total P separates those that will support large or slow growth rates and the allocation of the
populations of nitrogen-fixing Cyanobacteria as catalysing (P-rich) RNA molecules are inextricably
being low-N-high-P habitats, from those that sup- interlinked. The matching of evolutionary traits,
port non-fixing Microcystis (high N, high P) and from their molecular bases to the environments
those that seem never to be dominated by bloom- in which whole organisms function and interact,
forming genera (low-P lakes). now has a wide following. The recent book by
To be any more specific about predictions of Sterner and Elser (2002) on Ecological Stoichiometry
differentiated growth or the composition of the conveys and nourishes much of this excitement.
population structures that it might yield needs Besides providing an alternative perspective on
much more information. One component obvi- ecosystem function and a persuasive argument
ously lacking from the previous paragraph is the for unifying concepts in its understanding, the
carbon input and the solar-energy income that book invokes pertinent explorations of just what
is essential to its compounding with nutrient goes on inside the living cell to yield recognisable
resources into biomass. In developing a hypoth- structural stoichiometries in the first place and
esis about the deterministic importance of the what activities lead to (relatively modest) depar-
light : nutrient ratio in lakes, Sterner et al. (1997) tures therefrom.
showed that the relationship between the mean To bring us back specifically to phytoplank-
mixed-layer light level (equivalent to the calcu- ton, if these attractive theories of resource ratios
lation of I∗ in Section 3.3.3) in each of a num- and biological stoichiometry are to be helpful to
ber of lakes and its corresponding total P con- understanding how pelagic communities func-
centration is a good probabilistic predictor of tion, then it is important first to separate just
the C : P content of the seston and of the effi- what is cause and what is effect. It is necessary
ciency of resource use by the system as a whole. to emphasise the distinctions among the ratios
They further hypothesised that the seston C : P of algal cell quotas, the ratios of resource avail-
ratio influenced the pathways of secondary pro- ability and supply, and the competitive abilities
duction (essentially the food-web consumers of of algae to take up elements at low concentra-
primary product) and thence, biassed the intentions. Taking cell quotas first, we have to accept
sity of nutrient recycling, the strength of micro- that the ratio N : P = 16 is certainly not abso-
bial processing and, indeed, the structuring of lute, that it is subject to a margin of physiological
the entire ecosystem. variability, including in the complement of RNA,
and that there may indeed be systematic, inter- and a probable biomass N : P (5.3) indicative
phyletic differences in the optimal elemental bal- of N-limitation. The residual P concentration
ances. Nevertheless, the range of normality in the (arguably ∼0.7 µmol P L−1 ) confirms that P is
elemental composition of cells, from bacteria to certainly not a constraint and it is also suffi-
elephants, is relatively quite narrow, reflecting cient to support the activity of nitrogen-fixing
general similarities in the cell complements of Cyanobacteria. These could grow to the limit
protein and nucleic acid (Geider and La Roche, of the phosphorus capacity (47 µg chla L−1 ),
2000). From Chapter 4, it is plain that a factor of only to now be dominated by nitrogen-fixing
50% variation in either the N or the P content is Cyanobacteria having an intracellular N : P ratio
hardly exceptional, yet it yields a full range in N:P of ∼30. Note that, provided it is not itself
from 5 to 36. For the purpose of estimating stoi- limited by some other factor, nitrogen fixa-
chiometrically the phytoplankton-carrying capac- tion forces the nitrogen-deficient system to the
ity of the nutrients in a given habitat, less error capacity of its phosphorus supply (cf. Schindler,
attaches to the adoption of a mean complement 1977).
of 16 to 1 than to the correct estimation of
the base of bioavailable nutrients (Reynolds and Resource depletion and growth-rate regulation
Maberly, 2002). However, it is more straightfor- Now let us consider the role of interspecific com-
ward (and more illuminating) to examine the petition in the way different species might simul-
support resource by resource. To be able to form taneously satisfy their resource requirements to
a standing phytoplankton biomass equivalent to, sustain their growth rates. Here we bring into
say, 106 µmol C L−1 (1.27 mg C L−1 , ∼25 µg chl a sharp focus the uptake capabilities of the algae
L−1 ) ostensibly requires the supply of 16 µmol N themselves and the sorts of threshold concentra-
and 1 µmol P L−1 (i.e. 224 µg N, 31 µg P L−1 ). If tions at which they fail to be able to take up
the water body can fulfil only 10 µmol N and 0.1 specific nutrients as fast as maximum consump-
µmol P L−1 (note, N : P = 100), it is obvious that tion would demand (1–2 µmol N, 0.1 µmol P
the growth demand will first exhaust the phos- L−1 (Section 5.4.4). The corollary of this is that
phorus, at a rather smaller chlorophyll yield than if the concentrations of bioavailable N and P
25 µg chl a L−1 . Long before this maximum is are significantly greater than these thresholds,
reached, the biomass : P yield is stretched, accord- neither imposes a rate-limiting constraint upon
ing to Eq. (4.15), to 12.2 µg chla L−1 ; the biomass the growth of any alga. Several species could
has a probable content of carbon equivalent to grow simultaneously and, while each satisfies its
∼51 µmol C and 7.7 µmol N L−1 , but no more requirement, they are not in mutual direct com-
than the originally available 0.1 µmol P L−1 . We petition (sensu Keddy, 2001; see Box 4.1). Each per-
deduce a biomass N : P ratio of 77, correctly infer- forms to its capacity (or to some independent
ring the obviously severe phosphorus limitation regulation). At this stage, the ratio of the available
on the biomass. The magnitude of the eventual resources is quite irrelevant to the regulation of species-
P-limited quota constraint on the biomass is pre- specific growth rates.
dictable on the basis of the initial phosphorus This deduction arouses persistent controversy.
availability. That it was phosphorus, rather than Yet it is readily verifiable in the laboratory
nitrogen, that would impose the eventual limit through the measurement of the early exponen-
is implicit in the starting resource ratio. tial increase of a test alga in prepared media
Now, if the water body could fulfil 1 µmol offering nutrients in differing mutual ratios
P but only 1.6 µmol N L−1 (i.e., 31 µg P, 22.4 but at initially saturating concentrations. I am
µg N L−1 ; N:P 1.6), growth would be expected to unaware of any publication that draws attention
exhaust the nitrogen, for the production of 32 to this behaviour. However, I am most grateful
µmol biomass C L−1 (assuming the minimum C to Dr Catherine Legrand, of the University of
: N ratio quoted in Section 4.4) and which, stoi- Kalmar, Sweden, for her permission to reproduce
chiometrically, would have a phosphorus content a graph that she presented at the 1999 Meet-
equivalent to not less than 0.3 µmol P L−1 ing of the American Society of Limnology and
of the point I wished to illustrate; however, she

does not necessarily share my interpretation.)
In nature, phytoplankters inhabit dynamic
environments, far removed from the contrived
and quasi-steady states of the laboratory, and
algal consumption will, in many instances, take
one or another of the free resources to below
the concentration representing its competitive
thresholds. It is within this region of poten-
tial growth-rate limitation, that Tilman-type
resource-based competition is expected to be
strongly expressed. Thus, with external concen-
trations of (say) MRP falling to below 0.1 µmol
P L−1 , Asterionella may maintain a rate of repli-
cation close to the maximum that the temper-
ature and photoperiod may allow, yet Cyclotella,
once its reserves are depleted, is able to maintain
only a fraction of its resource-saturated growth
rate. Given initial parity, its abundance relative
to Asterionella is set to fall quickly behind. As
the MRP concentration falls yet further, Cyclotella
may cease to increase at all, while Asterionella
Figure 5.7 Initial increase in cell populations of
Alexandrium tamarense KAC01 in laboratory cultures artificial is still absorbing phosphorus that is now effec-
media with widely differing ratios of N : P: ν – 160; – 16; λ – tively denied to the ‘outcompeted’ Cyclotella. Very
1.6. Original data of Dr C. Legrand, replotted with soon, of course, the MRP is too depeleted to
permission. See text for further details. be able to support further growth of Asterionella
either, and the performance failure is still more
abrupt. All this is predicted by the species-specific
growth-rate dependence on P concentration
Oceanography, held in Santa Fé, NM, USA. Her (Fig. 5.6).
paper was concerned with the development of Against gradients of falling silicon concentra-
toxicity in a laboratory strain (KAC01) of the tion, Cyclotella can continue to absorb silicic acid
dinoflagellate Alexandrium tamarense in relation from concentrations already severely constrain-
to the nitrogen and phosphorus resources sup- ing Asterionella. The outcome of interspecific com-
plied. Replicate treatments were grown in arti- petition betwen other pairs of diatoms may be
ficial media, at 17 ± 1 ◦ C and subject to a 16-h verifiably predicted according to the model of
light : 8-h dark cycle, differing only in the con- Tilman et al. (1982). Another, quite separate, illus-
centrations at which nitrogen and phosphorus tration of resource competition was described by
were supplied. Media contained either 20 µM N Spijkerman and Coesel (1996b). Of three species
and 0.75 µM P (i.e., N : P = 160), or 120 µM N of planktic desmid grown in continuous-flow
and 7.5 µM (N : P = 16) or 12 µM N and 7.5 µM cultures under stringent P-limiting conditions,
P (N : P = 1.6). The results, summarised in Fig. one – Cosmarium abbreviatum – showed superior
5.7, refer to the increases in cell concentration affinity for phosphorus at low concentrations
of algae grown in the three media. They show over the other two (Staurastrum pingue, S. chaeto-
clearly that initial growth performances in the ceras), even though these have faster maximum
three media were indistinguishable, though, not P-uptake rates. The outcome of competition in
surprisingly, they diverged as the experiment pro- mixed chemostat cultures conformed to the pre-
gressed. (Note: When Dr Legrand gave her permis- diction that, at concentrations of <0.02 µmol P
sion to use her data, it was in her full knowledge L−1 , Cosmarium outcompeted the Staurastra but,
at higher delivery rates, their faster rates of missa), as well as in progressive changes in the
uptake and growth allowed the Staurastra to nutrient resources in the lake, Asterionella has
dominate the numbers of Cosmarium. If instead dominated in all but one of those years. Two
of low continuous supply, phosphorus was sup- trends over the period have been unmistakable.
plied in single daily doses of 0.7 µmol P L−1 , One is that the Asterionella maximum, which, in
superior uptake rates enabled the Staurastrum all years, has been contained ultimately within
species to sequester relatively more of the pulsed the capacity of the silicon availability (∼30 µmol
resource supply (velocity/storage then proving Si L−1 (Lund, 1950) has, on average, been com-
more advantageous than high affinity). ing earlier each year (by an average of 30 days
In none of these cases is the outcome over 40 years). Second, during the same period,
attributable to anything other than the abso- the MRP available to phytoplankton at the incep-
lute characteristics of the critical resource sup- tion of the annual growth increased from
ply to the needs and sequestration abilities of ∼0.1 to almost 1 µmol P L−1 . Early in the period,
the organism(s) concerned. In none is the out- Asterionella may have seem well suited to the high
come the direct consequence of the ratio of Si : P conditions, although, in reality, its dom-
resources available or of the rate at which crit- inance in any individual year also invoked the
ical resources are supplied. Thus, at this stage size of the inocula and its ability to grow on low
too, it is not the ratio of available resources that deter- daily light doses. By the time of the maximum
mines the outcome. The resource ratio is an inter- (and for a substantial period beforehand), MRP
pretative convenience in identifying which of two levels were below the limit of detection. As the
scarce resources is likely to be, or to become, lake has become more enriched with phospho-
limiting and it aids the understanding of simul- rus, the Asterionella has continued to dominate
taneous limitation of coexisting species by dif- the early growth stages but it has been able to
ferent resources, provided both are below their maintain an accelerating growth rate, all the way
respective critical thresholds. When the limiting to the division that finally reduces the silicon to
concentrations of both resources are exceeded, concentrations limiting its ability to take it up
the interspecific competition for those resources (Reynolds, 1997b). Under these limiting and now
is correspondingly diminished, as both satisfy relatively low Si : P conditions, should we not
their immediate needs without interference to expect resource competition to alter the outcome
the other, and the likeihood of the one excluding of the spring growth? Reynolds’ (1998a) simple
the other is minimised. The explicit prediction of model envisaged a typical standing population of
coexistence, inserted in the top right-hand cor- Asterionella of 4 × 106 cells L−1 at the time that
ner of Fig. 5.6d, is correct but the explanation is the silicon concentration is lowered to the point
different from that applying towards the bottom (8 µmol L−1 ) where its growth rate is increasingly
left-hand side. regulated by the rate of silicon uptake. Its next
There is a further perplexity over the gen- and possibly last doubling (it might take 4–7 days
eral assumption of resource competition among to complete) would require all of the remaining
species, which I have aired at length in cer- silicon in the water. At the same time, the sub-
tain earlier publications (Reynolds, 1997b, 1998a). dominant Cyclotella, at no more than 0.8 × 106
From a comparison of the interannual varia- cells L−1 , is experiencing neither phosphorus nor
tions in the dynamics and composition of the silicon limitation and maintains the maximum
phytoplankton through successive spring phy- growth rate that the temperature and the light
toplankton blooms in Windermere since 1945, regime will allow, as predicted by the Tilman
several dynamic characteristics have been recog- model. The difficulty is that before the altered
nised (Maberly et al., 1994; Reynolds and Irish, competitive basis can be expressed at the level of
2000). Despite interannual differences in temper- community composition, the silicon is effectively
ature, rainfall and stratification, in the size of exhausted. There is no advantage to better com-
the inocula and the rates of growth attained petitors for a non-existent resource: Asterionella
by each of several species of diatom (Asterionella still dominates, despite its (by now) competitive
formosa, Aulacoseira subarctica and Cyclotella praeter- inadequacies.
Table 5.2 Phytoplankters tolerant or indicative of chronically oligotrophic conditions in lakes and the
upper mixed layers of tropical oceans
Lakes
Cyanobacteria Synechococcoid picoplankton
Chlorophyceae Chlorella minutissima, Coenocystis, Coenochloris (Eutetramorus), Sphaerocystis, Oocystis
aff. lacustris, Willea wilhelmii, Cosmarium, Staurodesmus
Chrysophyceae Chrysolykos, Dinobryon cylindricum, Mallomonas caudata, Uroglena spp.
Bacillariophyceae Aulacoseira alpigena, Cyclotella comensis, C. radiosa, C. glomerata, Urosolenia eriensis
Sources: Pearsall (1932), Findenegg (1943); Reynolds (1984b, 1998a); Hino et al. (1998), Huszar et al.
(2003).
Oceans
Cyanobacteria Prochlorococcus, Synechococcus, Trichodesmium spp.
Bacillariophyceae Rhizosolenia, Bacteristrum, Leptocylindrus
Haptophyta Emiliana, Gephyrocapsa, Umbellosphaera
Dinophyta Amphisolenia, Dinophysis, Histoneis, Ornithocercus spp.
Sources: Riley (1957), Campbell et al. (1994), Karl (1999), Smayda and Reynolds (2001), Karl et al. (2002).
Chronic nutrient deficiencies ratios. The extent to which the N : P gradi-

There is, however, another way in which resource- ent underpins, or even correlates with, compo-
based competition shapes communities, at least sitional patterns in natural lakes has still to be
in oligotrophic waters in which the availability fully resolved. Increasing the phosphorus loads
of one or more resources waters is chronically relative to those of nitrogen (thus lowering N : P)
deficient and where species having high uptake does result in compositional shifts, often towards
affinities for the limiting nutrient(s) are likely dominance by nitogen-fixing Cyanobacteria, as
to thrive at the expense of inferior competitors. has been shown in numerous whole-lake fertil-
These species are also more likely to provide the isations by Schindler (see, e.g., 1977) and his co-
inocula in future seasons, so the competitive out- workers. The possible responses to simultaneous
come is magnified from generation to generation. enrichment with nitrate or ammonium (raising
In this way it is relatively easy to hypothesise or maintaining N : P) include frequent incidences
that the selective traits most favoured in chroni- of enhanced production of green algae (especially
cally oligotrophic systems – high affinity for lim- Chlorococcales and Volvocales) but they are fre-
iting nutrients (Sommer, 1984) and small organ- quently confounded by altered carbon dynamics
ismic size that is independent of high turbu- and altered trophic effects attributable to the
lent diffusivities for their delivery (Wolf-Gladrow activities of grazers (Chapter 6). Seasonal changes
and Riebesell, 1997; Section 4.2.1) – select for in resourcing (especially with respect to changing
the relative absence of large species, for the availabilities with depth) may prompt composi-
high incidence of nanoplankters (Gorham et al., tional shifts that may be influenced less by ratios
1974; Watson and Kalff, 1981) and, especially, than by such mechanisms as highly specialised
picoplankters, both in lakes (Zuniño and Diaz, affinities, alternative sources of nutrients (nitro-
1996; Agawin et al 2000, Pick 2000) and the open gen fixation, facultative bacterivory and mixotro-
sea (Chisholm et al., 1992; Campbell et al., 1994; phy) or vertical migration.
Karl, 1999; Karl et al., 2002; Li, 2002). Thus, dis- The idea that species that are simultaneously
tinctive groups of species come to characterise limited by different resources do not compete
severely P-deficient lakes and ultraoligotrophic but, rather, coexist successfully has a wide fol-
oceans (see Table 5.2). lowing among plankton ecologists. It conforms
It may well be the case that the lakes can to Hardin’s (1960) principle of competitive exclusion,
be described accurately as having high N : P which states a long-standing ecological tenet that
true competitors cannot coexist. One of its corol- cessful of the early colonist and pioneer species
laries (that, in a steady state, each truly coexist- of the pelagic succession have to be poised to
ent species occupies a distinctive niche Petersen invade or to exploit the favourable conditions
1975) has been advanced to account for the mul- that open to them. They may ascend to pre-
tiple species composition of natural phytoplank- eminence through an ability to grow faster than
ton species assemblages. The resource-based com- their rivals (to ‘outperform’ them but not to ‘out-
petition model appears to be powerfully support- compete’ them). Sooner or later, however, grow-
ive and, provided the limiting conditions persist ing demands impinge upon the supply of one
uninterrupted and for long enough, will lead to or more essential components, after which con-
the competitive exclusion of all but the fittest tinued dynamic success requires slightly more
species. However, models attempting to verify specialist adaptations for resource gathering.
this provision by simulating rather more than For instance, noticeably more ‘eutrophic species’
two species competing for more than two limit- might start to exclude oligotrophic ones on
ing resources seem to break down into chaos with the basis that the reserve of carbon dioxide is
unpredictable outcomes (Huisman and Weissing, depleted as a consequence of plankton growth.
2001). Another possibility is that sufficient growth is
accommodated for the light availability in the
Discontinuous nutrient deficiences surface mixed layer to become contested by supe-
The pattern of change in many pelagic sys- rior light-harvesters. A third likelihood, to which
tems is that, as a result of seasonal mixing, particular attention is now given, is that algae
storm episodes or periodic inflows, the levels that are able to penetrate deeper in the water
of all nutrients may be raised sufficiently to column gain access to nutrients located at depth
support the vigorous growth of various species in the increasingly segregated vertical structure.
of phytoplankton. The consequence is that the The surface mixed layer is an important
available resources are depleted, one or more entity, in limnology as in oceanography. The fre-
to a level that represents a threshold of lim- quency with which it is turned over (∼45 min-
itation and the ability of some or all species utes or less: see Section 2.6.5) is much shorter
to grow becomes subject to stress. The initial than the generation time of the plankters embed-
combination of relative resource-richness and ded within the layer. It is quite proper, when
high insolation supports strong growth of many considering planktic populations, to regard the
species in the surface mixed layer. Consequen- surface mixed layer as a single, isotropic envi-
tial resource depletion tends to proceed from the ronment. In cold or shallow waters exposed to
mixed layer downwards, leading eventually to the moderate wind stress, the surface mixed layer
progressive uncoupling of resources from light. extends to the bottom of the water column, or
The water column progressively segregates into to within a millimetre or two of the bound-
an increasingly resource-depleted upper layer ary layer therefrom. Beyond the density gradi-
and to deeper, increasingly light-deficient lay- ent separating mixed layer from deeper, denser
ers, wherein available nutrients persist pending water masses, many properties of the habitat
exploitation by autotrophs. can differ markedly from those of the surface
Such sequences of events are held to differ- mixed layer: renewal rate, temperature, insola-
entiate among the adaptive traits and species- tion, gas and nutrient exchange rates, and so on.
specific performances of phytoplankton, deeply Thus, the variability in its vertical extent can also
influencing the species selection and seasonal be a critical determinant of performance, alter-
shifts in species dominance. These progressions nately entraining and randomising the planktic
of composition and dominance are sufficiently population through a column of uniform and
striking, in some cases, to have been analo- increasingly low insolation, then disentraining
gised to classical ecological successions in terres- it through stagnating layers. The frequency of
trial plant communities (Tansley, 1939; Reynolds, the alternation can be critical too: segregation
1976a; Holligan and Harbour, 1977). The most suc- may last for a few hours (as in dielly stratifying
systems), a few days (polymictic and atelomictic form resistance are essential to embedding, so
systems), months on end (seasonal stratification) large size and low form resistance are advan-
or more or less continuously (as in meromic- tageous to ready disentrainment from weaken-
tic systems). The longer is the separation, the ing turbulence and to commencing controlled,
greater can be the differentiation with, instead directed excursions through the water column.
of one, unique environment, a continuous and The velocities achieved can nevertheless seem
widening spectrum of individualistic microhab- impressive. Smaller gas-vacuolate, bloom-forming
itats (Reynolds, 1992c; Flöder, 1999). Simultane- Cyanobacteria (including Microcystis, Gompho-
ous feedbacks, including the transfer of settling sphaeria, Gloeotrichia, Nodularia, Cylindrospermopsis,
biomass, cadavers and the faecal pellets of algal raft-forming Aphanizomenon and the species of
consumers to the uninsolated layers below, may Anabaena and Anabaenopsis which typically aggre-
further enhance the developing structural dis- gate into secondary tangles of filaments) whose
continuity. buoyant velocites may reach 40–100 µm s−1
Among the species attempting still to assem- normally sink three to six times more slowly
ble biomass and replicate their numbers, there (say 0.6–3.0 m d−1 ) (Reynolds, 1987a). The move-
is a progressive transfer of advantage from those ments of large freshwater dinoflagellates like Cer-
adept at exploiting the mixed-layer resource base atium hirundinella and Peridinium gatunense can
(high-sv−1 species, sustaining high replication cover 8–10 m in a single night (Talling, 1971;
rates) to those which are equipped to benefit Pollingher, 1988). The rates quoted for many
from the separation of the resource base. The marine species exceed 100 µm s−1 ; one or two
situation is reminiscent of the progressive ele- exceed 500 µm s−1 (Smayda, 2002), although
vation of productive terrestrial foliage from the the distances they travel are not given. Volvox
herb layer to the woodland canopy. In the case undertakes the longest reported circadian migra-
of the pelagic, however, it is the downreach of tion of any freshwater flagellate, traversing 17 m
the resource-gathering capacity that is respon- in either direction of the Cabora Bassa Dam,
sible for the functional separation rather than Mozambique (Sommer and Gliwicz, 1986), at an
the uplift of the light-harvesting apparatus! The average velocity scarcely under 1.5 m h−1 . In abso-
investment by (appropriately adapted) terrestrial lute terms, this is modest but, at the scale of the
plants is in building the mechanical connection. organism, progression at the rate of 1 to 3 colony-
Among the microscopic pelagic plants the adap- diameters per second is impressive.
tive investment is in migration. The question has to be asked whether these
Two interrelated sets of adaptations give the migrations do actually yield a harvest of nutri-
advantage to pelagic canopy species. One is the ents, sufficient to provide a dynamic advan-
power of motility: if the alga is to have any tage over non-migrating species. According to
prospect of covering the vertical distance sep- Pollingher (1988), patterns of movement are
arating light and nutrient resources, it must dominated by a strong, positive phototaxis in
be able to determine the direction of its move- the early part of the day (though, generally,
ment. Flagellate genera of the Cryptophyta, their movements will avoid supersaturating light
the Pyrrhophyta, the Chrysophyta, the Eugleno- intensities) but they show a decining photo-
phyta and the Chlorophyta would appear to responsiveness during the course of the solar
have the essential preadaptations, although the day. The quality of the light and temperature
buoyancy-regulating mechanism of gas-vacuolate gradients, the extent of nutrient limitation and
Cyanobacteria is just as effective a means of pro- the age of the population also influence pat-
pelling migratory movements. The second adap- terns of movement. The extent of dinoflagellate
tation, however, is the one that turns the ability migrations in a given lake are said to increase
to move into the ability to perform substantial with decreasing epilimnetic nutrients, provided
vertical migration, i.e. the size-determined capac- the segregated structure persists and the excur-
ity to disentrain from residual turbulence (Sec- sions into deeper water provide reward. It is inter-
tions 2.7.1, 2.7.2). Just as small size and maximun esting, too, how diminishing nutrient resources
should determine that less of the photosynthate direction enables an alga to recover vertical sta-
produced by buoyancy-regulating Cyanobacteria tion very quickly in the wake of disruptive mix-
ends in new cytoplasm and more goes to offset- ing events, when smaller flagellates or solitary
ting buoyancy, forcing organisms to sink lower buoyany regulating filaments take hours or days
in the water column (Sas, 1989). Note that short- to do so (Reynolds, 1984c, 1989b). Another is that
age of carbon, like shortage of light, means less in the face of weak wind- or convective-mixing,
photosynthate is produced, so organisms become the alga can be quite effective in self-regulating
lighter and float closer to the surface. This prin- its vertical position in order to balance its pho-
ciple has been demonstrated well in the obser- tosynthetic production and its resource uptake
vations and experiments of Klemer (1976, 1978; with the rate of cell growth and replication. In
Klemer et al., 1982, 1985) and Spencer and King this way the cell saves energy in fixing photo-
(1985). These self-regulated movements of large synthate, which, if it could not be made into
plankters certainly seem to open the access to proteins and new cell material, would otherwise
deep-seated nutrient stores. It is apparent, too, have to be voided from the cell. Organisms which
that their growth may be enhanced when con- do this very well, such as Microcystis, not only rise
ditions of near-surface nutrient depletion obtain to dominance but remain dominant for months
and the range of vertical migration extends to (and even years on end) when the appropriate
depths offering replenishment. conditions persist (Zohary and Robarts, 1989).
There are few studies that provide com-
pelling evidence that this is always the case Low insolation and growth-rate regulation
(Bormans et al., 1999). However, Ganf and Oliver Under conditions of short photoperiods and low
(1982) showed, through observation and careful aggregate insolation, the problem for phyto-
experimental translocation, that Anabaena fila- plankton is defined by the point that the alga
ments picked up substantial amounts of nutri- is no longer able to intercept and harvest suffi-
ent on their buoyancy-regulated excursions in cient light energy, or to invest it the recruitment
the Mount Bold Reservoir, South Australia. Raven of new protoplasm and daughter biomass, at a
and Richardson (1984) considered the extra nutri- rate that the temperature and the nutrient sup-
ents derived by a migrating marine Ceratium to ply will allow. Below this level, growth rate is,
be only weakly attributable to movement per se indeed, light-limited. The curves inserted in Fig.
(see Section 4.2.1) and much more to the encoun- 5.4 differentiate among plankters on the basis of
ters with unexploited nutrients in the (to them) their shape and their capacity for low-light adap-
accessible parts of the water column. Deep-water tation. The point to notice is that the species
reserves of phosphorus were shown to be within that are capable of the fastest rates of growth
the facultative swimming ranges of Ceratium in under relatively high insolation are not necessar-
Esthwaite Water, UK (Talling, 1971) and within ily the best adapted to live on small light doses.
the ‘vertical activity ranges’ of Ceratium (and, for The limnetic species that do this well include
a time, Microcystis) intercepted by sediment traps the diatom Asterionella and solitary filamentous
in Crose Mere, UK (Reynolds, 1976b). species of the Oscillatoriales (Planktothrix agardhii,
Thus, strong, self-regulated motility is con- Limnothrix redekei), in which the capability is cor-
sidered to offer significant advantages, provid- related with relative morpholgical attenuation
ing opportunities for the selective garnering of (high msv−1 : Fig. 5.5). There is often a high capac-
the diminishing resources of a structured envi- ity for auxillary and accessory pigmentation as
ronment. Adapted species are enabled access to well. Thus, their successful contention to per-
parts of the water column that other algae do form relatively well in poorly insolated, natural
not reach, or do not do so sufficiently quickly, mixed layers owes most to their extraordinary
or, having done so, cannot reverse their motion to abilities to open the angle of r on I (Fig. 5.4; α r
recover a position in the euphotic zone. It should in Fig. 5.5) and, thus, to lower the light inten-
not be assumed that this is the only advantage. sity at which growth rate can be saturated. In
The ability to swim strongly and in controlled the open, mixed-water column, this extends the
actual depth through which growth-saturating tons m−2 s−1 ), the light-limited Asterionella might
photosynthesis may be maintained and, in turn, still increase at a rate of 0.335 d−1 , which is
lengthens the aggregate of probabilistic photope- twice as fast as that of the temperature-limited
riod, tp , over that expected for unadapted species Planktothrix or the light-limited Chlorella. It is
in the same water layer. From the least-squares at once appreciable how subtle are the condi-
regression fitted to the data in Fig. 5.5, tions distinguishing among species performances
under low doses of light. We might also spec-
αr = 0.257(msv−1 )0.236 (5.8)
ulate that although there is an apparent dis-
it may be predicted that the stellate colony of cretion in favour of diatoms in cold, energy-
Asterionella generates a slope of α r = 0.86 (mol deficient, mixed layers, a little more vigourous
photon)−1 m2 , while for a 1-mm thread of Plankto- mixing (I∗ falls) or a less severe winter tem-
thrix agardhii, α r = 1.12. For the small spherical perature might favour filamentous Cyanobac-
cell of Chlorella, the slope is predicted to be only teria instead. Reduced mixing and better near-
0.39 (mol photon)−1 m2 . Analogous to the inter- surface insolation immediately favours faster
relationships among photosynthetic rate and the growing nanoplankters such as Chlorella.
onset of light saturation of photosynthesis (Eq.
3.5, Section 3.3.1), the lowest light dose that will
sustain maximal growth rate at 20 ◦ C is indicated Trait interaction and functional

by the quotient, r20 /α r . Thus, on the basis of the differentiation in phytoplankton
assembled data (see Table 5.1), we may deduce The real world of phytoplankton is a blend
that Chlorella growth will be saturated by a pho- of deficiencies of differing intensity and fre-
ton flux of 3.34 mol photons m−2 d−1 (equivalent quency, especially with respect to the availabil-
to a constant ∼39 µmol photons m−2 s−1 ), that of ity and accessibility of nutrient resources and
Asterionella by 2.07 mol photons m−2 d−1 (24 µmol the solar energy needed to process them. Spe-
photons m−2 s−1 ), and that of the Planktothrix by cialist adaptatations, both in terms of physio-
0.77 mol photons m−2 d−1 (9 µmol photons m−2 logical responses at the scale of the life cycle
s−1 ). and the traits distinguished at the evolution-
The inference is emphasised: at irradiance lev- ary scales, may increase the relative fitness of
els exceeding levels of 3.34 mol photons m−2 some species along particular gradients of envir-
d−1 , Chlorella is the ‘fittest’ of the three, and its onmental variability but none is well suited to
regression-predicted growth rate of 1.84 d−1 out- all conditions. For instance, we may suppose that
strips those of Asterionella (1.78 d−1 ) and Plank- the most competitive adaptation would be to
tothrix (0.86 d−1 ). However, at light levels equiv- enable the phytoplankter to self-replicate more
alent to 0.77 mol photons m−2 d−1 , Planktothrix rapidly than other species that might be present;
can still be argued to be able to maintain its max- hence, a morphology conducive to rapid surface
imum growth rate (0.86 d−1 ), and when that of exchanges of nutrients should be favoured, that
Asterionella is cut back to 0.66 d−1 and Chlorella is is, one that maintains a large sv−1 . The oppos-
severely light-limited at not more than 0.42 d−1 . ite trend of increasing size (and reducing sv−1 )
When the low temperatures of high-latitude win- carries advantages of motility, storage and per-
ters are taken into account, the impact of surface- sistence (see also Section 6.7), where the ability
to-volume relationships modify the relative fit- to influence vertical position, to gain access to
nesses of these organisms. At 5 ◦ C, the predicted nutrient resources unavailable to other species
resource-saturated growth rates for Chlorella, and to avoid consumption by herbivores offer
Asterionella and Planktothrix are, respectively, superior prospects of survival. One advantage has
0.375, 0.335 and 0.163 d−1 and the respective sat- been ‘traded’ against another, at the price of low-
urating fluxes are calculated to be 0.96, 0.39 and ered habitat flexibility: some environments will
0.15 mol photons m−2 d−1 . Thus, under an aver- be better tolerated, or even preferred, by a given
age irradiance (I∗ ) of 0.4 mol photons m−2 d−1 species than will others. Such differentiation is,
photon flux (equivalent to a constant 5 µmol pho- of course, the basis of patterns in the spatial
Figure 5.8 Comparison of

growth-rate performances of some
phytoplankters. In (a), the minimum
light intensity (I) and the minimum
soluble-phosphorus concentration
([Slim ]) required to saturate the
growth at 20 ◦ C of the named algae
are plotted against each other. The
algae are: Ast, Asterionella; Chlo,
Chlorella; Mic, Microcystis; Per,
Peridimium cinctum; Pla, Planktothrix
agardhii. In the other sub-figures,
selected isopleths of growth rate at
20 ◦ C (inserted numerals are ×
10−6 s−1 ) are constructed against
the same gradients, for (b) Chlorella,
(c) Asterionella and (d) Microcystis.
The contours are Redrawn with
and temporal distribution of species, whereby against axes representing steady-state concentra-
some are more clearly associated with particular tions of phosphorus and the photon flux of white
conditions than are others. The further adaptive light, at 20 ◦ C. The high levels of resource and
option for larger algae – that of shape distortion light required to saturate the most rapid growth
that increases the surface area bounding a given of Chlorella show well against the requirements of
cytovolume – provides not so much a compromise four others species (Fig. 5.8a). The sensitivity of
between small and large size but the enhanced Chlorella performance to both light and phospho-
ability to process resources into biomass in rela- rus relative to that of Asterionella or of the poorly
tively short periods of exposure to light. performing Microcystis against these two criteria
Based upon the growth rates of various is evident (Fig. 5.8b, c, d).
species of algae against chosen dimensions in
the foregoing sections, we are now able to
devise comparative graphical representations of Growth and reproductive strategies
the replicative performances of algae against the When growth under persistently low levels of
two key axes of resource availability and inso- light and nutrient are considered simultane-
lation. In Fig. 5.8a, growth-rate contours of sev- ously, the basis for some of the very interesting
eral algae are drawn in space defined by light patterns alluded to by Tilman et al. (1982) and
and phosphorus saturation of growth-rate poten- by Sterner et al. (1997) may be readily appreci-
tial. The result is broadly similar to those built ated. Now, for example, we may envisage circum-
on mean underwater light levels and KU val- stances under which growth rate in Asterionella
ues with respect to species-specific phosphorus is encountering (say) silicon limitation when the
uptake rates (Reynolds, 1987c) or of light sup- growth rate of Cyclotella is constrained by light,
ply and nutrient supply (Huisman and Weiss- and when the growth rate of Planktothrix is too
ing, 1995). The plots making up the rest of Fig. constrained by low temperature or low phosphor-
5.8 show species-specific replication-rate contours us to be able to take full advantage.
There are probably sufficient data to be able ing resources in short supply with their reten-
to simulate these interactions more rigorously. tion among a high survivorship. Unlike the obli-
This is less interesting to pursue than it is to gately fast-growing, r-selected category, resource-
abstract the generalities about the differing adap- (K-)selected species do not share the constraint of
tations shown by the algae considered here and maintaining a high surface-to-volume (sv−1 ) ratio.
the broad properties that underpin their strate- However, the acquisitive garnering of dimin-
gies for growth and survival. The use of the word ishing resources sometimes favours significant
‘strategy’ in the context of the evolution of life powers of migratory motility, for which a rel-
histories is open to criticism on etymological atively large size (with attendant penalties in
grounds, as it implies that their differention is reduced sv−1 , slow growth rate and impaired
planned or anticipated in advance (Chapleau et light-absorption efficiency, ε a ) is essential (see
al., 1988). In reality, different patterns for pre- Sections 2.7.1, 3.3.3). Microcystis aeruginosa pro-
serving and reproducing genomes have evolved vides a good example of this second type of
along with the organisms they regulate and, just strategy that identifies ‘winners’, or like Aesop’s
as certainly, have been shaped by the same forces fabled tortoise, the ‘good competitor’ in the sense
of natural selection. The patterns are distinctive, understood by most plankton ecologists (Kilham
separating life histories that, for example, permit and Kilham, 1980).
opportunistic exploitation of resources and pho- The ability to harvest and process energy from
ton energy (as does Chlorella in the example in low or diminishing irradiances or from truncated
the previous section) or, alternatively, may pro- opportunities at higher irradiance is favoured by
vide high adaptability to a low nutrient or to low small size or by attenuation of larger sizes (in
energy supply (as does Planktothrix). The compar- one or possibly two) planes. These traits repre-
ative efficiencies and flexibilities of investment sent a high photon affinity, which is not bound
of harvested energy and gathered resources into exclusively to either r- or K-selection, and to
species-specific biomass define the growth and which Reynolds et al. (1983b) applied the term
reproductive strategies of phytoplankton (Sandgren, w-selection.
1988a). There are clear similarities and apparent
So far, the discussion has identified three analogies in these broad distinctions with the
basic sets of strategic adaptations, involv- three primary ecological and evolutionary strate-
ing morphologies, growth rates and associ- gies identified among terrestrial plants (Grime,
ated behaviours. The first is the Chlorella type 1977, 1979, 2001). Grime’s concept was built
of exploitative or invasive strategy, in which around the tenability of habitats according to (i)
organisms encountering favourable resource the resources available and the levels of stress on
and energy fluxes can embark upon the life cycles that resource shortages might impose
rapid resource processing, biosynthesis and on plant survival and (ii) the duration of these
genomic replication (reproduction) that consti- conditions, pending their disruption or obliter-
tute growth. They necessarily have a high growth ation by habitat disturbance. Of the four possi-
rate, r, based on an ability to collect and con- ble permutations of stress and disturbance (Table
vert resources before other species do and, in this 5.3), one, the combination of continuous severe
sense (the one followed by most plant ecologists), stress and high disturbance results in environ-
they are ‘good competitors’. Curiously, plankton ments hostile to the establishment of plant com-
ecologists reserve this term for the ‘winners’ of munities is untenable. These are deserts! The
the competition, applying it to those species that three tenable contingencies are variously pop-
specialise in the efficient garnering, conserving ulated by plants specialised in either (a) rapid
and assembling the limiting resource base (K) exploitation of the resources available (‘com-
into as much biomass as it will yield. Thus, petitors’ in the original usage of Grime 1977,
the second set of strategic adaptations variously 1979), which he dubbed ‘ C-strategists’; or (b) tol-
combines high resource affinity and/or special- erance of resource stress, by efficient matching
ist mechanisms for obtaining scarce or limit- of the limited supply to managed demand, and
Table 5.3 Basis for evolution of three

primary strategies in the evolution of plants, C R
Monor
phytoplankton and many other groups of
organisms
Syn
Chlo Plg Fra
Habitat productivity Scq Ast
Habitat 100 Chla Monod Din Lim r
duration High Low Sth
Tab Aul
sv −1 / µm −1
Pla ag
Long Competitors, Stress-tolerant Aph
Cry
invasive (C ) (S )
Ana Cer
Short Disturbance- No viable strategy
Per
tolerant 10−1
ruderals (R )
Eud S
Source: Original scheme of Grime (1979; modified Mic
after Reynolds, 1988a; Grime, 2001).

Vol
10−2
101 102 103

msv −1
Figure 5.10 Morphological ordination of some species of

freshwater phytoplanton, against axes invoking maximal linear
dimension (m), surface area (s) and volume (v) of the
vegetative units, with the C-, S- and R-strategic tendencies.
The algae are: Ana, Anabaena flos-aquae; Aphan,
Aphanizomenon flos-aquae; Ast, Asterionella formosa; Aul,
Aulacoseira subarctica; Cer, Ceratium hirundinella; Chla,
Chlamydomonas; Chlo, Chlorella sp.; Cry, Cryptomonas ovata; Din,
Dinobryon divergens; Eud, Eudorina unicocca; Fra, Fragilaria
crotonensis; Lim r, Limnothrix redekei; Mic, Microcystis aeruginosa;
Monod, Monodus sp.; Monor, Monoraphidium contortum; Per,
Peridinium cinctum; Pla ag, Planktothrix agardhii; Plg, Plagioselmis
nannoplanctica; Scq, Scenedesmus quadricauda; Sth,
Figure 5.9 Grime’s model of tenable and untenable
Stephanodiscus hantzschii; Syn, Synechococcus sp.; Tab, Tabellaria
habitats, and noting the primary (C, S or R) life-history
flocculosa var. asterionelloides; Vol, Volvox aureus. Redrawn with
strategies required to secure survival. Redrawn, with
permission from Grime (2001).
to which he gave the label ‘ S-strategists’; or (c) respectively, C, S or R strategies, on the satisfying
tolerance of disturbance, through making good basis of agreement among the morphological
opportunity of transient habitats and interrupted properties, growth rates and life-history traits.
opportunities to process resources into biomass The distribution of phytoplankton species
(‘ R-strategists’). according to their individual morphologies
The three primary strategies of Grime’s CSR plotted against axes of sv−1 and msv−1 (Fig. 5.10).
model form the apices of a triangular ordination Just as with Grime’s (1979) scheme, species are
(Fig. 5.9), which representation readily allows not exclusively C or S or R in their strategic
the accommodation of numerous intermediates adaptations. Many species of phytoplankton
and trait-combinations. Reynolds (1988a, 1995a) show intermediate characters. Interestingly,
found only minor difficulties in analogising intermediacy in morphological and physiolog-
the r-, K- and w-selected groups to exemplifying, ical characters matches well the intermediacy
of their ecologies. The C–S gap is spanned by tion recognises that motility and large size
genera such as Dinobryon, Dictyosphaerium, are not necessary adaptations to function in
Coenochloris, Pseudosphaerocystis, Eudorina and, chronically very resource-depleted pelagic envi-
arguably, Volvox (Reynolds, 1983b), and by ronments. Indeed, resource gathering in spatially
Aphanocapsa and Aphanothece. The series spans continuous, rarefied environments is favoured
diminishing sv−1 ratios, maximum growth rates by small size, whereas the low levels of dif-
and low-temperature tolerance but increas- fuse biomass is an unattractive resource for
ing ability to exploit and conserve nutrient direct grazing by mesozooplankters (see Chap-
resources. Algae in the C–R axis include predom- ter 6). The adaptive strategies for surviving
inantly centric diatoms of varying tolerance of the ‘resource desert’ of the ultraoligotrophy of
turbidity and the Scenedesmus–Pediastrum element the oceanic pelagic are accorded the additional
of enriched shallow ponds and rivers). The R–S stress-tolerant category SS.
possibility is represented by the slow-growing, The original ascriptions of C, S and R cate-
long-surviving, acquisitive but highly acclimated gories to phytoplankton (Reynolds, 1988a) sepa-
species of density gradients, like Planktothrix rate quite satisfactorily on the plot of the areas
rubescens and Lyngbya limnetica. Certain (not projected by various species of phytoplankton
all) members of the genus Cryptomonas show a and the product of maximum dimension and
blend of the characteristics of all three primary surface-to-volume ratio (msv−1 ) Fig. 3.12). Near-
strategies in being unicellular, having cells of spherical forms align close to msv−1 [d × 4π (d/2)2
moderate size (1–4 ×103 µm3 ) and of interme- ÷ 4π (d/2)3 /3] = 6 but separate broadly in to
diate sv−1 (0.3–0.5 µm−1 ), and being capable of C and S species according to size, because the
intermediate replication rates (r20 ∼10 × 10−6 carbon and chlorophyll contents vary with v =
s−1 ; r0 ∼0.9 × 10−6 s−1 ). 4π (d/2)3 /3 but the light interception increases
It is right to point out that Grime’s CSR con- as a function of the disk area, a = π(d/2)2 .
cept of plant stategies is not universally accepted The morphological attenuation of the R species
and it has been subject of vehement and chal- pulls out the plot to much higher msv−1 val-
lenging debate (see Tilman, 1977, 1987, 1988; ues. Thus, we distinguish species that are capa-
Loehle, 1988, a.o.). Although there is much com- ble of rapid growth in benign, resource-replete
mon ground shared by the adversaries and, in environments, those that are able to go on
truth, the differences are more of perspective squeezing out increased biomass from diminish-
and emphasis (Grace, 1991), the differences have ing light income and those who are physiolog-
never been entirely resolved. The application to ically or behaviourally adapted to function in
plankton has not been so criticised and some spite of developing nutrient stress. The model
(Huszar and Caraco, 1998; Fabbro and Duiven- appears in various guises later in the book,
vorden, 2000; Gosselain and Descy, 2000; Kruk demonstrating the power and flexibility of the
et al., 2002; Padisák, 2003) but by no means strategy–process–ecosystem interactions. It even
all (Morabito et al., 2002), have found the argu- provides the bridge to the light : nutrient hypoth-
ments convincing and helpful to interpretation. esis (Sterner et al., 1997) in so far as the species
The applicability of a scheme devised for plant best adapted to cope with low doses of I∗ are most
species is not a barrier: it is now quite evi- able to cope with high particulate content in the
dent that the idea has a long pedigree among water and the C : P ratio of the seston available
other ecological schools (Ramenskii, 1938) and to secondary consumers.
has been applied successfully to the ‘violent’,
‘patient’ and ‘explerent’ strategies of zooplank- 5.4.6 Resource exhaustion and survival
ton (Romanovsky, 1985). The CSR model has been It is reasonable to assume that the growth of
applied to fungi (Pugh, 1980) and periphyton phytoplankters distinguished by efficient, high-
(Biggs et al., 1998). affinity resource-gathering capabilities may con-
An updated application to phytoplankton tinue until they deplete their growth-limiting
is set out in Box 5.1. A notable modifica- resource to near exhaustion. It was often and
Box 5.1 Summary of behavioural, morphometric and

physiological characteristics of growth and
survival strategies of freshwater phytoplankton
With little adjustment, the primary strategies (otherwise, functional types) of plants
devised by Grime (1979, 2001) are known to apply to other types of organism,
including phytoplankton. The application of the scheme to plankton (Reynolds,
1988a) required some modest adjustment but the relevant morphological and
physiological characteristics are, of course, peculiar to planktic algae. These are
listed below, following Reynolds (1988a, 1995a) but include the features of a new
subcategory (SS) to accommodate features of permanent stress-tolerant algae of
ultraoligotrophic oceans.
C strategists
Grime’s name Competitors
Reynolds’ (1995a) label Invasive opportunists
Dispersal Highly effective, cosmopolitan; mechanisms
sometimes obscure
Selection r
Cell habit Mostly unicellular
Unit sizes 10−1 –103 µm3
msv−1 6–30
Cell projection >10 m2 (mol cell C)−1

r 20 >10 ×10−6 s−1 ; >0.9 d−1
r 0 >2 × 10−6 s−1 ; >0.18 d−1
Q10 <2.2
Species experience low growth thresholds for light (Section 5.5), have generally low
rates of sinking (some are motile; Section 6.3) and are highly susceptible to grazing
zooplankton (Section 6.4). Representative genera Chlorella, Ankyra, Chlamydomonas,
Coenocystis, Rhodomonas.
R strategists
Grime’s name Ruderals
Reynolds’ (1995a) label Attuning or acclimating (also processing–
constrained)
Dispersal Widely distributed, mechanisms
sometimes obscure
Selection r and K (w in Reynolds et al., 1983b)
Cell habit Some unicellular, many coenobial
Unit sizes 103 –105 µm3
msv−1 15–1000
Cell projection 8–30 m2 (mol cell C)−1

r 20 >10 × 10−6 s−1 ; >0.85 d−1
r 0 0.08–2 × 10−6 s−1 ; >0.1 d−1
Q10 2.0–3.5
Species force very low growth thresholds for light (Section 5.6), sinking rates low to
high, most are non-motile (Section 6.2); some susceptible to grazing zooplankton
(Section 6.3). Representative genera Asterionella, Aulacoseira, Limnothrix, Plank-

tothrix.
S strategists
Grime’s name Stress-tolerators
Reynolds’ (1995a) name Acquisitive (also resource-constrained).
Dispersal Tendency to discontinuous distribution,
mechanisms better known
Selection Strongly K
Cell habit Some unicellular, many coenobial
Unit sizes 104 –107 µm3
msv−1 6–30
Cell projection <2.5 m2 (mol cell C)−1

r 20 <8 × 10−6 s−1 ; <0.7 d−1
r 0 <1 × 10−6 s−1 ; <0.09 d−1
Q10 >2.8
Species contend effectively with diminishing, or increasingly distant nutrient

resources, either through exploiting alternative sources (nitrogen fixation, phos-
phatase production, phagotrophy on bacteria or particulate organic material).
Most are motile, some are strongly so, sinking rates low but may undertake con-
trolled migrations over large vertical distances. Also referred to as resource ‘glean-
ers’ (Anderies and Beisner, 2000). Representative genera Microcystis, Anabaena,
Gloeotrichia, Ceratium, Peridinium, Chrysosphaerella, Uroglena.
SS strategists
Grime’s name (not applicable)
Reynolds’ (1995a) Chronic-stress tolerant
name
Dispersal Cosmopolitan
Selection Ultimately K
Cell habit Exclusively unicellular
Unit sizes ≤4 µm3
msv−1 6–8
Cell projection ∼50–60 m2 (mol cell C)−1

r 20 unknown, probably >20 ×10−6 s−1 ; >1.8
d−1
r 0 Possibly up to 0.5 × 10−6 s−1 , 0. 4 d−1
Q10 Not known, probably ∼2
This newly separated group of species tolerant of chronic nutrient stress accom-
modates the prokaryotic picoplankters that dominate the rarefied environments
of the tropical seas and which, increasingly, have been shown to be active in the
open waters of the world’s largest and most oligotrophic lakes (Reynolds et al.,
2001). They are non-motile but have very low sinking rates. Their small size is the
key to living on very dilute nutrient sources. It would leave them very vulnerable
to grazing by filter-feeders, except that they inhabit environments that fail to sus-
tain filter-feeding zooplankton. Representative genera Prochlorococcus (in the sea;
Cyanobium, Cyanodictyon are considered to be limnetic analogues).
commonly supposed by many plankton ecolo- of reproductive and resting propagules has been
gists that nutrient exhaustion is followed by mass mainly confined to studies on particular phyloge-
clonal mortalities. This view was perhaps encour- netic groups (Sandgren, 1988a). It is not inappro-
aged by numerous observations of ‘bloom col- priate to give a brief perspective at this point.
lapse’, of diatoms running out of silicon (e.g. Resting stages come in a variety of forms and
Moed, 1973) or the photolysis of surface scums of are stimulated by a variety of proximate events
Cyanobacteria (e.g. Abeliovich and Shilo, 1972). and circumstances, and their success in ‘carrying
These relatively impressive eventualities apart, forward’ biomass and genomes is also quite vari-
however, phytoplankters are rather better pre- able. Among the simplest resting stages are the
pared than this to be able to avoid sudden death contracted protoplasts produced in such centric
and disintegration. Depletion of one of the essen- diatoms as Aulacoseira (Lund, 1954) and Stephano-
tial resources usually leads to a cellular reserve of discus spp. (Reynolds, 1973a). These form quite
the others and the cell may be able to use stored freely in cells falling into aphotic layers and may
carbohydrate, polyphosphate or protein reserves be prompted by microaerophily and low redox,
to maintain some essential activity. However, it which conditions may be tolerated for a year
is quite clear that it is better for the cell to lower or more. The contents pull away from the wall,
its metabolism and to close down those processes abandon the central vacuole and shrink to a tight
not directly associated with actually staying alive. ball, a micrometre or so in diameter. Individual
Earlier chapters have emphasised the mecha- cells or filaments containing resting stages lit-
nisms for internal communication of nutrient- ter the surface sediments. If seeded sediment is
uptake activity of the membrane- transport sys- placed under low light in the laboratory, Aulaco-
tem (Section 4.2.2), the activation of inhibitory seira will ‘germinate’ and produce swathes of new
nucleotides (such as ppGpp) in response to falling filaments in situ. Germination in nature may be
amino-acid synthesis (Section 4.3.3) and the sus- only a little less spectacular but it always depends
pension of nuclear division (Section 5.2.1). Each upon the resuspension of filaments and cells by
represents a step in the biochemical procedure by entrainment from sediments accessible to tur-
which the cell senses its environmental circum- bulent shear. Thus, formation and germination
stances and organises its appropriate defences to of the resting stages is governed by the activity
enhance its survival prospects. These may include or otherwise of its photosynthetic capacity. Per-
the inception of a ‘cytological siege economy’ haps 5–20% of the sedimenting population may
and the structural reorganisation of the proto- form resting stages. The percentage of these that
plast into resting cells, with or without thickened return to the plankton is probably small but they
walls. can provide quantitatively important inocula to
The biological forms of most kinds of rest- future populations (Reynolds, 1988a, 1996b).
ing cell are well recognised by plankton ecolo- The Cyanobacterium Microcystis has the abil-
gists and, in many cases, so are the environmen- ity to control its own vertical migrations through
tal attributes which induce them. Equally, the regulating its buoyancy and, in warm latitudes,
implicit benefit of survival of resting stages is it may move frequently (perhaps dielly) between
widely accepted as a means to recruit later pop- sediment and water, very much as part of
ulations from an accumulated ‘seed bank’. They its vegetative activity (May, 1972; Ganf, 1974a;
need to recognise and respond to their reintro- Tow, 1979). In temperate lakes, Microcystis is fre-
duction into favourable environments or to ame- quently observed to overwinter on the bottom
liorating conditions by embarking upon a phase sediments (Wesenberg-Lund, 1904; Gorham, 1958;
of renewed vegetative growth. However, it has to Chernousova et al., 1968; Reynolds and Rogers,
be stated that, in marked contrast to the efforts 1976; Fallon and Brock, 1981; and many others
that have been made to observe and under- reviewed in Reynolds, 1987a). There is a mas-
stand the mechanisms generating the spatial sive autumnal recruitment of vegetative colonies
and temporal patterns of phytoplankton occur- from the plankton to depth (Preston et al., 1980),
rence, detailed information on the significance where they enter a physiological resting stage.
No physical change occurs (they do not encyst) colonies apparently need the low temperatures
and chlorophyll, as well as a latent capacity and low oxygen levels for their maturation. Ulti-
for normal, oxygenic photosynthesis, is retained mately, they also require low oxygen levels and
(Fallon and Brock, 1981). Curiously, the cells simultaneous low insolation to persuade them
also remain gas-vacuolate. Despite being (ini- to initiate the formation of the new colonies
tially) loaded with glycogen, other carbohydrates, that recolonise the water column in the follow-
proteinaceous structured granules and polyphos- ing year (Reynolds and Bellinger, 1992; Brunberg
phate (Reynolds et al., 1981), they would be buoy- and Blomqvist, 2003). The completion of this
ant but for the precipitation of iron hydroxide cyclical process depends on interactions among
on the colony surfaces, which acts as ballast light, temperature and sediment oxygen demand.
and causes the organisms to sink (Oliver et al., Whereas upwards of 50% of the colonies consti-
1985). Once on the sediments, in very weak light tuting the previous summer maximum number
and at low temperatures, they experience con- of colonies may settle to the sediments, ≤10%
siderable mortalities, although some cells live might contribute to the re-establishment of a
on under these conditions, apparently for several summer population the following year (Preston
years in some cases (see Livingstone and Cambray, et al., 1980; Brunberg and Blomqvist, 2003;
1978). The surviving cells function at a very low Ishikawa et al., 2003).
metabolism and are tolerant of sediment anoxia It may be noted that Microcystis colonies also
(and consequent re-solution of the attached iron) survive in microenvironments created by down-
but there are, by now, too few of them to lift the wind accumulations of surface scums on large
erstwhile colonial matrix back into the water col- lakes and reservoirs, especially where warm sum-
umn. mers, high energy inputs and high upstream
Reinvasion of the water column follows a nutrient loadings are simultaneously prevalent.
phase of in-situ cell division, in which clusters of Good examples come from the reservoirs of the
young cells are formed, constituting a pustule- Dnieper cascade (Sirenko, 1972) and the Hart-
like structure that buds out of the original, beespoort Dam in South Africa (Zohary and
‘maternal’ mucilage matrix, until it is released Robarts, 1989). The conditions in these thick,
or it escapes into the water. The process was copious ‘crusts’ or ‘hyperscums’ are effectively
described originally by Wesenberg-Lund (1904), lightless and strongly reducing (Zohary and Pais-
but the information was largely ignored. The Madeira, 1990) but, save those actually baked dry
‘nanocytes’ found by Canabeus (1929) and, later, at the surface, Microcystis cells long remain viable
‘rediscovered’ by Pretorius et al. (1977), seem to and capable of recovering their growth.
refer to the young, budding colonies. Sirenko Many species respond to the fabled ‘onset
(pesonal communication quoted in Reynolds, of adverse conditions’ by producing morphologi-
1987a) has viewed the entire sequence, claim- cally distinct resting propagules. Among the best
ing that the potential mother cells are identi- known are the cysts of dinoflagellates, which
fiable in advance by their larger size and more are sufficiently robust to persist as a fossils of
intense chlorophyll fluorescence. The process has palaeontological significance (for a review, see
also been reproduced under controlled condi- Dale, 2001). Some 10% of the 2000 or so marine
tions in the laboratory (Cáceres and Reynolds, species are known to produce resting cysts. In
1984), using material sampled from autumnal some instances, they are known, or are believed,
sediment. It requires the exceedence of a tem- to be sexually produced hypnozygotes. The cell
perature and insolation threshold and it occurs walls in many species contain a heavy and com-
more rapidly while anaerobic conditions perplex organic substance called dinosporin, chem-
sist. These conditions have to be mirrored in ically similar to sporopollenin of higher-plant
natural lakes of the temperate regions before pollen grains. Some species deposit calcite. In
Microcystis colonies begin to be recruited to the the laboratory, cyst formation may, indeed, be
water column in the spring. Sediments have to induced by nutrient deprivation and adverse
retain colonies through the winter period, where conditions but the regular, annual formation
of cysts in nature (coastal waters, eutrophic tal adequacy but which are tenanted briefly by
lakes) possibly occurs in response to cues that vegetative populations.
anticipate ‘adverse’ conditions rather than the In contrast, nostocalean Cyanobacteria pro-
actual onset of those adversities. The protoplasts duce their asexual akinetes in rapid response
of newly formed cysts usually contain conspicu- to the onset of physiological stress. Akinetes are
ous reserves of lipid and carbohydrate, accumu- the well-known ‘resting stages’ of such genera as
lated during stationary growth (Chapman et al., Anabaena, Aphanizomenon and Gloeotrichia (Roelofs
1980). The number of cysts produced by fresh- and Oglesby, 1970; Wildman et al., 1975; Rother
water Ceratium hirundinella in autumn has been and Fay, 1977; Cmiech et al., 1984). These, too,
estimated from the sedimentary flux to account have typically thickened external walls, within
for ≤35% of the maximum standing crop of vege- which the protoplast remains viable for many
tative cells (Reynolds et al., 1983b). The success in years. Livingstone and Jaworski (1980) germinated
recruiting vegetative cells from excysting propag- akinetes of Anabaena from sediments confidently
ules in the following spring is, in part, propor- dated to have been laid down 64 years previously.
tional to the abundance of spores retained from On the other hand, rapid akinete production has
the previous year (Reynolds, 1978d; Heaney et al., been stimulated in the laboratory by the sort of
1981). carbon : nitrogen imbalance that occurs as a con-
The excystment of vegetative cells from cysts sequence of surface blooming, and from which
was first described by Huber and Nipkow (1922). conditions an effective means of escape is offered
Much detail has been added from such land- (Rother and Fay, 1979). Moreover, substantial ger-
mark micrographic investigations as those of mination can take place shortly (days rather than
Wall and Dale (1968) and Chapman et al. (1981). A months or years) after akinete formation, pro-
naked flagellate cell, or gymnoceratium, emerges vided the external conditions (temperature, light
through an exit slit and soon acquires the distinc- and, possibly, nutrients) are suitable (Rother and
tive thecal plates of the vegetative cell. Heaney et Fay, 1977). Reynolds (1972) observed that Anabaena
al. (1981) noted a sharp, late-winter recruitment akinetes were regularly resuspended by wind
of new, vegetative cells of Ceratium to the plank- action in a shallow lake but failed to germi-
ton of Esthwaite Water, UK, after the water tem- nate before a temperature or insolation thresh-
perature exceeded 5 ◦ C, and coincident with an old had been surpassed. In other years, vegetative
abrupt increase in the proportion of the empty filaments surviving the winter were sufficient to
cysts recoverable from the bottom sediments of explain the growth in the following season. These
the lake. thresholds could be important to the distribu-
Among the Volvocales, sexually produced tions of individual species. The current spread
zygotes of (e.g.) Eudorina (Reynolds et al., 1982a) of Cylindrospermopsis raciborskii from the tropics
and Volvox (Reynolds, 1983b) have the robust to continental lakes in the warm temperate belt
appearance of resting cysts and, indeed, serve may be delimited by a germination threshold
as perennating propagules between population temperature of 22 ◦ C (Padisák, 1997). The akinetes
maxima. Deteriorating environmental conditions of Gloeotrichia echinulata are able to take up phos-
may trigger the onset of gametogenesis but for- phate through their walls and colonies germi-
mation of the eventual resting stages cannot be nating the following year can sustain substantial
claimed certainly to have been consequential on growth even when limnetic supplies are small
resource starvation. Among the Chrysophyceae, (Istvánovics et al., 1993).
there has evolved an opportunistic perennation As suggested above, regenerative strategies are
strategy, involving zygotic and asexual cysts that not uniform among the phytoplankton, neither
are produced early in the growth cycle, when con- is the production of spores and resting stages
ditions are supposedly good (Sandgren, 1988b). exclusively brought on by ‘adverse conditions’.
This pattern of encystment apparently ensures However, the existence of resting propagules of
the production of resting stages during what a given species are likely tolerant of more severe
often turn out to be short phases of environmen- conditions than vegetative cells and they do
GROWTH OF PHYTOPLANKTON IN NATURAL ENVIRONMENTS 217
increase the probability of survival through diffi- lary of this is that attestably rapid phases of pop-
cult times and also perhaps raise the scale of the ulation increase, independent of recruitment by
infective inoculum when favourable conditions importation from horizontally adjacent patches
return. or from germinating resting stages, are indicative
of yet higher simultaneous rates of cell replica-
tion.
5.5 Growth of phytoplankton in
natural environments Growth rates from episodic events
Generically, these accumulative phases fall into
two categories. One of these is the annually recur-
The rates of cell replication and population
rent and broadly reproducible event, such as the
growth that are achieved in natural habitats have
spring increase of phytoplankton in temperate
long been regarded as being difficult to deter-
waters, in response to strong seasonally varying
mine. This is primarily due to the fact that
conditions of insolation (see Section 5.5.2). The
what is observable is, at best, a changing den-
second is the stochastic event, when, perhaps, a
sity of population, expressed as species rate of
sharp change in the weather, resulting in the
increase (rn , in Eq. 5.2) which falls short of the
fortuitous stagnation of a eutrophic water col-
rate of cell replication because of unquantified
umn, or the relaxation from coastal upwelling,
dynamic losses of whole cells sustained simul-
or the deepening of a nutrient-depleted mixed
taneously. The net rate of change can be nega-
layer with the entrainment of nutrient-rich met-
tive (−rn ) without necessarily signifying that true
alimnetic water, or some abrupt consumer fail-
growth has failed, merely that the magnitude of
ure through herbivore mortality, leads to the
rL , the rate of loss noted in Eq. (5.3), exceeds
realisation of potential respondent growth. In
that of replication, r . The problem of patchi-
this second category, the phases of increase
ness and advection (Section 2.7.2) provides the
may be brief and sensing them, accurately and
further complication of compounded sampling
with reasonable precision, requires the close-
errors, in which even the observed rate of popula-
interval sampling of well-delimited populations.
tion change (±rn ) may prove an inadequate base.
The study of in-situ increase rates of phytoplank-
From the other direction, the true replication
ton in Bodensee (Lake of Constance), assembled
rate cannot be estimated from measurable photo-
by Sommer (1981), was one that satisfied these
synthetic or nutrient-uptake capacities, unless it
conditions. The research based on the large (1630
can be assumed with confidence that the actual
m2 ), limnetic enclosures in Blelham Tarn, English
rate of growth is constrained by the capacity fac-
Lake District (variously also referred to as ‘Blel-
tor concerned.
ham Tubes’, ‘Lund Tubes’ (Fig. 5.11), being iso-
There are ways around these problems and
lated water columns of ∼12–13.5 m in depth and
there are now several quite reliable, if somewhat
including the bottom sediment from the lake;
cumbersome, methods for estimating growth
for more details, see Lund and Reynolds (1982),
rates in situ. Some of these approaches are high-
carried out in the period 1970–84, has similarly
lighted below, through the development of an
provided many insights into phytoplankton pop-
overview of dynamic trait selection in natural
ulation dynamics. Examples of specific increase
habitats.
rates noted from either location are included in
Table 5.4.
5.5.1 Estimating growth from observations The evident interspecific differences are
of natural populations partly attributable to the time period of obser-
On the same basis that replication rates cannot vation, and the seasonal changes in water tem-
be sustained at a faster rate than cell division can perature and in the insolation attributable to
be resourced, it is clear that the observable rates seasonally shifting day length and vertical mix-
of population increase cannot exceed the rates of ing. In some instances, these environmental vari-
recruitment through cell replication. The corol- ations are reflected in intraspecific variability in
Figure 5.11 The Blelham

Enclosures: the positions of the
three butylite cylinders (A, B and C),
each measuring about 1630 m2 of
water surface and a similar area of
bottom sediment, are shown in
relation to the bathymetry of the
lake. The line X–Y was set up as a
permanent transect with shallow
(‘S’) and (‘D’) sampling stations. For
further details, see Lund and
Reynolds (1982).
increase rate. Where species are common to both physiological activity of numerically scarce phy-
locations, maximal performances are similar; toplankters are needed to answer this question.
species are either fast-growing or slow-growing One of the best-known and most precise tech-
in either location. niques for estimating the species-specific growth
The observed rates of increase are also plausi- of sub-dominant populations is to estimate the
ble in terms of the dynamic behaviours of the frequency of dividing cells. This works best for
algae in culture. Allowing for winter tempera- algae whose division is phased (i.e. it occurs at
tures and short days, only half of which might certain times of day or night) and it may need
be passed in the photic zone, a vernal growth close-interval sampling (every 1–2 h) of the field
rate of 0.15 d−1 for Asterionella is perfectly explica- population. It works especially well with algae
ble. For small, unicellular species such as Ankyra (e.g. desmids, dinoflagellates, coccolithophorids)
and Plagioselmis to be able to double the popu- that have complex external architecture which
lation at least once per day in summer (when has to be reproduced at each division and often
they can manage it twice in grazer-free, contin- requires several hours to complete. Then the
uously illuminated culture) also seems to be a numbers of cells before and after the division
reasonable observation. The growth-rate perfor- phase is increased by a number that should agree
mances of the bloom-forming Cyanobacteria and with, or be within, the increment deduced from
the dinoflagellate Ceratium are about half those the frequency of dividing cells. Pollingher and
noted in culture at 20 ◦ C (cf. Table 5.1). Serruya (1976) gave details of the application
of this method to the seasonal increase of the
Frequency of cell division dominant dinoflagellate in Lake Kinneret, now
Relatively rapid growth rates, sustained over the called Peridinium gatunense. During the period of
equivalent of several cell divisions, lead assuredly its increase (usually February to May), the num-
to the establishment of populations making up ber of cells in division on any one occasion was
a significant part of the biomass, if not actu- found to be variable between 1% and 40%. They
ally coming to dominate it. It is equally proba- showed that the variability was closely related
ble that the same species may be relatively inac- to wind speed. While daily average wind veloci-
tive for the quite long periods of their scarcity. ties exceeded 8 m s−1 , the frequency of dividing
Is their increase prevented by lack of light, or cells (FDC) was always <10%. This accelerated to
lack of resources, or losses to grazers, parasites 30–40% during the spring period of weak winds
or to the consignment to the depths? Obviously, (and, hence, weak vertical advection) averaging
more precise means of investigating the in-situ <3 m s−1 . Successful recruitment of new cells
Table 5.4 Some maximal in-situ rates of increase (rn d−1 ) of some species of freshwater phytoplankton reported from
Bodensee (Sommer, 1981) and from large limnetic enclosures in Blelham Tarn (Reynolds et al., 1982a; Reynolds, 1986b,
1998b), together with some reconstructed rates of replication (r ) where available (see text)
Blelham Enclosure
Bodensee
Phasea (rn ) (rn ) (r )
Ankyra judayi MS 0.86 1.09
Plagioselmis nannoplanctica V 0.17
ES 0.56 0.71
Cryptomonas ovata V 0.15
ES 0.46
MS 0.89 0.61 0.62
Dinobryon spp. ES, MS 0.45 0.27
Eudorina unicocca ES 0.43 0.48
Pandorina morum MS 0.52
Coenochloris fottii ES, MS 0.64 0.43
Peridinium cinctum MS 0.18 0.16
Ceratium hirundinella MS 0.17 0.13
Anabaena flos-aquae MS 0.41 0.34 0.34
Aphanizomenon flos-aquae MS 0.43
Microcystis aeruginosa MS 0.24 0.24
Planktotrix mougeotii V 0.06
ES, MS 0.33 0.33
Asterionella formosa V 0.15 0.24
MS 0.36 0.34 0.50
Fragilaria crotonensis V 0.10
MS 0.27 0.24 0.58
LS 0.10
Aulacoseira granulata MS 0.43
Closterium aciculare LS 0.18
Staurastrum pingue LS 0.13 0.28
a
‘Phase’ refers to part of the year: V, vernal, or spring bloom period, temperatures 5 ± 2 ◦ C; ES,
early-stratification phase (temperature 11 ± 3 ◦ C); MS, mid-stratified (temperature 17 ± 3 ◦ C); LS,
◦
late-stratified period (temperature generally 12 ± 2 C).
divided off at this rate yields maximum rates of reservoir near Madrid to be between 0.13 and
population increase equivalent to 0.26–0.34 d−1 . 0.16 d−1 . More recently, Tsujimura (2003) has
The typical net rate of population increase of Peri- estimated in-situ growth rates from FDC among
dinium in Kinneret over a sequence of 20 con- cell suspensions of Microcystis aeruginosa and M.
secutive years was found to be 0.22 ± 0.03 d−1 wesenbergii from Biwa-Ko (prepared by ultrasoni-
(Berman et al., 1992). cation of field-sampled colonies). In both species,
Heller (1977) and Frempong (1984) estimated the frequency of diving cells varied between 10%
FDC of Ceratium in Esthwaite Water to be between and 15% in offshore stations and between 15%
2% and 10%, occasionally 15%, sufficient to and 40% at inshore stations, with the average
explain in-situ seasonal increase rates of 0.09–0.14 duration of cytokinesis varying from 25 h to
d−1 . Alvarez Cobelas et al. (1988) estimated growth 3–6 h. Growth rates of 0.34 d−1 thus appear
rates from afternoon FDC peaks in the popula- sustainable in the near-shore harbour areas of
tion of Staurastrum longiradiatum in a eutrophic Biwa-Ko, whence they are liable to become more
widely distributed in the circulation of the lake 1996b); just the case of Asterionella formosa in Blel-
(Ishikawa et al., 2002). ham Enclosure B in 1978 is highlighted here.
The inflatable collar of the enclosure was
Frequency of nuclear division lifted on 2 March of that year, isolating part of
For phytoplankton species that are less amenable the lake population comprised almost wholly of
to the tracking of cell division, the principle may Asterionella, then itself already actively increasing,
be extended to the monitoring the frequency at a concentration of ∼630 cells mL−1 ). Over the
of karyokinesis (Braunwarth and Sommer, 1985). next 19 days, the population increased exponen-
The success of this method relies on the good tially at an average rate of 0.147 d−1 , then more
fixation of field samples followed by careful slowly to its eventual maximum (a total of 24,780
staining with the DNA-specific fluorchrome, 4,6- cells mL−1 , though by then including 1950 cells
diamidino-2-phenylindole (DAPI) (Coleman, 1980; mL−1 judged to be dead or moribund) on 4 April.
Porter and Feig, 1980). This precise and sensi- The decline in the standing population was slow
tive method has been applied to a natural Cryp- at first but accelerated enormously as warmer
tomonas population (Ojala and Jones, 1993), the and sunnier weather mediated the thermal strat-
results being broadly predictable on the basis of ification of the Tarn and the enclosure. Nutrients
growth rates under culture conditions. Like any were added to the enclosure each week, by disper-
methods based upon the events in the cell cycle, sal into solution across the enclosure surface, in
its prospects for measuring replication accurately measured doses respectively designed to restore
are high (cf. Chang and Carpenter, 1994). the levels of available resource to 300 µg N, 20 µg
There is also keen interest in sensing the DNA P, 100 µg TFe and 1000 µg SiO2 (466.7 µg Si) L−1 .
replication itself. Since the groundbreaking study None of these fell to growth-limiting thresholds.
of Dortch et al. (1983), microbial ecolologists have However, there was no artificial relief for either
been debating the validity of DNA to cell carbon high pH or probable carbon limitation. The con-
as an index of the rate of DNA replication. As an sumption of silicon was calculated as the sum
indicator of the capacity for protein synthesis, of the observed decline in the initial concentra-
the RNA : DNA ratio is already in use as a barom- tion on 2 March, aggregated with the Si added,
eter of the cell growth cycle in marine flagellates and averaged out across the whole volume. The
(Carpenter and Chang, 1988; Chang and Carpen- conversion to Asterionella cells between additions
ter, 1990) and bacteria (Kemp et al., 1993; Kerkhof was calculated using contemporaneous routine
and Ward, 1993). measurements of the Si content of cells sampled
from the growing population (consistently within
Growth from the depletion of resource the range 51.8–61.1 pg Si cell−1 ). As there was
In contrast to monitoring growth-cycle indica- no other significant diatom ‘sink’ at the time,
tors, methods for reconstructing growth rates the consumption was assumed to be equal to
from resource consumption are unsophisticated its deposition in new Asterionella frustules. The
in approach and notoriously imprecise. However, rate of growth, r (Si) , was approximated from the
methods invoking uptake of resources deployed numbers of new cells that the observed silicon
in specific structures, such as silicon for diatom depletion could have sponsored. Estimates were
frustules (Reynolds, 1986a) or sulphur for pro- comparable with the observed rates of increase
tein synthesis (Cuhel and Lean, 1987a, b), offer (rn ) and with the rates of growth reconstructed
more promise. Reynolds and Wiseman (1982) by correcting for the simultaneous loss processes
were able to combine the advantages of the spa- (discussed fully in Chapter 6). Simultaneous sink-
tial constraints of enclosure, offered by the Blel- ing losses were ‘monitored’ in two ways: using
ham tube, with frequent serial sampling of the the flux of settling cells into sediment traps
plankton and careful accounting of the amounts placed near the bottom of the enclosure; and
of replenishing sodium silicate, in order to mea- using a technique of coring and subsampling
sure the true growth rate of a diatom population. the semi-liquid superficial deposit (see Reynolds,
Several datasets were collected and presented 1979a). The possible losses to grazers were esti-
(partly also in Reynolds et al., 1982a; Reynolds, mated from contemporaneous measurements of
Table 5.5 Comparison of the observed rates of increase in a natural population of Asterionella in a large limnetic
enclosure with the growth rate as estimated by silicon uptake. The difference is equated with the rate of loss of live cells
(rL ) and which may, itelf be compared with simultaneous observations of settling rates (rs ) and estimates of the rates of
loss to grazers (rg ) and to death (rd )
2 Mar–21 Mar 21 Mar–4 Apr 4 Apr–25 Apr 25 Apr–15

May
obs rn 0.147 0.065 −0.032 −0.242
calc rSi 0.154 0.072 0.012 0
calc rL 0.007 0.007 0.044 0.242
obs rs 0.009–0.027 0.016–0.021 0.032–0.126 0.150–0.189
est rg 0.001–0.009 0–0.011 0.001–0.035 0.002–0.040
est rd <0.001 <0.002 <0.031 <0.064
calc r 0.157–0.184 0.081–0.099 0.001–0.160 0–0.051
population, it is possible to calculate a budget for

Table 5.6 Comparison of the observed biomass (as
cells per unit area) of Asterionella in Enblosure B,
the entire phase of production. These are entered
Spring 1978, with reconstructed production (from in Table 5.6, but are calculated for the full water
silicon uptake), the total eventually sedimented, the column (mean depth, 11.0 m). The entries are
intercepted flux and the estimated loss to grazers at subject to wide margins of error but the agree-
the time ment is encouraging: the lower ends of the con-
fidence intervals are mutually compatible. From
Cells (10−9 m−2 ) the aggregate loss of silicon from solution (equiv-
Maximum standing crop 272.6 (±19.9) alent to 18.788 g Si m−2 ), the number of Asteri-
Production from Si 335.0 (±27.7) onella cells that could have been produced was
uptake estimated to have been between 307.3 and 362.7
Estimated consumption 10.35 (±9.10) × 109 cells m−2 . Had they all been in suspen-
by grazers sion simultaneously, the concentration would
Sedimentary flux 341.2 (±12.3) have been between 27.94–32.97 × 103 mL−1 (cf.
Recruitment to 319.0 (±58.0) observed maximum, 24.78 × 103 mL−1 ; 272.6 ×
sediments 109 cells m−2 ). The numbers that were recruited
to the sediment, the settling flux that was inter-
Source: Data of Reynolds and Wiseman cepted and the range of fatalities to grazing are
(1982) and Reynolds et al. (1982a), as each inserted in Table 5.6.
presented in Reynolds (1996b).
filter-feeding by zooplankton (chiefly of Daphnia 5.5.2 The spring increase in temperate

galeata, though activity was minimal until late lakes: the case of Windermere
April: Thompson et al., 1982). It is appropriate now to consider an example
The various outcomes are tabulated and com- of the role of growth in the development of
pared in Table 5.5. At first, almost all the silicon phytoplankton maximum, taking the case of a
investment is realised in observable additions familiar and annually recurrent event, observed
to the extant population. Moreover the slowing over a sequence of several consecutive years and
rate of increase is explained almost wholly by a responding to a substantial degree of interan-
declining rate of replication, for whatever reason nual environmental variability. The example of
this might be. However, the rates of loss mount the Asterionella-dominated spring increase in Win-
and rates of recruitment through cell division dermere is chosen for the length and detail of
slow, until the former exceed the latter and the the observational record and because the year-
population goes into decline. In this particular to-year variations in the dynamics of growth
Figure 5.12 The seasonal change

in the standing population of live
Asterionella cells in the upper 7 m of
the South Basin of Windermere
(heavy line), averaged over the
period 1946 to 1990 inclusive,
together with the envelope of 95%
confidence (shown by hatching) and
the maximal and minimal values
recorded (the lighter lines). Modified
from an original figure in Maberly et
al. (1994) and redrawn with
and population increase (and subsequent decline) sures, commenced in 1991, see Reynolds and
have been intensively analysed. Windermere is a Irish, 2000). By the late 1980s, autumn–winter
glacial ribbon lake in the English Lake District, concentrations of MRP increased from a pre-1965
covering 14.76 km2 . It is not far short from being average of 2–2.5 µg P L−1 (0.06–0.08 µM) to ∼8 µg
two lakes, a shallow morainic infill separating P L−1 in the North Basin and to about 30 µg
the lake into two distinct but contiguous basins. P L−1 in the South Basin. In both basins, autumn–
The North Basin holds nearly two-thirds of the winter DIN levels more or less doubled over the
total volume (314.5 × 106 m3 ) and has a mean same period, from ∼350 to ∼700 µg N L−1 (25–
depth of 25.1 m (maximum is 64 m); the mean 50 µM), but SRSi levels have remained steady (at
depth of the smaller South Basin is 16.8 m (max- 0.9–1.1 mg Si L−1 ; 32–39 µM; 1.9–2.3 mg SiO2 L−1 ).
imum 42 m) (data of Ramsbottom, 1976). The This information permits us to establish the
southward water flow through the lake is gen- physical–chemical characters of the habitat of
erated mainly from the catchment in the central the spring bloom. It is a relatively clear (ε wmin
uplands of the Lake District. The mean annual ∼ 0.31 m−1 ) – and soft(alkalinity <0.26 meq L−1 )
discharge from Windermere (437 × 106 m3 a−1 ) – water, barely mesotrophic lake, experiencing
corresponds to a theoretical mean retention time a mostly cool, temperate, oceanic climate, but
of 0.72 a. The upper catchment is based upon incurring, between 1965 and 1991, substantial
hard and unyielding volcanic rocks, the lower anthropogenic enrichment. For almost 200 years,
foothills comprising younger and slightly softer Asterionella has been a conspicuous member of
Silurian slates. Both are poor in bases. Thin soils, the plankton in both basins and, during at least
mostly cleared of the natural woodland covering the last 60 of those, the almost unchallenged
and replaced with rather poor, leached grazing dominant species of the spring-bloom period.
land, contribute little in the way of bases or nutri- Maberly et al. (1994) carried out a thorough
ents to the lake. Nutrient loads have increased statistical analysis of the (then) complete run of
substantially over the last 50 years or so, through data collected from (mostly) weekly samplings of
the increased use of inorganic fertilisers on the the South Basin, initiated by J. W. G. Lund in
land. The sheep population that the added fer- 1946 (see Lund, 1949) and maintained, with only
tilty supports has roughly doubled over the same detailed methodological variation, over the 45
period. However, it is the increase in the human years to 1990. A simplified version of their main
population, together with the introduction (in figure is reproduced here, as Fig. 5.12, to illus-
1965) of secondary sewage treatment, which has trate the reproducibility of the main features of
most affected the phytoplankton-carrying capac- the development. The heavy line represents the
ity of Windermere (for full details, reconstructed 45-year mean of the standing population (on a
loadings and the effects of the ‘restoration’ mea- logarithmic scale) as a function of the day of
the year. The narrowness of the 95% confidence general trends over the full period are towards
interval (shown by hatching) attests to the strong smaller overwintering inocula but faster rates of
interannual comparability of the growth, even exponential rise in the spring. Of particular inter-
though the boundaries of the extreme records est is the fact that, although the size of the max-
over the 45 years (delimited by the lines either imum crop seems not to have increased over the
side of the mean plot) cover 2 or more orders 45 years, the date of its attainment has tended to
of magnitude of variation. Maberly et al. (1994) be reached earlier in the year, as a consequence of
diagnosed several cardinal characteristics of the the trend towards sustained faster growth rates.
growth curve, including the size of the extant The maxima of Asterionella in the North Basin
population at the beginning of the year (mean of Windermere have been similar in magnitude
9.7 (×/÷ 3.85) cells mL−1 ; range 0.6–330 mL−1 ); to those observed in the South Basin in the cor-
the maximum (mean 3940 (×/÷ 2.28) cells mL−1 responding years but, typically, have always been
range 330–11500 mL−1 ), the date of its achive- reached one to two weeks later. The greater mean
ment (day 124 ± 16.7 (17 April–21 May)), the start depth of the North Basin appeared to be the
(day 52 ± 24 (28 January–17 March)) and end most likely reason for this relative delay. However,
of the period of rapid exponential increase (day the maxima here are also now reached signifi-
106 ± 17 (30 March–3 May)); the mean rate of cantly earlier (∼10 d) in the year. The proximal
increase achieved (0.0925 (± 0.0357) d−1 ), as well explanation is, again, that a faster average rate
as the date of the commencement of the steep of increase is maintained but why this should
post-maximum exponential decrease (day 142 ± have changed when light is alleged to be the rate-
16 (6 May–9 June)). regulating factor is not immediately apparent.
The source of the bloom is essentially the Circumstantial evidence and some simple
standing stock in the water at the turn of the modelling reveal a complex factor interaction at
year (Lund, 1949; Reynolds and Irish, 2000). Inter- work. Starting either from the premise of sus-
annual variations in this survivor stock (mean tainable growth rates (Reynolds, 1990) or photo-
6.6 (×/÷ 4.83) cells mL−1 ) are influenced by the synthetic behaviour (Neale et al., 1991b), it may be
size of the late summer maximum of the pre- shown that the long-term average of the under-
vious year and the extent of its net dilution by water light integral, I∗ , sets much lower con-
autumnal wash-out. There is a modest increase in straints on Asterionella growth than does water
the standing crop detectable throughout the first temperature, SRSi or SRP concentration. The time
few weeks of the year, so there is, in no sense, track of I∗ in Fig. 5.13 is reconstructed from long-
any part of the year when growth is not possi- term (42-year) records of daily integrals of irra-
ble. This is close to, or just prior to, the time diance (I0 ) and wind run, and is hypothesised to
of greatest nutrient availability in Windermere, express the intergal daylight period experienced
so the restriction on biomass increase has long by entrained algae. Comparison of the modelled
been supposed to be physical. The lowest water growth rates, r, as determined by each of the con-
temperatures are generally encountered in late straining factors in turn, shows that, initially, the
January (days 14–28, weeks 3 or 4, of the year) least are those set by I∗ . They move from about
but usually remain <7 ◦ C until week 12 (day 84 0.04 to 0.08 d−1 during the first 50 days (7 weeks)
and several weeks after the inception of the main of the year, before accelerating up to 0.21 d−1 by
exponential increase phase). On the other hand, week 15 (days 98–105). Depending upon the size
Talling’s (1957c; see Section 3.3.3) extrapolations of the starting inoculum, this is sufficient to take
of column-integrated photosynthetic production the number of formed cells into the range 103 to
in the lake show that light, especially in a fully 104 cells mL− 1 and incipient Si exhaustion. The
mixed water column, strongly indicate that light supportive capacity of the initially available sili-
is the main growth-regulating factor. The early con is 14–19 × 103 cells mL−1 , if it is assumed that
growth of Asterionella in Windermere has been SRSi is drawn down to extinction and shared at
observed to be relatively weak in the stormiest 60–65 pg Si cell−1 produced. Only for the purpose
winters predominated by south-westerly winds of modelling is it assumed that there is no con-
and stronger in anticyclonic winters. However, comitant loss of cells from suspension, or that
there is no demand for silicon from any other

agency. In fact, observable populations of 9–10 ×
103 Asterionella cells mL−1 will not have formed
without using most of the available SRSi, at least
down to the critical half-saturation level of 23 µM
(see Section 5.4.4).
The key deduction is that, prior to 1965, the
initially available phosphorus (say 2.5 µg P L−1 )
would have had to have been shared among a
maximal population of 10 × 103 cells mL−1 , each
having a mean residual quota of ≤2.5 pg cell−1 .
By halving the quota again, the next cell divi-
sion will submit the growth rate to P-limitation
(see Section 5.4.4). It follows that the diatom-
supportive capacity of Windermere is, indeed,
set by the available silicon. However, the rate of
its assimilation and conversion into Asterionella
biomass is closely regulated first by the light
income availability but, as soon as this begins
to be relieved by lengthening days and weaker
vertical mixing, phosphorus availability starts to
squeeze the attainable growth rate instead. The
final twist in this changing factor interaction
comes with the recent phosphorus enrichment
of the lake. As first noted by Talling and Heaney
(1988), enrichment with phosphorus relieves the
growth-rate constraint, which now continues to
the limits permitted by I∗ , until the ceiling of
silicon exhaustion is reached, now rather earlier
in the year. The sketches in Fig. 5.14 summarise
the shifting date of maximal attainment. The sili-
Figure 5.13 Features of the environment of the North con limit remains inviolable. Interactions among
Basin of Windermere, during the first half of the year, interannualy varying constraints have influenced
influencing the spring increase of Asterionella. The top panel the extent of capacity attainment. Over a num-
shows the water temperature (θ ◦ C), averaged for each week
ber of years of enrichment, it became the case
of each of 43 consecutive years (1946 to 1988 inclusive, with
that the Asterionella maximum failed to exhaust
the 95% confidence interval). The second panel shows similar
averages for I0 , the insolation at the water surface, and for I∗ , all the phosphorus in solution, instead leaving it
the average light level in the mixed layer. Typical (not available to be exploited by other phytoplankton.
statistically averaged) changes in the content of SRSi (soluble As will be argued later (Section 8.3.2), this is a
reactive silicon) are shown in the third panel, while the fourth defining stage in the eutrophication of temper-
reflects long-term increases in SRP (soluble reactive ate lakes.
phosphorus) levels by providing typical trend lines for two There is much more to this story, and to the
separate periods (1945–65 and 1980–85). The lower panel next one about the rapid post-maximum collapse
shows plots of modelled growth rate capacities of θ , I∗ , SRSi
of the spring bloom. Both concern the magnitude
and SRP. The sustainable growth rate is that of the limiting
of the loss terms making up rL in Eq. (5.3), discus-
component, which, clearly, is I∗ until week 16 or 17, when
diminishing SRP or SRSi levels become critical. Revised from sion of which is developed in Chapter 6. All the
Reynolds (1990) and redrawn with permission from Reynolds time that the population is increasing, the repli-
(1997a). cation rate is sustaining losses of formed cells
to mortality – to such physical agents as out-
wash and settling beyond the resurrecting limits
behaviour of other kinds of phytoplankton. The

main illustrative point is that species have to
make the best of the environments in which they
find themselves and, often, they must press their
specific traits and adaptations to perform ade-
quately under environmental conditions verging
on the hostile. In this way, we might interpret
the ability of Asterionella to dominate the ver-
nal plankton of Windermere as being dependent
upon certain attributes. The first is so obvious
that it is easily overlooked: it is there! The suc-
cess of a species in a habitat is a statement
that the habitat is able to fulfil its fundamen-
tal survival requirements and, of the species that
have arrived there, this species will be, relatively,
the most efficient in exploiting the opportunity
offered. Absence of a species does not inform a
deduction that the habitat is not suitable; it may
just not have had the opportunity to grow there.
No species dominates a habitat just because the-
ory argues it to be the most suitable. However,
to be able to to outperform other species at a
given point in space and in time must suggest a
favourable combination of inoculum and a rela-
Figure 5.14 Reynolds’ (1997b) proposed explanation for tively superior exploitative efficiency under the
the advancing date of the Asterionella maximum in conditions obtaining.
Windermere. (a) On the plot of accelerating standing crop, The fossil record shows that Asterionella domi-
the ceiling capacity of the silicon resource K(Si) occurs only a nance is a relatively new phenomenon in Winder-
little short of the point where the supportive capacity of the mere, having occurred only since the nineteenth
phosphorus would have been exhausted, although recent century (Pennington, 1943, 1947). Previously,
phosphorus enrichment has raised this ceiling further. Thus in
Cyclotella species had dominated a more olig-
(b), the likelihood of a phosphorus-limited growth rate is
otrophic period in the lake’s history (Haworth,
delayed and relatively fast growth rates can be maintained for
longer. The silicon ceiling, which has remained stable, is thus 1976). Asterionella is able to grow faster than other
reached sooner. Changing nutrient availability may be cited diatoms under the poor vernal underwater light
for the advance which may have little or nothing to do with conditions and faster than Planktothrix at low
climate change. Redrawn with permission from Reynolds temperatures. It also manages to carry over sub-
(1997b). stantial winter populations from which spring
growths can expand. Asterionella does not have
of entrainment and to the biological demands
matters exclusively to itself – Aulacoseira species
of grazing and parasitic consumers. The spring
(A. subarctica, A. islandica) overwinter well, though
bloom in Windermere, as elsewhere, is sustained
they grow less rapidly than Asterionella; flagellates
to within the limits that light and available nutri-
such as Plagioselmis grow relatively well in win-
ents can support, net of ongoing sinks, recognis-
ter anticyclones (with frosts, sunshine and weak
ing that the latter may be more or less critical in
wind-driven vertical mixing) (Reynolds and Irish,
commuting the size of the maximum crop below
2000).
that of the chemical capacity.
Small interannual variations in these environ-
5.5.3 Selection by performance mental features may not make decisive intrasys-
The example of Asterionella in Windermere may, tem differences in outcomes but they may assist
or may not, have direct analogues to other us to understand the differences in timing, the
diatom blooms in other systems or even to the magnitude of crops and the species dominance of
populations elsewhere. We have shown, through dence between 18 and 20 ◦ C. There also seems
the comparison of growth responses and their to be a common threshold light exposure (∼4 h
sensitivities to environmental deficiencies, how d−1 on the analogue scale) applying at all tem-
the dynamic performances differ among species peratures. The information sits comfortably with
in experiments. Can we now discern differences our understanding of growth sensitivites in the
of performance in nature that will confirm – laboratory. It may also be noted that the repli-
or help us to recognise – the traits that select cation rates in the field do not differ widely
for some species and against others at a given from the maximum resource- and light-saturated
location? If so, how much does this tell us about rates observed in culture at 20 ◦ C, if the appro-
the ways in which natural communities are put priate adjustments for temperature and photope-
together and shape trophic relationships? riod are applied.
The answers to these questions are clouded Analogous deductions can be drawn from the
by the usual problems of accurate measure- data for the other algae represented in Fig. 5.15.
ment of population dynamics in the field (see The ability of another R-strategist alga, Asteri-
Section 5.5.1). Work with captive wild popula- onella, to adapt to function on low average insola-
tions of phytoplankton in the Blelham enclo- tion is confirmed by observed growth rates of up
sures, growing within a defined space, subject to 0.15 d−1 on a low aggregate daily photoperiods
to well-characterised and, in part, artificially con- (≤4 h d−1 on the contrived scale) at temperatures
trolled conditions, subject to separately quanti- from 5 to 15 ◦ C. Collected data for Cryptomonas
fied loss rates of cell loss and, above all, sam- spp. (mostly C. ovata) and Planktothrix mougeotii
pled at high frequencies (3–4 days), does provide also confirm that growth rates are maintained
some insights. From data collected from numer- by photoadaptation to low aggregate insolation
ous growth phases, observed in three enclosures (with thresholds of 1–3 h d−1 ) but they are not
over 6 years, Reynolds (1986b) assembled a series so fast as the most rapid performances of the
of in-situ replication rates for each of a number diatoms. The most rapid growth rates observed in
of common species. Summaries are shown in Fig. the enclosures have been attained by C-strategists
5.15. Each datum point is calculated from a mini- such as Ankyra, which, on several occasions, has
mum of three serial measurements on an increas- been observed to self-replicate at >0.8 d−1 (dou-
ing population and is corrected for the contem- bling its mass in less than a day!). These short-
poraneous estimates of loss rates to sinking and lived episodes have been possible in warm, clear,
grazing (details to be highlighted in Chapter 6). usually water that is restratifying and supplied
These points were then plotted against a common with nutrients well in excess of growth-rate lim-
scale of analogous insolation, this being the prod- iting concentrations. Lack of carbon, self-shading
uct of the day length (, from sunrise to sunset, or increased vertical mixing contribute to a slow-
in hours) and the ambient ratio of Secchi-disk ing growth rate in these instances: note the
depth to mixed depth (zs /zm , with the proviso apparent threshold at ∼8 h d−1 on the contrived
that solutions >1 are treated as 1). Finally, the scale of daily photoperiod. The plots for the C–S
points are grouped according to the approximate species Eudorina unicocca seem to point to an even
contemporaneous water temperatures. greater photoperiod response, little influenced
The plot does reveal an encouraging level by the (somewhat narrow range of) water tem-
of intraspecific consistency of performance and peratures available. The two strongly S-strategist
significant interspecific differentiation. Taking Cyanobacteria (Anabaena and Microcystis) are gen-
the observations on Fragilaria, for instance, erally slow-growing (<0.36 d−1 ); as a function of
replication rates in the field, between 13 and photoperiod, there is an intermediate threshold
17 ◦ C, reveal a common dependence upon the of 4–6 h d−1 on the artificial scale.
aggregate-by-analogy photoperiod, with a slope Many other observations on the growth per-
that appears steeper than (two) observations formances of phytoplankton have emerged from
applying to temperatures between 9 and 11 ◦ C, yet the studies using the Blelham enclosures. Some
less steep that the indicated photoperiod depen- relate to the dynamics of loss and the way these
Figure 5.15 Approximations of the daily specific growth

interact with differential growth rates in influ-
rates (r = rn + rG + rS ) reconstructed from detailed
encing community assembly and succession, to
observations on the dynamics of populations of named
phytoplankters in Blelham enclosures, plotted against the which reference will be made in subsequent
contemporaneous products of the length of the daylight chapters. Of particular interest to the question
period (, in h) and the ratio of the Secchi-disk depth to the of selection by growth performances is the
mixed depth (hs /hm ), being an analogue of I∗ (in instances collective overview of species-specific devel-
where hs /hm > 1, hs /hm is put equal to 1). Curves are fitted opment in relation to chemical factors. The
to data blocked according to contemporaneous enclosures have been subject to differing levels
temperatures, as stated. The algae are: Anaba, Anabaena of fertilisation, and to variation in the frequency
flos-aquae; Ankyr, Ankyra judayi; Aster, Asterionella formosa;
and the scale of nutrient supplied. Against
Crypt, Cryptomonas ovata; Eudor, Eudorina unicocca; Fragi,
the naturally soft-water, relatively P-deficient
Fragilaria crotonensis; Micro, Microcystis aeruginosa; Plank,
Planktothrix mougeotii. Modified and redrawn with permission water of Blelham Tarn, it is not surprising
from Reynolds (1986b). that manipulations of the phosphorus and the
carbon content of the enclosed water should
have yielded the most satisfying outcomes.
Reynolds (1986b) contrasted the yield of phy-
toplankton, in terms of biomass and species
composition, through six enclosure–seasons
Urogl/Dinob Aster Eudor subject to P loads between 0.3 and 2.5 g P

6 m−2 . Many species occurred in all sequences
but in differing absolute and real proportions.
Some were relatively much more frequent at
0
high rates of P fertility and some were rela-
Coeno Plank Ankyr
6 tively more abundant at low rates. Using more
enclosure-years (making 18 in total) but in lesser
detail and in respect of a small number of repre-
0 sentative species, Reynolds (1986b) summarised
Anaba Crypt Staur the apparent preferences of certain specific
6 ascendencies to reveal the patterns shown in Fig.
5.16. Whereas Asterionella dominated for a time
in every enclosure in every year and the inci-
0
dence of dominant populations of Planktothrix,
Cerat Fragi Micro
6 Cryptomonas, Fragilaria and Microcystis was not
clearly correlated with the eightfold variations
in phosphorus supplied, the growth of Eudo-
0 rina, Ankyra and the desmid Staurastrum pingue
showed preferences for phosphorus richness. On
1.0 2.0 1.0 2.0 1.0 2.0
P loa / g P m−2 the other hand, the chrysophytes Uroglena and
Dinobryon, as well as the colonial chlorophyte
Figure 5.16 The frequency of years (out of 18) in which Coenochloris, conformed to supposition (Table 5.2)
the annual areal load of phosphorus added to a Blelham by flourishing well down the scale of phosphorus
enclosure fell within the ranges stated (0–0.5, 0.5–1.0 g P fertility. Population development of Anabaena
m−2 , etc.) and the number of years in each category during and Ceratium also shows bias towards the less
which the named phytoplankters produced a dominant or
enriched conditions.
large sub-dominant population. Species abbreviations as in Fig.
Following a similar approach, increase phases
5.15, plus: Cerat, Ceratium hirundinella; Coeno, Coenochloris fottii;
Dinob, Dinobryon spp.; Staur, Staurastrum pingue; Urogl, Uroglena of the same group of species were blocked as a
cf. lindii. Redrawn with permission from Reynolds (1986b). function of the pH of the water in which they
grew (see Fig. 5.17). The water in Blelham Tarn
has a rather low bicarbonate alkalinity (<0.4 meq
L−1 ) and, in the enclosures, it is isolated from
Urogl /Dinob 4 Eudor Figure 5.17 The frequencies with

12
which the increase phases of
Aster 8 dominant phytoplankton species in
the Blelham enclosures were halted
2 in the pH ranges indicated (pH
4 Volvo
Plank 7.0–7.5, 7.5–8.0, etc). Species
Ankyr 2
abbreviations as in Figs. 5.15 and
Cerat 2
Coeno 4 5.16, plus: Coeno, Coenochloris fottii.
Anaba 4 Redrawn with permission from
Fragi 4 Reynolds (1986b).
Micro 4
Crypt 8
4
Staur
0 0
7 8 9 10 11 7 8 9 10 11
pH pH
Figure 5.18 Simultaneous

changes in the concentrations of
vegetative cells of the dinoflagellate
Ceratium hirundinella in the Blelham
Enclosures A, B and C, during 1980.
Cyst production, noted in the figure,
generally marks the termination of
vegetative growth. Redrawn with
permission from Lund and Reynolds
(1982).
terrestrial replenishment (Fig. 5.11). Thus, the organism so much as those of them that were
inorganic carbon supply to phytoplankton in there were able to contend effectively with the
these experiments depended upon internal recy- conditions imposed. Yet, in many ways, these rela-
cling, augmented by whatever dissolved from the tive performances do, most probably, distinguish
air at the water surface. Thus, rising pH is a use- sufficiently among the traits and adaptabilities
ful surrogate of carbon deficiency in the enclo- of a number of common types of plankter for
sures (experiments in 1978 attempted to relieve their basic ecological preferences and sensitivi-
the deficiency with additions of bicarbonate; they ties to be recognised and identified in further,
did increase carbon but did not reduce pH). The more focussed testing.
species-specific distributions of histograms in Fig. We see that enrichment with nutrient seems
5.17 peak in range 9.0–9.5, simply because that is to be beneficial to most species, raising the ceil-
also the range reached most frequently during ing of attainable biomass and, in many instances,
the maxima encouraged by fertiliser additions; releasing the growth rate from the restriction
it is not necessarily indicative of any algal pref- of nutrients. This may not be universally so,
erence for this range. However, it is evident that for some of the inherently slow-growing, self-
Fragilaria, Cryptomonas, Eudorina, Ankyra, Ceratium regulating species, like Ceratium hirundinella, have
and, especially, Anabaena, Microcystis and Stauras- adaptations for supporting their growth require-
trum pingue were all able all to function at pH ments under nutrient-segregated conditions and
levels up to 1 point higher. The observation growth rate is not necessarily enhanced by nutri-
matches those of Talling (1976) concerning dif- ent abundance. The growth-rate performances
ferential insensitivity to carbon dioxide shortage; achieved by Ceratium in each of the three enclo-
all these species are known or suspected for the sures during 1980 were ultimately comparable
efficiency of their carbon-concentration mecha- (see Fig. 5.18: 0.092 d−1 in Enclosure A, 0.098 d−1
nisms (Section 3.4.3). The apparent failure of the in Enclosure B and 0.105 d−1 in Enclosure C), even
chrysophytes Uroglena and Dinobryon to maintain though the non-physical growth conditions were
growth at pH levels ≤9 aroused the suspicion quite disparate. That the yields were quite differ-
that these algae might be obligate users of carbon ent is influenced by the length of time that cell
dioxide (Reynolds, 1986b), as indeed, has since division was maintained in situ. This was partly
been verified in the laboratory (Saxby-Rouen et al., influenced by resource availability and, as is now
1998). known, by resource renewal in the graded Enclo-
These response patterns require careful inter- sure C (Fig. 5.11; Reynolds, 1996b). The major
pretation and their subjectivity to the experi- influence, however, is the source of excysting
mental design must be taken into account. The inocula. The uniformly deep sediments of Enclo-
species responding to the conditions contrived sures A and B supported many fewer surviving
are, almost exclusively, the ones that are already cysts than those of Enclosure C, whereas recruit-
well established in the lake and/or the exper- ment of ‘germinating’ gymnoceratia (see Section
imental enclosures. The observed performances 5.4.6) was also relatively stronger in Enclosure
are not necessarily those of the nature’s ‘best-fit’ C. It is interesting, indeed, that Lund (1978) had
Figure 5.19 The maximum

fraction of the summer standing
phytoplankton biomass in Blelham
enclosures contributed by
nitrogen-fixing species of Anabaena
and Aphanizomenon in each
enclosure–year, plotted against the
corresponding areal loading rates of
nitrogen and phosphorus. Their
relative abundance generally
coincides with low nitrogen and
moderate phosphorus availability
but not on a low N : P ratio per se.
Redrawn with permission from
Reynolds (1986b).
regarded the enclosures as being somehow hos- sus P field of Fig. 5.19, corresponding to load-
tile to Ceratium, for the alga had never become ings of 6–10 g N m−2 (0.45–0.71 mol m−2 ) and
so numerous as in the lake outside. As Fig. 5.18 0.19–1.2 g P m−2 (6–38 mmol m−2 ), but with no
demonstrates, there is nothing about these enclo- clear dependence on N : P (actual ratios 12–118).
sures that interferes with growth. Isolation of At higher loads of N and P (but with ratios still
the water from that of the lake, except during in the order of 10–20), the species were hardly
winter opening, evidently made it difficult for represented at all. Thus seasonal dominance by
Ceratium to invade in numbers. The point about these species has failed to come about under
Enclosure C is that the presence of shallow sed- conditions in which neither nitrogen nor phos-
iments assisted the success of perennation and phorus was likely to have been limiting plankton
reinfection of the water column in the spring. growth. What can be said is that, at the low nitro-
This pertinent biological observation was, obvi- gen concentrations limiting non-N-fixing species,
ously, quite peripheral to the design and purpose the yields of Anabaena and Aphanizomenon are
of the experiment. broadly proportional to phosphorus loadings up
The behaviour of Anabaena, a supposed indi- to ∼1 g (or ∼30 mmol) P m−2 . At higher levels
cator of carbon-deficient eutrophic waters, and of N and P, there will always be faster-growing
its apparent preference for less-enriched condi- species – such as Ankyra, Chlorella, Plagioselmis,
tions may also be explained from more detailed Cryptomonas, even Eudorina – poised to outper-
analysis. Suspecting a performance influenced form them.
by the ratio of available nitrogen to available Finally, additional information is available
phosphorus, the maximum fractional abundance from the enclosure work to amplify the specific
of Anabaena spp. (plus the smaller amounts of growth performances under persistent and rela-
Aphanizomenon) in any given enclosure year was tively low phosphorus concentrations. Coenochlor-
plotted against the coordinates of the nitrogen is fottii featured prominently in a number of
and phosphorus supplied (Fig. 5.19). Anabaena the sequences (it was then referred to Sphaero-
spp. featured in most annual sequences observed cystis schroeteri). It was relatively more common
in the enclosures but only occasionally did it in the summer periods when phosphorus was
produce dominant populations. Moreover, when strongly regulating biomass and growth (see
levels of DIN had fallen much below 100 µg especially Reynolds et al., 1985) but absolutely
N l−1 (∼7 µM), populations developed substan- larger populations and sustained growth rates
tial heterocyst frequencies (maximum, 7% of were observed at greater nutrient availabilities
all cells). In terms of N and P loads, how- (Reynolds et al., 1983b, 1984). With the impos-
ever, Anabaena and Aphanizomenon abundances ition of artificial mixing, the alga was again
are clustered within one corner of the N ver- found to be tolerant but only for so long as
the water was clear and the ratio of Secchi-

disk depth to mixed depth (zs /zm ) was near or
greater than 1. Noting rather smaller numbers
of Oocystis aff. lacustris, Coenococcus, Crucigeniella
and the tetrasporalean Pseudosphaerocystis lacus-
tris (formerly Gemellicystis neglecta) showed sim-
ilar responses to imposed variations in nutri-
ents and insolation, Reynolds (1988a) grouped all
these non-motile, mucilaginous colonial Chloro-
phyceae in a single morphological–functional
group of non-motile, light-sensitive, mucilage-
bound species. It is their behavioural traits with Figure 5.20 Differing demographic behaviours in
respect to threshold light levels that tend to exploiting favourable growth opportunities: species grow
exclude them from turbid or deep-mixed water either rapidly and invasively (1) or more slowly to build a
columns and to give them a common association conserved, acquisitive population (2). Slow accumulation of
with oligotrophic lakes. For the common chrys- biomass may be offset by the recruitment of pre-formed
propagules from a perennial seed bank (3). Modified after
ophyte species of Dinobryon, Synura and Uroglena,
Reynolds (1997a).
the apparently similar association with nutrient
poverty in the enclosures is not due to any intol-
erance of high nutrients but to an unrelieved higher-biomass-achieving K-selected trait of many
dependence upon the supply of carbon dioxide, larger algae, represented by Curve 2 in Fig. 5.20,
which, in the soft-water confines of the Blelham which, initially, lags behind the performance of
enclosures, is readily outstripped by demand more r-selected species (Curve 1) is characteris-
(Saxby-Rouen et al., 1998). tic of the performances of S- and CS-strategists.
However, it is also clear that some of the self-
5.5.4 Temporal changes in performance regulating S-strategists, such as Ceratium and
selection Microcystis, are obliged to grow so relatively slowly
There is clearly a good match between how the that eventual abundance in the plankton is influ-
various species studied in the controlled field enced by the recruitment of sufficient peren-
conditions of the Blelham enclosures and the nating propagules at the initiation of the next
sorts of trait characteristics and strategic adapta- period of growth. This very strong K-selected fea-
tions identifiable among the properties revealed ture is represented by Curve 3 in Fig. 5.20.
in earlier sections (especially pp. 31–34). In par- We may venture further than this by superim-
ticular, a distinction is to be made between the posing the triangular ordination of species traits
manner in which species develop their popula- (e.g. of Fig. 5.9) and overlaying this on a notional
tions in response to what are perceived by them plot to describe the interaction of mixing and
to be favourable conditions. On the one hand, nutrient availability in the near-surface waters
there are species that specialise in rapid, inva- (see Fig. 5.21). The loops and arrows are inserted
sive growth, building up stocks at the expense to show how temporal seasonal variations in the
of freely available resources and high photon coordinates of nutrient availability and mixing
fluxes, and which, by analogy to the terminol- might select for particular performances and rel-
ogy of Grime (1979, 2001), we have defined r- evant traits and adaptations. These are notional
selected C-strategists (see Box 5.1). Besides the and unquantified at this stage of the develop-
examples of Ankyra and Chlorella, Asterionella and ment and the representation is qualitative but
some of the other freshwater diatoms that are they illustrate some general points that need to
tolerant of intermittent and poor average inso- be made. The two large loops (marked a and b)
lation (R-strategist traits) are also strongly r- reflect the transitions that might be observed
selected over other R-type species such as Plankto- over a year in a seasonally stratified (not nec-
thrix agardhii. The slower growth but often essarily temperate) lake. At overturn, there is a
at the extremes. The converse is that variability is

good for maintaining high diversity as more spe-
cific performances are accommodated. The con-
strained cycles of (say) an enriched water col-
umn subject to variable stratification or of persis-
tently mixed, resource-cycling systems may also
be represented in this scheme (respectively, c
and d). Other factors notwithstanding, the effects
of population growth should follow the direc-
tion of the arrow (marked e) as nutrients are
withdrawn from the water and the increasing
biomass reduces light penetration and the rel-
ative mixed depth is increased.
The traces provide adequate summaries of
changes in seasonal dominance in given lakes
and, to an extent, they may reflect longer-
Figure 5.21 Notional representation of Grime’s CSR
triangle on axes representing relative nutrient abundance and term changes in phytoplankton in response to
water-column mixing, showing the adaptive traits most likely nutrient enrichment or restoration measures.
to be selected by changing environmental conditions. The In the naturally eutrophic, nutrient- and base-
loops represent time tracks of selective pressures acting rich kataglacial lakes of northern Europe and
through the year in (a) oligotrophic lakes, (b) eutrophic lakes, North America (see, for instance, Nauwerck, 1963;
(c) and (d) in smaller, enriched systems; (e) is the anticipated Lin, 1972; Kling, 1975; Reynolds, 1980a), the
course of autogenic succession. Redrawn from Reynolds annual cycle of phytoplankton dominance fea-
(1988a).
tures (i) vernal diatoms (which may include any
or all of Asterionella formosa, Fragilaria crotonen-
rapid rightward shift on the mixing axis with sis, Stephanodiscus rotula, Aulacoseira ambigua from
an upward drift as dissolved nutrients are redis- the R apex of the triangle), followed by (ii) a
persed from depth. The best-performing species burst of readily grazeable, plainly C-strategist
here and throughout the bloom period are likely nanoplankton (e.g. chlorellids, Ankyra, Chlamy-
also to show the traits and growth responses of R- domonas, Plagioselmis), and/or (iii) by populations
strategist species: the limits of their morphologi- of colonial Volvocales (e.g. Eudorina, Pandorina –
cal and behavioural adaptations are more suited best classed as CS strategists) and increasingly
to coping with low average insolation. The onset more S-strategist Anabaena spp., Microcystis or Cer-
of themal stratification is represented by a lurch atium). The cycle is completed by (iv) assemblages
to the left, where conditions of low relative mix- of diatoms (Fragilaria, Aulacoseira granulata) and
ing and high relative nutrients obtain and which, desmids (Closterium aciculare and several species of
initially, are open to exploitation by fast-growing Staurastrum).
C-strategist species. Their activity depletes the In the nutrient- and base-deficient lakes of
resources and may lead, eventually, to the par- the English Lake District (Pearsall, 1932), the
titioning of availability and to the dependence oligotrophic subalpine lakes of Carinthia (Find-
upon increasingly effective S-strategist adapta- enegg, 1943) and the more oligotrophic lakes of
tions to access them. Note that the more severe New York and Connecticut studied by Huszar
and ongoing is the insolation or resource defi- and Caraco (1998), as well, in all probability,
ciency, the closer are the trajectories to the apices similar lakes throughout the temperate regions
of the CSR triangle, where the relevant adapta- (Reynolds, 1984a, b), the vernal plankton is typ-
tions become ever more important. As a conse- ically dominated by Cyclotella–Urosolenia diatom
quence, the few that have them are alone able to associations (R or CR strategists). These may be
perform successfully at all. The potential diver- replaced typically by such chrysophytes as Dino-
sity of surviving species is finally ‘squeezed out’ bryon, Mallomonas or Synura (RS strategists) and/or
by colonial Chlorophyceae (CS strategists), then within 4–6 months before being short-cut back
by S-strategist dinoflagellates (Peridinium umbon- to an earlier stage. Even in temperate lakes and
atum, P. willei or Ceratium spp.) and, finally, by R or reservoirs subject to extreme fluctuations in
SR desmids such as Cosmarium and Staurodesmus. mixed depth on scales of 5–50 days, alternations
Between the oligotrophic and eutrophic sys- between phases of increase and dominance
tems are ranged the lakes of intermediate sta- by R species (Stephanodiscus, Synedra) and C–S
tus (mesotrophic lakes), as well as several deep groupings (cryptomonads, Chlamydomonas, Oocys-
alpine lakes of central Europe (Sommer, 1986; tis, Aphanizomenon) are clearly distinguishable
Salmaso, 2000; Morabito et al., 2002) Here, the ver- (Haffner et al., 1980; Ferguson and Harper,
nal phase is characterised by R-strategist diatoms 1982). Again, in nutrient-rich lakes where the
featuring perhaps Aulacoseira islandica or A. sub- alternations result in the lake being either
arctica, as well, perhaps, as Asterionella, Fragi- predominantly mixed or stratified, so the dom-
laria or Cyclotella radiosa. There is a late-spring inating species would be (respectively) R species
phase of C strategists (e.g. Plagioselmis, Chrysochro- (as in Embalse Rapel, Chile: Cabrera et al., 1977)
mulina), followed either by a phase of colo- or C species (as in Montezuma Well, Arizona:
nial Chlorophyceae–Chrysophyceae or, especially Boucher et al., 1984). These possibilities comply
in deeper lakes, S-strategist Ceratium or Peri- with the track marked (c) in Fig. 5.21. Examples
dinium, or by the Cyanobacteria Gomphosphaeria of enriched shallow or exposed lakes that are
or Woronichinia, or again perhaps by potentially more or less continuously mixed seem to be dom-
nitrogen-fixing Anabaena solitaria, A. lemmerman- inated by the K-selected R strategists (Planktotrix
nii or Aphanizomenon gracile. Late-summer mixing agardhii, Limnothrix redekei, Pseudanabaena spp.:
may favour R-strategist diatoms (notably includ- Gibson et al., 1971; Berger, 1984, 1989; Reynolds,
ing Tabellaria flocculosa or T. fenestrata), desmids or 1994b) are represented in Fig. 5.21 by track (d).
the filamentous non-diatoms (such as Mougeotia, It is not yet possible to apply the same
Binuclearia, Geminella). The outstanding algae of approach to temporal changes in the marine
the deep mesotrophic systems, however, are the phytoplankton with a similar level of inves-
RS-strategist Planktothrix rubescens/mougeotii group tigative evidence, as the resolution of temporal
which both tolerate winter mixing and exploit changes is less clear. On the other hand, it is
deep stratified layers in summer. a testable hypothesis that similar performance-
The cycles in Fig. 5.21 are not tracked at an led drivers, influenced by similar morpholog-
even rate, neither are they precisely identical ical adaptations to analogous liqiud environ-
each year. Progress may proceed by a series ments, govern the spatial and temporal differ-
of lurches, whereas interannual variability can ences in the growth of phytoplankton in the
divert the sequence to differing extents. However, sea. There is good supporting evidence that this
the growth and potential dominance of phyto- may be the case. Smayda (2000, 2002) has shown
plankton adheres closely to the model tracking that the wide diversity among the dinoflagel-
(Reynolds and Reynolds, 1985). The cycle may be lates may be rationalised against an ecological
completed in less than a year: the description of pattern that invokes morphology. Whereas the
Berman et al. (1992) of the periodicity of phyto- smaller, non-armoured adinophytes (such as Pro-
plankton of Lake Kinneret follows a mesotrophic rocentrum) and gymnodinioids (Gymnodinium, Gyro-
path before stalling in summer deep on the left- dinium, Heterocapsa spp.) that are characteristic of
hand (nutrient) axis of Fig. 5.21. It does not really shallow, enriched coastal waters have unmistake-
move on until wind-driven mixing or autumn ably C-like morphologies and growth-rate poten-
rains relieve the severe nutrient (nitrogen and tial, the larger, armoured and highly motile cera-
phosphorus) deficiency. The cycle may also be tians have clear S tendencies. In the open ocean,
recapitulated: Lewis’ (1978a) detailed description Smayda (2002) distinguishes among dinoflagel-
of the seasonal changes in the plankton of Lake lates associated preferentially with fronts and
Lanao, Philippines, could reasonably represented upwellings (Alexandrium, Karenia) and those of
by track (b) in Fig. 5.21 but it would be completed post-upwelling relaxation waters (Gymnodinium
catenatum, Lingulodinium polyedrum). S-strategist supreme over vast areas of ultraoligotrophic

dinoflagellates are also prominent in the oligo- ocean, as archetypes of the newly proposed SS
trophic, stratified tropical oceanic flora, where strategy.
self-regulation, high motility and photoadapta- Seasonal changes in the plankton flora of
tive capabilities distinguish such dinoflagellates the English Channel, described by Holligan and
as Amphisolenia and Ornithocercus. The buoyancy- Harbour (1977), show a clear tendency for ver-
regulating adinophytes of the genus Pyrocystis are nal diatom–dominated (supposedly R-strategist)
most remarkable in showing parallel adaptations assemblages to be displaced by more mixed
to limnetic Planktothrix rubescens and in similarly diatom-dinoflagellate (CR?) associations (Rhizosol-
constituting a mid-water shade flora, deep in the enia spp.; Gyrodinium, Heterocapsa, Prorocentrum) in
light gradient of the tropical ocean. early summer and by green flagellates (Carteria,
It is interesting to speculate on the range Dunaliella, Nannochloris) or (S?) ceratians (Ceratium
of adaptations among the planktic diatoms of fusus, C. tripos) in mid to late summer. In enriched
the sea. Most are non-motile and (presumably) near-shore areas, the haptophyte Phaeocystis, in
reliant upon vertical mixing for residence in its colonial life-history stage, may dominate the
the near-surface waters. Thalassiosira nordenski- early summer succession, in a manner strongly
oldii, Chaetoceros compressus and Skeletonema cost- reminiscent of the abundance of volvocalean CS
atum are all chain-forming diatoms featuring strategists in eutrophic lakes.
in the spring blooms of North Atlantic shelf A satisfying aspect of the performances of
waters. The attenuate forms of species of Rhizosol- phytoplankton in both the sea and fresh waters is
enia, Cerataulina, Nitzschia and Asterionella japonica the superior influence of morphological and (pre-
are conspicuous in the neritic areas. All these sumably) physiological criteria over phylogenetic
diatoms can be accommodated in the under- affinities. This is a powerful statement attest-
standing of R-strategist ecologies. However, there ing to evolutionary adaptations for relatively spe-
are also centric diatoms such as Cyclotella capsia cialised lifestyles and for the radiative potential
that occur predominantly in shallow eutrophic latent within all major phyletic divisions of the
coastal waters and estuarine areas, along with, photosynthetic microorganisms.
arguably, C-strategist green algae (Dunaliella,
Nannochloris), cryptomonads, the dinoflagellate 5.5.5 Modelling growth rates in field
Prorocentrum, the euglenoid Eutreptia and the There has for long been a requirement for robust,
haptophytes Chrysochromulina and Isochrysis. In predictive models of phytoplankton. Nowadays,
another direction of adaptive radiation, the very the element of stochasticity of environmental
large, self-regulating Ethmodiscus rex, belonging to events is appreciated as a near-insuperable bar to
the tropical shade flora, shows the typical prop- accurate predictions of sufficient precision. How-
erties of an S-strategist species. ever, considerable use can be made of the regres-
Many coccolithophorids are small and are sions fitted to growth performances in the labora-
regarded, somewhat ‘uncritically’ (Raymont, tory and the philosophy of strategic adaptations
1980), as nanoplankton. Emiliana huxleyi, Gephyro- to drive predictive solutions to which of certain
capsa oceanica and Cyclococcolithus fragilis are, kinds of alga will grow in particular water bodies
indeed, small C-like species of open water, and under which conditions.
which they inhabit with nanoplanktic flagel- This section is not intended to provide a guide
lates, including the prasinophyte Micromonas, or a review of different modelling approaches.
the chlorophytes Carteria and Nannochloris, the These are available elsewhere (Jørgensen, 1995,
cryptophytes Hemiselmis and Rhodomonas and 1999). The purpose here is to refer to some of
the haptophytes Pavlova and Isochrysis. The the approaches addressed specifically to mod-
nitrogen-fixing, vertically migrating Cyanobac- elling growth and performance of phytoplank-
terium Trichodesmium displays strong S-strategist ton and to promote the use of models that
characteristics. The picoplanktic Cyanobacteria invoke them. It is worth first repeating the obvi-
Synechococcus and, especially, Prochlorococcus, reign ous, however, that different models attempt to
do different things. These may be in-and-out uously overrides the others. This was the case in
‘black-box functions’, such as Eq. (4.15), where an the models of growth of filamentous Cyanobac-
input (in this case, biologically available phos- teria in an enriched, monomictic lake (Jiménez
phorus) generates a yield (in this case, phyto- Montealegre et al., 1995) or deep in the light
plankton chlorophyll) on the basis of pragmatic gradient of the Zürichsee (Bright and Walsby,
observation, without any attention to the explica- 2000). A most promising modern development
tive processes. These internal linkages may be has come through the exploration of linkages
investigated, imitated and submitted to empiri- (stimulus, responses) and the probabilistic analy-
cal model explanations, for instance, those which sis of effects (likely reaction) through artificial
link the generation of phytoplankton biomass neural networks (ANNs) (see Recknagel et al.,
to photosynthetic behaviour in the underwater 1997). Like the nerve connectivities they resem-
light field. The various explanative equations of ble, these models can be ‘trained’ against real
(say) Smith (1936), Talling (1957c), Pahl-Wostl and data in order to predict outcomes with modi-
Imboden (1990) predict, with accuracy, precision fied variables. Recknagel’s (1997) own application
and increasing detail, the photosynthetic car- to interpret the variability in the phytoplank-
bon yield as a function of light and respiration. ton periodicity in Kasumigaura-Ko in Japan pro-
They are, nevertheless, restricted in their effec- vides an excellent indication of the power of this
tiveness to cases where anabolic processes are approach. Its further development is at an early
simultaneously constrained by some other fac- stage but the use of ‘supervised’ and ‘unsuper-
tor (carbon or nutrient supply). What is then vised’ learning algorithms to interpret field data,
needed is the further sets of precisely quan- through ‘self-organising maps’ of close interre-
tifiable algorithms that will describe these fur- lationships (Park et al., 2003), promises to over-
ther processes (many of which are available) and come some of the difficulties experienced with
their incorporation into a supermodel to simu- other compound models. Prediction of ‘top-line’
late the interactions among all the components. outcomes based on ‘bottom-line’ capacities is gen-
In a third type of model, the broad function of erally difficult without knowledge of interme-
the system (‘the box’) is predicted from a knowl- diate processes. The fundamental truth is that
edge of the fundamental mechanisms and limi- algal growth rate is not a continuous function
tations (such as genomic information and energy of nutrient supply or uptake, or of the ability to
efficiency), as elegantly employed in Jørgensen’s fix carbon in the light. Below the ‘threshold’ val-
(1997, 2002) own structural–dynamic models of ues discussed in Section 5.4.5, growth cannot for
ecosystem organisation and thermodynamics. long exceed the weakest capacity. On the other
Each of these approaches, even when applied hand, capacity in excess of the threshold sat-
directly to phytoplankton ecology, has its inher- urates processing: it does not make organisms
ent weaknesses and these have long been recog- grow faster.
nised (Levins, 1966). The first type simulates an So, how can the growth rates of natural pop-
indirect relationship with accuracy and some pre- ulations in the field be modelled? The approach
cision but lacks general applicability. The sec- advocated by Reynolds and Irish (1997) was to
ond has a small number of variables and yields suppose that the photautotrophic plankter does
accurate and often precise information but only not grow anywhere better than it does in the
under very conditioned circumstances. The third contrived culture conditions in the laboratory.
achieves accuracy and applicability through gen- Given that the best growth performances of
erality, at the cost of all precision. given species occur under ideal culture condi-
Modelling philosophy and (certainly) comput- tions, that they are consistent and that between-
ing power has moved on. Several attempts to species differences in growth rate are systematic
compound specific process models of the sec- (Reynolds, 1984b), an ‘upper base line’ for simu-
ond type into more comprehensive growth sim- lating natural population growth could be pro-
ulations have been rather unsuccessful, except posed (Reynolds, 1989b). Three equations were
where one or other of the components contin- invoked to predict attainable growth rate in the
field. In the best traditions of Eppley’s (1972) Thus, the solution to Eq. (5.9) incorporated in
model of phytoplankton growth, two of the to the growth-rate model is:
equations set the growth potential to water tem-
perature. The first equation (5.5) predicts repli- r(θ,I ) d−1 = [rθ (24 h m )−1 ] · ln 2I 0 ·

cation rate at a standard 20 ◦ C as a function of 0.257(msv−1 )0.236 · rθ−1 ε−1 (5.13)
algal morphology. The second equation (5.6) pro-
vides the information to adjust specific growth Equation (5.13) is thus the third of the three
rate to other temperatures. These predictions are equations written into the model that eventu-
applied to an inoculum (or, in reiterations) to the ally became known as PROTECH. It remained
incremented standing crop to simulate day-on- under development and testing for several years,
day accumulation. This has to be linked, through but this central core has remained intact. An
a loop in the model logic, to an inventory of important adjustment in respect of dark respi-
resource supply, which checks that a given daily ration was incorporated into the model that was
increment is sustainable and, if so, to permit the eventually published (Reynolds et al., 2001). This
growth step to be completed. A further feedback followed the important steps of sensitivity test-
loop deducts the consumption from the pool of ing, authentication and validation (Elliott et al.,
available resources. 1999a, 2000a). It has been used to make realistic
Insofar as the depth-integration of light inten- reconstructions of phytoplankton cycles of abun-
sity and the duration of daylight impose, almost dance and composition (by functional type) in
everywhere on the surface of the planet, a sub- lakes and reservoirs (Elliott et al., 2000a; Lewis
ideal environment with respect to continuously et al., 2002). It has been applied to simulate
light-saturated cultures, the sensitivity of species- succession in undisturbed environments (Elliott
specific growth sensitivity to insolation is written et al., 2000b) and to investigate the minimum
into the third of the model equations. The orig- size of inoculum for the growth rate still to
inal model supposed that the insolation-limited enable an alga to attain dominance (Elliott et al.,
specific growth rate is in proportion to the frac- 2001b). Vertical mixing can be used as a vari-
tion of the day that the alga spends in the light: able to disturb community assembly (Elliott et
al., 2001a) and to evaluate selective impacts of
r(θ,I ) d−1 = rθ t p /24 (5.9) intearctions between variations in mixing depth
and in surface irradiance (Elliott et al., 2002). PRO-
where the daily sum of photoperiods, tp , comes
TECH models exist with differing physical drivers
from:
(PROTECH-C, PROTECH-D), that work in coastal
tp = h p / h m (5.10) waters (PROTECH-M) and which are dedicated to
(specific) rivers (RIVERPROTECH). Versions have
where hm is the mixed depth and hp is the height been prepared for numerous UK and European
of the light compensated water column, which, lakes and reservoirs, with accumulating success.
following Talling (1957c; see also Section 3.5.3) is At time of the writing, summary papers are still
solved as: in press.
h p = ln(I 0 /0.5I k )ε −1 (5.11)
where I 0 is the daily mean irradiance immedi- 5.6 Summary

ately beneath the water surface (in µmol photon
m−2 s−1 ) and ε is the coefficient of exponential Cell replication and population growth are con-
light attenuation with depth. The onset of light sidered as a unified process and accorded expo-
limitation of growth, Ik is here related specifically nential logarithmic units with the dimensions of
to the alga via Eq. (5.12): time. The observed rates of population change in
I k = rθ αr−1 (5.12) nature (±rn ) are net of a series of in-situ rates of
loss (rL , treated in Chapter 6). However, the true
where α r = 0.257 (msv−1 )0.236 (Eq. 5.8). rates of replication (r ) must always be within
SUMMARY 237
(and are sometimes well within) the least rate species having weaker affinities and may outper-
that is physiologically sustainable on the basis of form them by building up larger inocula than
the resources supplied. Cell replication is regu- potential competitors. Chronic or eventual defi-
lated internally and cannot occur without the ciencies in nitrogen may have an analogous selec-
prior mitotic division of the nucleus. Nuclear tive effect, although, subject to the fulfilment of
division is prevented if the cell does not have other criteria (available phosphorus, high insol-
the resources to complete the division. If it does, ation and a supply of relevant trace metals;
replication can proceed at a species-specific max- see Chapter 4), dinitrogen fixers may experience
imum rate. selective benefit.
In general, division rates are strongly reg- In the frequent cases in which nutrient deple-
ulated by temperature. Species-specific division tion is experienced first in the near-surface
rates at 20 ◦ C (r20 ) are correlated with the surface- waters of the upper mixed layer and proceeds
to-volume (sv−1 ) ratios of the algal units, as downwards, high affinity may be less help-
are the slopes of the temperature sensitivity of ful than high mobility. The beneficial ability
growth. The light sensitivity of growth is sub- to undertake controlled downward migration
ject to physiological photoadaptation but, ulti- with respect to the relevant ‘nutricline’ or to
mately, the shape of the alga influences its effec- perform diel or periodic forays through an
tiveness as a light interceptor. Species which offer increasingly structured and resource-segregated
a high surface-to-volume ratio through distortion medium, is conferred through the combination
from the spherical form (such that the maxi- of self-regulated motility and large organismic
mum linear dimension, m, is rather greater in size. The greater is the isolation of the illumi-
one plane than in one or two of the others) nated, nutrient-depleted, upper column from the
and show high values of the product msv−1 are dark but relatively resource-rich lower column,
indicative of probable tolerance of low average the greater is the value of such performance-
insolation. Given that the potential daily photon maintaining adaptations.
harvest becomes severely constrained by vertical In the face of silicon shortages, diatoms
mixing through variously turbid water to depths are unable to perform at all, whereas the per-
beyond that reached by growth-saturating levels formances of non-diatoms is normally consid-
of downwelling irradiance, enhanced light inter- ered insensitive to fluctuations in supply. The
ception becomes vital to maintaining growth. amounts of silicon required vary interspecifi-
Below a threshold of about 1–1.5 mol photons cally, as do the affinities for silicic-acid uptake.
m−2 d−1 , growth-rate performances are main- Because diatom nuclei divide before the silicic
tained relatively better in plankters offering the acid needed to form the daughter frustules is
combination of relevant morphological preadap- taken up, it is possible for otherwise resource-
tation and physiologically mediated photoadap- replete and growing populations to experience
tative pigmentation. big mortalities as a consequence of sudden
Growth rates of phytoplankton species are var- encounter with Si limitation. With other nutri-
iously sensitive to nutrient availability, though ents, planktic algae are able to ‘close down’
not at concentrations exceeding 10−6 M DIN vegetative growth and to adopt a physiologi-
or 10−7 M MRP. Above these ‘critical’ levels, cal (or perhaps morphological) resting condition,
growth rates are neither nitrogen nor phospho- which improves the survival prospects for the
rus limited. Neither, it is argued, are they depen- genome.
dent upon the ratio of either of these resources The environments of phytoplankton may be
to the other. However, differing species-specific classified, following Grime (1979), upon their
nutrient-uptake affinities influence the potential ability to sustain autotrophic growth, in terms
growth performances when nutrient availabili- of the production that the resources will sus-
ties fall to, or are chronically below, the ‘critical’ tain and the duration of the opportunity. Most
levels. Species with high affinity for phosphorus algae will grow under favourable, resource-
are able to maintain a faster rate of growth than replete conditions during long photoperiods. The
species that perform best rely on an early pres- resource-gathering constraints against large size
ence, rapidity in the conversion of resources but the smallest picoplanktic sizes of photoau-
to biomass and a high frequency of cell divi- totrophs have these provinces of the aquatic envi-
sion and recruitment of subsequent generations. ronment very nearly to themselves. It is proposed
By analogy with Grime’s (1979) functional clas- here that they be henceforth referred to as ‘ SS
sification of plants, these algae are considered strategists’.
to be C strategists; they are typically small Many algae have adaptations and biologies
(<103 µm3 ), usually unicellular, have high sv−1 that represent intermediate blends of C-, S- or R-
ratios (>0.3 µm−1 ) and sustain rapid growth strategist adaptations and lifestyles. The spatial
rates (r 20 > 0.9 d−1 ). Algae whose growth per- and temporal distributions of particular types
formance is relatively tolerant of and adaptable and species of phytoplankton, and the opportuni-
to progressively shorter photoperiods and aggre- ties of replication that lead to population devel-
gate light doses are comparable to Grime’s rud- opment, are shown to be closely correlated with
eral (R) strategists: their sizes are varied (103 –105 the extent of their C, S or R attributes. Though
µm3 ) but all offer favourable msv−1 ratios (range best demonstrated among the freshwater phyto-
15–1000). Algae whose growth performances are plankton, the functional–strategic approach of
maintained in the face of diminishing nutri- Grime appears to hold just as well for the ecolo-
ent availability are equipped to combat resource gies of the marine plankton. In both the sea
stress. Their conservative, self-regulating S strate- and fresh waters, morphological and (presum-
gies are served by the properties of large size ably) physiological criteria are better predictors
(104 –107 µm3 ) and motility but at the price of of ecology than are phylogenetic affinities. In a
low sv−1 (<0.3 µm−1 ), low msv−1 (<30) and slow concluding section, this finding is reversed to
rates of growth (r 20 <0.7 d−1 ). show how functional properties of phytoplank-
Neither large size nor motility offer any ton and their respnses to environmental drivers
advantage to survival in chronically resource- can be used to predict the structure of ascen-
stressed environments of the ultraoligotrophic dent phytoplankton communities on the basis of
oceans and the largest lakes. Moreover, the their likely strategic growth responses and not
extreme resource rarification also offers a respite the stochasticity of the processes befalling indi-
from direct consumption. Not only are there vidual species.
Chapter 6
Mortality and loss processes in phytoplankton
was almost wholly and precisely compensated

6.1 Introduction by simultaneous bulk loss rates (Forsberg, 1985),
when the rates of grazing or sedimentation
This chapter considers the sinks and, more par- might only rarely explain the disappearance of
ticularly, the dynamic rates of loss of formed the equivalent of the day’s new product, it
cells from phytoplankton populations. Several became clear that some further separation of the
processes are involved -- hydromechanical trans- ‘losses’ was necessary, together with some refine-
port, passive settlement and destruction by her- ment of the terminology. Here, adjustments to
bivores and parasites -- which, separately or in the photosynthate content that the cell is unable
concert, may greatly influence the structuring to deploy in new growth or to store intracellu-
of communities and the outcome of compet- larly and which must be dispersed through accel-
itive interactions among phytoplankton. More- erated respiration, or photorespiration, or secre-
over, these same processes may contribute power- tion as glycollate or other extracellular product,
fully to the biogeochemical importance of pelagic are considered to be ‘physiological’. The adjust-
communities, through their role in translocating ments are as much to protect the intracellular
bioproducts from one point of the planet’s sur- homeostasis as to supply any other component of
face to another. the pelagic system. On the other hand, success-
Before expanding upon these processes, how- fully replicated cells in growing populations are
ever, the opportunity is taken to emphasise that continuously but variably subject to physical or
the losses considered in this chapter are those biological processes that deplete the pelagic con-
that affect the dynamics of populations. The centrations in which they are produced. Detract-
(sometimes very large) loss of photosynthate pro- ing from the numbers of new cells added to
duced in excess of the cell’s ability to incorpo- the population, these losses are ‘demographic’
rate in biomass is not considered here. The topic and, as such, are the proper focus of the present
is covered in a different context in Chapter 3 chapter. Its objective is to establish the quantita-
(see especially Section 3.5.4). The emphasis is nec- tive basis for estimating the drain on the poten-
essary as the term ‘loss rates’ was applied col- tial rate of recruitment, provided by cell repli-
lectively to the dynamics of almost all measur- cation, that is represented by the counteracting
able photosynthetic production that did not find processes which, effectively, dilute the recruit-
its way into increased producer biomass (Jassby ment of phytoplankton biomass. In the sense of
and Goldman, 1974a). It had been supposed by Eq. (5.3), the task is to quantify the magnitude
many workers at the time that the realised short- and variability in the rates of dilution of finished
fall was attributable to grazing and sedimenta- cells (rL ).
tion of biomass. However, with the demonstra- As already suggested, the principal loss pro-
tion that, very often, production in some systems cesses are hydomechanical dispersion (wash-out
240 MORTALITY AND LOSS PROCESSES IN PHYTOPLANKTON
from lakes, downstream transport in rivers, patch the residual population. The essential question is
dilution at sea), sedimentation and consumption whether the inflow simply replaces the original
by grazers. Attention is also accorded to mor- volume by direct displacement or by a flushing
talities through parasitism (a specialised con- action, in which the inflow volume mixes exten-
sumption) and physiological death and wastage. sively with the standing volume, displacing an
Although highly disparate in their causes, each equivalent volume of well-mixed water. In the lat-
process has the effect of diluting the locally ran- ter case, the original volume and the algal pop-
domised survivors. Hence, each is describable ulation that it entrained will have been depleted
by an exponent, summable with other loss and less completely and will now be, on average, less
growth terms. Just as Eq. (5.1) explains the rate dilute.
of population change, δ N /δt, by reference to
the first-order multiplier, ern t , and where, from 6.2.1 Expressing dilution
Eq. (5.3), it may be asserted that rn = r − rL , it The mathematics of dilution are well estab-
may now be proposed that: lished. Dilution of the standing volume and
its suspended phytoplankton is described by
rL = rW + rS + rG + r . . . (6.1)
an exponential-decay function. Until Uhlmann’s
where rW , rS , rG . . . are the respective exponents (1971) consideration of the topic, there had been
for the instantaneous rates of biomass loss due few attempts to express the dilution of phyto-
to wash-out, sedimentation, grazing, etc. It is plankton by displacement. He was quick to see
accepted that these terms, either individually or and to exploit the opportunity to put losses in the
in aggregate, may raise rL > r , in which case, rn same terms as recruitment (cf. Eq. 5.3) and simply
is negative and symptomatic of a declining pop- sum the instantaneous exponents. As depletion
ulation. rates to settlement and some forms of grazing
The following sections will consider the mag- succumb to analogous exponential functions, it
nitudes and variabilities of the loss terms. is perhaps helpful to rehearse the logic that is
invoked.
In the present case of dilution by wash-out,
6.2 Wash-out and dilution we suppose that a population of uniform, non-
living, isopycnic particles (N0 ) is fully entrained
The hydraulic displacement and dispersion of and evenly dispersed through the body of a brim-
phytoplankton is best approached by considering full impoundment of volume, V. The introduc-
the case of algae in small lakes or tidal pools tion of a volume of particle-free water, qs , in
in which the volume is vulnerable to episodes unit time t, displaces an equal volume into the
of rapid flushing. Inflow is exchanged with the outflow from the impoundment. Thus, the the-
instantaneous lake volume and embedded plank- oretical retention time of the impoundment, tr ,
tic cells are removed in the outflowing volume is given by V/qs . The outflow volume will carry
that is displaced. In this instance, the algae thus some of the suspended particles. From the ini-
removed from the water body are considered tial population (N0 ), particles will be removed in
‘lost’. It may well be that the individuals thus the proportion −qs /V. After a given short time
‘lost’ will survive to establish populations else- step, t, the population remaining, Nt , is calcula-
where. Indeed, this is an essential process of ble from:
species dispersal. The balance of the original pop-
N t = N 0 (1 − qs t/V ) (6.2)
ulaton that remains is, of course, now smaller
and, occupying the similar volume of lake, on During a second time period, of identical length
average, less concentrated. The predicted net rate to the first, an equal proportion of the original
of change in the depleting population may be off- might be removed but only if the remainder orig-
set or possibly compensated by the simultaneous inal population has not been intermixed with
rate of cell replication but, for the moment, we and diluted by the inflowing water. If there is
shall consider just the effects of biomass loss on mixing sufficient to render the residue uniform
WASH-OUT AND DILUTION 241
at t, then, at t2 , we should have: critical factor in the population ecology of phyto-

plankton. At the mesoscales of patch formation,
N 2 = N t (1 − qs t/V )
where recruitment by growth is pitched against
= N 0 (1 − qs t/V )(1 − qs t/V ). dilution through dispersion, it is possible to con-
Thus, after i such periods of length t, sider V as the patch size and qs as the rate of its
horizontal diffusion as being critical to the main-
N i = N 0 (1 − qs t/V )i (6.3) tenance of patchiness (the models of Kierstead
and Slobodkin, 1953; Joseph and Sendner, 1958;
and after one lake retention time (tw )
see also Section 2.7.2). In lakes much smaller
N tw = N 0 (1 − qs t/V )tw/t (6.4) than 10 km2 in area, wherein the maintenance
of large-scale developmental patches is largely
This is, of course, an exponential series, which
untenable (that is, critical patch size usually
quickly tends to
exceeds the horizontal extent of the basin and
N tw = N 0 e−1 (6.5) any small-scale patchiness is very rapidly aver-
aged) the proportion of qs to V assumes increas-
where e is the natural logarithmic base. Equa-
ing importance. While tw > 100 days, small dif-
tion (6.5) has a direct mathematical solution
ferences in hydraulic throughput may remain
(Ntw = 0.37N0 ) which predicts that, at the end
empirically inconsequential in relation to poten-
of one theoretical retention time, the volume
tial growth rate: rw is <0.01 d−1 . Its doubling to
diluted by flushing retains 0.37 of the original
0.02 d−1 is still small in relation to the rates of
population. Had the same volume simply been
growth that are possible. However, when the lat-
displaced by the inflow, the retained population
ter are themselves severely limited by environ-
would have fallen to zero. These possibilities rep-
mental conditions, dilution effects can become
resent the boundaries of probable dilution, lying
highly significant. In the instance of Planktothrix
between complete mixing with, and complete dis-
agardhii considered in Section 5.4.5, performing
placement by, the inflow volume.
at its best to grow under the mixed conditions
Supposing the tendency is strongly towards
of a temperate lake in winter, its replication rate
the flushing of algae by inflow, Eq. (6.5) may be
of 0.16 d−1 will be insufficient to counter out-
used to estimate the residual population at any
wash losses when tw ≤ 7 days (rw ≥ −0.16 d−1 ).
given point in time, t, so long as the same rates
If it is growing less well, the sensitivity to flush-
of fluid exchange apply:
ing clearly increases. Temperate lakes regularly
N t = N 0 e−t/tw experiencing retention times less than about
= N 0 e−q s t/V (6.6) 30 days seem not to support Planktothrix popu-
lations. In the persistently spring-flushed Mon-
To now derive a term for the rate of change tezuma’s Well, Arizona, studied by Boucher et al.
in the standing population that is attributable (1984) (tw < 9 d), the distinctive phytoplankton
to outwash (rw in Eq. 6.1) is quite straightfor- comprises only fast-growing nanoplanktic species.
ward, provided that the time dimension of the In the English Lake District, some of the lakes
inflow rate (s−1 , d−1 ) is compatible with the other have volumes that are small in relation to their
terms. largely impermeable, thin-soiled mountainous
catchments, and which episodically shed heavy
rw = qs /V (6.7)
rainfall run-off (1.5--2.5 m annually). In Grasmere
(tw ∼ 24 d annual average but, instantaneously,
6.2.2 Dilution in the population ecology of ranging from 5 d to ∞), Reynolds and Lund (1988)
phytoplankton showed that the phytoplankton had almost to
At first sight, the magnitude of hydraulic dis- recolonise the water column after wet weather,
placement rates (qs ) relative to the scale of stand- while it required a long dry summer for Anabaena
ing volume in large lakes and seas seems suffi- to become established. Wet winters also keep
ciently trivial for outwash to be discounted as a phytoplankton numbers low but, in the relatively
dry winter of 1973, the autumn maximum of physics of river flow and on the adaptations
Asterionella persisted to merge with the spring and population dynamics of river plankton (pota-
maximum. moplankton) developed quickly during the mid-
1980s. These were reviewed and synthesised by
6.2.3 Phytoplankton population dynamics Reynolds and Descy (1996) in an attempt to
in rivers assemble a plausible theory that would explain
The most highly flushed environments are rivers. the paradoxes about potamoplankton. The prin-
Larger ones often do support an indigenous cipal deduction is that rivers are actually not
phytoplankton, usually in at least third- or very good at discharging water. Not only is their
fourth-order affluents, and sometimes in very velocity structure highly varied, laterally, verti-
great abundance (perhaps an order of magni- cally and longitudinally, but significant volumes
tude greater concentration than in many lakes; I (between 6% and 40%) may not be moving at all.
have an unpublished record of over 600 µg chla A part of this non-flowing water is, depending
L−1 measured in the River Guadiana at Mourão, on the size of the stream, explained in terms of
Portugal, under conditions of late summer flow). boundary friction with banks and bed but a large
The ability of open-ended systems, subject to per- part is immobilised in the so-called ‘fluvial dead-
sistent unidirectional flow, to support plankton zones’ (Wallis et al., 1989; Carling, 1992). These
is paradoxical. It is generally supposed to be a structures are sufficiently tangible to be sensed
function of the ‘age’ of the habitat (length of the remotely, either by their differentiated tempera-
river and the time of travel of water from source ture or chlorophyll content (Reynolds et al., 1991;
to mouth), for there is no way back for organ- Reynolds, 1995b). They are delimited by shear
isms embedded in the unidirectional flow. On the boundaries across which fluid is exchanged with
other hand, the wax and wane of specific popula- the main flow. The species composition of such
tions in given rivers seem fully reproducible; they plankton they may contain is hardly different
are scarcely stochastic events. Moreover, some from that of the main flow but the concentra-
detailed comparisons of the mean time of travel tion may be significntly greater. It has also been
through plankton-bearing reaches of the River shown that the enhancement factor is a function
Severn, UK, with the downstream population of the fluid exchange rate and the algal growth
increment would imply rates of growth exceeding rate: the longer cells are retained, the greater is
those of the best laboratory cultures (Reynolds the concentration that can be achieved by grow-
and Glaister, 1993). Downstream increases in the ing species before they are exchanged with the
phytoplankton of the Rhine, as reported by de flow (Reynolds et al., 1991). Each dead-zone has its
Ruyter van Steveninck et al. (1992), would require own V and qs characteristics and its own dynam-
specific growth performances paralleling any- ics. The analogy to a little pond, ‘buried in the
thing that could be imitated in the laboratory. river’ (Reynolds, 1994b), is not entirely a trite one.
It was also puzzling how the upstream inoc- Reynolds and Glaister (1993) proposed a model to
ula might be maintained and not be themselves show how the serial effects of consecutive fluvial
washed out of a plankton-free river (Reynolds, dead-zones contribute to the downstream recruit-
1988b). ment of phytoplankton. The recruitment is, nev-
These problems have been raised on many pre- ertheless, sensitive to changes in discharge, fluid
vious occasions and they have been subject to exchange and turbidity. It proposes a persistent
some important investigations and critical anal- advantage to fast-growing r-selected opportunist
yses (Eddy, 1931; Chandler, 1937; Welch, 1952; (C-strategist) phytoplankton species or to process-
Whitton, 1975). However, it was not until rel- constrained (CR-strategist) ruderals, as later con-
atively more recently that the accepted tenets firmed by the categorisation of potamoplankters
advanced by the classical studies of (such as) of Gosselain and Descy (2002).
Zacharias (1898), Kofoid (1903) and Butcher (1924) The model does not cover the many other
could be verified or challenged. New, quantita- fates that may befall river plankton or influ-
tive, dynamic approaches to the study of the ence its net dynamics, especially consumption
SEDIMENTATION 243
by filter-feeding zooplankton and (significantly) not exceeding the quotient hw /ws (Section 2.6.2):
zoobenthos, including large bivalves (Thorp and
t = h w /w s s (6.8)
Casper, 2002; Descy et al., 2003). Neither does it
cover explicitly the issue of the perennation of The point that has been made at length and on
algal inocula. However, in reviewing the avail- many previous occasions is that planktic algae
able literary evidence, Reynolds and Descy (1996) do not inhabit a static medium but one that is
argued for the importance of the effective mero- subject to significant physical movement. Forcing
plankty to centric diatoms whose life cycles of its motion by buoyancy, tide, wind, Coriolis
conspicuously include benthic resting stages effect is resisted by the viscosity of the water. This
(Stephanodiscus and Aulacoseira, both common in resistance is largely responsible for the charac-
potamoplankton generally, have proven survival teristically turbulent motion that predominates
ability in this respect). They also cited the remark- in surface waters of the sea, in lakes and rivers
able studies of Stoyneva (1994) who demonstrated (see Section 2.3). Moreover, the turbulent veloci-
the role of macrophytes as shelters and substrata ties so overwhelm the intrinsic velocities of phy-
for many potamoplanktic chlorophyte species. toplankton sinking that the organisms are effec-
The presence of such plants, in headwaters and tively entrained and randomised through tur-
in lateral dead-zones, provides a constant source bulent layers. However, it may be emphasised
of algae that can alternate between periphytic again that turbulent entrainment does not over-
and planktic habitats. This is little different from come the tendency of heavy particles to sink rel-
the proposition advanced by Butcher (1932) over ative to the adjacent medium and, in bound-
60 years before. His prognoses about the sources ary layers and at depths not pentrated by turbu-
of potamoplankton remain the best explanation lence, particles are readily disentrained and more
to the inoculum paradox and the one aspect still nearly conform to the behaviour expressed in
awaiting quantitative verification. Eq. (2.16).
Following Humphries and Imberger (1982),
the criterion for effective entrainment ( ) is set
6.3 Sedimentation by Eq. (2.19) (i.e when w s < 15[(w )2 ]1/2 ) and is
illustrated in Fig. 2.16. The depth of the mixed
6.3.1 Loss by sinking layer over which it applies (hm ) may be calculated
Most phytoplankters are normally heavier than from the Monin--Obukhov and Wedderburn for-
the water in which they are dispersed and, there- mulations. It may often be recognisable from the
fore, tend to sink through the adjacent medium. vertical gradient of density (δρw ≥ 0.02 kg m−3
The settling velocity (ws ) of a small planktic alga m−1 ) (Section 2.6.5) or, casually, from inspection
that satisfies the condition of laminar flow, with- of the vertical distribution of isotherms.
out frictional drag (Re < 0.1), may be predicted by The estimation of sinking loss rates from a
the modified Stokes equation (2.16; see Sections fully mixed water column (H) or a mixed layer
2.4.1, 2.4.2): (hm ) applies logic analogous to the dilution by
wash-out of a fully dispersed population of parti-
w s = g(ds )2 (ρc − ρw )(18ηφr )−1 m s−1 (2.16) cles subject to leakage across its lower boundary.
Moreover, as sinking loss is the reciprocal of pro-
where ds is the diameter of a sphere of identical longed suspension, it is relatively simple to adapt
volume to the alga, (ρ c − ρ w ) is the difference Eqs. (2.20--2.25) to the sequence of steps traced in
between its average density and that of the water, Eqs. (6.3--6.6) and to be able to assert that popula-
and φ r is the coefficient of its form resistance tion remaining in the column, hm , at the end of
owing to its non-sphericity; η is the viscosity of a period, t, of sustained and even sinking losses
the water, and g is the gravitational force attrac- is predicted by:
tion. In completely still water, particles may be
expected to settle completely through a column N t = N 0 e−t/t
of water of height hw within a period of time (t ) = N 0 e−ws t/ h m (6.9)
ws = 0.3m d−1 Figure 6.1 Biomass-specific

ws = 1.0 m d−1 sinking loss rates of phytoplankters
from mixed columns as a function of
ws = 3.0 m d−1
rate / d−1
their depth and intrinsic settling

rates (w s ). Three instances are
inserted to show the greater
tolerance of shallow mixing by
slow-sinking algae. Redrawn with
depth / m
Whence, the rate of change in the standing taining net intracellular carbon accumulation
population that is attributable to sedimentation and its deployment in growth are increased
(rs in Eq. 6.1) is: accordingly (Section 3.3.3; see also Sverdrup,
1953; Smetacek and Passow, 1990; Huisman
rs = w s / h m (6.10) et al., 1999). It must be recognised that, in qual-
The equation expresses the sinking loss rate sus- itative terms, larger non-motile plankters expe-
tained by a population dispersed in a mixed rience mixing that is ‘too shallow’ for growth to
layer. overcome their sinking velocity (because hm is rel-
It may be deduced that the continued resi- atively small in relation to w s in Eq. 6.10) or is ‘too
dence of non-motile particles in the pelagic is deep’ (because hm is relatively much larger than
dependent not only on having maximum entrain- hp in Eq. 5.10; see also Section 3.5.3 and Fig. 3.18)
ability (low ws ) but also on the settling velocity (O’Brien et al., 2003; see also Huisman et al.,
being small in relation to the mixed depth, hm . As 2002).
discussed in Section 2.6, the depth of mixing is
an extremely variable quantity. Disentrainment 6.3.2 Mixed depth and the population
is not a disadvantage for a swimming organism, dynamics of diatoms
especially not a large one, but non-motile organ- Included among the larger non-motile plankters
isms are highly vulnerable to variations in mixed are some of the larger freshwater desmids and,
depth (see Fig. 6.1). The growth rate of an alga especially, the diatoms of the seas and of inland
with an intrinsic sinking rate of 3.5 µm s−1 (or waters. The additional ballast that is represented
∼0.3 m d−1 ) may be able to exceed the leak- by the complement of skeletal polymerised silica
age of sinking cells across the base of a 10-m merely compounds the density difference compo-
mixed layer (rs ∼ −0.03 d−1 ) but, in just 2 m nent, (ρ c − ρ w ), in Eq. (2.16). Accepting that most
(rs ∼ −0.15 d−1 ), the sinking loss rate may become species of phytoplankton are required to be either
unsustainable. Species with greater settling rates small or motile or to minimise excess density if
experience proportionately more severe loss rates they are to counter the inevitability of mixed-
from any given mixed layer. Thus, they require layer sinking losses, it is striking how poorly the
yet deeper mixed layers to sustain positive diatoms represent all three attributes. Yet more
net growth. On the other hand, greater mix- perplexing is the fact that the planktic diatoms
ing depths quickly begin to impose constraints of freshwaters are relatively more silicified that
of inadequate photoperiod-aggregation (see Sec- their marine cousins (effectively raising ρ c ; Sec-
tions 5.4.1, 5.5.3), and the difficulties of sus- tion 1.5.2 and Fig. 1.9), while the density of many
SEDIMENTATION 245
non-saline inland waters (ρ w ) is less than that the seasonal variations in the vertical distribu-
of the sea. Thus, the density difference of some tion of Asterionella formosa in the North Basin
freshwater diatoms may exceed 100−200 kg m−3 of Windermere (Fig. 2.21). The build-up in num-
(cf. Table 2.3). How are we to explain how this bers during the month of April and, especially,
group of organisms, so relatively young in evolu- towards the maximum in May reflect the general
tionary terms, became so conspicuously success- decline in vertical diffusivity. In the end, a near-
ful as a component of the phytoplankton of both surface concentration maximum is reached, fol-
marine and fresh waters, when it has not only lowed by a rapid decline. In this instance, recruit-
failed to comply to Stokes’ rules but has actually ment through growth was impaired by nutrient
gone against them by placing protoplasts inside deficiencies (Lund et al. cited critical silicon lev-
a non-living box of polymerised silica? There els but phosphorus is now seen likely to have
is no simple or direct answer to this question, been the more decisive; see Section 5.5.2). How-
although, as has been recognised, sinking does ever, it is quite clear from the isopleths that
have positive benefits, provided that subsequent the decline in concentration in the upper 10
generations experience frequent opportunities m or so is extremely rapid. It is compensated,
to be reintroduced into the upper water col- to an extent, by a temporary accumulation in
umn (Section 2.5). In general, however, many of the region of the developing pycnocline. This
the ecological advantages of a siliceous exoskele- behaviour is entirely consistent with elimination
ton were experienced first among non-planktic through sedimentation from the mixed layer and
diatoms. As the diatoms radiated into the plank- slow settlement through the weak diffusivity of
ton, morphologies had to adapt rapidly: siliceous the metalimnion, revealed in the case of non-
structures mutated into devices for enhancing living Lycopodium spores (Fig. 2.20). Particles con-
form resistance and entrainability within turbu- tinue to settle through the subsurface layers for
lent eddies (Section 2.6). As was demonstrated many weeks after the population maximum and,
in the case of Asterionella in the experiments indeed, after the surface layer has become effec-
of Jaworski et al. (1988) (see also Section 2.5.3), tively devoid of cells.
the configuration of the structures is overriding. Heavy sinking losses are not exclusive to
Despite order-of-magnitude variations in colony nutrient-limited diatom populations. The sensi-
volume and dry mass, as well as an approxi- tivity of the population dynamics of diatoms to
mate twofold variation in cell density, the sink- the onset and stability of thermal stratification in
ing behaviour of Asterionella remains under the Crose Mere, a small, enriched lake in the English
predominating influence of colony morphology. north-west Midlands, rather than to nutrient lim-
The corollary must be that the advantage of itation, was shown by Reynolds (1973a). Diatoms
increased form resistance, and its benefits to such as Asterionella, Stephanodiscus and Fragilaria
entrainability, is greater than the disadvantage of were lost from suspension soon after the lake
increased sinking speed incumbent upon coeno- stratified in late spring, even though the con-
bial formation. The counter constraint, however, centrations of dissolved silicon and phosphorus
is that these diatoms are continuously dependent remained at growth-saturating levels. Lund (1966)
upon turbulence to disperse and to randomise had already argued for the positive role of tur-
them within the structure of the surface-mixed bulent mixing in the temporal periodicity of
layer. As predicted by Eq. (6.10), positive popula- Aulacoseira populations. He showed, in a field-
tion recruitment is always likely to be sensitive to enclosure experiment in Blelham Tarn (inciden-
the absolute mixed-layer depth (Reynolds, 1983a; tally, the one that inspired the construction of
Reynolds et al., 1983b; Huisman and Sommeijer, the renowned tubular enclosures in the same
2002). lake) that the periodicity could be altered readily
The impact of this interplay between settle- by superimposing episodes of mechanical mixing
ment and population dynamics of diatoms on by aeration (Lund, 1971).
their distribution in space and time is elegantly A little later, the eventual Blelham enclosures
expressed in the study of Lund et al. (1963) of (Fig. 5.11) were the site of numerous quantitative
Figure 6.2 Instances in the loss

from suspension of Asterionella cells
in Blelham Enclosure A during 1980,
in response to intensifying thermal
stratification (shown by the
temperature plots). Algal
concentration is sampled in 1-m
integrating sampler (Irish, 1980) and
counted as an average for a 1- or
2-m depth band. The vertical arrows
represent the depth of Secchi-disk
extinction on each occasion.
studies of the fate of phytoplankton popula- m d−1 (equivalent to 11.8 µm s−1 ). The data are
tions. One early illustration, cited in Reynolds plotted in Fig. 6.3.
(1984a), shows the depletion by settlement of The accelerated sinking loss was contributed,
a thitherto-active Asterionella population, follow- in part, by an accelerated sinking rate. This was
ing the onset of warm, sunny weather and the not unexpected. Reynolds and Wiseman (1982)
induction of a stable, near-surface stratification, had noted the altered physiological condition
and despite the availability of inorganic nutri- of the cells at the time, both in the plankton
ents added to the enclosure each week (Fig. 6.2). and in the sediment traps, drawing attention
In a further season-long comparison of loss pro- to the contracted plastids and ‘oily’ appearance
cesses in these enclosures (Reynolds et al., 1982a), of the contents. They attributed the changes to
during which changes in extant numbers, ver- the sudden increase in insolation of cells caught
tical distribution, growth (as a function of sili- in a stagnating and clarifying (cf. Fig. 6.2) epil-
con uptake) and sedimentary accumulation rates imnion at the same time that temperature and
into sediment traps and at the enclosure bot- light intensity were increasing. They suggested
tom were all independently monitored, a steady that the changes were symptomatic of photoin-
‘leakage’ of Asterionella cells was demonstrated hibition. When Neale et al. (1991b) made simi-
over the entire cycle of net growth and attri- lar observations on diatom populations in other
tion (see Table 5.5 and Reynolds and Wiseman, lakes, they made the similar deduction. A positive
1982). In the Blelham Enclosure B, Asterionella feedback implied by the sequence of more insola-
increased at a rate of 0.147 d−1 during its main tion → more stratification → more inhibition →
phase of growth, net of sinking losses calcu- faster sinking rates → faster sinking loss rates has
lated to have been ∼0.007 d−1 . The sedimenting a satisfying ring of truth. However, a modified
cells intercepted by the traps were calculated to interpretation would see the accelerated sinking
have been sinking at an average rate of (w s = ) rate as a withdrawal of the vital mechanism of
0.08 m d−1 (just under 1 µm s−1 ) through a water reducing sinking rate (see Section 2.5.4) for the
column (hm = ) 11.7 m. As the population reached very positive purpose of escaping the high levels
its maximum, the net rate of increase slowed of near-surface insolation.
(to 0.065 d−1 ) but the sinking loss rate remained The bulk ‘production-and-loss budgets’ com-
steady (−0.007 d−1 ). However, shortening of the piled by Reynolds et al. (1982a) for phytoplank-
mixing depth led to an accelerated rate of sink- ton populations in the Blelham Enclosures and
ing loss (to −0.044 d−1 ; from a mixed depth of exemplified in Table 5.6 offer a clear account of
now only 7.5 m, a faster sinking rate of w s = 0.33 the the fate of the total production. In the exam-
m d−1 is also implied). More remarkably, as the ple given, 81% (confidence interval, 70--95%) of
epilimnion shrank to 4 m, the loss rate then rose the Asterionella formosa produced in Enclosure B
to −0.242 d−1 , sustained by a sinking rate of 1.02 in the spring of 1978 constituted the observed
SEDIMENTATION 247
recently recruited material) was 92% at the begin-

ning of April. By the end of the month, it had
fallen to 67%, to <2% by the end of July and to
zero by the first week in September.
As part of the same investigation, Reynolds
and Wiseman (1982) compared the rates of
production, sedimentary fluxes and sediment
recruitment of several other species forming
major populations in the same enclosures. Of
the estimated summer production of another
diatom, Fragilaria crotonensis, at least 49% (sta-
tistically, possibly all) of the production was
recruited to the sediments. In contrast, sedimen-
tation could explain the fate of no more than
4% of the observed population maxima of Ankyra,
Chromulina or Cryptomonas. Intermediate between
the extremes of heavy diatoms and nanoplanktic
unicells, sediment and trap recoveries of Eudo-
rina accounted for 55 ± 15% of the maximum
standing crops. For Microcystis, the sedimentary
behaviour was strongly seasonal, increasing from
8% to 100% through the autumn.
From the measurements of the production
and eventual fate of phytoplankton in confined,
Figure 6.3 Net increase and attrition of an Asterionella
flat-bottomed Blelham enclosures, at least, the
population cells in Blelham Enclosure B during spring 1978.
(a) Changes in the instantaneous areal cell concentrations in assertion that most of the larger diatoms are
the water to 3, 5 and 11 m; (b) changes in the silicon-specific destined to be lost to sedimentation is strongly
replication rate (r Si ) and the net rate of population change supportable. Scaling up to larger and deeper sys-
(rn ); the hatched areas correspond to the rate of population tems, subject to significant horizontal diffusive
loss, almost wholly to sinking. Redrawn with permission from transport, the deduction requires some caution.
Reynolds (1997a). In a 2-year study of sedimentary fluxes in the
South Basin of Windermere (maximum depth
42 m), Reynolds et al. (1982b) found good, order-of-
maximum. Around 95% (confidence interval, 72- magnitude agreement between the annual fluxes
-123%) of the production was recruited intact to into deep sediment traps and the maximal stand-
the sediment. The proportion of the cells pro- ing crops of five species of planktic diatom (Asteri-
duced that was estimated to have been lost to onella formosa, Aulacoseira subarctica, Cyclotella
herbivores was probably <6% (see Sections 6.4.2, praetermissa, Fragilaria crotonensis, Tabellaria floccu-
6.7). The use of the adjective ‘intact’ is taken losa var. asterionelloides) and two of desmid (Cosmar-
to include cells that may well have been dead ium abbreviatum, Staurastrum cf. cingulum). Inter-
by the time they reached the sediment surface. estingly, the magnitude of the fluxes (in num-
Judged from weekly recoveries from sediment bers of cells m−2 d−1 ) varied with the size of
traps placed about ∼1 m above the sediment extant poulations but measurable fluxes to depth
(and to which preservative was added), Reynolds persisted through most of the year. This is pre-
and Wiseman (1982) observed that the propor- sumed to reflect the relative proportion of the
tion of live cells was always greater than 89% particle settling rates to the vertical distance to
throughout the course of the population rise and be traversed; this also fits with the observations
decline. The proportion of live cells in the superfi- of Lund et al. (1963) for the North Basin and the
cial sediment (supposedly dominated by the most distribution of population isopleths plotted in
Fig. 2.21. Incidentally, the proportion of live Aste- netic diatoms are subject. In most other phyto-
rionella cells trapped fell from around 95% at the plankton, greater proportions are either eaten
time of the May population maximum to just 3% or decompose long before they reach the sedi-
in August and September. In the 100-or-so days ment. The deduction concurs with the studies of
that it takes some diatoms to settle through 40 losses made by Crumpton and Wetzel (1982) and
m, many must perish, leaving only the empty with that of Hillbricht-Ilkowska et al. (1979) in
frustules to continue downwards. Jezioro Mikolajske, Poland, on the seaonal vari-
In contrast to the diatoms, the sedi- ations in the main sinks of limnetic primary
mentary fluxes of three colonial chlorophyte products.
species (Coenochloris fotti, Pseudosphaerocystis lacus- The sensitivity of marine diatom dynamics
tris, Radiococcus planctonicus), three Cyanobacte- to mixed-layer depth is not so clearly defined.
ria (Anabaena flos-aquae, Woronichinia naegeliana, On the one hand, net population increase is
Pseudanabaena limnetica) and the dinoflagellate dependent upon an enhancement in insolation
Ceratium hirundinella were 1--3 orders of magni- above thresholds which may be lower than for
tude smaller than the potential of the maximum many other marine species (Smetacek and Pas-
standing crop. All these species either sink very sow, 1990) but the diminution of the mixed layer
slowly or they have sufficient motility to avoid in the sea to the 1--3 m that may be critical to net
being sedimented for long periods. Cryptomon- diatom increase is inconclusively documented.
ads and nanoplankters were virtually unrecorded Nevertheless, oceanic diatom populations expe-
in any trap catches; they are presumed to rience considerable sinking losses that may be
have been subject to loss processes other than sustained only at or above certain levels of pro-
settlement. ductivity. It is inferred that these are dependent
These various findings supported the earlier upon adequate physical and chemical support
deductions of Knoechel and Kalff (1978), who had (Legendre and LeFèvre, 1989; Legendre and
applied a dynamic model to compare the effects Rassoulzadegan, 1996; see also Karl et al., 2002).
of measured rates of growth, increase and set- As to the question posed by Huisman et al.
tlement in order to calculate sinking loss rates (2002) about the long-term persistence of sink-
of planktic populations in Lac Hertel, Canada. ing phytoplankton, we have shown that there
Their calculations showed that the rates of sink- are obvious short-term benefits in being able
ing loss were sufficient to explain most of the dis- to escape surface stagnation and resultant dam-
crepancy between growth and the contemporane- aging levels of insolation in the near-surface
ous rate of population change, be it up or down. waters (Reynolds et al., 1986). Provided there is
They were also able to provide quantified sup- an opportunity for surviving propagules to be re-
port for the idea that, whatever fate may befall established within the photosynthetic range, the
them (nutrient, especially silicon, exhaustion, sooner may the longer-term benefit of population
grazing, parasitism), planktic diatoms remain re-establishment be realised. Particle aggregation
crucially sensitive to the intensity and extent and, especially, the formation of ‘marine snow’
of vertical mixing. Other workers who espoused (Alldredge and Silver, 1988) may contribute effec-
this explanation for the seasonal fluctuations in tively to accelerated sinking and to the escape
diatom development and abundance in limno- from high-insolated surface layers. Aggregation
plankton include Lewis (1978a, 1986), Viner may also serve to provide microenvironments
and Kemp (1983), Ashton (1985) and Sommer that slow down the rate of respirational con-
(1988a). sumption and resist frustular dissolution of sil-
There is now also ample evidence to sup- icon (Passow et al., 2003). The mechanisms of
port the qualitative contention of Knoechel and accelerated sinking may also add to the longevity
Kalff (1975) that sedimentation is a key trig- of clone survival and facilitate the improved
ger to the seasonal replacement of dominant prospect of population re-establishment when
diatoms by other algae. It is also plain that sed- more suitable growth conditions are encoun-
imentation is the principal loss to which lim- tered.
SEDIMENTATION 249
6.3.3 Accumulation and resuspension of ticulate organic matter, the exuviae and exc-
deposited material reta of aquatic animals and a rain of sedi-
As has already been discussed, settling is not menting phytoplankters, especially of non-motile
exclusively a loss process in the population diatoms.
dynamics of phytoplankton: the recruitment of Several studies have attempted to focus on
resting propagules to the bottom deposits is the nature of the freshly sedimented material
recognised to constitute a ‘seed bank’ from which in lakes and its immediate fates. For a time,
later extant populations of phytoplankters may the newest recruited material remains substan-
arise (see Section 5.4.6). For this to be an effec- tially uncompacted and floccular, like a fluff, on
tive means of stock perennation and mid- to the immediate surface. It comprises live or mori-
long-term persistence in a given system, how- bund vegetative cells, often bacterised or beset
ever, there has to be a finite probability of set- with saprophytic fungal hyphae, and resembles
tled material both surviving on the sediments on a smaller scale, the structure of ‘marine snow’
and, thence, of re-entering the plankton. The (Alldredge and Silver, 1988; see Section 6.3.2).
species-specific regenerative strategies of phyto- As its substance diminishes, however, it does
plankton -- roughly their ability to survive at become slowly compressed by later-arriving mate-
the bottom of the water column and the means rial. At the base of the semifluid layer, the same
of ‘escape’ to the overlying water column -- are materials are progressively lost to the permanent
extremely varied, ranging from the conspicuous sediment (Guinasso and Schink, 1975): compact-
production of morphological and/or physiologi- ing, losing water, perhaps leaching biominerals,
cal resting stages, with an independent capac- the first stages of sediment diagenesis and forma-
ity for germination, regrowth and reinfection of tion are engaged.
the water column (as in the case of Microcystis Accordingly, the manner in which strictly
or Ceratium), through a range of resting cysts ordered, laminated sediments might flow from
and stages whose re-establishment in the water the sequenced deposition of specific phytoplank-
depends upon still-suspended or resuspended ton populations seems obvious. However, direct
propagules encountering tolerable environmen- sampling of the semifluid layer from intact cores
tal conditions (as is true for akinetes of nosto- of the sediment water interface (Reynolds and
calean Cyanobacteria, certain species of volvo- Wiseman, 1982, used a syringe inserted into pre-
calean and chrysophyte resting cysts and the dis- drilled plastic tubes fitted to a Jenkin surface-
tinctive resting stages of centric diatom), to those mud sampler, as described in Ohnstad and Jones,
that seem to make virtually no such provision at 1982) reveals that sedimenting material under-
all (see Section 5.4.6). goes a kind of sorting process. Once recruitment
In most instances, the settlement of vege- to the semifluid layer from the water column is
tative crops should be regarded as terminal. effectively complete, its presence in the semifluid
Vegetative cells sinking onto deep, uninsolated layer is found to decay exponentially. Moreover,
sediments have little prospect but to slowly the rates of dilution from the semifluid layer
respire away their labile carbohydrates, pend- are not uniform but vary interspecifically, accord-
ing depth. Resting cysts may remain viable for ing to size and shape (Haworth, 1976; Reynolds,
many years (64 a is a well-authenticated claim 1996b): long cells of Asterionella, filaments of
of viability of Anabaena akinetes: Livingstone Aulacoseira and chains of Fragilaria are diluted less
and Jaworski, 1980) but without the mechanical rapidly from the semifluid layer than centric uni-
resuspension of the resting spores in insolated, cells of Cyclotella or Stephanodiscus.
nutrient-replete water, the reinfective potential Relative persistence in the surface layer
remains unrealised. Once settled to the bottom improves the prospect of live specimens being
of a water column, the most likely prospect is restored to suspension in the water column, sup-
progressive burial by the subsequent sedimen- posing that the physical penetration of adequate
tary recruitment of further particulate material, resuspending energy obtains. In general, friction
including fine, catchment-derived silts and par- in the region of the solid sediments creates a
velocity gradient and a boundary layer of reduced greater than 5 m, sedimenting material accumu-
water velocities, in which freshly settled plank- lates and builds up in layers, undergoing dia-
ton can accumulate (see Section 2.7.1). Resuspen- genesis under substantially anoxic conditions.
sion is thus dependent upon the application of Neither live vegetative cells nor most resting
sufficient turbulent shear force to compress the spores enjoy much prospect of return to suspen-
boundary layer to the dimensions of the set- sion and regeneration. In contrast, similar mate-
tled particles or even beyond the resistance of rials settling onto shallow sediments are liable
the unconsolidated sediment to penetration by a to resuspension. The viable fractions (vegetative
shear force, by then competent to entrain and cells, resting spores) may well fulfil their infec-
resuspend it (Nixon, 1988). Quantitative obser- tive potential and contribute directly to the estab-
vations confirm the intuition that shallow sed- lishment of extant, vegetative populations. This
iments are rather more liable to resuspension has been many times observed in the case of
than sediments beneath a substantial column of Aulacoseira populations (Lund, 1954, 1966, 1971)
water, although the actual depth limits vary with and is inferred on other occasions involving
sediment type and the energy of forcing (Hilton, other species (Carrick et al., 1993; Reynolds et al.,
1985). In many small lakes, sediments at a depth 1993a). For the non-viable detritus, including
greater than 5 m beneath the water surface are empty diatom frustules, redeposition is the most
protected from wave action and from most wind- likely consequence but with a finite proportion
generated shear. In the short to mid term, resus- settling into deeper water. This is precisely the
pension may require physical forcing of seismic mechanism of the process of ‘sediment focusing’
proportions, or depend upon disturbance by bur- (Lehman, 1975) whereby fine particulate material
rowing invertebrates or foraging behaviour of is moved progressively away from lake margins
fish or diving animal (Davis, 1974; Petr, 1977). In and towards greater basin depths (Hutchinson,
contrast, shallow sediments (substantially <5 m) 1941; Likens and Davis, 1975; Hilton, 1985)
may be rather more routinely exposed to resus-
pension of sediment and, incidentally, the redis-
persion of sediment interstitial water that may 6.4 Consumption by herbivores
be relatively enriched, with respect to the open
water, with nutrients released in decomposition Sharing an apparently refugeless, open-water
(see also Section 8.3.4). In the Blelham enclo- habitat with a variety of phagotrophic animals,
sures (see Fig. 5.11), very little resuspension of phytoplankton is generally vulnerable to severe
live phytoplankton, resting spores or even empty physical biomass losses and, at best, to the
diatom frustules was observed from the universal dynamic drain on the potential recruitment
deep sediments of Enclosures A or B but it was of biomass. In fact, there are many types of
observed on numerous occasions in the graded consumer, each with differing food preferences
Enclosure C (Reynolds, 1996b). Moreover, distur- and habitat demands, making for an extremely
bance or removal of the semifluid sediment from wide range of possible outcomes. The subject of
the shallow-water station, CS (Fig. 5.11, depth food and feeding is, indeed, a broad one, and
∼4.5 m), occurred at such times, whereas, the rather beyond the remit of the present chapter,
deeper station, CD (depth ∼12.5 m) was exempt the focus of which will remain trained on the
from this. In the wake of such resuspension dynamic consequences for the producers. How-
events, material was perceived to resettle uni- ever, even this modest ambition must take some
formly at both stations. Over a series of resus- account of the biologies of the consumers and
pensions, a net transport of once-settled material how their impacts fluctuate in time and space.
from shallow areas to deep sites was deduced. What follows here is necessarily selective,
So far as the accumulation of sediment- emphasising those aspects of zooplanktic biol-
ing phytoplankton is concerned, near-permanent ogy which relate to phytoplankton dynamics and
deposition follows analogous patterns to non- to the shaping of pelagic ecosystems. Numerous
living particulate matter. At depths typically books and monographs describing the biology
CONSUMPTION BY HERBIVORES 251
and ecology of particular zooplankton groups approach in preference to a taxonomic one.

are available. Of the more general accounts, There are important differences in the composi-
none had equalled those of Hutchinson (1967) or tion, ecological function and key selective mech-
Raymont (1983), until the recent publication of anisms between the nano-/micro-planktic and
Gliwicz’s (2003a) superb overview, to which the meso-/macro-planktic components and, indeed,
reader is happily referred. The emphasis here among the principal types of mesozooplanktic
is on crustacean herbivory, with only acknowl- association.
edgement of the part played by small herbiv-
orous fish in cropping phytoplankton in (usu- Protistan microzooplankton
ally) small tropical lakes (see Fernando, 1980; In terms of numbers, the most abundant and
Dumont, 1992). The present account also recog- most common zooplankters, both in lakes and
nises that planktic primary products are con- in the sea, belong to the category of nano-/
sumed not only as the particulate foods of her- microzooplankton. This includes all planktic het-
bivores but also as the dissolved substrates of erotrophs in the size range 2--200 µm, with
aquatic microorganisms. the exception of the bacteria, actinomycetes and
moulds (Sorokin, 1999). Rather than be pedan-
6.4.1 The diversity of pelagic phagotrophs tic about the nano--micro separation, it is con-
and their foods venient to follow Sorokin’s (1999) usage of the
Zooplankton comprises small animals suspended word ‘microzooplankton’ to apply to all het-
in the water. Some (nanozooplankton and micro- erotrophic protistans and metazoans smaller
zooplankton, all <200 µm) are truly planktic than 200 µm. This then includes representatives
in the sense of being fully embedded in the of protistan groups already listed as phytoplank-
eddy spectrum (Section 2.3.4) but even most ton in Table 1.1 (especially Chrysophyta and Dino-
larger forms of mesoplankton (0.2--2 mm) are phyta). As pointed out in Section 1.3, many of
too small and too weak to escape entrainment these are photoautotrophs with a phagotrophic
by open-water currents. A characteristic of all capability (that is, that they are mixotrophic) but
zooplankters is that they are partly or wholly the mesoplanktic, colourless marine dinoflag-
phagotrophic -- much or all of their organic ellates, such as Noctiluca and Oxyrrhis, are
carbon and energy requirements are satisfied obligate phagotrophic consumers, feeding on
by feeding on live or detrital organic parti- nanoplankters, including the haptophyte Prym-
cles. Another feature of zooplankton, which, nesium (Tillmann, 2003). Freshwater mixotrophs
broadly, is shared with phytoplankton, is the in the nanoplanktic size range include Chromu-
cosmopolitan distribution of many genera and lina, Chrysococcus and Ochromonas. Chrysochromu-
even some species. This may seem obvious in lina spp. and Prymnesium spp. fulfil this role in
contiguous seas but it applies no less to the the sea (Riemann et al., 1995). They ingest par-
plankton of inland waters. The main difference ticles in the picoplanktic size range, including
between the zooplankton of seas and lakes is bacteria and algae. The nanoheterotrophs also
the much poorer phylogenetic representation in include numerous small flagellated protists, clas-
the latter. Whereas certain life-history stages of sified in Table 6.1 as Zoomastigophora. These
certain species from almost every animal phy- have well-developed cytostomes and they are able
lum occur in the marine plankton, limnoplank- to ingest substantial food particles, up to about
ton mainly comprises protists, rotifers and crus- 5--8 µm in size. Free-living bodonids and pro-
taceans. Major groups and representative plank- tomonadids are represented in the nanoplank-
tic genera are summarised in Table 6.1. ton of lakes and seas, where they can be effec-
In briefly surveying the diversity of zooplank- tive consumers of nanophytoplankton. The group
ton within the context of its quantitative impacts includes the choanoflagellates and bicosoecids
on the phytoplankton and the flow of carbon that are usually attached to the surfaces of larger
through the pelagic towards its larger metazoan plankters such as diatoms. At this scale, deep
beneficiaries, it is useful to adopt a functional within the viscous range of the eddy spectrum
Table 6.1 Zooplankton in marine and freshwater habitats
Phylum: Zoomastigophora
Four orders of free-living and epiphytic or epizoic flagellates.
Includes: Bodo, Monas, Peranema (marine and freshwater);
Bicosoeca, Salpingooeca and Monosiga
Phylum: Dinophyta
Several families of marine dinoflagellate are mixotrophic or primarily
heterotrophic.
Includes: Dinophysis, Noctiluca, Oxyrrhis, Protoperidinium
Phylum: Rhizopoda (Sarcodina)
Four main divisions.
Order: AMOEBINA
Naked, lobose protists.
Plantic genera include: Asterocaelum, Pelomyxa (freshwater)
Order: FORAMINIFERA
Amoeboid protists with non-calcareous shells.
Includes: Globigerina (marine); Arcella, Difflugia (freshwater)
Order: RADIOLARIA
Marine planktic sarcodine protists having central capsule and usually a
skeleton of siliceous spicules.
Includes: Acanthometra
Order: HELIOZOA
Mostly freshwater sarcodines with axopodia and, typically, vacuolated
cytoplasm and a siliceous skeleton.
Includes: Actinophrys
Phylum: Ciliophora
Class: CILIATA
Non-amoeboid protists that possess cilia during part of their life cycle:
several planktic orders, including:
Order: HOLOTRICHA
Uniformly ciliated.
Includes: Colpoda, Prorodon, Pleuronema and freshwater Nassula
Order: SPIROTRICHA
Ciliates posessing gullet and undulating membrane.
Includes many common genera of marine and freshwaters:
Euplotes, Halteria, Metopus, Strobilidium, Strombidium, Stentor, Tintinnidium
Order: PERITRICHA
Ciliates, usually attached to surfaces. Cilia reduced over body and
confined to oral region.
Includes: Epistylis, Vorticella, Carchesium
Class: SUCTORIA
Ciliophorans lose cilia in adult stage. Possess one or more suctorial
tentacles.
Includes: Acineta
Phylum: Porifera
Amphiblastula larvae temporally in marine plankton.
Table 6.1 (cont.)
Phylum: Coelenterata
Subphylum: Cnidaria
Coelenterates with stinging nematocysts. Several orders have genera
that live, or appear, in (mostly marine) plankton
Class: HYDROZOA
Order: LEPTOMEDUSAE
Hydrozoan coelenterates with horny perisarc. Medusa stage in plankton.
Includes: Obelia, Plumularia
Order: ANTHOMEDUSAE
Hydrozoan coelenterates with horny perisarc that does not cover polyp
base. Medusa stage in plankton.
Includes: Hydractinia
Order: HYDRIDA
Hydrozoan coelenterates without a medusa.
Includes: Hydra, young hydroids of which disperse through
freshwater plankton
Order: TRACHYLINA
Hydrozoan coelenterates with a relatively large medusa and minute
hydroid stage.
Includes: freshwater Limnocnida, Craspedacusta
Order: SIPHONOPHORA
Large, free-moving colonial hydrozoans.
Includes: Velella, Physalia
Class: SCYPHOZOA
Cnidaria that exist mostly as medusae. Several orders.
Includes: Aurelia, Cyanea, Pelagia
Subphylum: Ctenophora
Swimming coelenterates lacking nematocysts
Class: TENTACULATA
Ctenophores with tentacles (‘sea gooseberries’).
Includes: Pleurobrachia
Class: NUDA
Ctenophores lacking tentacles.
Includes: Beroë
Phylum: Platyhelminthes
Acoelomate metazoans (flatworms, many parasitic) with a few free-living
representatives in the marine plankton.
Class: TURBELLARIA
Includes: Convoluta, Microstomum
Phylum: Nemertea
Flattened unsegmented worms with a ciliated ectoderm and eversible
proboscis. Several orders, one with planktic genera.
Order: HOPLONEMERTINI
Pilidium larvae of several genera are dispersed as marine plankton. Some
genera remain bathypelagic, beyond the continental shelf, in their adult
stages.
Includes: Pelagonemertes
Table 6.1 (cont.)
Phylum: Nematoda
Unsegmented round worms. Some shelf-water species have been
reported but these may not be truly planktic.
Phylum: Rotatoria
Acoelomate metazoans with planktic genera widespread in the sea and
in lakes. Most planktic forms belong to one order, mainly in fresh waters.
Order: MONOGONONTA
Sub-order: Flosculariacea
Free-swimming, soft-bodied.
Includes: Conochilus, Filinia.
Sub-order: Ploima
Free-swimming, usually with firm lorica but some illoricate.
Includes: Asplanchna, Brachionus, Kellicottia, Keratella, Notholca,
Synchaeta, Trichocerca
Phylum: Gastrotricha
Minute unsegmented acoelomate metazoans. May be encountered in
plankton of small freshwater bodies.
Includes:Chaetonotus
Phylum: Annelida
Coelomate segmented worms. One class has planktic representatives.
Class: POLYCHAETA
Several planktic genera, one of which is very well adapted to a pelagic
existence. Trochospere larvae of some polychaetes are also temporarily
resident in the plankton.
Includes: Tomopteris
Phylum: Crustacea
Large group of segmented, jointed-limbed arthropods, with many

planktic representatives.
Class: BRANCHIOPODA
Free-living small crustaceans with at least four pairs of trunk limbs,
flattened, lobed and fringed with hairs (phyllopods).
Two orders have planktic genera.
Order: ANOSTRACA
Branchiopodans lacking a carapace, phyllopods numerous.
Includes Chirocephalus, Artemia
Order: DIPLOSTRACA
Branchiopodans with a compressed carapace enclosing fewer than 27
pairs of phyllopods. Two sub-orders, one of which (Cladocera) includes
several genera important in the plankton of sea and in lakes:
Evadne, Podon (marine); Sida, Diaphanosoma, Holopedium, Bosmina,
Daphnia, Ceriodaphnia, Moina, Simocephalus, Chydorus, Bythotrephes,
Leptodora (freshwater)
Class: OSTRACODA
Free-living small crustaceans with a bivalve shell, few trunk limbs, none
being phyllopods.
Includes: Gigantocypris (marine), Cypris (freshwater)
Table 6.1 (cont.)
Class: COPEPODA
Free or parasitic crustaceans lacking carapace or any abdominal limbs.
Of some seven orders, two provide most free-living planktic forms. A
third is well represented in the benthos and species are encountered in
the pelagic.
Order: CYCLOPOIDEA
Copepods with short antennules of <17 segments.
Includes: Oithona (marine); Mesocyclops, Tropocyclops (freshwater)
Order: CALANOIDEA
Copepods with long antennules of >20 segments. Major group of
mesozooplankters in the sea and in many lakes.
Includes: Calanus, Temora, Centropages (marine); Eudiaptomus,
Eurytemora, Boeckella (freshwater)
Order: HARPACTICOIDEA
Benthic copepods with similar thoracic and abdominal regions and very
short first antennae.
Includes: Canthocamptus (freshwater)
Class: BRANCHIURA
Crustacea, suctorial mouth, capacace-like lateral expansion of the head.
Temporary parasites of fish.
Includes: Argulus (freshwater and estuaries)
Class: CIRRIPEDIA
Crustacea, sedentary and plated as adults; wholly marine; several orders,
one of which (Order THORACICA: barnacles) whose cypris larvae are
dispersed in the marine plankton.
Includes: Balanus
Class: MALACOSTRACA
Mostly larger crustacea, usually with distinct thoracic and abdominal
regions. Thorax generally covered by a firm carapace; most abdominal
segments bear appendages. Five major orders, two of which have
typically planktic genera or ones whose juvenile stages are passed in the
pelagic.
Order: LEPTOSTRACA
Primitive malacostracans with large thoracic carapaces. One genus is
present in the bathypelagic beyond the continental shelf.
Includes: Nebaliopsis (marine)
Order: HOPLOCARIDA
Benthic malacostracans with shallow carapace fused to three thoracic
somites. Larvae are temporary entrants to the plankton of warm seas.
Includes: Squilla
Order: PERACARIDA
Malacostracans in which the carapace is fused with no more than four
thoracic segments. Several suborders include:
Sub-order: Mysidacea
Peracaridans with well-formed carapace. Mostly marine, ‘opossum
shrimps’ are typically planktivorous in the deep layers of shallow seas.
Some relict or invasive species in fresh waters.
Includes: Leptomysis, Gastrosaccus (marine); Mysis (freshwater)
Table 6.1 (cont.)
Sub-order: Cumacea
Mostly benthic in marine sublittoral sand and mud, whence animals are
entrained into pelagic samples.
Includes: Diastylis
Sub-order: Isopoda
Peracaridans with a carapace covering three or four thoracic segments
only. Most are not planktic. An exception is:
Eurydice (marine)
Sub-order: Amphipoda
Peracarida with no carapace. Body laterally compressed. Most of the
shrimp-like animals are littoral or benthic but some euplanktic genera.
Includes: Apherusa (marine); Macrohectopus (freshwater genus of
Lake Baykal)
Order: EUCARIDA
Malacostracans in which the capace is fused to all thoracic segments.
Two main sub-orders:
Sub-order: Euphausiacea
Eucarida in which the maxillary exopodite is small and the thoracic limbs
do not form maxillipeds. Large meso- and macroplankters, including krill,
important as a food for whales.
Includes: Euphausia, Nyctiphanes (marine)
Sub-order: Decapoda
Eucarida in which the maxillary exopodite is large (the scaphognathite)
and the first three pairs of thoracic limbs are modified as maxillipeds and
the next five as ‘legs’. These are the lobsters, prawns and crabs. Zoea,
megalopa and phyllosoma larvae are temporary entrants to the marine
plankton.
Includes: Carcinus, Palinurus
Phylum: Arthropoda
Class: HEXARTRA (INSECTA)
Larvae of two orders show distinct adaptations to planktic existence.
Order: MEGALOPTERA
First (especially) and second instars of alder flies are dispersed in the
limnoplankton.
Includes: Sialis
Order: DIPTERA
Larvae of the Culicidae are typically associated with the littoral of lakes
and many feed in very small, still bodies of water. Larvae of the subfamily
Chaoborinae (or Corethrinae) are adapted for a larval life in the open,
deeper waters of small lakes and lagoons.
Phylum: Mollusca Includes: Chaoborus, Pontomyia
Unsegmented coelomates with a head, a ventral foot and a dorsal

visceral hump, developed to varying extents. Of the five major classes,
only two have typically planktic representatives.
Class: GASTROPODA
Slugs and snails. Molluscs with distinct head and tentacles.
Table 6.1 (cont.)
Order: PROSOBRANCHIATA
Gastropods in which adults show torsion. Some marine genera
dispersed by pelagic trochophore larvae.
Includes: Patella
Order: OPISTHOBRANCHIATA
Gastropod line showing secondary ‘detorsion’ and shell reduction.
Several marine genera of pteropods or ‘sea butterflies’.
Includes: Limacina, Clio
Class: LAMELLIBRANCHIATA
Bilaterally symmetrical molluscs more or less enclosed in a hinged bivalve
shell. Certain genera dispersed by pelagic trochophore or veliger larvae.
Includes: Ensis, Ostrea [trochophore] (marine); Dreissenia [veliger]
(freshwater).
Class: CEPHALOPODA
Bilaterally symmetrical molluscs, with well-developed head, and foot
modified into a crown of tentacles. Some marine pelagic species release
paralarvae into the plankton.
Includes: Loligo
Phylum: Chaetognatha
Slender, differentiated coelomates with distinctive head, eyes and
chitinous jaws. Chaetognathes (arrow worms) are exclusive to the
marine plankton, where they are carnivorous.
Includes: Sagitta
Phylum: Ectoprocta
Unsegmented sedentary coelomates, usually colonial. One division is

exclusive to marine habitats and reproductive propagules are dispersed
as trochophore-like cyphonautes larvae (freshwater ectoprocts produce
dispersive statoblasts).
Order: PHYLACTLAEMATA
Marine ectoprocts producing planktic cyphonautes larvae.
Includes: Flustra
Phylum: Phoronidea
Small group of unsegmented tubicolous coelomates with affinities to the
ectoprocts. The free-swimming actinotrocha larvae are encountered in
the marine plankton.
Includes: Phoronis
Phylum: Echinodermata
Large group of coelomates in which adults show radial symmetry.
Modern genera are exclusively marine, usually littoral or benthic but
many are dispersed by planktic larvae. One genus known to have
planktic adults.
Class: ASTEROIDEA
Starfish and seastars; pentagonal free-living benthic and littoral
echinoderms. Many genera dispersed as pelagic bipinnaria larvae.
Includes: Asterias
Table 6.1 (cont.)
Class: OPHIUROIDEA
Brittle stars; discoid free-living benthic and littoral echinoderms with five
radial arms. Many genera dispersed as pelagic pluteus larvae.
Includes: Ophiura
Class: ECHINOIDEA
Sea urchins; globular or discoid armless echinoderms. Many genera
dispersed as pelagic pluteus larvae.
Includes: Echinus
Class: HOLOTHUROIDEA
Sea cucumbers; sausage-like, armless echinoderms. Many genera

dispersed as pelagic auricularia larvae. One adult is bathypelagic in the
South Atlantic Ocean.
Includes: (larval) Holothuria, adult Pelagothuria
Phylum: Chordata
Coelomate animals with notochord and gill slits and possessing a dorsal
hollow central nervous system. One ‘protochordate’ subphylum and the
vertebrate subphylum have planktic representatives.
Subphylum: UROCHORDA
(TUNICATA).
Unsegmented, boneless chordates, notochord restricted to larval tail.
Class: ASCIDIACEA
Sedentary tunicates. Motile, appendicularia larvae (‘ascidian tadpole’) has
well-developed notochord. They enter and are briefly resident in the the
marine plankton but do not feed.
Includes: Ciona, Clavelina
Class: THALIACEA
Salps; pelagic tunicates of warm seas. Circumferential muscle bands are
used to pump water through body, providing food and propelling animals
forward. Tadpoles develop into zooids that are eventually set free as
young salps.
Includes: Doliolum, Salpa, Pyrosoma
Class: LARVACEA
Neotenic pelagic tunicates in which the appendicularian tadpole
becomes the sexual form. These are planktic but live within a secreted
‘house’, which itself resembles a salp, and which is equipped with filter.
The larvacean’s movements produces the filtration current.
Includes: Oikopleura
Subphylum: VERTEBRATA
Chordates that develop an articulated, bony or cartilaginous backbone.
Of the eight or so extant classes, only one is considered to have typically
planktic representatives.
Class: ACTINOPTERYGII
Bony fishes. Many types of pelagic, marine demersal and limnetic littoral
feeding fish grow either from pelagic eggs and larvae or from larvae
hatched from eggs on the sea bottom. Initially, at least, the larvae are
truly microplanktic.
Compiled from several sources: Hardy (1956), Borradaile et al. (1961), Donner (1966), Reynolds (2001b).
and where feeding relies mainly on encounters ingesting so many gas vesicles that they became
of food organisms by consumers, the availabil- irreversibly buoyant. The remarkable ability of
ity of a substratum to which to attach does Nassula to ingest long filaments of green algae
not necessarily improve feeding efficiency. The (Binuclearia) and Cyanobacteria (Planktothrix) by
heterotrophic nanoflagellate genera Bodo, Monas, sucking them in, spaghetti-like, and coiling them
Bicosoeca, Monosiga are represented by species in intracellularly (Finlay, 2001, and personal com-
both the marine and the fresh water plankton. munication) provides a striking case of the feed-
Other microplanktic protists include sar- ing adaptations of this animal.
codines and ciliophorans, which may range in
size between 10 and 200 µm. The radiolarians and Multicellular microzooplankton
foraminiferans are mainly marine (though the Metazoans contribute to the composition of both
latter are represented in fresh waters). Most are the marine and freshwater microzooplankton
phagotrophic on detritus and picoplankters but but, despite some common features, the differ-
radiolarians also harbour algal symbionts. Amoe- ences in the principal organisms represented are
bae are not major players in the plankton, gen- substantial. In the sea, they are dominated by
erally contributing <5% of the nanoheterotroph larval crustaceans (especially the nauplii of resi-
biomass (Sorokin, 1999) but dense populations dent copepods), rotifers and larvaceans as well as
are now recognised in the region of hydrother- the larvae of other groups, such as molluscs and
mal vents. echinoderms (see Table 6.1). There is a wide range
Ciliophorans are often numerous and well of food preferences (nanoplanktic autotrophs,
represented in the microplankton of lakes and heterotrophs and algae) as well as detrital parti-
seas by some common and highly cosmopolitan cles. Cilia around the oral region, aided by mus-
species (Finlay and Clarke, 1999; Finlay, 2002). cular contractions around the feeding apparatus,
These include species of the naked spirotrichs or (in the case of rotifers and nauplii), mandibu-
(such as Coleps, Strombidium, Strobilidium) and lar mouthparts may be used to handle captured
holotrichs (Prorodon, Pleuronema), as well as the particles into the digestive tract but the main
loricate Tintinnidium. In freshwater, vorticellids feeding strategy is still largely encounter. Larger
are epiphytic on large algae and use their oral microplankters may have some limited influence
cilia to move particles into the cytostome. Micro- over orientation and exploration of the local envi-
zooplankters feed principally on nanoplanktic ronment but, mostly, they are still too small
autotrophs and heterotrophs, thus fulfilling a key to overcome the problem of viscosity. The lar-
linkage in the transfer of carbon through the vacean, Oikopleura, perhaps comes closest to gen-
microbial food web (Sherr and Sherr, 1988) (see erating and filtering a significant feeding current
also Section 3.5.4). Ciliates may become the dom- through its secreted ‘house’ (see Table 6.1).
inant animal in micraerophilous or anoxic sea The multicellular microzooplankton of lakes
water (such as in the Black Sea, at depths >30 m) is often dominated by rotifers and copepod
or in the metalimnia of eutrophic lakes (Fenchel nauplii. Feeding among the freshwater rotifers
and Finlay, 1994). Other, mainly benthic or deep- has been studied extensively; the comprehensive
water ciliates are frequently encountered in reviews of Donner (1966) and Pourriot (1977) con-
plankton samples from shallow water columns. tinue to provide helpful guides. Production of the
Among fresh waters, planktic ciliates that most common pelagic rotifers in lakes (Keratella,
feed on larger or more specialised foods some- Brachionus, Synchaeta, Polyarthra) responds well to
times rise to become, usually for short periods, abundant populations of nanophytoplankton but
dominant consumers of algae or cyanobacteria there is an evident selectivity which is influ-
or flagellates (Dryden and Wright, 1987). Some enced by the size of the food organisms (Gliwicz,
cases have been reported (Reynolds, 1975; Heaney 1969). For instance, the relatively robust Keratella
et al., 1990) in which Nassula effectively removed quadrata experiences an upper size limit of inges-
the biomass of floating Anabaena from the sur- tion of 15--18 µm but, for the smaller K. cochlearis,
face layer of a lake, in one case the animals it is only 1--3 µm. Thus, it is hardly surprising
that, in the experiments conducted by Ferguson or littoral in distribution. Cyclops vicinus and Meso-
et al. (1982), both species coexisted and were, cyclops leuckarti are species that are common and
simultaneously, able to increase while in the widely distributed in the plankton of larger lakes.
presence of a phytoplankton with abundant pop- Adult calanoids are more cylindrical, have
ulations of Chlorella, Rhodomonas and Cryptomonas. long antennules which are extended laterally to
However, only K. quadrata flourished among a sub- ‘hang’ on a rotational current generated by the
stantially unialgal Cryptomonas nanoplankton. beating of the paddle-like thoracic limbs. The ani-
mal moves abruptly to a new position by a single
Freshwater mesozooplankton beat of the antennules (Strickler, 1977). Calanoids
The mesozooplanktic element (0.2--2 mm) are able feed on small (∼4-µm) algae and bac-
embraces what are, to many people, the most terised particles which are filtered by bristle-like
familiar components of the zooplankton. These setae on the maxillae from the same currents
animals are big enough not just for their move- generated by the swimming appendages (Van-
ments to escape the constraints of viscosity but derploeg and Paffenhöfer, 1985). The filtration
for them to be able to exploit turbulence and cur- rates may reach 10--20 mL per individual adult
rent generation to optimise contact with poten- per day (Richman et al., 1980; Thompson et al.,
tial foods (Rothschild and Osborn, 1988). The 1982), although most reported averages are an
groups of animals that are principally involved order of magnitude smaller. In addition, animals
are really quite diverse and, beyond some shared feed on larger algae and ciliates, up to 30 µm
adaptations, they are fitted to mutually distinct in size, which are actively captured and manip-
life modes and fulfil different ecological roles. ulated with the maxillae and maxillipeds (and
Any residual notion that zooplankton just eat probably other appendages too). Prior to making
phytoplankton must be rejected as being crassly its strike, the animal will have detected and delib-
oversimplistic! The common adaptations include erately oriented itself towards its quarry. The inci-
the tendency to transparency (to minimise their dence of successful encounters is high (Strickler
visibility to planktivorous fish, crustacea and and Twombley, 1975). Whether or not calanoids
other potential predators) and the ability to should still be regarded as being primarily herb-
propel themselves through the water. Individual ivorous, they have a demonstrable potential to
adaptations concern their means of movement control the numbers of ciliates in the plankton
and, especially, their means of gathering their (Burns and Gilbert, 1993; Hartmann et al., 1993).
required food intake. The two feeding modes afford to calanoids an
In most fresh waters, the mesozooplankton enhanced dietary flexibility, while the measure
comprises crustaceans from two main classes (see of electivity allows them to survive lower concen-
Table 6.1): copepods (from two orders in par- trations of food.
ticular, the cyclopoids and the calanoids) and The cladocera are specialised filter-feeders.
branchiopods (of the sub-order Cladocera). Adult The body is much modified from the basic crus-
(final-instar) planktic cyclopoids are faintly pear- tacean form: a short thorax and abdomen car-
shaped and streamlined with the paddle-like ries a compressed, shell-like carapace that forms
thoracic legs held under the body. They have a chamber in which four to six pairs of flattened,
short biramous antennules and several caudal setae-bearing trunk limbs (‘phyllopods’) beat
rami of varying lengths. Cyclopoids swim gently rhythmically. Their motion draws water through
by reciprocation of the antennules and antennae the variable carapace gape and the setae filter
but they can ‘pounce’ as well, through the simul- particles from the inhalant current (Fryer, 1987;
taneous use of the thoracic limbs. They are rap- Lampert, 1987). The animals swim by beating
torial feeders (they seize their food) on a range the large, biramous antennae. Several cladoceran
of nano- and micro-planktic particles including families are represented in the fresh-water plank-
algae, rotifers and detrital particles. There are ton. The Sididae, which have six pairs of phyl-
numerous species in several genera; many are lopods and strong, branched antennae, are found
confined to ponds, some are primarily benthic mostly in vegetated margins and ponds. The
Holopedidae are represented by a single genus, Moina species are most closely associated with
which has a reduced carapace and is instead small water bodies with a tendency to offer suit-
embedded in a mass of jelly. Bosminids are plank- able habitat conditions on a temporary basis
tonic in ponds and small lakes and have a wide but explosive growth phases afford a dynamic
geographical distribution. The macrothricid and advantage when favourable conditions obtain
chydorid cladocerans have relatively small anten- (Romanovsky, 1985). In the open pelagic of perma-
nae, and are mainly feeders on hard surfaces. Chy- nent larger lakes, dominance among the daph-
dorus sphaericus is extremely common in the bot- niids is contested by such species as D. cucul-
tom water and among weeds of small lakes and lata, D. galeata, D. hyalina and D. pulicaria (Gliwicz,
it is frequently encountered in the plankton of 2003a).
the water bodies in which they are present. In
two other families, the polyphemids and the lep- Marine mesozooplankton
todorids, the carapace is small and covers little The principal groups of mesoplanktic herbivores
more than the brood pouch; planktic species of in the sea are the calanoids, the cladocerans
Bythotrephes and Leptodora are macroplanktic (2-- and the thaliacean tunicates (the salps and their
20 mm) predators and do not comply with the allies) (Sommer and Stibor, 2002). In fact, the
generalisation about cladoceran filter-feeders. calanoids are the most familiar and such tem-
There is one further family, the Daphni- perate shelf-water species as Calanus finmarchius,
idae, whose species can be extremely promi- Acartia longiremis, Temora longicornis and Cen-
nent in limnoplankton and which plays a major tropages hamatus have long been regarded as the
role in regulating the structure and function main food organisms of commercially important
of lacustrine ecosystems. Daphniids have five surface-feeding fish, like herring (Clupea haren-
pairs of phyllopods within the carapace and, like gus) and mackerel (Scomber scombrus). Accordingly,
the sidids, have powerful swimming antennae. they have been studied in some detail (e.g. Hardy,
They are efficient and more-or-less obligate filter- 1956; Cushing, 1996). A long-enduring under-
feeders. They can remove all manner of foods on standing of a three-link food chain (phytoplank-
the filtering setae, within defined size ranges. The ton -- zooplankton -- fish) places the calanoids at
upper limit is set by the width of the carapace the fulcrum between the harvest of fish the pri-
gape (which is species-specific and varies with the mary producing phytoplankton. The relative pro-
size of the animal). The lower limit is governed by portions of the respective annual production by
the spacing of the phyllopod setae (Gliwicz, 1980; these three components also fitted well with the
Ganf and Shiel, 1985). As will be further explored contemporaneous appreciation of the pyramidal
below, individuals are able to filter such relatively Eltonian relationship, with an approximate 10%
large volumes of water that, under favourable transfer of the energy acquisition at each trophic
conditions, it is likely that a significant popula- level being transferred to the one above (Elton,
tion of maturing daphniids may be able to filter 1927; Cohen et al., 1990). This paradigm was
the entire volume of a lake in a day or less. The seriously challenged by the realisation of how
implications for their food organisms, and for the large a proportion of pelagic photosynthate is
other organisms using the same food resource, transferred, as dissolved organic carbon, through
and for those other microplanktic feeders that microbes and their microplanktic consumers to
can be themselves be ingested by the daphniids ciliates (Williams, 1970; Pomeroy, 1974; Porter
are formidable and complex. et al., 1979; Sherr and Sherr, 1988) (see also Sec-
There is considerable further differentiation tion 3.5.4).
of the habits, predilections and dynamics within On this basis, ciliate-consuming calanoids are
the Daphniidae and within the type genus, Daph- already the fifth stage in the food chain. How-
nia. Simocephalus, Ceriodaphnia and some of the ever, the close coupling of the components and
larger Daphnia species (D. lumholtzi, D. magna) are the functional integrity of microbial food webs
more common in ponds or at the weedy mar- (e.g. Šimek et al., 1999) are known to achieve a
gins of lakes than in the open water of lakes. high ecological efficiency of energy transfer (10--
35%: Gaedke and Straile, 1994a). Even so, the real a relatively rich tissue content of ribosomal
problem very often relates to the low supportive RNA, bringing with it a consequent high cell-
capacity of the nutrient resources available and phosphorus content. This is, incidentally, the
the low biomasses that can ever be maintained. major factor influencing the relatively low C : P
Selectively browsing calanoids are simply the stoichiometries (averaging about 80 : 1) that are
most efficient harvesters of the carbon flux. More- typical among these (Gismervik, 1997) and other
over, a relatively low fecundity and modest rate planktic cladocera (Elser et al., 2000).
of investment in egg production enables them to The third main group of marine meso-
satisfy their minimum food requirements at low planktic animals comprises the pelagic, free-
POC concentrations, in the range of 5--80 µg C living tunicates, especially the salps, pyrosomans
L−1 (Hart, 1996) (as algae, this is roughly equiv- and doliolids that represent the Thaliacea (see
alent to a chlorophyll content in the range 0.1-- Table 6.1). These near-transparent, gelatinous ani-
1.6 µg chla L−1 ). Except when food resources are mals have a low body mass, comprising little
truly limiting, the dynamics of calanoid growth more than an open barrel-shaped tube, a filter-
are most likely to be governed by temperature ing gill and circumferential muscle bands whose
(Huntley and Lopez, 1992). It is even possible systematic contraction and relaxation refresh the
that the distinctively oceanic calanoids (such as current of water across the filter screen. All
Acartia clausi, Centropages typicus) function at yet pelagic tunicates are filter-feeders, straining per-
lower resource availabilities than the shelf-water haps the entire nanoplanktic--microplanktic size
species. range of particles (Sommer and Stibor, 2002).
In the more enriched coastal waters, receiving In additon, they are, collectively, ubiquitous
nutrient inputs from rivers or as a conequence of components of the pelagic fauna, from coasts
deep-oceanic upwellings, several resource-driven to the deep sea; however, they are perhaps best
effects are evident. There are absolutely more of known for their presence in the plankton of
the biomass-constraining nutrients (N, P, Fe), at warm, ultraoligotrophic oceans. The low body
once raising the potential to support more pri- mass requires absolutely modest resources for
mary producers and to maintain a higher algal maintenance, while they expend little energy in
biomass. The additional nutrients may also allevi- keeping their near-isopycnic structures from sink-
ate the dependence of large algae upon turbulent ing. The architecture and physiology of these ani-
diffusivity to fulfil their nutrient demands (cf. mals is substantially geared to function at very
Riebesell and Wolf-Gladrow, 2002) (see also Sec- low concentrations of assimilable POC.
tion 4.2.1). There may well be a lowering effect on
the Si--N and Si--P relationships, which some con- Planktivorous macroplankton, megaplankton
sider relevant to changes in the species composi- and nekton
tion of the phytoplankton, although these really Although this section has so far, taken a ‘bottom-
depend on the absolute nutrient levels (Sommer -up’ view of the structure of the zooplankton,
and Stibor, 2002). in emphasising the nature of the resource base
The combination of these effects results and the evolutionary adaptations of the main
in potentially greater concentrations of high- pelagic groups to be able to exploit it, it is
quality, primary foods that will support direct only half the story. For, as many authorities
filter-feeding. It is not just the fact that cladocer- tirelessly point out (e.g. Gliwicz, 1975, 2003a;
ans, such as Evadne, Podon and Penilia, can filter Banse, 1994), active net production of each of the
more water and, thus, harvest more food than components of the plankton of seas and lakes
calanoids (Sommer and Stibor, 2002). They also is regulated by its consumers. Thus, the abun-
have faster rates of metabolism and growth, with dance of ciliates in the open plankton might be
proportionately more of their greater energy expected to increase on an abundant resource of
intake (60--95%) being invested in partheno- nanoflagellates but the capacity to do so may be
genetic reproduction (Lynch et al., 1986; Stibor, severely constrained by the numbers of calanoids
1992). At the physiological level, this requires or other predators (see, e.g., Thouvenot et al.,
2003). Equally, the growth of microphytoplank- tant euphausids. The latter are entirely pelagic
ton in a lake might be constrained by the feed- throughout their lives; they live in all the oceans
ing of herbivorous Daphnia but the vulnerability but the key place of Euphausia frigida and E. tri-
of Daphnia to consumption by planktivorous fish cantha in the southern oceans is renowned for
might not only reduce their impact upon the their being the main food of the great baleen
phytoplankton but the latter would be allowed whales. Quantitatively less important are the car-
to increase to something like its ungrazed nivorous crustacean larvae of decapods (crabs,
potential. Such dynamic ‘cascading’ interactions lobsters) and of the mantis shrimps, Squilla spp.
(Carpenter et al., 1985) may now seem to be self- celebrated by Hardy (1956) as being the ‘most
evident phenomena but they were not formally beautiful of larvae’. Among fresh waters, non-
described before the publication of now-classic vertebrate planktivores other than mysids occur
quantitative studies of Hrbáček et al. (1961). Since mainly in inshore waters, and include larval
then, trophic cascades and their manipulation megalopterans, hemipterans (e.g. Notonecta) and
have been studied in great detail (see Carpen- dipterans.
ter and Kitchell, 1993) and exploited as the basis Among the larger pelagic animals (fish,
of system management by what has become squid), size and muscular strength take them
known as biomanipulation (Shapiro et al., 1975; out of the plankton and into the swimming nek-
see also Section 8.3.6). For the moment, however, ton. For these to be truly pelagic planktivores,
I seek only to make the point that the effects the ability to sample large volumes of water
of mesoplanktic herbivory and microphagy on and strain from this food items of only 1--2 mm
the dynamics and standing crops of the primary is essential and demanding of a very efficient
producers are subject to cascading impacts of means of food filtration. In the herring and some
planktivory. Again, the point is elegantly made other closely related clupeoid species, for exam-
in Gliwicz’s (2003b) observation that the struc- ple, this capacity is provided by gill rakers --
tural composition and size distribution of the comprising numerous, long and slender close-
zooplankton is very different between systems set bristles borne on the gill arches. The bask-
with and without the presence of zooplanktiv- ing shark, Cetorhinus, is a relatively very large
orous fish. The point is especially pertinent, as elasmobranch but its gill arches are analogously
zooplankters have nowhere to hide in the open set with many close-set, flattened strips that
water: survival depends on not being seen or function analogously to the rakers of bony fish
eaten by planktivorous cosumers. Apart from (Greenwood, 1963). The direct sustenance of a
the protection that comes from fortuitously low creature attaining a length of 10--12 m and a
predator densities, much depends upon either body mass of several tonnes through feeding on
being less visible (which exacts a premium on animals individually ten thousand times smaller
zooplankton that might grow larger) or less acces- and 10−9 of its body mass is a truly impressive
sible to visual predators by descending to the example of emergy gain through trophic link-
lightless depths. The need to return to the surface age. In another, fresh-water case, the produc-
waters to feed (at night!) invokes an unavoidably tive basis of the fisheries of the meromictic rift
high energetic cost. valley Lake Tanganyika, founded mainly on two
Besides adult fish such as herring and mack- species of planktivorous clupeid (Stolothrissa tan-
erel (see above), marine consumers of mesozoo- ganicae, Limnothrissa miodon) and their endemic
plankton include invertebrates of the next two centropomid predator (of the genus Lates: Lowe-
size divisions -- macroplankton (2--20 mm) and McConnell, 1996, 2003), has been shown to be
megaplankton (>20 mm). Among the most sig- quantitatively dependent upon the pelagic food
nificant of these are the chaetognaths (arrow web of the lake (Sarvala et al., 2002). Yet more
worms), ctenophores (sea combs and sea goose- remarkable is the fact that the annual stim-
berries) and the polychaete Tomopteris. There is ulus for production in this oligotrophic lake
also a variety of crustaceans at this scale -- the relies heavily upon the recycling of nutrients
mysids, the pelagic amphipods and the impor- during the period of increased wind action and
monimolimnetic deepening and the enhanced 6.4.2 Impacts of filter-feeding on

production of picoplanktic cyanobacteria. Bacter- phytoplankton
ial consumption of primary DOC yields between Moving on from qualitative description of the
25 and 130 g C m−2 a−1 to the pelagic food web, structural components of the phagotrophic
which, in turn sustains the producton of up to plankton, consideration is now given to the quan-
23 g copepod C m−2 a−1 . Much of this (16 g C m−2 titative impacts of their feeding on the pro-
a−1 ) is eaten by planktivores in generating the net ducer mass. It is conceptually easier to deal first
annual production of 1.1 g C m−2 a−1 Stolothrissa with the impacts of filter-feeders. Although this
plus Limnothrissa and 0.3 g C m−2 a−1 Lates. method of food gathering is far from universal,
Adults of many families of fish will eat zoo- it can be the most striking and complete in its
plankton when it is sufficiently abundant to be impact. Moreover, its effects are relatively easy to
an attractive and satisfying resource. The major- model. These are good enough reasons to explain
ity of adult pelagic fish (and of many that inhabit the additional fact that a large literature on filter-
shallower margins) are carnivorous on other fish feeding has accumulated.
and/or on other large prey. However, many of Supposing that, to the potential planktic
these species produce large numbers of small consumer, the relatively most abundant food
eggs that give rise to juvenile stages, which are resource is the nanoseston -- algae, large bacteria,
initially mesoplanktic. They feed on microplank- detrital particles measuring 2--20 µm across -- and
ton (Sarvala et al., 2003), often in direct compe- that particles are generally well dispersed within
tition with and exposed to predation by adult the medium, then the development of some
planktivores (e.g. O’Gorman et al. 1987). Accord- means to sieve and to concentrate such parti-
ing to the systematic simulation modelling of cles is likely to be favoured by evolution (Gliwicz,
Letcher et al. (1996), metabolic growth capacity, 2003a). The coupling of a filter and the means
rather than foraging ability or resistance to of generating a water current across is a char-
starvation, is the leading bottom--up component acteristic of the feeding apparatus of many zoo-
in larval survival. Predator size is a powerful plankters, including ciliates and rotifers. How-
influence on survival but has only a weak effect ever, it is at the mesoplanktic scale, of crus-
on the variability in the composition of the taceans and tunicates, that filter-feeding has a
available prey. Letcher et al. (1996) also deduced significant impact on the availabilty of food in
that whether young fish died through starvation the entire medium, or at least beyond the imme-
or predation usually depended most on the diate environment of the individual animal. This
availability of their smallest prey organisms. difference is most due to viscosity. Where the
Reference should also be made to pelagic smallest turbulent eddy is in the order of 1
squid (Loligo spp.) whose juvenile hatchlings (or mm or so, the typical microplankter (<200 µm
paralarvae) are released into the open water. in length) experiences a wholly viscous environ-
They are free-living and self-propelling, using ment in which to move itself and, more impor-
rhythmic contractions of the mantle to force tantly, to influence the encounter with particles
a series of alternating bursts of water flow ≤20 µm. An analogy that comes to mind is of
and recovery. Being barely 2 mm in length, a human trying to collect bananas while both
paralarvae are, unmistakeably, initially (albeit are immersed in a swimming pool filled with a
briefly) mesoplanktic (Barón, 2003). liquid having the consistency of molasses or set-
On the basis of this brief survey, it is clear ting concrete. Scaling upwards, the same human
that the precise structure of the pelagic web would be rather more successful in picking out
of consumers is highly variable and subject to kidney beans from the same liquid but now dis-
dynamic forces. So far as the impacts upon the persed in a vessel the size of a bath tub. It is only
phytoplankton is concerned, outcomes hinge on by being big enough and strong enough to over-
the numbers and sizes of the herbivores present come frictional drag and to generate turbulent
and the sustainability of the feeding modes currents that it is possible to increase the rate of
available. particle contact (Rothschild and Osborn, 1988).
Quantitatively, the best studied of the three

characteristic mesozooplanktic groups of filter-
feeders is the Cladocera and, especially, those of
the fresh-water genus Daphnia. Collectively, these
also illustrate well the factors that most influence
the dynamic impacts on the microplanktic food
organisms: how many feeders there are, what and
how much food they remove, and what the conse-
quences might be, both for themselves (in terms
of biomass increase) and for the food (in terms of
how much more cropping it can withstand). Each
of these components is pursued exhaustively in
the available literature. Much of this amplifies
the findings of some of the earliest of the mod-
ern investigations. Indeed, many of these were of
such enduring quality that they provide an ideal
base for this section.
Filtration rates
Certainly for the larger individual filter-feeders,
the basic quantity of interest is the volume of
water that animals are able to strain per unit
time. The usual means of its determination is
Figure 6.4 Functional responses of (a) filtering rate and
to measure the rates of depletion from known
(b) feeding rate of filter-feeding animals with respect to food
concentrations of radio-labelled ingestible foods.
concentration. The arrow defines the incipient limiting
To obtain sensible results in a short period of concentration, as defined by McMahon and Rigler (1963).
time, it is necessary to make the measurements Redrawn with permission from Gliwicz (2003a).
on suspensions of known numbers of animals,
which, in turn, necessitates that the per capita
volumes of water processed are nearly always mals that are ‘starved’ for some hours before
mean values. Moreover, as the individual volumes introduction into a suspension of radio-labelled
filtered vary conspicuously with the size of the food organisms. In this case, what is measured
animal, it is necessary always to pre-sort the ani- is strictly the ‘particle clearance rate’. If all the
mals beforehand. Possible methodological short- above conditions are satisfied, the mean indivi-
comings attach to the effects of handling on the ual clearance rate should coincide with (but not
animals’ performances. Another potential diffi- exceed) the mean volume of water processed per
culty that must be addressed is that the removal unit time, which is the mean individual ‘filtra-
of radio-labelled particles is incomplete or is tem- tion rate’ (F). The rate at which food particles are
porary (i.e. food is selected or rejected, leading captured is obviously dependent upon the filtra-
to an underestimate of the volume filtered). Fur- tion rate but the ‘feeding rate’ (or ‘ingestion rate’)
ther, allowance has to be made for the possibilty is a proportion of the food concentration in the
that, once satiated, the individual may slow its inhalant volume. Moreover, it has been demon-
filtration rate. strated (McMahon and Rigler, 1963) that, at high
Experience has shown that all these are real concentrations of food, clearance rates of Daph-
constraints. Consensus has clarified their magni- nia slow down as the animals are sated for less
tude and also spawned a terminology. All mean- filtering effort. The relationships between filtra-
ingful measurements take place at known tem- tion and feeding as a function of food concentra-
peratures and within short, timed exposures tion are sketched in Fig. 6.4. Below the ‘incipient
to known concentrations of size-categorised ani- limiting food concentration’, Daphnia is likely to
Table 6.2 Individual filtration rates, Fi (in mL d−1 ), for various planktic animals as reported in, or derived from
relationships in, the literature
Species Fi References
Rotifers
Generally 0.02–0.2 Pourriot (1977)
Brachionus calyciflorus 0.1–0.2 Halbach and Halbach-Keup (1974);
Starkwether et al.(1979)
Freshwater calanoids
Eudiaptomus gracilis (12 ◦ C) 0.6–1.8 Kibby (1971)
Eudiaptomus gracilis (20 ◦ C) 1.3–2.5 Kibby (1971)
Eudiaptomus gracilis (adults, 17 ± 3 ◦ C) 0.5−10.7 Thompson et al. (1982)
(copepodites, 17 ± 3 ◦ C) 0.5–6.7 Thompson et al. (1982)
Diaptomus oregonensis (adults, 2.4–21.6 Richman et al. (1980)
temperatures various)
Freshwater cladocerans
Bosmina longirostris <3.0 Thompson et al. (1982)
Chydorus sphaericus 0.5–2.6 Thompson et al. (1982)
Daphnia galeataa
(<1.0 mm, 17 ± 3 ◦ C) 1.0–7.6 Thompson et al. (1982)
(1.0–1.3 mm, 17 ± 3 ◦ C) 3.1–19.3 Thompson et al. (1982)
(1.3–1.6 mm, 17 ±3 ◦ C) 3.1–30.7 Thompson et al. (1982)
(1.6–1.9 mm, 17 ± 3 ◦ C) 14.0–60.0 Thompson et al. (1982)
Daphnia spp.
(<1.0 mm, 15–20 ◦ C) <5.0 From Burns’ (1969) regression
(1.0–1.3 mm, 15–20 ◦ C) 3.6–10.4 From Burns’ (1969) regression
a
The organism was reported as D. hyalina var. lacustris. Under current nomenclature, the identity D.
galeata is to be preferred (D. G. George, personal communication).
be hungry (‘food-limited’) and to filter-feed as fast The superior filtration capacity of Daphnia
as it is able. (especially larger individuals) is explicit in the
It was another of Frank Rigler’s colleagues, relationship that Burns (1969) demonstrated
Carolyn Burns, who made some of the first accu- between the filtration rates (Fi ) of four species
rate measurements of the filtration rates in Daph- of Daphnia (D. magna, D. schoedleri, D. pulex, D.
nia. Her work (Burns 1968a, b, 1969) has been well galeata) and their carapace lengths (Lb ). The plots
supported by subsequent determinations by oth- in Fig. 6.5b and the entries in Table 6.2 are calcu-
ers (e.g. Haney, 1973; Gliwicz, 1977; Thompson lated from her regressions. The measurements of
et al., 1982), to the extent that her original Thompson et al. (1982) on D. galeata are especially
detailed findings are used for this quantitative well predicted. At 15 ◦ C, the hourly filtration rate
development. What remains remarkable is the is given by
enormous capacity of the feeding current created
by the rhythmic beating of the thoracic limbs to F i = 0.153L b 2.16 (6.11)
draw water into the carapace chamber and across
the filtering setae on the third and fourth tho- and at 20 ◦ C
racic limbs. Comparison with other zooplankters
is made in Table 6.2. F i = 0.208L b 2.80 (6.12)
Figure 6.5 Filter-feeding in

Daphnia spp. (a) The maximum
volume of water filtered by animals,
also as a function of body length, at
two temperatures (data of Lampert,
1977a); (b) The size of the largest
particle available to an individual
animal as a function of its body
length, Lb , according to Burns
(1968a). Figure redrawn with
Possibly the most pertinent deduction is the fact range of potential foods of filter-feeding cladocer-
that as few as 20 large Daphnia per litre of lake ans increases with the maturity of the consumer.
water, or 200 neonates L−1 , is sufficient a popula- Whereas a 1.0-mm Daphnia is probabilistically
tion to process daily the entire volume in which restricted to foods <25 µm in size, a 2.0-mm ani-
they are suspended. More generally, the aggre- mal can take food particles up to 50 µm across.
gate volume of water that is potentially filtered Conversely, for smaller animals the food availabil-
each day ( Fi ) is equivalent to: ity may be rather more restricted and the feed-
ing rate falls below the potential of the filtration
F i = (N 1 · F i1 ) + (N 2 · F i2 ) + · · · · ·(N i · F ii ) (6.13)
rate. Those items that are too large are simply
where F ii is the filtration rate and Ni is the stand- rejected and do not enter the filter chamber. As
ing population of the ith species size category. Gliwicz (1980) recognised, the filtering daphniid
is very easily able to regulate the size of food par-
Food availability ticle that reaches the phyllopods through its con-
The lower end of the size range of particles trol over the carapace gape. Obviously it cannot
available to the filter-feeder is determined by accept particles above a size-specific maximum
the filter itself. In the case of the daphniids, (Gliwicz and Siedlar, 1980) but it may, however,
it is set by the spacing and orientation of the self-impose a lower maximum, perhaps to avoid
setules, the short branches besetting the filter- ingesting mucilaginous colonies. Thompson et al.
ing setules fringing the third and fourth tho- (1982) observed sharply reduced filtering rates
racic phyllopods (Gliwicz, 1980; see also Gliwicz, of D. galeata when the phytoplankton was dom-
2003a, and references cited therein). Comparing inated by Microcystis colonies of varied size. The
this character with the filterable particles recov- possibility that the behaviour was in some way
ered by rotifers, calanoids and other cladocerans attributable to toxicity of the phytoplankter
(Geller and Müller, 1981; Reynolds, 1984a), there seemed a less likely explanation than the direct
seems to be little interphyletic variation, 0.2--2.0 observation that the rate of phyllopod beating
µm being a general value. In fact, the efficiency was impaired only if a Microcystis colony became
of retention of the smallest particles may be low- ensnared in the apparatus or blocked the median
ered in Daphnia, owing to some leakage from the chamber. At this point, the post-abdominal claw
filter, from the median chamber and the food was flexed and used to scrape out the blocking
groove, before they get to the animal’s mouth. mucilage. Thompson et al. (1982) argued that so
Fewer particles in the range 1--3 µm are retained many animals were spending so long purging the
than in the range 5--10 µm (Gliwicz, 2003a). filtering apparatus that the aggregate filtration
The upper size limit of particle that can be rate became severely depressed.
ingested is, according to Burns (1968a), strongly It is not just the size and texture but the
correlated to animal size (Fig. 6.5b). The relation- shape of the phytoplankton that influences the
ship makes the important statement that the proportion of food strained from the filtration
Figure 6.6 Schematic

representation of food selectivity in
Daphnia pulicaria (vertical arrows
indicate food items retained, side
arms show those lost or rejected) at
three critical points in the feeding
process. (a) Encounter and intake
into the filter chamber; (b) capture
from phyllopods to food groove; (c)
ingestion of particles at the mouth.
From an original representation of
Hartmann and Kunkel (1991),
redrawn with permission from
Gliwicz (2003a).
current. This topic was elegantly investigated and All these facets contribute to feeding rates
summarised by Hartmann and Kunkel (1991). The that are below (in some instances, well below)
diagram in Fig. 6.6 distinguishes the progress to the potential of the aggregate filtration rate.
ingestion of three differing algal shapes. Small Reynolds et al. (1982a) proposed a food-specific
sphaeroid and cuboidal unicells are easily drawn correction factor (here designated as ω ) in order
in to the median chamber and efficiently com- to relate removal rates to aggregate filtration rate
pacted by the phyllopods, and most of that food (F). As removal of algae from the water is recip-
will be ingested. Slender algal cells are more dif- rocated by the random redispersion of the sur-
ficult to orientate and compress and there is vivors in ostensibly the same volume of water,
some loss from the median chamber and some the reaction of the extant algal population corre-
rejection from the food groove. Long filaments sponds to another exponential series, analogous
are really quite difficult. However, as observed to dilution (cf. Eq. 6.7). The analogous grazing loss
by Nadin-Hurley and Duncan (1976), Daphnia is rate term, rG , inflicted upon the algal population
able to arrange foods into spaghetti-like bun- is given by
dles (though not without significant rejection
losses). rG = ω ( F i )/V (6.14)
Thompson et al. (1982) found that the removal of the status of a number of cyanobacterial genera
nanoplanktic unicells and Cryptomonas from the has been puzzling. Work with pure cultures led
feeding current of D. galeata is highly effective, Rippka et al. (1979) to take a strongly reductionist
so that the value of ω is close to 1 and rG is view of cyanobacterial diversity, proposing that
little different from that predicted directly from a small number of named genera could accom-
the filtration rate. The value of ω falls to ∼0.3 modate the few essential distinctions among the
for eight-celled Asterionella colonies and rapidly unicellular strains (size, shape of cells and planes
from ≤0.3 to zero for Fragilaria colonies com- of cell division). Since then, new techniques and
prising more than 6--13 cells per colony. Large a new generation of researchers have scarcely
Daphnia are capable of feeding on small Eudorina added to the range of known picocyanobacteria.
colonies, short Planktothrix filaments and young Most known inland-water forms conform to being
Microcystis colonies while they are still quite small short rods (Synechococcus), ellipsoids (Cyanobium)
(m generally <50 µm) but have great difficulty or spheroids dividing in one or two planes
with larger colonies. Thus, the rate of removal, (Cyanothece) (Komárek, 1996; Callieri and Stockner,
rG , varies among the algae in mixed populations, 2002). Moreover, the similarity of the cells of
even when they are simultaneously subject to those of various species of colonial genera
the activities of the same set of filter-feeders. inhabiting mildly eutrophic waters (such as
Combined with the age--size structure of the Aphanocapsa, Cyanodictyon and Synechocystis) has
filter-feeding populations, precise values of rG are been noted on many occasions and attribution
scarcely easy to calculate and should carry such to the same or close genetic lines has been
a wide margin of error that, for most purposes, proposed. The experiments of Komárková and
it is acceptable to work with approximations. It Šimek (2003) imitated transformations of grow-
should also be borne in mind also that if vertical ing strains of colonial Aphanocapsa and Synechocys-
migration takes the main filter-feeding compo- tis into unicellular suspensions, and back, stimu-
nents of the zooplankton to depths beyond the lated by the presence or absence of herbivorous
visibility of fish, and they return to the alga-dense brine shrimp Artemia (or to chemicals in medium
surface waters only during darkness, then Eq. in which Artemia had been present).
(6.14) is yet more difficult to evaluate. The maxi- Interestingly, this behaviour mirrors the
mum filtration rates of Daphnia are no higher by known morphological anti-predator defences
night than by day (Thompson et al., 1982). (lengthened spines, bulbous heads) that are
Before concluding this section, it is also inducible in cladocerans and rotifers exposed
important to make reference to the adaptations to water in which fish or even Chaoborus lar-
that algae invoke to make themselves less palat- vae have been present (see Gliwicz (2003a) for a
able to the filter-feeding consumers with which detailed overview of the literature). Chemosen-
they do come in contact. In mathematical terms, sory perception in locating and selecting prey
these help them to reduce the instantaneous is probably vital in the viscous world of micro-
value of ω . Hessen and van Donk (1993) observed zooplanktic consumers and the capability may
a tendency of some species of alga to main- well be widespread among the planktic protists
tain larger coenobia (sensu more cells per coeno- (Weisse, 2003). It is reasonable to expect that
bium) in the presence of Daphnia than without. chemoperception works in the opposite direc-
They showed experimentally that growing clones tion and that predator detection and reaction
of Scenedesmus maintain higher proportions of is similarly influenced. The ciliate Euplotes is
colonies comprising eight or more cells in media known to react to predator-specific chemical fac-
in which Daphnia had been present but since tors produced by its amoeboid predators by pro-
removed. Their findings have been repeated by ducing giant, uningestible cells (Kusch, 1995). The
Lampert et al. (1994). chemical nature of the substances involved is
Another interesting discovery is the reaction not well known. They are collectively referred
of picoplanktic cyanobacteria to the presence of to as kairomones, though their mutual similari-
grazers. On strict grounds of cell morphology, ties owe to the ability of potential prey to sense
the threat of potential predators, and not to any (1983), working with Eudorina, found that larger
known chemical likenesses. The production of colonies escaped ingestion and uncropped colony
kairomones is distinct from the production of growth and recruitment were such to explain the
overtly toxic substances by potential prey, whose dynamics of population increase. It could be said
effects on would-be consumers may be quite gen- that, by removing smaller algae, larger units are
eral or incidental unintentional in their effects positively selected by moderate aggregate filtra-
on consumers. The distinctions blur somewhat tion rates.
in the interesting interrelationship between the
toxin-producing mixotrophic haptophyte Prymne- Algal removal and grazer nutrition
sium and its phagotrophic dinoflagellate preda- Even a broad picture of the effects of filter-
tor, Oxyrrhis. According to Tillmann (2003), when feeding on the resource cannot be completed
Oxyrrhis was introduced into dense cultures of P- without reference to the dynamic limits of filter-
limited Prymnesium, its initial feeding rate was feeding activity, or to the way that this changes
quickly depressed (to <0.1 cell grazer−1 h−1 ) through time. The answers to both problems
below that of control animals fed on similar- depend upon the relation of nutrition, growth
sized cryptophytes (2.75 cells grazer−1 h−1 ), a and recruitment of the consumers and the quan-
direct response to toxicity. Poisoned Oxyrrhis cells tity and quality of the food available. The food
then lysed and were attacked by phagotrophic requirements of Daphnia pulex were exhaustively
Prymnesium, reversing the direction of the carbon investigated by Lampert (1977a, b, c). He first
flow! The Prymnesium-free medium also invoked derived equations to describe the maximum
lysis in Oxyrrhis, though the effect reduced when hourly assimilation rates of Daphnia, as a func-
the Prymnesium culture was more dilute or the tion of the length (Lb ) and mass (w b ) of individ-
Oxyrrhis were more abundant. ual animals at various temperatures. At 15 ◦ C, a
Many green algae, Cyanobacteria and some 0.8-mm neonate assimilates up to 2.4 µg C d−1 ,
chrysophytes are normally invested in mucilagi- whereas a 2.1-mm adult is accumulating at 15.7
nous sheaths. Among many other functions that µg C d−1 . The amount of food needed to supply
mucilage might serve (see Box 6.1), the pack- such requirements varied with the quality of the
age increases the size of the algal particle and food. The highest assimilation efficiencies (∼60%)
decreases the likelihood of its entry into the fil- came on such readily filterable and digestible
ter chamber of Daphnia, or its retention on the algae as Asterionella and Scenedesmus. Thus, to sati-
filter or its successful ingestion. Thompson et al. ate the metabolic capacity requires the feeding
(1982) found depressed values of ω for Daphnia to yield upwards of 4.0 and 26.2 µg C d−1 (to the
feeding on Eudorina colonies, except those newly smaller and larger animal respectively). On the
released daughters being <25 µm in diameter. other hand, the respirational expenditure for the
Even if they are ingested, mucilaginous colonies same individuals (respectively, 0.6 and 4.3 µg C
are resistant to digestion. They are not only d−1 ) defines the minimum daily intake that will
capable of viable passage (see Canter-Lund and just maintain metabolism, below which they will
Lund, 1995 for examples) but they are said to lose weight and eventually die of starvation.
use the opportunity of exposure to high nutri- The volume of water that can be filtered
ent concentrations to absorb and store them and defines the external food concentrations that will
to use them effectively after release from the sustain the minimum and saturation require-
anus (Porter, 1976). Gliwicz (2003a) considered ments of the Daphnia. From the appropriate
that, despite doubts about the extent of diges- entries in Table 6.2, it is unlikely that a 0.8-mm
tive resistance, the net profit of nutrient uptake D. pulex might filter more than 5 mL of lake water
by algae surviving gut passage might well com- each day, while the 2.1-mm animal may be capa-
pensate. This might explain the frequent observable of procesing 30 mL. Then the minimum con-
tion that the numbers of mucilaginous colonies centration of filterable food necessary to fulfil the
often increase when the density of filter-feeding neonate’s minimum requirement is thus 0.6 µg
Cladocera is high. However, Reynolds and Rodgers C per 5 mL, i.e. ∼0.12 µg C mL−1 ; however, a
Box 6.1 On the sticky question of mucilage
Sheaths or investments of mucilage are produced characteristically around the

cells of many kinds of planktic algae and bacteria, often binding them in colonial
structures (as in chroococcoids such as Microcystis, in sulphur bacteria, among
representatives of several orders of chlorophyte, chrysophyte and haptophyte).
Mucilage also covers filamentous Cyanobacteria (Anabaena, Planktothrix) and some
unicellular algae, including certain diatoms. Mucilaginous threads are trailed by
Thalassiosira filaments and, possibly, by many other kinds of planktic alga (Padisák
et al., 2003a). It is a feature of both marine and freshwater species. Mucilage
investments vary in texture, from being robust and readily visible under the light
microscope (e.g. Microcystis wesenbergii) to being, as in some desmids, so tenuous
to require negative staining in (e.g.) Indian ink preparations to reveal the existence
as a translucent halo around the organism (John et al., 2002). On other occasions,
the extent of a mucilaginous sheath is identifiable under the microscope by the
numbers of bacteria and detrital particles that cling to the perimeter.
Despite the ubiquitous occurrence of algal mucilage and the fortuitous assis-
tance it gives to microscopists attempting to deduce the identity of the organisms
they encounter, there are surprisingly few general accounts in the literature that
consider the functions and benefits that mucilage might provide or that question
how the many other species of alga seem to manage quite well without it. The
probability is that there is no consensus answer anyway. There are certainly sev-
eral measurable benefits that the presence of mucilage imparts to the organisms
that produce it. These are considered in detail at appropriate points in this book.
However, it has not been resolved that any of these is a primary or an original
function of mucilage production or merely opportunistic adaptive applications of
some ancient trait.
Mucilages (the plural is probably reasonable) are hygroscopic lattice-like poly-
mers of carbohydrate and substances resembling acrylic. In the literature, the prod-
uct is sometimes referred to as mucus, though this term is generally applied to
similar polysaccharides produced in many groups of animals (especially coelenter-
ates, molluscs, annelids and many kinds of vertebrate). The elemental composition
(C, H, O) of the secretions involves little of intrinsic value and, in this sense, may
be considered as a by-product of metabolism. The observation has been made
(Margalef, 1997) that there is little difference biochemically between producing
mucilage and any other unused extracellular photosynthetic derivative, save that
mucilage is not released in solution. The possibility that mucilage production orig-
inated as a mechanism for regulating the accumulation of photosynthate in cells
that cannot be assimilated into amino acids and proteins has some resonance with
statements suggesting that mucilage-bound algae are more common in nutrient-
poor waters than in enriched systems and that organisms that produce variable
amounts of mucilage (such as Phaeocystis) produce more when nutrient (especially
phosphorus) concentrations are depleted (Margalef, 1997). The production of
gelatinous polysaccharides has been observed among marine phytoplankton pop-
ulations in the photic layer that have become ‘aged by nutrient deficiency’ (Fraga,
2001, citing Vollenweider et al., 1995; Williams, 1995). Margalef (1997) makes the
point that mucilage production is more frequent over long days and in shallow
mixed layers. This, too, might be indicative of causal imbalances in the intracellular
metabolism of carbon, nitrogen and phosphorus.
Mucilage certainly has a high water content and, in some cases, it must disperse
rapidly into the medium (whilst being replaced by secretion from the proximal side).
In spite of the high density of the polysaccharide, the mean density of the mucilage
in live Microcystis colonies is within 0.07% of the density of the water in which they
are suspended (Reynolds et al., 1981) (see Section 2.5.2). This has for long nurtured
the supposition that mucilage contributed to the suspension of phytoplankton by
reducing average density. This it certainly can do, but it is not effective in reducing
sinking rate unless the overall dimensions comply with Eq. (2.17) (Hutchinson,
1967; Walsby and Reynolds, 1980).
Other functions proposed to be fulfilled by mucilage include the following.
Streamlining
Almost in direct contradiction to the principle of reducing sinking rate, a large
mucilage investment enhances floating and sinking responses to self-regulated buoy-
ancy changes in colonial Cyanobacteria such as Microcystis and Woronichinia, making
controlled migrations in natural water columns feasible (See pp. 68, 81).
Nutrient storage
The mucilage has been supposed to provide a repository for the concentration
and storage of essential nutrients (e.g. Lange, 1976). No mechanism for this has
been suggested; it is not clear how outward diffusion gradients and progressive
dilution and dissipation of the mucilage effects could be countered.
Nutrient sequestration and processing

In nutrient-dilute environments, encounter with sufficient limiting nutrients is an
empirically demonstrable problem (Wolf-Gladrow and Riebesell, 1997) (see Sec-
tion 4.2.1). It is possible that a mucilaginous coat provides a cheap mechanism for
increasing the size of the algal target whilst simultaneously providing a microenvi-
ronment wherefrom the rapid uptake of the nutrients across the cell wall (Section
4.2.2) maintains a yet more dilute than the exterior environment and a help-
ful inward gradient. No compelling demonstration of this nutrient scavenging has
been offered. For cells producing phosphatases designed to work externally (Sec-
tion 4.3.3), there is a need to confine the activity close to the sites of intracellular
uptake, which function could arguably be fulfilled by a mucilaginous boundary layer.
To be valid, however, the entry of organic solutes must be faster than the loss of
phosphatase. Again, no compelling experimental evidence is available to verify this.
Metabolic self-regulation
Nutrient-deficient cells may be prevented from completing their division cycle
(Vaulot, 1995) (see Section 5.2.1) but they cannot stop photosynthesis. Margalef
(1997) proposed that sheaths slow down diffusion and minimise unnecessary
metabolic activity. Although this idea fits with some of the field observations and
also matches to the one well known to algal culturalists, that mucilage is usually lost
quickly in laboratory strains, it is not clear that colonial species (e.g. of Coenochloris:
Reynolds et al., 1983b) fail to attain maximum rates of growth under favourable
supplies of nutrients and light. This they do without apparent loss or dimnution of
the colonial form.
Defence against oxygen

Sirenko and her co-workers (reviewed in Sirenko, 1972) demonstrated a low-
redox microenvironment is maintained within the mucilage of several species of
Cyanobacteria, apparently through the production of sulphydryl radicals. They have
argued that this helps to protect against oxidative processes and leads to tolerance
of high external concentrations of oxygen (similar comments were also made by
Gusev, 1962, and Sirenko et al., 1969).
Defence against metal poisoning

The selective permeability of mucilaginous envelopes might provide a defence
against the uptake of toxic cations in the acidic environments tolerated by some
desmids (Coesel, 1994). This idea has been investigated by Freire-Nordi et al.
(1998), applying electron paramagnetic resonance to compare the decay rates of
hydrophobically labelled tracers in normal sheathed cells of Spondylosium and in cells
divested of their mucilage by ultrasound. Decay was slower in cells with mucilage,
because of a suspected interaction with –OH groups in the polysaccharide. They
concluded that such interactions could play a decisive role in uptake selectivity.
Defence against grazing

Mucilaginous sheaths reduce the palatability of algae by making them too large
for microplankters to ingest, more difficult for mesoplanktic raptors to grasp and
less filterable and more mechanically obstructive for cladocerans (see Sections
6.4.2, 6.4.3). There is also recent evidence that free-living picocyanobacteria are
stimulated to form colonial structures in response to the presence of herbivores
and to the chemicals in the water that herbivores have recently vacated (Komárková
and Šimek, 2003) (see Section 6.4.2).
Defence against digestion

If mucilaginous algae fail to avoid ingestion, they may resist digestion during the
period of their passage through the guts of some (but not all) consumers. The
original observations of Porter (1976) have been verified by others, including
Canter-Lund and Lund (1995). The simultaneous scavenging of nutrients by viable
algae during passage and their deployment is a bonus function of mucilage (see
Section 6.4.2).
What may be concluded? There is no single clear function of mucilage, and not
all those suggested could be considered valued judgements. The buoyant prop-
erties, the reducing microenvironment, the selective permeability and the grazing
deterrence are backed by good empirical, experimental verification. These may
all be different, positively selected adaptations to what may have been originally
homeostatic mechanisms for balancing cell stoichiometry.
concentration of 0.8 µg C mL−1 is necessary to sat- (Asterionella and, especially, Cryptomonas, Ankyra).
urate its maximum assimilation. The correspond- With maturing individuals producing abundant
ing minimal and saturating concentrations for parthenogenetic eggs and more animals recruit-
the 2.1-mm adult are quite similar, at 0.14 and ing freely to the standing stock, the aggregate
0.87 µg C mL−1 . daily filtration quickly appreciated to some 600-
The figures make no allowance for wastage or -800 mL L−1 d−1 (Thompson et al., 1982; see
rejection (Fig. 6.6) or for the slightly different out- also top frame of Fig. 6.8). During this particu-
comes at higher temperatures. A further aspect lar phase, the mean body length of a cohort of
that was appreciated by Lampert (1977c) and individuals, feeding mainly on growing Ankyra
developed by Lampert and Schober (1980) and Gli- cells and at water temperatures close to 20 ◦ C,
wicz (1990) is that, for any filter-feeder, there is increased from 0.8 to 1.7 mm in 13 days and,
an approximate upper threshold concentration simultaneously, recruited a fivefold increase of
of edible, filterable foods when the maximum neonates of the next generation. Applying the
physiological demands of growth and reproduc- contemporaneous direct estimates of Thompson
tion are satisfied. This corresponds to the inflec- et al. (1982) to Eq. 6.14, there was, over the same
tion point shown in Fig. 6.4. Of rather greater period, a 12-fold increase in the collective filtra-
significance, however, is the threshold food con- tion volume, Fi . Reynolds (1984a) used these
centration that must be surpassed before an indi- measurements to express the resource-replete
vidual animal can balance respiration and main- growth in Daphnia filtration as an exponent: from
tenance and at which growth is zero. A slightly (ln 12) / 13 days, the specific rate of increase in Fi
higher threshold than this is necessary to main- is 0.1911 d−1 at ∼20 ◦ C. In contrast, recruitment
tain the population, allowing some maturation of Daphnia had been negligible in April, when
and egg production to offset mortalities (Lampert water temperatures were still below 8 ◦ C, despite
and Schober, 1980). an apparent abundance of food at that time (Fig.
To quantify the dynamic variability in the 6.8, top panel: over 2 µg C mL−1 ). Supposing
relationship between food and feeders was a a geometric scaling of temperature dependence
major objective of the studies in the Blelham of growth and recruitment capacity between 8
enclosures (Section 5.5.1, Fig. 5.11). In the early and 20 ◦ C, Reynolds (1984a) approximated a rate
part of the programme to measure loss rates of increase in the resource-saturated maximum
of primary product (Reynolds et al., 1982a), phy- aggregate filtration of 0.0159 K−1 .
toplankton was allowed to develop free of the The rate of change in the aggregate filtration
constraint of nutrient supplies and the herbi- capacity (F) is not the same thing as the popula-
vore populations developed in an environment tion growth rate for Daphnia, although there are
that was substantially free of predators. The analogies and some shared information. Growth
deep-water enclosures (A, B in Fig. 5.11) were rates of Daphnia spp. under comparable condi-
devoid of any significant numbers of fish, while, tions noted in Gliwicz (2003a: ∼0.2 d−1 ) are of
fortuitously, the restriction of enclosure open- similar magnitude. However, the seasonal fluc-
ing to just a few weeks in winter prevented tuations in Fi provide a ready and reasonably
the inward migration of chaoborid larvae from sensitive indicator of the impact of grazing on
their shallow overwintering sites (Smyly, 1976). the food resources and the consequences of food
Thus, it has been assumed that the herbivore depletion on the survival of Daphnia. Of the
periodicities that were observed were susbstan- examples from the Blelham enclosures plotted
tially driven from the bottom up, being regulated in Fig. 6.8, reference has already been made to
mainly by temperature and variable food avail- events in Enclosure A during 1978 (A78). Sub-
ability. An example is included in Fig. 6.7. Once ject to satisfaction of the temperature constraint,
water temperatures exceeded 7--8 ◦ C, daphniids the episodes of increase in Fi (at one point, to
rapidly established themselves as the most signif- >1 L L−1 d−1 ) were observed in the presence of a
icant consumers, responding quickly to an abun- finite resource base of filterable algae (defined in
dance of algae of filterable size food resources the study as being individually <104 µm3 ; mean
Figure 6.7 Seasonal variations in

the abundances of the main
zooplanktic species in the Blelham
Enclosure A, during 1978. The
numbers of Daphnia in each of five
size categories are shown separately
(I, < 1.0 mm, 17 ± 3 ◦ C; II, 1.0−
1.3 mm; III, 1.3−1.6 mm; IV, 1.6−
1.9 mm; V, > 1.9 mm). Redrawn
d = 26 µm). These increases were not sustained, response of the aggregate filtration rate could
however, as the calculated aggregate filtration scarcely be better correlated.
capacity periodically collapsed, owing, on each Overall, there is manifestly no reciprocity
occasion, to severe reductions in the numbers between food and feeders, whereas, under
of filter-feeders present. In almost all of these the contrived conditions, the latter plainly
instances, the herbivore collapses followed severe tracked the former. Baldly, without a resource,
depletion in the filterable algal mass. In Enclo- filter-feeding is unsustainable. Filter-feeders can
sure B in 1979 (B79), aggregate filtration rate was increase, provided the minimum threshold con-
generally reduced and confined to a narrower centration of edible, filterable foods is exceeded.
time window than in A78, owing to the dom- The evidence from Fig. 6.8 is that the threshold
inance of the algal biomass by Planktothrix (in is pitched between 0.10 and 0.13 µg C mL−1 . Sub-
spring) and Microcystis (in summer). Both were stantially greater concentrations than this will
considered to be inaccessible food resources to support growth of filter-feeders. Once the aggre-
the filter-feeders on the grounds of size. In B81, gate rate of loss of algal foods to filter-feeders
grazing was generally modest throughout a year (rG ) is in excess of the net rate of their recruit-
when filterable foods were conspicuously low ment (r , or r net of all other simultaneous loss
throughout the year. In A83, filterable foods were rates), rapid exhaustion of the food resource (to
abundant in the spring but not in summer; the <0.10 µg C mL−1 ) is inevitable. This is followed,
Figure 6.8 Seasonal variations in

the phytoplankton biomass (as
carbon) in Blelham enclosures
during selected years and its
approximate fractionation between
filterable (hatched) and
non-filterable (unshaded) size
categories. The bold dashed line is
the aggregate daily filtration rate
generated by the zooplankton,
which, at times, exceeded 1.0 (sensu
1 litre per litre of enclosed water
volume per day). Note that the
filtration volume mainly reflects the
numbers of filter feeders present
and that it responds to the
availability of suitable foods
(abundance and filterability).
Reynolds (1986b).
almost as inevitably, by starvation and death, par- 16.5 s. At food concentrations below this thresh-
ticularly of the neonates (Ferguson et al., 1982). old level, animals of both sizes will fail to ingest
sufficient food to satisfy their minimum needs.
Food thresholds and natural populations In reality, several potential foods may be present
For completeness, the lower and upper thresh- but, if their combined carbon content (supposing
old concentrations impinging on the ability of D. it to be wholly filterable and assimilable) is sub-
galeata populations to survive or to saturate their stantially below the threshold concentration, the
capacity for growth are expressed in terms of the animals will starve.
populations of some common fresh-water plank- While it is proposed that these thresholds
ters (Table 6.3). The concentrations cited should set firm boundaries within which the mutual
be interpreted as defining the fufillment or sat- dynamics of phytoplankton and zooplankton
uration or otherwise of the Daphnia’s nutritional interact, it is also fair to say that, in the real
requirements, were the food resource monospe- world, they are not often easily recognisable.
cific. For example, were Cryptomonas ovata the There are several reasons for this and they need
only food available, a population of 175 mL−1 to be acknowledged. First, not all filter-feeding
would be the minimum that will could meet the zooplankton need experience the same threshold
feeder’s basic maintenance needs at 15 ◦ C. In the concentrations. Second, there is the subsidiary
case of the 0.8-mm neonate, filtering up to 5 mL issue that larger individuals of any filter-feeding
of water during each 24 h, it would be expected species can ingest larger prey. This means that
to encounter 875 algal cells in a day, at an average the lower metabolic threshold can be fulfilled by
frequency of one every 99 s. A 2.1-mm adult, feed- a wider size range of foods, so that the resource
ing in the same suspension, should be capable of base of the larger filter-feeder is greater than
ingesting 5250 algae during the day, or one every that of the smaller one. Moreover, the algal food
Table 6.3 Equivalent concentrations of selected algal and bacterial foods representing the minimum and saturating
threshold requirements of Daphnia maintenance, growth and reproduction
Populations equivalent to
Volume Cell C
Food species (µm3 ) (pg) 0.1 µg C mL−1 0.8 µg C mL−1
Cryptomonas ovata cell 2710 569 175 1 406
Scenedesmus quadricauda 1000 225 444 3 560
(four-cell coenobium)
Asterionella formosa cell 645 85 1 176 9 411
Plagioselmis nannoplanctica cell 72 15 6 667 53 333
Synechococcus cell <30 <7 >14 285 >114 280
Ankyra judayi cell 24 5 20 000 160 000
Free-living bacteria – 0.013 7.7 × 106 6.2 × 107
resource of all filter-feeders may be supplemented imum threshold is some 4--7 × 106 mL−1 . Even in
to a degree by a resource of suitably sized partic- these instances, any further inroad into the fil-
ulate organic matter (detritus) and by free-living terable resource leaves filter-feeders severely food
bacteria. Finally, predation of planktic herbivores limited. Almost all the other components of the
distorts the impression of community structure, microbial food web, including the nanoflagellates
function and the outcome of dynamic processes. and ciliates, as well as fine, suspended detritus,
Gliwicz (1990) compared the body-growth will have been eliminated by daphniids feeding
rates of a number of Daphnia and Ceriodaphnia in significant numbers (Porter et al., 1979; Jürgens
species as a function of food concentration. Some et al., 1994; Sanders et al., 1994; Wiackowski et al.,
modest interspecific differences in the lower 1994).
thresholds were evident (in certain cases, these Do these thresholds not then determine that
could be as low as 0.03 µg C mL−1 ), with the the abundance of zooplankton in pelagic systems
best survivorship being noted among the larger having a biological supportive capacity substan-
species of D. magna and D. pulicaria. Given that tially greater than 0.1 mg C L−1 would tend to
larger animals also filter more water and ingest alternate between glut and self-inflicted dearth
larger particles, we may deduce that, potentially, of the largest filter-feeding species present? In
adults of the larger species, together with mature the case of Blelham Enclosure A in 1978 pre-
adults of the intermediate size classes, enjoy a sented above, the surges and collapses in the
wider filterable resource base and may be better Daphnia population would appear to conform to
adapted to survive instances of periodic starva- this supposition. Elsewhere, of course, the zoo-
tion than juveniles or adults of small species. plankton dynamics are moderated by planktivory
On balance, ecological evidence is supportive (and especially by fish) which tends to damp
of the physiological deduction. Daphniids of all the fluctuations between famine and plenty.
kinds seem to be relatively scarce in the plankton Furthermore, the zooplankters that are both
of lakes in which the carbon supportive capac- more visible and more rewarding to the plank-
ity is generally ∼0.1 mg C L−1 . Moreover, devel- tivore are the larger ones, and, other things
opment in waters where the supportive capacity being equal (see Sections, 6.4.5, 8.2.2, when inter-
may only coincide with filterable POC concentra- actions and entire pelagic resources are dis-
tions >0.05--0.1 mg C L−1 . If fulfilled exclusively cussed), selection is against large animals and in
by algal plankton, the corresponding threshold favour of the survival of smaller animals and of
is equivalent to a chlorophyll concentration of 1- smaller species that reproduce at smaller body
-2 µg chla L−1 . If substituted by bacteria, the min- sizes (Gliwicz, 2003a, b). As a consequence, the
survival prospects of each species in the zoo- tive and the effects of filter-feeding zooplankton
plankton and the resultant structure of the com- on phytoplankton are less intensively regulated
munity are poised between the influences of food than they are above it.
limitation and herbivore predation. Zooplankton The elegant study of the feeding selectivity
ecologists have, for long, recognised the interac- of young roach (Rutilus rutilus) undertaken by
tion between, on the one hand, the ability to Townsend et al. (1986) revealed the progressive
feed, assimilate, grow and reproduce under con- switching between benthivory and planktivory
ditions of resource stress and, on the other, the in response to the abundance of planktic crus-
vulnerability to predation. The balance of these taceans in the range 10--50 mg C L−1 . The implied
counteracting influences was expressed in the threshold is not a fixed one, for it depends upon
elegant size-efficiency hypothesis (SEH) of Brooks the availability and accessibility of benthos, as
and Dodson (1965). Planktivory intervenes in the well as upon the numbers of fish competing for
tendency of the large-bodied species of zooplank- it. Its existence explains tacit upper limits on
ton to monopolise the resources to the exclusion the concentrations of filter-feeding zooplankton
of small-bodied species (including, it might be (equivalent to as few as 10--20 large Daphnia L−1
added, those of the microzooplankton). Selective but 100--200 small-bodied species: Kasprzak et al.,
predation of large-bodied species favours larger 1999), that could be set as much by their attrac-
populations of small-bodied species that repro- tivity to predators as by the sustainability of their
duce when they are still small. The relative abun- phytoplanktic food resources. The consequences
dance of large- and small-bodied species is thus impinge on the structure of planktic food webs,
strongly influenced by the intensity of predation the predominant energy pathways in aquatic sys-
by planktivores (cf. Hrbáček et al., 1961). tems and the role of the phytoplankton in sus-
Although the broad thrust of the SEH remains taining either. These issues are the subject of
wholly acceptable, it is important to recog- Section 8.2.
nise Gliwicz’s (2003b) distinction between the
dynamic effects of the immediacy of predation 6.4.3 Selective feeding
and the gradual debilitation forced by food short- In fresh waters, mesoplanktic filter-feeding
ages. The relationship is further complicated by clearly has a lower threshold of viability, that
the fact that, in shallow or marginal areas of falls in the range 0.03--0.1 µg filterable C mL−1
water bodies, predators may easily switch their (30--100 mg C m−3 ). This value is set by the biol-
foraging efforts to the frequently more rewarding ogy of the most successful cladocerans and not
resources of the littoral and sub-littoral benthos. by some physiological override. After all, tuni-
In this way, the intensity of planktivory on sparse cates survive in oceanic waters supporting chloro-
zooplankton populations may be less than might phyll concentrations habitually in the range 0.1--
be deduced from the abundance of predatory 0.3 µg chla L−1 (say, 5--15 mg C m−3 ). In fairness,
species. Thus, it may be only in the true pelagic, it must be added that the filtering rate to body
well away from the influences of shores and sedi- mass in tunicates has to be high and their rates of
ments, that planktivory exerts the expected con- growth are modest compared with those of clado-
tinuous constraint on the growth and recruit- cerans (Sommer and Stibor, 2002). The only viable
ment pattern in the zooplankton. Elsewhere, the foraging alternative in the rarefied, low-biomass
effects of planktivory on the zooplankton may be worlds of the open pelagic, of both marine and
a more casual or more opportune constraint, at fresh-waters, is to be able to locate and select prey
least until the zooplankton offers a sufficiently of high nutritive value and to be successful in its
abundant and attractive alternative food refuge. capture.
The corollary of this deduction is that there is The point has been made above that this abil-
a further conditional threshold for the feeding ity is particularly developed among the plank-
pressure exerted on the phytoplankton that is tic copepods. The majority of cyclopoids are
dependent upon herbivore engagement. Below it, exclusively raptorial, feeding on algae in their
the benthic alternatives remain the more attrac- juvenile naupliar stages, then becoming more
carnivorous in the later copepodite and adult There is thus considerable complementarity
stages, when they actively select rotifers and in the respective dynamics, spatial and tem-
small cladocerans. Adult calanoids have the abil- poral distributions of calanoids and daphniids.
ity to filter-feed but they have the further facility Though they frequently coexist in mesotrophic
to supplement the spectrum of available foods by and mildly eutrophic lakes, the obligately filter-
seizing, grappling and fragmenting microplank- feeding cladocerans are unable to satisfy their
tic algae. When both options are available, adult metabolic needs at chronically low food con-
diaptomids supplement the yield of filter-feeding centrations. Thus, calanoid dominance is rela-
with larger items that they capture, and usu- tively common among oligotrophic lakes, usu-
ally ingest more food than could have been sup- ally in association with an underpinning micro-
plied contemporarily through filter-feeding alone bial food web. If the supportive capacity of
(Friedman, 1980). The benefit of doing so presum- POC (be it detrital, bacterial or algal) is signif-
ably increases inversely to deficiency in the con- icantly >0.1 mg C L−1 , Daphnia should be able
centration of nanoplanktic particles available. to thrive and, for so long as the food concen-
Moreover, upward extension of the size range of tration satisfies the demand, to reproduce and
ingestible foods to 30 µm or more permits adults recruit new individuals rapidly. The condition
of Eudiaptomus to feed on algae as large as Cos- may persist until, potentially, the Daphnia popu-
marium and Stephanodiscus (Gliwicz, 1977; Rich- lation is processing such large volumes of water
man et al., 1980). However, it is the selection of each day that not just the recruitment of detri-
ciliates (e.g. Strickler and Twombley, 1975; Hart- tus, bacteria and filterable algae but also the com-
mann et al., 1993) that most broadens the main ponents of the entire microbial food web are
diet of calanoids. This may be critical to their sur- exhausted (Lampert, 1987; Weisse, 1994). Other
vival in very oligotrophic marine and fresh-water things being equal, such occurrences are fol-
systems, wherein it also proves to be the deci- lowed by significant mortalities of cladocerans
sive trophic linkage in the carbon metabolism of and, to a lesser extent, other mesozooplank-
lakes and seas of low biomass-supporting capac- ters too (Ferguson et al., 1982). These observa-
ity (Cole et al., 1989; Weisse, 2003) (see also tions fit well with established features of lim-
Section 8.2). netic zooplankton ecology, especially those relat-
The undoubted effectiveness of calanoid for- ing to shifts in dominance. Cladocerans gener-
aging under these circumstances combines the ally respond positively to anthropogenic eutroph-
animals’ well-developed ability to select the sizes ication (Hillbricht-Ilkowska et al., 1979), bringing
of food attacked and ingested with their impres- greater amplitude of fluctuations in the biomass
sive facility to be able to chemolocate (De Mott, of Daphnia and its foods (McCauley et al., 1999;
1986), then orientate towards and strike out at Saunders et al., 1999).
their potential prey organisms (Strickler, 1977; Not all cladoceran families are obligate filter-
Alcaraz et al., 1980; Friedman, 1980). The abil- feeders. Some of the littoral-dwelling chydorids
ity of calanoids to survive on modest rations and macrothricids have thoracic limbs that are
is a further factor contributing to their fre- adapted to scrape periphyton from the surface
quent dominance of the oligotrophic mesozoo- of leaves, etc. Chydorus sphaericus is frequently
plankton (Sterner, 1989). To judge from Hart’s seen among larger microphytoplankton in small,
(1996) data (see P. 262), calanoids are able to eutrophic lakes, often clutching onto colonies
draw sufficient food to satisfy their maintenance such as Microcystis and scraping epiphytes from
demands from POC concentrations that are an the mucilage (Ferguson et al., 1982; Ventelä et al.,
order of magnitude more dilute than those tol- 2002). Though the bosminids have filtering phyl-
erated by cladocerans. Their demands for energy lopods and use them for this purpose for much of
and reproductive investment may well be satu- the time, they are also capable of supplementing
rated at food concentrations (0.08 µg C mL−1 ) that nanoplanktic particles from the water by seizing
would hardly keep Daphnia from slowly starving larger individual prey items. To do this, they
to death. use the (fixed) antennules, abdominal claw and
carapace gape, fragmenting the potential food of between 0.1 and 0.2 µg C (µg C)−1 d−1 .
to pieces of more ingestible size (De Mott, 1982; The imposition of selective herbivory on the
Bleiwas and Stokes, 1985). Selectivity of food is relevant size ranges of specific foods may be
aided by chemoreception (De Mott and Kerfoot, approximated from the sum of the individual
1982). number--mass products of consumers per unit
Neither are the advantages of food selectivity volume. By analogy to Eq. (6.13),
confined to the exploitation of scarce resources
G = (N 1 · G 1 ) + (N 2 · G 2 ) + · · · (N i · G i ) (6.15)
in oligotrophic conditions. In waters rendered
turbid through the repeated resuspension of where Gi is the ingestion rate and Ni is the stand-
clay and fine silt particles (1--20 µm), as many ing population of the ith species-size category of
or more of the food-sized particles are inert and selective grazer. Finally, a rate of grazing on the
near useless and greatly diminish the benefits of food species affected, to be set against other expo-
filter-feeding. Selective feeding provides a more nents of change for the food species affected.
productive food return for the foraging effort.
rG = (G )/V (6.16)
This is a functional explanation for the common
observation that the zooplankton of turbid By way of an example, Ferguson et al. (1982)
lakes, even quite eutrophic ones, is typically observed populations of Eudiaptomus gracilis in
dominated by calanoids or small cladocerans planktivore-free enclosures of up to 20 adults
(Allanson and Hart, 1979; Arruda et al., 1983). L−1 . Supposing an average length of 1.5 mm
High densities of large or mucilaginous algae in and an individual mass equivalent to 9.3 µg C,
productive waters also constitute a nuisance to food intake may be approximated to be <2 µg
large, obligate filter-feeders such as Daphnia, espe- C animal−1 d−1 , with the population demand
cially when the algae are close to the upper limit- peaking at <40 µg C L−1 d−1 . In terms of the
ing size. Although Daphnia species exercise their principal food ingested at that time, Asterionella
limited powers of food selection (varying the (85 pg C cell−1 ), the maximum demand could
carapace gape), they do less well than bosminids have been equivalent to a loss of not more than
that filter-feed on abundant nanoplankton and 470 cells mL−1 d−1 . For smaller populations of
bacteria nourished by organic solutes released smaller animals at lower temperatures, the food
from the algae. Cascading reactions among the demand would normally have been rather lower
heterotrophic nanoflagellates are also noted than this maximum (perhaps 10--20 µg C L−1
(Ventelä et al., 2002). An analogy comes to mind d−1 is a more likely optimum). Numbers may
of small animals grazing or browsing among be sustainable on the relatively smaller daily
trees, which are, nevertheless, of sufficient size rations of just 1--2 µg C L−1 d−1 . Relative to algal
and density to exclude larger animals. In the standing crops, the removal of 470 cells mL−1
end, these same larger algae may share part of d−1 is scarcely sustainable by anything under
the role of planktivorous fish in defending the 500 cells mL−1 doubling each day but, to a
entire community against total elimination by standing crop of 10,000 cells mL−1 , it represents
overwhelming Daphnia filtration rates. an exponential loss rate of rG < 0.05 d−1 .
The demand made on the resources by selec-
tive feeders is more difficult to gauge than is 6.4.4 Losses to grazers
that of less specialised filter-feeding. The carbon The combined effects of selection and filter-
intake required by calanoids saturates at close feeding impose differential rates of removal upon
to the minimum requirements of filter-feeding the species composition of the phytoplankton.
cladocerans. In consideration of a small (0.8-mm) Some (mostly smaller species, generally <50 µm
and large (2.1-mm) Daphnia, the supposed mini- in maximum dimension, but often mainly
mum daily carbon requirements (0.6 and 4.3 µg nanoplanktic species) are removed primarily by
C respectively) service projected body masses cladoceran filter-feeding but also by other crus-
equivalent to 3.2 and 42 of µg C respectively taceans. Others, generally the smaller microphy-
(Box 6.2), to provide daily carbon-specific intakes toplankters (20--100 µm in maximum dimension),
Box 6.2 Carbon equivalents of planktic components
In interrelating observations on the biomass and production of plankton with the

distribution and flow of organic carbon through pelagic ecosystems, it is helpful to
have a ready base for interconversion. Many such conversions are available in the
literature and some have been invoked in this book. The selection of conversions
below is the one used throughout this book. The veracity of the relationships
is not always well known and, for any particular species, something better can
generally be found. On the other hand, for generalisations and order-of-magnitude
flux estimates, the following statements will be found to be helpful.
Free-living bacteria
Conversions linking carbon content and biovolume were explored by Lee and
Fuhrman (1987); more general averages (0.01–0.02 pg C cell−1 ) are given in Sorokin
(1999). In calculations here, I follow the determination adopted by Ferguson et al.
(1982) for freshwater bacteria in a eutrophic reservoir: 1 × 106 cells / 13 ng , or
carbon content of bacteria, 0.013 pg C (cell)−1 .
Non-diatomaceous phytoplankton
As developed in Section 1.5 the carbon content of most phytoplankters (save for
diatoms) conforms well to the regression C = 0.225 v0.99 (Fig. 1.8) where v is
the cell volume. Thus, the carbon content of planktic algae is ∼0.225 pg C µm-3 .
(Individual cells range from >1 pg to 10 ng C.)
Diatoms
The large vacuole and high ash content of the cell wall make it difficult to rely on
the above relation. A more reasonable estimate is to take C = 50% of the ash-
free dry mass. The ash content can be approximated on the basis of cell volume
according to the regression in Fig. 1.9. Table 1.5 suggests an average Si of 0.1 pg
Si (or 0.21 pg ash) per µm3 of cell volume. In this case, 0.15pg C µm−3 is also a
helpful approximation.
Nanoflagellates
Assumed to conform to the relationship for phytoplankton. Thus, carbon content
of planktic algae is ∼0.225 pg C µm−3 (individual cells in range 1–900 pg C).
Ciliates
Pending better information, individual cells probably range from 1 to 100 ng. Herein,
‘small planktic ciliates’ are considered <30 µm, with a carbon content of ∼3 pg C;
‘large ciliates’ (60 µm) may easily comprise 50 ng C.
Rotifers
Examples from Bottrell et al. (1976), assuming carbon is 50% dry weight. All means
subject to margin of at least ±30%.
Polyarthra spp. 35 ng C (individual)−1
Keratella cochlearis 50 ng C (individual)−1
Kellicottia longispina 100 ng C (individual)−1
Brachionus calyciflorus 110 ng C (individual)−1
Filinia longiseta 150 ng C (individual)−1
Keratella quadrata 150 ng C (individual)−1
Asplanchna priodonta 250 ng C (individual)−1
Cladocera: Daphnia
Bottrell et al. (1976) collected, presented and pooled a wide range of regressions
relating body mass to body length in Daphnia, of the form ln (Wanimal ) = ln a + b
ln (Lanimal ). With Lb in mm, W (in µg) is generally well predicted by the slope b =
2.67 and the intercept ln a = 2.45. Subtraction of ln 2 (0.693) gives the prediction
ln[C ] = 2.45 − 0.693 + 2.67 ln(L b ).
Some examples:
Lb = C (individual)−1 =
0.8 mm 3.2 µg C
0.95 mm 5.0 µg C
1.3 mm 11.7 µg C
1.6 mm 20.3 µg C
2.1 mm 42 µg C
2.24 mm 50 µg C
4.0 mm 234 µg C
Cladocera: Bosmina
Bottrell et al. (1976) pooled several regressions relating body mass to body length
in Bosmina, of the form ln (Wanimal ) = ln a + b ln (Lb ). With Lb in mm, W (in µg)
is generally well predicted by the slope b = 3.04 and the intercept ln a = 3.09.
Subtraction of ln 2 (0.693) gives the prediction
ln[C ] = 3.09 − 0.693 + 3.04 ln(L b ).
Some examples:
L = C (individual)−1 =
0.3 mm 0.3 µg C
0.7 mm 3.7 µg C
1.0 mm 11.0 µg C
Copepoda
Bottrell et al. (1976) pooled several regressions relating body mass of copepods
to body length of the form ln (Wanimal ) = ln a + b ln (Lb ). With Lb in mm, W (in
µg) is generally well predicted by the slope b = 2.40 and the intercept ln a =
1.95. Subtraction of ln 2 (0.693) gives the prediction
ln[C ] = 1.95 − 0.693 + 2.4 ln(L b ).

Some examples:
Lb = C (individual)−1 =
0.1 mm 0.014 µg C
0.3 mm 0.2 µg C
1.0 mm 3.7 µg C
1.5 mm 9.3 µg C
2.0 mm 18.5 µg C
2.5 mm 31.7 µg C
Satapoomin (1999) published relationships of mass and carbon content for several
marine species of copepods. For body lengths of 0.5 to 1.4 mm, Centropages carbon
contents of 0.5–16.6 µg C (individual)−1 and Temora 1–33 µg C (individual)−1 are
found.
Euphausiids
Lindley et al. (1999) presented regressions relating body mass of some euphausians
to various allometric length measures, in the form log (Wanimal ) = a + b log (Lc ),
where Lc is the body length from the rostrum to the telson tip. Putting a = 0.508
and b = 2.723, and the typical carbon content of 40% dry mass, [C] is predicted
as:
ln[C ] = 0.508 − 0.4 + 2.723 log(L c ).
Some examples:
Lc = C (individual)−1 =
2 mm 8.5 µg C
6.3 mm 193 µg C
20 mm 13 620 µg C
are taken mainly by calanoids. Larger micro- feeding is normally capable of clearing the water
plankton may be immune from either, although of small algae and can almost wholly account for
even these may be liable to specialist predators, the demise of extant populations.
which may include rotifers and protists. Because it still seems to generate surprise
In the size range, 20--50 µm, phytoplankton among some limnology students, it needs to be
may be liable to attack by both main kinds of emphasised again that the simultaneous pres-
crustacean. The dynamic rates of their removal ence of grazers and grazed species is in no way
from their respective populations should be for- incompatible. In fact, with the one being depen-
mally restated by combining Eqs. (6.14) and dent upon the other, no other possibility is ten-
(6.16): able. Referring back to Eqs. (5.3) and (6.1), graz-
ing is tolerated by the grazed population so long
as r > rL and rn remains positive. True, the
rG = [ω ( F i ) + (G )]/V (6.17) rate of loss, rL , at least for some species, may be
very largely attributable to the rate of removal
In relative terms, selective feeding is the more by grazers, rG . Moreover, rising removal rates,
prevalent loss process where filter-feeders fail especially from recruiting populations of clado-
to operate. However, as has been suggested, the cerans, can easily exceed declining rates of algal
dynamic effect of filter-feeding, where it is sustain- recruitment. These are the circumstances of the
able, is potentially much greater than selective rapid elimination of the food species. Certainly,
feeding in absolute terms. Only cladoceran filter- under these conditions, loss to grazers becomes
Table 6.4 Derived rate constants for two consecutive development phases of the nanoplanktic alga Ankyra judayi in
Blelham Enclosure A during the summer of 1978
Period Observed rn rS a rG b Back-calculated r

12 Jun–3 Jul 0.497 0–0.143 0.193–0.622 0.690–1.262
3 Jul–10 Jul 0.122 c
0.644–1.416 >0.766
10 Jul–24 Jul −1.043 0–0.082 0.595–1.341 0–0.380
14 Aug–28 Aug 0.493 0–0.344 0.418–2.022 0.911–2.859
28 Aug–4 Sep −0.150 0.046–0.068 0.253–0.556 0.149–0.474
4 Sep–18 Sep −0.643 0 0.244–0.714 0–0.071
a
Estimated from sediment traps set at depth.
b
Calculated from aggregate filtration rate only, assuming no rejection (ω = 1).
c
No traps set.
the main fate of the phytoplankton population in The upper sinking loss rates predict a bulk sedi-
question. mentation rather greater than was actually mea-
In mesotrophic and in mildly eutrophic lakes, sured during the same experiments (Reynolds
grazing losses are evidently heavy among species and Wiseman, 1982). Thus, the additions to the
vulnerable to filter-feeding. Reynolds et al. (1982b) one precise statistic, the observed net rate of
commented on the poverty of nanoplankton and increase, rn , should err on the low side.
small microplankton that were recoverable in The exponents are used to back-calculate the
deep sediment traps in Windermere. In the much bulk additions or subtractions in each phase.
shorter water columns of the experimental lim- Thus, supposing
netic enclosures in Blelham Tarn, too, abundant
N t = N 0 exp (r − rS − rG )t (6.18)
growths of Cryptomanas, Plagioselmis, Chromulina
and, especially, Ankyra that, at various times, then the increment or decrement is given by:
were stimulated (Reynolds, 1986b) contributed lit-
tle to the sedimentary flux. Within the confines N t − N 0 = N 0 {[exp (r − rS − rG )t] − 1} (6.19)
of the enclosures, some reasonable approxima- whence the number of cells produced (P) is cal-
tions of the rates of grazing removal were possi- culated as
ble. The entries in Table 6.4 track the changes in
the rates of increase and decrease through two P = [r /(r − rS − rG )] N 0 {[exp (r − rS − rG )t] − 1}
consecutive peaks of Ankyra in Enclosure A in (6.20)
1978 (during which weekly fertilising averted the By analogy, the number of cells sedimented (S)
likelihood of nutrient limitation and the main and grazed (G) are calculated:
grazers, Daphnia galeata, were almost uncon-
strained by predators). The estimates necessar- S = [rS /(r − rS − rG )] N 0 {[exp (r − rS − rG )t] − 1}
ily carry large error margins: these are properly (6.21)
shown, although it may be confidently stated that
G = [rG /(r − rS − rG )] N 0 {[exp (r − r S − rG )t] − 1}
the lower ends of the ranges shown are more real-
(6.22)
istic in each case. The true rate of replication (r )
of Ankyra under the general conditions of light In either of the depicted instances, Ankyra cells
and temperature obtaining during these times were first detected in near-surface water at con-
could scarcely have much exceeded 1.0--1.1 d−1 centrations in the order of 20--40 cells mL−1 .
in the earlier phase whereas a rate between 0.9 In July, this built to a maximum of over
and 1.0 would have applied during the second. 300, 000 cells mL−1 , still mainly confined to the
Table 6.5 Calculated production (P) of Ankyra cells and losses attributable to sedimentation (S) and grazing (G) in
each of the periods noted in Table 6.4
Period P (cells m−2 × 10−9 ) S (cells m−2 × 10−9 ) G (cells m−2 × 10−9 )
12 Jun–3 Jul 905–1655 0–188 253–816
c
3 Jul–10 Jul 5520–11084 4641–10204
10 Jul–24 Jul 0–556 0–120 871–1962
a 6425–13295 0–308 5765–12982
RW82b <14
14 Aug–28 Aug 1466–4601 0–554 673–3255

28 Aug–4 Sep 513–1631 158–234 871–1914
4 Sep–18 Sep 0–31 0 106–309
a 1979–6263 158–788 1650–5478
RW82 4–5
a
is the total over the wax and wane period.
b
RW82 is the direct measurement of Reynolds and Wiseman (1982) over the whole population.
epilimnion (1531 × 109 m−2 ). The contempora- zooplankton interactions. It is generally true that
neous development of a Daphnia population (see neither functions independently of the presence
Fig. 6.7) was shown by gut-content analysis to of the other, even if there is no direct link-
be largely sustained on Ankyra (Ferguson et al., age between the components. Excepting large
1982). Solution of Eq. (6.22) indicates that not colonial phytoplankters (like Microcystis or Volvox)
fewer than 4891 × 109 Ankyra cells m−2 would whose consumers are more likely to be epiphytic
have been eaten by then. Thus, not fewer than than planktic, or those microzooplankters whose
6422 × 109 cells m−2 would have been produced diet may be exclusively bacterial, most phyto-
(for comparison, the potential production of cells plankton is liable to become the food of some
over the same period and at a sustained rate of zooplankton at some time. Removal of primary-
0.69 d−1 would have led to the production of 49 producer biomass as the food of herbivores
× 1015 cells m−2 !) By the end of the population, inevitably means that phytoplankton biomass is
which was also rapidly cleared by grazers, losses smaller than it would have been had there been
to Daphnia of >5765 × 109 cells m−2 accounted no grazing. The converse might be that there
(Table 6.5) for 90% of the inferred production. would be a smaller consumer biomass if the con-
In the second phase of Ankyra growth, which sumers were denied access to primary foods and
built to a maximum of 671 × 109 cells m−2 , that these were not supplemented by an energet-
the deduced production was not less than 1979 ically equivalent organic carbon source.
× 109 cells m−2 , of which > 1650 × 109 m−2 If that much ‘states the obvious’, it has to be
would have been cropped by grazers, whereas said that there are few other popular assump-
the directly measured sedimentary export was tions about the interactions between phytoplank-
<5 ×109 m−2 (<0.3% of the total production). ton and zooplankton that are passed unchal-
lenged. For instance, there is no clear reciprocity
6.4.5 Phytoplankton–zooplankton between the abundance of phytoplankton and
interactions of zooplankton (see the examples in Fig. 6.8),
In order to draw a general perspective on although there may be an element of the ‘track-
the effects of consumers on the phytoplankton ing’ by zooplankton numbers of fluctuations in
and on the export of primary production to edible biomass. There is little evidence, either,
the aquatic food webs, it is helpful to formu- that grazing by zooplankton necessarily con-
late some overview of nature of phytoplankton-- trols the dynamics of the phytoplankton in any
also fittest to exploit opportunities. In this sense,

either the best competitors (sensu Keddy, 2001) or
the most exploitative animals will be favoured to
succeed.
This generalisation is well exemplified by
the two major groups of herbivores -- calanoids
and cladocerans -- in lake plankton. The daph-
niid cladocerans, especially, are efficient in gath-
ering and concentrating uniformly small food
particles from dense suspensions and, also, in
then turning carbon resource into new daphniid
biomass. Subject to the satisfaction of an overid-
ing condition of food adequacy, daphniids are
exploitative and expansive, and tend to domi-
nate overwhelmingly all other zooplankters, if
not by eating them, then by eating out their
resources. Their sensitivities relate to these same
adaptations: having the potential to grow to
Figure 6.9 Season-long comparison between (a) the
numbers of phytoplankton species attaining maximal large numbers and to relatively large individual
population within a given week of the year and (b) the body sizes makes them disproportionately vul-
contemporaneous zooplankton biomass. There is no nerable to planktivorous predators. Their restric-
correlation, in spite of the fact that the phytoplankton tion to filterable foods leaves them vulnerable
development was never nutrient-limited and the zooplankton to diminished and to chronically low concentra-
was almost entirely unpredated. Data from Ferguson et al. tions of algae and, perhaps surprisingly, also to
(1982) for Blelham Enclosure A 1978, and redrawn with an abundance of algae of varied or of larger sizes,
that interfere with systematic filtration of the
resources. In stark contrast, the calanoid cope-
predictable way. To judge from the example in pods fill in many of the gaps left by cladocerans
Fig. 6.9, there is no correlation between the num- through superior economy on a small resource
bers of phytoplankton species achieving their base and through greater discrimination in feed-
respective population maxima and the biomass ing on large food particles. Analogous principles
of unpredated herbivores. probably apply in the open sea, save that con-
ditions of resource deficiency are more general
Competitive interactions and more persistent. Cladoceran filter-feeding is
Part of the complexity of these otherwise obvi- seen to bring smaller returns than are earned by
ous relationships is due to a wide range of inter- calanoid selectivity.
specific interactions between key consumers and The interactions between the abundance and
key primary producers; another part is the conse- species structure of the phytoplankton and the
quence of forcing by the consumption at higher zooplankton are similarly contrasted according
trophic levels. There are also some effects trans- to whether the phytoplankton is cropped mainly
mitted in the opposite direction, whereby phy- by calanoids or cladocerans. Using small enclos-
toplankters benefit from the presence of zoo- ures (mesocosms) placed in a mesotrophic lake,
plankters. Consideration reveals some interesting Sommer et al. (2001) observed the effects on
discernable patterns in the linkages among the the composition of the summer phytoplank-
main food organisms and the main groups of ton of experimental adjustments to the relative
feeders. Gathering sufficient food to support sur- abundances of Daphnia and Eudiaptomus. The sup-
vival and reproduction is a general problem for pression of small phytoplankton by dominant
most animals, while those that are most efficient Daphnia and the suppression of larger species
in converting extra energy into extra biomass are by Eudiaptomus occurred much as predicted,
although the authors noted that neither was able through bacteria and consumers. This might also
to graze down the phytoplankton to the low con- apply to undigested foods rejected (but not nec-
centrations frequently observed in cladoceran- essarily undamaged) by so-called ‘sloppy’ filter-
dominated lakes in spring (Lampert, 1978; feeding. Cladocerans produce a rather diffuse
Sommer et al., 1986) (see also Fig. 6.8). Yoshida and amorphous waste which seems more recy-
et al. (2001) also undertook mesocosm experi- clable in the immediate vicinity of its release. It
ments on an oligotrophic lake plankton, testing seems likely that, even then, bacterial mineralisa-
the effect of fertility and macrozooplankton on tion plays a large part in regenerating nutrients
community structure. Raising the fertility alone that, once again, become available to phytoplank-
led to a higher biomass of algae in the nano- and ton. Probably of greater importance are the quan-
microplanktic size ranges and to more numerous tities of nitrogenous or phosphatic metabolites
heterotrophic nanoflagellates. The response to that may be excreted across the body surfaces of
higher populations of Eodiaptomus japonicus con- zooplankters. These are soluble and also readily
formed to a cascading impact, with fewer ciliates bioavailable to algae (Peters, 1987). The restora-
and bacteria but more heterotrophic nanoflagel- tion of nutrients to primary producers is one of
lates. With Daphnia galeata, algae, nanoflagellates the ways in which pelagic ecosystems contribute
and microzooplankton were all suppressed. to their own self-regulation and maintain their
With enrichment raised to the level of eutro- own resource base (e.g. Pahl-Wostl, 2003).
phy, when nutrient availability can support high The excretory wastes will always be domi-
levels of producer biomass, cladoceran feeding nated by the metabolites in excess. It has been
may eventually select against smaller algae and pointed out (Hessen and Lyche, 1991; Elser et al.,
in favour of the larger forms that are often dif- 2003) that animals will retain the amino acids
ficult for large Daphnia to ingest. Although these that are deficient in their food and metabolise
may not present such a problem to diaptomids, those that are in excess. Thus, Daphnia that meet
their relatively weak growth potential may delay their carbon intake requirements from planktic
any substantial control on the dynamics of the foods that are (say) P but not N deficient, will
larger algae. Under these conditions, there may retain relatively more of the P content than of the
be limited scope for microbial pathways to re- N content, so that the N : P ratio of their excretory
establish and for them to be exploited by abun- products is likely to be yet higher than that of the
dant microzooplankton (including rotifers, such food intake. Thus, the nutrients that are recycled
as Keratella and Polyarthra) or by small cladocer- do not necessarily assist in the correction of defi-
ans, either feeding selectively (Bosmina, Chydorus ciencies in the growth environment of the food
spp.). or by filtration in the nanoplanktic size organisms but rather accentuate them (see also
range (perhaps involving small daphniids, such Sterner, 1993) (see also Section 5.4.5).
as D. cucullata or Ceriodaphnia spp.). These organ-
isms are frequently prominent in the plankton of Bottom--up and top--down processes in
small eutrophic lakes (Gliwicz, 2003a) where they oligotrophic systems
represent a third kind of interactive structure in During the late 1980s, there developed vehement
the plankton. debate about whether phytoplankton biomass
was controlled mainly by nutrients or by its graz-
Feedbacks ing consumers. Both sides recognised that algal
The nature of phytoplankton--zooplankton rela- biomass was not some continuous function of
tionships is not without benefits passing in the the nutrients available and that herbivory was
reverse direction. Perhaps the most important of capable of reducing phytoplankton biomass to
these is the return of limiting nutrients to the very low levels. The contentious issues were the
medium as a consequence of the elimination of twin possibilities that consumers might contin-
digestive wastes. This is not always a close loop, uously regulate producer biomass at artificially
in so far as copepod faecal pellets presumably low levels or that low producer biomass is always
fall some way before critical nutrients are cycled attributable to grazing. Most now seem to accept
that steady-state relationships are rare and that, sive consumers are accommodated, the potential
within the bounds of normal variability, many oscillations become more complex but, at each
outcomes are possible, under the influence of sev- step, the same supportive capacity is just shared
eral factors that are not confined to resource sup- among more species representing more trophic
ply and consumption. Although the debate has levels. The assembled linkages making up this
receded, the vocabulary is permanently enriched food chain might well process and transport large
(or, possibly, just stuck with) the adjectives masses of carbon (in the example, as much as 4-
‘bottom--up’ and ‘top--down’. Now these refer less -5 mg C L−1 a−1 ) but without ever raising the
to ongoing controls and much more to the pro- aggregate biomass of the participating compo-
cesses themselves -- the extent to which the bio- nents of the microbial food web above the sup-
logical structure and function are shaped by the portive capacity. As it cannot be accumulated in
resources or by the impact of consumers. More- the standing crop, the carbon thus processed is
over, the terms are no longer applied only to phy- returned directly to the pool of dissolved carbon
toplankton but are now used freely in the context dioxide pool and/or exported in the sedimentary
of an implicit hierarchy of trophic levels. flux of faecal pellets and cadavers, to be returned
So far as the original debate about controls less directly through local or global circulations
is concerned, it is helpful to regard the issue in (Legendre and Rivkin, 2002a).
terms of supportive capacity. A low resource base These continuous severe constraints demand
(whether determined by low phosphorus concen- that the trophic components, from the produc-
trations in a lake, or very low iron concentrations ers and heterotrophic consumers right through
in the sea, or low concentrations of combined to juvenile fish, are powerfully selected by their
inorganic nitrogen in either, providing the addi- functional strengths and adaptations to deal
tional resources needed to fix atmospheric nitro- with the rarefied resources (Weisse and MacIsaac,
gen are also rare) is absolutely inescapable. Thus, 2000). The overall control is within anyone’s
on the basis of stoichiometry, a low concentra- understanding of ‘bottom--up’ regulation. Yet lit-
tion of biologically available phosphorus (BAP) of tle other than trophic interaction controls the
(say) 0.3 µg L−1 (10−8 M) cannot reasonably be relative masses of the components at any given
expected to support more than a relatively small moment: small oscillations in the effectiveness
biomass (in this case, perhaps 12--15 µg biomass of calanoid feeding cascade through heavier pre-
C L−1 ). Invested exclusively in phytoplankton, an dation on ciliates, better survivorship among the
equivalent maximum concentration of chloro- nanoflagellates and harder cropping of the bacte-
phyll a can be predicted (in this example, 0.2-- rial mass. Just as easily, fish feeding on calanoids
0.3 µg chla L−1 ), as can the adequacy to meet might trigger an effect upon ciliate numbers,
the minimum requirements of fresh-water filter- depress the numbers of nanoflagellates, with
feeders (here, it fails). Photosynthetically fixed cascading top--down effects on the balance of
carbon that cannot be turned into biomass may algal and bacterial masses (e.g. Riemann and
be wholly or part respired, with the balance Christoffersen, 1993).
being excreted as DOC. This is, of course, useable
substrate to bacteria, which, given their higher Bottom--up and top--down processes in
affinity for the phosphorus, are sufficiently com- enriched systems
petitive to be able to coexist with phytoplankton, Much the same model applies in oligotrophic
yet within the same capacity limitation on their lakes and seas, where top--down mechanisms reg-
aggregate biomass. Both bacteria and small algae ulate the interspecific and interfunctional com-
are liable to become the food of nanoplanktic position of the food web, even when the resource-
phagotrophs and, thus, to be cropped down, but limited carrying capacity sets a powerful bottom-
the same constraint on the total biomass per- -up constraint on behaviour. Where the bottom--
sists. In the three-component system, consumer up resource constraint is less severe, rather more
abundance might alternate with food abundance latitude is available to fluctuations among the
but only within the biomass capacity. As succes- components, with more opportunities to alter-
native species and a greater variety of possible algae are more susceptible than large ones to
responses. Raising the supportive capacity to (say) grazer control. Seasonal changes in the domi-
150 µg biomass C L−1 (against 3 µg BAP L−1 nance of the phytoplankton, in the direction of
or 10−7 M) opens the field to a greater aver- small to large algae, may be partly attributed to
age biomass of more algal species, to alternative this mechanism (Sommer et al., 1986).
means of harvesting and consuming them and, It is not always easy to separate the impact
so, to a wider variety of consumers. Now the cas- of bottom--up resource control from top--down
cading effects influence not just how much car- predator control on the abundance of Daphnia.
bon is resident in which functional level but may We can point to the impact of low or zero fish
play a strong part in selecting the survivors. populations, not just in limnetic enclosures, but
Increasing the carrying capacity by another in natural lakes following some mass mortality
order of magnitude (say, to 1.5 mg biomass of fish. Following the mortality of a major plank-
C L−1 , against ∼30 µg BAP L−1 ) would take the tivore in Lake Mendota (cisco, Coregonus artedii),
supportable system to well within the bounds Vanni et al. (1990) observed a big increase in
of direct herbivory by filter-feeders and to the the Daphnia pulicaria over the previously more
range of the best-known incidences of top--down abundant D. galeata, enhancing grazing pressure
control of phytoplankton. The point is now well on the phytoplankton and bringing an extended
made about the efficiency of direct consumption phase of clear water.
and assimilation of a sufficient number of Daph- In this context, events in the mesotrophic
nia feeding on phytoplankters up to 30--50 µm North Basin of Windermere illustrate inter-
is capable of grazing down the food resource annual fluctuations in the predominance of
(including most of the heterotrophic bacteria, predation- and resource-driven forcing. George
flagellates and ciliates too) to extremely low lev- and Harris (1985) noted that striking inter-
els indeed, leaving the water very clear (Hrbáček annual differences in the numbers of Daphnia in
et al., 1961; Lampert et al., 1986; Sommer et al., the plankton during June and July were inversely
1986) (see also Fig. 6.8). Reynolds et al. (1982a) correlated with the temperature of the water in
claimed this situation is unsustainable, at least June. The correlation with the biomass increase
in the pelagic, unless alternative sources of food in the young-of-the-year (YOY) recruits of perch
can be exploited, (say) in the littoral or sub- (Perca fluviatilis) is weak, even though the latter
littoral benthos. In the Blelham enclosures, no is broadly correlated with summer water temper-
such alternative was available; the consequence atutes (Le Cren, 1987). The later analyses of Mills
was the massive mortality of Daphnia, especially and Hurley (1990) confirmed only a small depen-
among the most juvenile cohort (Ferguson et al., dence on annual perch recruitment. The stronger
1982) and a collapse in top--down pressures. Algae influence of the physical conditions on both the
were able to increase again, at the expense of zooplankter and planktivore prompted deeper
nutrient resources, before supporting the resur- studies that related the biological fluctuations
gence of the next Daphnia episode. to subtle year-to-year variations in the annual
What happens in many small, enriched lakes weather patterns. These are themselves driven by
is that the tendency to stage these wide fluctu- significant year-to-year fluctuations in the posi-
ations in the biomass of grazed and grazer is tion of the northern edge of the North Atlantic
damped by planktivory. With many species of Drift Current (the ‘Gulf Stream’) (George and
juvenile fish and the adults of some feeding obli- Taylor, 1995) and, especially, to interannual oscil-
gately or opportunistically on zooplankton, the lations in the average atmospheric pressure dif-
intensity of planktivory may fluctuate seasonally, ference between southern Iceland and the Azores
in response to temperature, to the recruitment of (the now well-known North Atlantic Oscillation
young fish and to continued food availability (e.g. or NAO) (Hurrell, 1995). Throughout Europe, lim-
Mills and Forney, 1987). There are, nevertheless, nological behaviours are now being shown to be
some top--down effects evident as the larger zoo- correlatives of the NAO (Straile, 2000; George,
plankters are selectively removed and the smaller 2002).
So far as the linking mechanism in the variable magnitude and duration. However, if
dynamics of Daphnia in Windermere are con- the duration of the opportunity is sufficiently
cerned, Reynolds (1991) showed that the years of extended, the control switches to the consumer,
good recruitment coincided with the persistence becoming strongly ‘top--down’.
of Asterionella in the plankton, as a direct con-
sequence of the cool, windy weather associated
in this region with the dominance of Atlantic Intervention in food-web interactions
airstream. In warmer years, the earlier onset of As a footnote to this section, the nature of the
thermal stratification robs the putative Daphnia existing or recent interreactivity of the food-web
population of the one food source that is capa- components may be exposed not just by catas-
ble of sustaining its growth and recruitment in trophic interventions (fish kills, spills of toxic
this lake. Indeed, it was shown that the size substances, etc.) but also through the impacts
of the Asterionella crop was less important than of successful invaders. The historic overexploita-
was the last date in the year that its numbers tion of native piscivorous lake trout (Salvelinus
in the surface water exceeded the threshold of namaycush) in Lake Michigan and its susceptibil-
adequacy to sustain the requirements of Daphnia ity to attack by the invasive sea lamprey (Petromy-
(some 1300 cells mL−1 , or close to 0.1 µg C mL−1 ) zon marinus) led to a catastrophic decline in the
(cf Table 6.3). Thereafter, other filterable algae top carnivore niche during the 1950s and to an
were generally too few in number to make up the unopposed niche for the invasive planktivore the
shortfall. alewife (Alosa pseudoharengus) during the 1960s
Clearly, peak Daphnia numbers in any indi- (Christie, 1974). A rigorous programme of lam-
vidual year will have been influenced by size of prey eradication and salmonid restocking, helped
the overwintering stock, its early rate of recruit- by several severe winters, brought a substantial
ment and to the feeding choices of the stand- reduction in the alewife populations (Scavia and
ing fish stock. The point is that the years of suc- Fahnenstiel, 1988). There were significant reper-
cessful recruitment of the Daphnia are strongly cussions in the lower trophic levels, notably an
influenced by the coincidence of bottom--up and increase in large-bodied Daphnia pulicaria, abso-
top--down forces blending in a fortuitous manner. lutely and relative to small-bodied zooplankton
The food resource and the exploitative potential (Evans and Jude, 1986). Water clarity improved as
of the Daphnia are the most reliable of the many phytoplankton was cropped more heavily, aggre-
stochastic variables that bear upon the quanti- gate abundances actually paralleling the changes
tative strength of either their precise timing or in the alewife numbers (Brooks et al., 1984).
whether these coincide significantly or at all. Growing strength in the recruitment and
Thus, great interannual variations in the net pro- near-shore activity of yellow perch (Perca
duction and recruitment to pelagic communities flavescens) has since obscured one set of cascad-
of particular locations in seas and lakes come to ing effects and superimposed another. There
depend upon the interaction of small and more have been other spontaneous changes among
subtle interannual variations. This is, in essence, the lower trophic levels triggered by other
the ‘match--mismatch hypothesis’ advanced by alien introductions, notably the mesoplank-
Cushing (1982) to explain interannual fluctua- tic carnivorous cladoceran Bythotrephes, which
tions in the breeding success of fish of com- has established itself through much of the
mercial importance. It is an extremely important Laurentian Great Lakes (Lehman and Cáceres,
principle of population and community ecology. 1993) and by the zebra mussel, Dreissenia poly-
Its workings can be explained retrospectively and morpha. Though essentially benthic, the huge
its outcomes can be anticipated on a probabilistic aggregate filtration capacities represented by
basis but precise outturns cannot be predicted. well-established populations of the bivalve have
The sketches in Fig. 6.10 summarise a series of measurable effects on the planktic food web of
possibilities of bottom--up tracking responses to lakes as large as Lake Erie (Beeton and Hageman,
a routinely pulsed resource opportunity but of 1992; Holland, 1993).
Figure 6.10 (a) The (bottom–up)

responses of a consumer to a
regularly pulsed supply of resource
but of varying magnitude relative to
its requirement and, thus, of
exploitative opportunity. In (b), the
resource availability is absolutely
greater and of longer duration; the
opportunity is such to allow the
consumer to control resource
availability (from the top–down).
Reynolds (1991).
Food-chain length consumers were to achieve a transfer efficiency of

The examples given attest to the importance of even 20%, the rate of dissipation of the potential
the web of interacting consumers in affecting the energy (mainly as heat) through a producer →
biomass and species composition of the producer herbivore → carnivore sequence means that the
base. In later sections, the role of the food web in investment in carnivores is probably already less
regulating of ecosystem functions, resource stor- than 30 g C m−2 a−1 . Extension of this argument
age and recycling, and the yield to commercial suggests that the available food base becomes so
fisheries will be evaluated. The growth in appre- diluted that it is impossible to support more than
ciation of the complex mechanisms has neces- four or five links. Energy-based food webs are sup-
sitated a substantial rethink on the number of posed to be constrained by the number of times
viable linkages in the web and the flows of mate- is moved up to the next trophic level (Cohen et al.,
rials and energy across them. Conventional eco- 1990).
logical energetics developed and perpetuated the Recent studies (Spencer and Warren, 1996;
case that food-chain length is determined by the Vander Zanden et al., 1999; Post et al., 2000a; Post,
stability of the key components and by the avail- 2002a) using experimental enclosures and stable-
able resource base and the usable energy influx. isotope analyses, have shown the need to adopt
From a given solar area-specific energy input, an alternative metric for describing food-web
investment in carbon bonds and the high cost structure. This is based on the strength of compo-
of its transfer up the food chain, there is plainly nent interaction (which reflects the food choices
little ‘left’ after just a few links. In Section 3.5.2, that are actually made) and the measurement of
it was argued that the net photosynthetic pro- dominant food-chain linkages. Neither resource
duction of 50--800 g C m−2 a−1 represented an availability nor dynamical constraints play a con-
energy investment of 2--32 MJ m−2 a−1 , or less trolling role in these functional measures. On the
than 2% of the PAR flux available. If successive other hand, it is the size of the ecosystem and the
Figure 6.11 Comparison of the

changes in the standing crops of
vernal Asterionella populations ( t) in
Crose Mere, UK, during consecutive
years, in relation to the absolute
concentrations of infecting spores
and sporangia of the chytrid parasite
Zygorhizidium affluens ( d). The
percentages of host cells infected
and the water temperatures are also
plotted. Redrawn from Reynolds
(1984a).
totality of its resources that have emerged as the poorly known, although the work of few spe-
crucial determinant of chain length in aquatic cialist investigators shows that it should not be
food webs (Post et al., 2000b). Whereas just two or underestimated, for the effects of fungal attack
three trophic levels can be accommodated in very and bacterial lysis that have been reported are
small aquatic systems, from the phytotelmata of indeed impressive.
plant foliage to small pothole lakes (10−3 to 102
m3 ), the linkages increase and become succes- 6.5.1 Fungal parasites
sively more varied and adaptable in small lakes Most fungal parasites of phytoplankton belong to
(105 to 108 m3 ), large lakes (1012 to 1014 m3 ) and the order Chytridiales or are biflagellate mem-
oceans (1016 to 1017 m3 ) (Post, 2002b). bers of the Phycomycetes (see the review of
Canter, 1979). Moreover, infections can ascend
to epidemic proportions, becoming the proximal
6.5 Susceptibility to pathogens and cause of death of large proportion of the extant
hosts (Canter-Lund and Lund, 1995). However, it
parasites is often the case that separate instances of infec-
tion but ones that involve the same species of
The existence of pathogenic organisms and para- hosts and parasites nevertheless result in quite
sitic fungi and protists capable of infesting plank- different mortality outcomes (see Fig. 6.11). Some
tic algae and causing their death has been known of the most exhaustive investigations of the
for a long time. The range of the relationships epidemiology and ecology of fungal infections
that they strike with their planktic ‘hosts’ is of phytoplankton have been carried out in the
wide. On the one side, it may be a form of her- English Lakes, where chytrid infections of domi-
bivory, where the animal lives within the body nant diatoms had been observed from the outset
of the plant, like the rhizopods which inhabit of the research on phytoplankton dynamics. This
fresh-water Microcystis colonies (Reynolds et al., early work established the identities of some of
1981) or the rotifers such as Cephalodella and Hert- the parasites and the general correlation of the
wigia, which (respectively) eat Uroglena and Volvox dynamics of their infectiveness with the environ-
from the inside out (Canter-Lund and Lund, 1995; mental conditions governing other aspects of the
van Donk and Voogd, 1996). At the other, there dynamics of their hosts. They also showed that, in
are obligately parasitic fungi (Sparrow, 1960; the case of Asterionella in Windermere, parasitism
Canter, 1979; Canter-Lund and Lund, 1995) and was not merely a constraint on the dynamics of
pathogenic bacteria and viruses (Bratbak and Hel- the spring bloom but a significant control on size
dal, 1995; Weisse, 2003). The role played by these and species composition of the smaller autumn
agents in reducing host biomass is, in general, bloom (Canter and Lund, 1948, 1951, 1953).
SUSCEPTIBILITY TO PATHOGENS AND PARASITES 293
The most conspicuous fungal parasites on particular, hosts need to be numerous and spores
Asterionella belong to the genera Rhizophydium have to be infective for the parasite to take hold.
and Zygorhizidium, which also infect many other Of special interest is the fact that, under condi-
groups of fresh-water phytoplankters. It is gen- tions of low light or darkness, infective zoospores
erally difficult to distinguish among individ- are not attracted to potential host cells of Asteri-
ual species except as a consequence of their onella as they are in good light. They rarely attach
host specificity. In the presence of multi-species to the host cells and are quite ineffective in the
planktic assemblages, the incidence of infec- dark.
tive chytrids will usually be confined to just The effects of light on the infectivity and
one of them: ‘fungal parasites are excellent tax- development of the parasite contribute to the
onomists’ (J. W. G. Lund, personal communica- complexity of the circumstances of epidemic ini-
tion). Individual genera are separated on the basis tiation. The condition of both hosts and parasites
of sporangial structure, dehiscence, spore char- formed a key part of Bruning’s (1991a, b, c) sys-
acters and mycelial development. Zygorhizidium tematic investigation of the response of infectiv-
species constitute epidemics on mucilage-bound ity to light and temperature. Under low levels
chlorophytes such as Coenochloris, Chlamydocapsa of light, with algal growth constrained, fungal
and Pseudosphaerocystis. Rhizophydium species also zoospores of Rhizophydium are only weakly infec-
occur on Eudorina, on desmids and on the cysts of tive. On the other hand, the greatest production
Ceratium. Cyanobacteria such as Anabaena are var- of zoospores (as spores per sporangium) occurs at
iously attacked by chytrids of the genera Blasto- ∼2 ◦ C. Production is light saturated at ∼100 µmol
cladiella, Chytridium and Rhizosiphon. Other species photons m−2 s−1 over a wide range of tempera-
of Chytridium are found on Microcystis, on certain tures. Regression equations were devised to deter-
desmids (including Staurodesmus) and diatoms mine the development times of sporangia and
(including Tabellaria). Pseudopileum infects chrys- the survival time of infective spores. Thence, the
ophytes such as Mallomonas. Podochytrium is rep- limiting frequency of hosts (Asterionella) could be
resented by species that parasitise diatoms. In calculated. By plotting the (‘threshold’) concen-
all these instances, infective spores are dispersed trations required to facilitate (a) a positive infec-
in the medium, are attracted to and attach tion development rate of growing hosts and (b)
themselves the cells of host organisms, where spore suvival, Bruning (1991c) discerned the envi-
they establish absorptive hyphae into the host ronmental conditions most likely to lead to epi-
cells, enlarge and develop new sporangia. The demic infections. The plot of epidemic thresholds
host cells are almost always killed. Epidemics of host densities of <200 cells mL−1 (Fig. 6.12a)
will destroy a large number of host cells and shows a plateau at temperatures above 7 ◦ C and
impinge on the dynamics and perennation of the at light intensities down to ∼15 µmol photons
hosts. m−2 s−1 , which values are mildly limiting to host
Elucidation of the host--parasite relationships growth. This area extends into the lower left cor-
that impinge upon the ecology of phytoplankton ner of low light and temperature but it is oth-
could progress only so far on the basis of observa- erwise bounded by steep gradients of increas-
tion. Real advances came once it had become posing thresholds, where low light or low tem-
sible to isolate the fungi and to maintain them in perature preclude epidemic development (areas
dual-clonal cultures (Canter and Jaworski, 1978). shown in black). The minimum threshold, i.e the
In a series of remarkable experiments and obser- conditions most amenable to epidemic, occurs at
vations, Canter and Jaworski (1979, 1980, 1981, ∼11.0 ◦ C and ∼19 µmol photons m−2 s−1 (which
1982, 1986) did much to explain the life cycles conditions might support a host growth rate of
of Rhizophydium planktonicum emend., Zygorhizid- ∼0.66 d−1 ). The contours in Fig. 6.12b show the
ium affluens and Z. planktonicum, host range, host- host density necessary to parasite survival. The
-parasite compatibility, zoospore behaviour in lowest threshold values, indicating optimum con-
relation to light and darkness and the variabil- ditions for the persistence of the parasite within
ity governing the onset of fungal epidemics. In the host population, are encountered under
Figure 6.12 Effects of light and

temperature on the occurrence of
Rhizophydium on Asterionella.
Contour lines represent the
thresholds of host concentration in
cells mL−1 for (a) the development
of epidemics and (b) for the survival
of parasites. Redrawn with
permission from Bruning (1991c).
conditions of saturating light and low temper- around the prey organism (often a centric
atures. Respite from parasites under low light or diatom). They fall empty and transparent and
in darkness may assist host survival. the whole becomes surrounded by mucilage,
Other parasites and hosts possibly strike anal- as the animal contracts to an ornately spined
ogous relationships but the details must be cyst. Digestion of the food proceeds until one or
expected to differ. Looking again at the plots in more small amoebae emerge from the cyst, to
Fig. 6.11, the observations on Zygorhizidium are recommence the life cycle. Towards the end of a
compatible with the Bruning model for Rhizo- period of abundant hosts, the contents of some
phydium infectivity but the better Asterionella sur- maturing digestion cysts, instead of releasing
vivorship in 1967 as compared to 1968 clearly amoebae, are transformed into a rounded resting
benefitted from the later intervention of the par- spore. The animal can survive in this condition
asite into the host growth cycle. for many months until hosts once again become
It is also appropriate to recall that host-- abundant.
parasite relationships may be confounded by fac- Vampyrella is another interesting amoeboid
tors other than the ‘normal’ behaviours of the rhizipod that uses its slender pseudpodia to
proximal players. One of these is host hypersen- attach to the cells of algae (usually filamentous
sitivity (cf. Canter and Jaworski, 1979) to infec- chlorophytes such as Geminella) but not to wrap
tive spores, where algal cells die so soon after around its prey. Instead, Vampyrella dissolves a lit-
infection that sporangia fail to develop and tle area of host cell wall and extracts (‘sucks out’)
the progress of the epidemic becomes stalled. the contents of the host cell into its own body. It
Another is the remarkable phenomenon of may suck out another host, perhaps simultane-
hyperparasitism: Zygorhizidium affluens, for exam- ously, before turning itself into a digestion cyst
ple is itself frequently parasitised by another (though, in this genus, it is not spined). The cysts
chytrid, Rozella sp. (Canter-Lund and Lund, 1995). eventually break to release new infective amoe-
bae or a resting spore. Species that attack other
6.5.2 Protozoan and other parasites algae also seem to excise similar holes in the host
The dividing line between a consuming cell wall. It also appears that individual vampyrel-
phagotroph and true parasite is a fine one. lid species are usually consistent in their choice
Whereas the typical feeding mode in amoebae of hosts (Canter-Lund and Lund, 1995).
involves the pseudopodial engulfing of food Canter-Lund and Lund (1995) also gave
organisms and intracellular digestion from details of an aberrant, plasmodial organism
a food vacuole, the amoeboid Asterocaelum that attaches to Volvox colonies and sets up a
comprises of little more than a few long, fine mycelial-like network, consuming individual host
pseudopodia. According to the description of cells as it grows inexorably through the colony.
Canter-Lund and Lund (1995), these are wrapped These will extend beyond the host confines,
SUSCEPTIBILITY TO PATHOGENS AND PARASITES 295
presumably, to facilitate transfer to another. Then, as reports of minimal abundances of

Canter-Lund and Lund (1995) believe the alga to virus particles in various aquatic environments,
be one of the slime moulds, or Myxomycetes. ranging between 104 and 108 mL−1 (Torella and
Most of the non-fungal parasites (and many Morita, 1979; Bergh et al., 1989; Bratbak and
of the fungi too) escape without mention in Heldal, 1995) began to accumulate, more interest
many ecological studies of phytoplankton. This in the ecological importance of viruses has been
may reflect a true rarity but it seems unlikely registered.
that the profusion of parasites is confined to the Modern methods of ennumeration invoke
sites where their presence has been diligently plaque assays, epifluorescence microscopy and
sought. Thus, it is difficult to form an objective flow cytometry, using DNA-specific stains. Viruses
view of the role of losses to parasitic attack in are more numerous in lakes than in the sea and
the population dynamics of phytoplankton gen- their abundance in lakes is said to broadly corre-
erally. In terms of carbon pathways and sinks at lated with chlorophyll content and the numbers
the ecosystem scale, the probability is that para- of free-living bacteria; the likely causal connec-
sites interfere little, as infected hosts are con- tion is trophic state. According to Maranger and
sumed as are the uninfected algae, while the Bird (1995), viral abundance among the Québec
survivorship of host propagules may be such to lakes they studied was most closely correlated
prevent local elimination from participation in to bacterial production. This they considered
subsequent bursts of growth and recruitment. to be indicative of the greater dependence of
At the level of individual species, however, par- fresh-water microbiota on allochthonous organic
asites can exact a high cost in terms of domi- materials.
nance. From being the regular summer dominant Viruses are responsible for phage-induced
of the phytoplankton in Esthwaite Water from mortality of bacterial and cyanobacterial hosts
the 1950s to the early 1980s (Harris et al., 1979; but the literature considered by Weisse (2003)
Heaney and Talling, 1980b), Ceratium was all but shows enormous variability (from 0.1% to 100%).
eliminated by a sequence of parasitic epidemics He points out that differing assumptions and
(Heaney et al., 1988). An infection of the overwin- conversions distort a clear picture. Regulation of
tering cysts by a Rhizophydium species, may have the microbial community is yet harder to demon-
been the ‘last straw’ as it reduced the algal inocu- strate. Virus--host interactions are host-specific
lum to extremely low numbers. It was not until and dynamic. Hosts may be virus-resistant;
the mid-1990s that Ceratium began to become viruses may be dormant for long periods between
once again abundant in the lake (although, as infective opportunities. Viruses may even ful-
at 2003, still in subdominant numbers). fil a positive role in microbial food webs by
releasing and recycling nutrients and dissolved
organic matter from lysed cells, so stimulating
6.5.3 Pathogenic bacteria and viruses new bacterial production (Thingstad et al., 1993)
The existence of viral pathogens (see also Fig. 6.13). Much remains to be elucidated
(‘bacteriophages’: Spencer, 1955; Adams, 1966; about the net role of viruses in the dynamics of
‘phycoviruses’: Safferman and Morris, 1963; phytoplankton.
‘cyanophages’: Luftig and Haselkorn, 1967; Algal-lysing bacteria have also been isolated
Padan and Shilo, 1973) infecting bacteria (includ- from a variety of fresh-water habitats -- including
ing Cyanobacteria) and eukarytoic algae has sewage tanks -- where they attack a wide range
long been recognised. In some cases, they had of algae. A few affect planktic genera but there is
been isolated from field material and have been no known host specificity. Most of the identified
found to be capable of lysing laboratory strains species are gram-negative myxobacteria that pro-
of host organisms. Until recently, however, the duce a variety of hydrolytic enzymes, such as cel-
occurrence of the organisms had been regarded lulases, capable of lysing other microorganisms
as rare and their isolation in the laboratory (Shilo, 1970; Daft et al., 1975; Atlas and Bartha,
had scarcely been attempted (Weisse, 2003). 1993).
A qualification to this statement is necessary

in the context of cells sinking ‘irretrievably’ into
deep water. Treated as ‘losses to sinking’, settling
cells surely constitute a dynamic sink but it is not
settlement that is lethal. Unless the settling cells
are able to enter a physiological resting state (as
do the vegetative cells of several species of centric
diatom and the many types of dormant propag-
ules, considered in Section 5.4.6), cells leaving
the photic layers are likely just to respire them-
selves away. In deep water columns, especially
during late autumnal mixing at high latitudes,
much of the residual phytoplankton biomass in
suspension becomes similarly unsustainable on
the light energy available. Cells die and become
Figure 6.13 Diagram of the relationships among microbial liable to decomposition. This may be adjudged to
components of the plankton (including viruses) and the pool be non-beneficial to the survival of given species
of dissolved organic matter (DOM). HNF, heterotrophic but the saprophytic oxidation of cell material
nanoflagellates. Redrawn with permission from Weisse
and the re-solution of its mineralised compon-
(2003).
ents are key aspects of pelagic resource recyc-
ling. Whether material recycling occurs substan-
tially in the water column or mainly after the
bulk of the algal biomass has once settled to the
6.6 Death and decomposition bottom is critical to the rate of resource renewal.
Settled material may resuspended before it is
It has been implicit throughout the present chap- finally decomposed and thus play some part in
ter that the mortality of primary producers is the restoration of reusable resources to organ-
explained principally in terms of biomass losses isms still in the trophogenic water layers. If not,
to other agents or to other compartments of the carbon and nutrient cycling may be governed by
ecosystem. Spontaneous death -- the failure of the redox transformation and solute diffusion.
organism to maintain its basic metabolic func- It is relevant to refer here to the rates
tions -- as a consquence, for instance, of resource of phytoplankton disintegration and decompo-
exhaustion or light deprivation, is rarely consid- sition. There are reasonably attested allometric
ered as an issue. It has to be admitted that there exponents of basal metabolism (Section 5.4.1)
is some difficulty in discerning satisfactorily the to describe the cell-specific consumption of cell
boundaries between loss of vigour and physiolog- carbon in self-maintenance through dark peri-
ical close-down, or between dormancy and struc- ods, generally falling within the order of magni-
tural breakdown. From my own experiences of tude 0.01 to 0.1 d−1 . The stoichiometric demand
tracking the dynamics of populations through of oxidising 1--10% of the cell carbon each day
temporally sequential cycles, I am bound to a approximates to 0.03--0.3 mg oxygen (mg cell C)−1
view that unless an algal cell wall is entirely d−1 . For night-time algal respiration to consume
devoid of contents, which are not manifestly the air-equilibrated oxygen content of a water
those of some invasive parasite or saprophyte, column (say 8--9 mg O2 L−1 ) requires the pres-
then it is not safe to assume that it is dead and ence of algae at a density equivalent to ∼30
incapable of physiological revival. Equally, when mg cell C L−1 , or roughly 600 µg chlorophyll
the other ‘traditional’ causes of mortality are L−1 . This may give some comfort to the many
fully quantified and attributed, the proportion managers of lakes, reservoirs and fishponds who
that is otherwise unaccounted and, so, ascribed become perplexed about the consequences of oxy-
to cell death, is usually small. gen depletion by phytoplankton crops: their fears
AGGREGATED IMPACTS OF LOSS PROCESSES 297
are probably exaggerated. However, in water bod- sure of homeostatic function in the face of
ies already experiencing a significant biochemi- chaotic enviromental variability.
cal oxygen demand (BOD), the additional burden
of a sudden physiological collapse and decompo-
sition of phytoplankton might, indeed, push the 6.7 Aggregated impacts of loss
oxygen dynamics to levels beyond the survival of processes on phytoplankton
fish and other animals.
The principal agents of decomposition are composition
saprophytic bacteria and other microbial het-
erotrophs. The main groups of decomposer organ- The foregoing sections demonstrate that the rates
isms, the organic carbon compounds that they of biomass loss sustained by natural populations
oxidise (various carbohydrate polymers, proteins, of phytoplankton are comparable with and, at
fatty acids and lipids) and the relevant enzymes times, exceed the scale of the anabolic processes
that are produced are the subject of a useful that lead to the increase in biomass. Loss pro-
review by Perry (2002). The rates of algal decom- cesses are important in determining whether and
position have been reported in at least two dozen when given species will increase or not. Moreover,
studies that have been published since the care- different species-specific partitioning of losses
ful experiments of Jewell and McCarty (1971). must play a major selective role in influencing
Mostly these conform closely to the original day-to-day variations in the assembly and species
deductions and statements. The labile materials dominance of the whole community. To be clear,
that are readily respired and which account for loss processes affect only extant populations, the
about one-third of the carbon content of the presence of which is subject to the two over-
healthy cell will probably have been oxidised riding conditions that no alga will be conspic-
prior to death. Much of the carbon remaining in uous in the phytoplankton unless (i) it has had
the cadavers takes up to a year to oxidise but a both the opportunity and the capacity to exploit
fraction, comprising mostly structural polymers, the resources available and (ii) that the extent to
including cellulose-like compounds, decomposes which its specific light, temperature and resource
at only a few percent per year. First-order kinet- requirements for recruitment by growth are met
ics describe these processes quite comfortably: is at least as favourable as, if not better than, that
depending upon temperature, the exponents for of any of its rivals. Within these twin constraints,
oxidative carbon decay generally vary between seasonal variations in the magnitude and parti-
0.01 and 0.06 d−1 . tioning of species-specific losses contribute to the
As a footnote to this section, it is proper to net dynamics of waxing and waning populations,
point to some recent findings concerning pro- which are the driving mechanisms of phytoplank-
grammed cell death. Among microorganisms, ton periodicity.
perpetuation of the genomic line may be assisted Put in its most basic qualitative terms, inten-
by the sacrifice of quantities of vegetative cells. sive direct filter-feeding may well suppress the
Examples include the differential material invest- net increase rates of nanoplankters and of small
ment in the survivorship of overwintering cells microplankters and favour the recruitment of
of Microcystis at the expense of many that die larger algae instead. Similarly, contraction in the
during the benthic life-history stage (Sirenko, depth of the entrainment by fully developed tur-
quoted by Reynolds et al., 1981) and the reported bulence immediately selects in favour of small or
unequal allocation of biomass between daugh- motile species at the expense of larger non-motile
ter cells of dividing Anabaena, to impose a self- species, settling downwards following the loss of
control on population growth rate (Mitchison suspension. In many instances, the differential
and Wilcox, 1972). Carr (1995) believed that the sensitivity to losses has an intrinsic seasonality
genetically programmed process of cell destruc- of its own. In one of the earliest attempts to bud-
tion, or apoptosis, is probably employed widely get the fate of primary products in a lake, Gliwicz
among microbes in order to achieve some mea- and Hillbricht-Ilkowska (1975) deduced that most
Figure 6.14 Comparative

sensitivities of net increase rates of
three phytoplankters (Ast,
Asterionella; Chlo, Chlorella; Mic,
Microcystis), each growing at their
maximal rates under 12-h : 12-h
light–dark alternation, to (a) dilution
(q being the volume displaced each
day from a water body of volume V),
(b) sinking losses (hm being the
thickness of the mixed layer) and (c)
filter-feeding (F being aggregate
volume filtered each day). Redrawn
of the spring production in small, eutrophic, tem- suggested in Section 6.1), the processes balance
perate Jezioro Mikolajske was eliminated by sed- out, so that biomass carry-over and its mean
imentation. The summer production largely sus- standing crop are estimable outcomes of the sys-
tained the zooplankton while, in the autumn, tem that supports them. At the level of species
most of the primary product was submitted to and species-specific dynamics, however, loss pro-
decomposition. The observations dovetail with cesses discriminate among them and play a quan-
the perspective of alternative survival strategies tifiable part in selecting among the species aspir-
of phytoplankton (see p. 210 and Box 5.1). Thus, ing to increase their populations. The large inter-
the vernal plankton is selected broadly by the (R-) specific differences in the sensitivity of phyto-
trait of exploitation of turbulent mixing and its plankton to potential loss mechanisms are exem-
tolerance of low light and is vulnerable to density plified in Fig. 6.14. This has been devised to illus-
stratification. Early-summer (C-) plankton is effi- trate the discriminatory simultaneous effects of
cient in quickly turning resources to biomass but separate alterations in flushing rate (qs /V), mixed-
is usually vulnerable to herbivorous consumers. layer depth (hm ) and the aggregate filtration
Late-summer plankton has more of the conser- rate (Fi ) of the zooplankton community upon
vative (S-) characteristics of resisting sinking and three contrasted quite different phytoplankton
grazing to the limits of its (albeit diminishing) species: Chlorella (representative of C strategists),
sustainability. Asterionella (R) and Microcystis (S). Each has been
Self-evidently, populations increase, stagnate accorded a positive rate of replication, being that
or decrease in consequence of the relative quan- generated for a 12-h day of saturating light inten-
tities of the positive and the negative exponents sities and nutrient supplies at 20 ◦ C. Despite hav-
applied. Over substantial periods (≥1 year was ing the slowest potential rate of growth, which
AGGREGATED IMPACTS OF LOSS PROCESSES 299
prevents it from increasing at all in the face of Asterionella A

moderate flushing, Microcystis nevertheless out-
Asterionella B (Apr)
performs Asterionella, under conditions of near-
Asterionella B (Jul)
surface stratification, and Chlorella when subject
to intense grazing pressure. Asterionella B (Sep)
Reference back to Eq. (5.1) reminds us that Fragilaria A (season)
rates apply to standing inocula of varying sizes Ankyra A (Jul)
(N0 ) and affect realised populations (Nt ). A large, Ankyra A (Aug)
dominant population of Asterionella can still be
Chromulina B
dominant the next day, even when it has been
subject to greater net loss rates than all other Cryptomonas A (Aug)
species present on the preceding day, simply by Eudorina A (May)
virtue of the large ‘inoculum’ carried over. How- Eudorina B (May)
ever, if uninterrupted, this process of attrition Microcystis A (season)
in the face of net gain by another is certain to
0 50 100%
lead, eventually, to its replacement. This prin-
ciple is developed later (see Section 7.3.1). For KEY
G S O
the present, it is sufficient to distinguish the
dynamic processes from their effects. Most plank-
Figure 6.15 Graphical representation of the loss budgets.
ton biologists will accept the differences in the Each horizontal bar represents the gross cell production of
vulnerabilities of different kinds of phytoplank- the named populations in the Blelham enclosures during
ton to commonly imposed constraints. Some, 1978. Appropriate shading (as shown on key) represents the
however, seem to have difficulties in accepting minimum proportions that were grazed (G), sedimented (S)
that it is possible for small algae and grazers to or were otherwise lost from suspension (O); intermediate
coexist, and for Daphnia still to starve when fil- shading represents the mutual overlapping of confidence
terable algae and bacteria are present in water intervals about the determinations. Modified and redrawn
with permission from Reynolds et al. (1982a).
samples submitted to microscopy.
This may also be the right place to empha-
sise that it is the properties of the actual organ- of the surface mixed layer that is critical to sus-
isms in relation to the contemporaneous environ- pension.
mental conditions that are decisive. These may, The quantitative studies of phytoplankton
at times, outweigh behavioural differentiations losses in the Blelham enclosures have been men-
anticipated by taxonomic identity alone. Thus, tioned on several occasions in the development
although it is perfectly justifiable to suggest of this chapter. Among the outcomes were the
that mature Microcystis colonies are quite unman- interspecific comparisons of season-long loss par-
ageable and reputedly unsuitable as food for titioning of phytoplankton losses. These are cited
most crustacean zooplankters, a substantial pre- to provide an appropriate closing illustration of
existing population of filter-feeders is quite able the differential sensitivities of phytoplankton to
to suppress the onset of Microcystis dominance. the various processes. The horizontal bars in
This is attributable to the successful removal Fig. 6.15 each represent the total production of
by grazers of small colonies in their recruiting the species named (calculated as P, according to
(infective) stages faster than they can grow (say, Eq. (6.20)) in the named location and during the
rG ≥ 0.3 d−1 ). Neither is there any paradox that specified phase of growth or attrition. The min-
diatoms should increase in number when a large, imum fractions of each product that were lost
deep lake or the sea begins to stratify and inocula to grazing, settlement and ‘other’ sinks (lysis of
are entrained over subcritical depths (Huisman Microcystis in Enclosure A, gametogenesis of Eudo-
et al., 1999), yet populations decline abruptly when rina) are shown. Despite the width of the confi-
a small one stagnates (Reynolds, 1973a; Reynolds dence intervals, the contrasted fates of specific
et al., 1984). In both cases, it is the absolute depth production are immediately apparent from the
representations. Not less than 80% of the ver- of whole cells). These processes are the subject
nally produced Asterionella cells in Enclosure A of Chapter 6.
and B were eliminated through sinking, with Wash-out, especially in continuously and
any balance being lost to grazers. On the other intermittently flushed systems, sedimentation of
hand, >87% of the cells of Ankyra produced in disentrained cells and grazing by filter-feeders
Enclosure A, >95% of the Chromulina produced can each account for major losses of phytoplank-
in Enclosure B and probably 100% of the Crypto- ton biomass. The dynamic effects of each can
monas produced in either were removed by her- be described by exponential-decay functions,
bivorous zooplankters. Grazing also eliminated analogous to dilution, and be expressed in the
between 71% and 98% of the Asterionella produced same, summable units as cell replication, which
in Enclosure B during July but <78% of the Frag- is a boon to simulation models of phytoplankton
ilaria that were produced in Enclosure A. Prior population dynamics. The processes discriminate
to gametogenesis, Eudorina seemed to have been among species and, hence, contribute to the
less susceptible to grazing (<5% and <9% of the selection of species in given environments.
products of A and B respectively). Zygotic spores In theory, hydraulic flushing affects all
and spent male colonies did settle out at the entrained species to the same absolute extent but
end of the population though the latter were faster-growing species are better adapted to main-
already in an advanced state of decomposition tain themselves against dilution. Slow-growing,
and disintegration. The Microcystis produced in larger algae tend not to be represented in the
Enclosure A remained loss-resistant until a signif- phytoplankton of rivers (potamoplankton) unless
icant surface scum formed overnight (on 20/21 the rivers are long and have high retentivity.
September) and many cells were destroyed by Species which, for reasons of size or den-
lysis. Autumnal sedimentation as viable colonies sity difference, are most readily disentrained
accounted for the fraction (∼44% of the total pro- from turbulent layers and are simultaneously
duction) that survived bloom formation. Of the unable to swim or regulate their densities, are
Microcystis produced in Enclosure B, a large pro- vulnerable to sinking losses in water columns
portion left the enclosure following a mechanical experiencing weakened mixing and reduced
collapse of the buoyancy collar holding up the mixing depth. Sinking losses of planktic diatoms
enclosure wall (7 September 1978). The residue, may considerably exceed the rates of recruitment
plus a modest recruitment through new growth, by replication. The losses are not necessarily
wholly sedimented as viable colonies. For further permanent and it is suggested that, in certain cir-
details of these populations, see Reynolds et al. cumstances (e.g. dielly stratifying systems at low
(1982a). latitudes), rapid sinking can be positively ben-
eficial if later mixing is sufficiently vigourous
and regular to return diatoms to the water
surface.
6.8 Summary Several modes of herbivory characterise the
phagotrophic exploitation of the plankton. Bac-
The fate of gross planktic primary production is teria and picophytoplankton normally occur in
divided between consumption and excretion at concentrations that offer poor energetic return
the level of physiological homeostasis (basal res- for foraging effort. They are, however, harvestable
piration, maintenance, voiding of unassimilable by the protists (nanoflagellates and microciliates)
primary photosynthate, considered in Chapter that constitute a microplanktic--microbial food
3). The materials accumulated and invested in web, which is itself cropped by mesoplanktic
the growth and replication of new cells experi- (typically calanoid) consumers that feed effi-
ence further losses to mortality and removal of ciently on selected items of food. Nanoplanktic
fully formed individuals (through the wash-out, and microalgae, if they can be sustained in
sinking, consumption by animals, disruption by sufficient concentrations (above a threshold of
parasites and pathogens and physiological death ∼0.1 mg C L−1 ), represent an alternative food
SUMMARY 301
resource that is available to direct consumption generally favours the larger, more efficient
by filter-feeders. herbivorous filter-feeders. Fish predation selects
Filter-feeding is an efficient and effective against large, visible filter-feeders but, presum-
way of foraging, provided the foods are of fil- ably, only when the zooplankters also occur in
terable size and sufficiently concentrated above sufficient numbers to attract consumer attention
threshold levels. While these conditions are away from other, more rewarding food sources.
satisfied, cladoceran filter-feeders can flourish, Phytoplankton--zooplankton relationships vary
reproduce and recruit, potentially quite rapidly, with trophic state and with the presence and
new consumers to the assemblage. Two or more predilections of fish.
generations may be recruited over a period of a Algal populations are susceptible to epi-
month or so, each time raising several-fold the demics of fungal parasites. Far from being
rate of grazing loss experienced by ingestible stochastic events, the susceptibility of hosts to
species of phytoplankton. However, the aggregate infection and the conditons favouring epidemics
filtration results in increasingly unsustainable are now broadly predictable.
loss rates to the recruiting algae and which, once Descriptions of the destructive impacts of
surpassing the rate of cell replication, can thus virus attacks on algae, cyanobacteria and bac-
bring about the collapse of the food source. All teria have been appearing in the literature for
filterable particles, including the components of over 40 years but it is barely more than a decade
the microbial loop, can be very quickly cleared since their general abundance in waters (104 and
from the water, leading to its high clarity. Other 108 mL−1 ) has been appreciated. The ecological
things being equal, starvation and mortality role played by so many potential pathogenic
of the herbivores results in a respite for the organisms is still not wholly clear.
producers, which may increase in mass before Quantifiable effects of loss processes are
the next cycle of consumptive tracking. assembled and compared at the end of the
The presence of facultatively or obligately chapter, showing how potential changes in algal
planktivorous fish may damp down considerably dominance respond to seasonally varying rates
the potential fluctuations in feeding pressure of population depletion experienced by different
on the phytoplankton. Resource competition species.
Chapter 7
Community assembly in the plankton:

pattern, process and dynamics
7.1 Introduction 7.2 Patterns of species

composition and temporal
In the pelagic, as in the great terrestrial ecosys- change in phytoplankton
tems, space is occupied by numbers of organ-
isms of various species forming distinct pop- assemblages
ulations fulfilling differing roles. Of course,
these assemblages of species reflect autecological 7.2.1 Species composition in the sea
aspects of preference and tolerance but they also Despite the mutual contiguity of the marine
show many synecological features of the mutual water masses on the planet, the composition
specific interactions and interdependences that and abundance of the phytoplankton show broad
characterise communities. The numbers of organ- and significant variations, both in space and
isms, the relative abundances of the species, through time. To provide some sort of review
their biological traits and the functional roles and then to offer a classification of the patterns
that they fulfil all contribute to the observ- are formidable objectives. In setting out to ful-
able community structure. In the plankton and fil the present requirement, I am aware, on the
in other biomes, the challenge to explain how one hand, of the large body of mainly miscel-
these structures are put together, how they are laneous information culled from the detailed
then regulated and how they alter through time, records of particular sea cruises and, on the
falls within the understanding of community other, of the several monumental syntheses that
ecology. have heroically attempted to sort, to classify
This chapter considers the structure of phyto- and to draw generalities from the data (such as
plankton assemblages among a broad range of Raymont, 1980; Smayda, 1980). It is, of course,
pelagic systems, in the sea and among inland attractive to turn to the latter publications and
waters, seeking to identify general patterns and to précis and paraphrase the accounts. It may
common behaviour. In the second main section seem that this is the option that has been fol-
(7.3), the processes that govern the assembly lowed but the present account is closer to the
of communities and shape their structures are intentions of Margalef (1967) in being ordered
traced in detail. Because some of the terminology more about the main habitat constraints. This
has been used uncritically in the literature, some- permits its emphasis to be directed to the struc-
times erroneously and often confusingly, their tural elements of the organisation of marine
usage in the current work is explained in a sep- phytoplankton. In no sense does it provide either
arate text (Box 7.1). a catalogue of the species to be found in
SPECIES COMPOSITION AND TEMPORAL CHANGE 303
Box 7.1 Working definitions of some terms used in

community and ecosystem ecology
Niche The preferred use of this term refers to exploitable resources requiring
particular organismic abilities or adaptations for their utilisation. In an evolu-
tionary context, these abilities may be acquired by some species (who may
become specialist in retrieving this part of the resource pool) and, perhaps
to the extent that they do so better than others, whom they may eventu-
ally exclude through competition. With sufficient division of the resource pool
(niche differentiation), however, many species are able to coexist in the same
assemblage (see Tokeshi, 1997). More generalist species that may broadly share
the same spectrum of resources (wide niche) constitute a guild or functional
group. They may also coexist, until the resource diminishes to constraining pro-
portions and becomes the subject of competition and potential exclusion of all
but the superior competitor (Hardin, 1960).
Power Power is work per unit time. The usual unit is the watt (W = J s−1 ).
In the context of power delivery of primary production to the pelagic food
web, it is more convenient to express power in units of kJ a−1 . Salomonsen’s
(1992) calculations of the volume-specific primary production in the plankton
is typically within the range 100–1000 of kJ m−3 a−1 .
Emergy Emergy is the total amount of energy invested directly or available indi-
rectly to a system but not all of which can deliver ecologically significant power
(Odum, 1986). Odum (1988) went on to compare the transformity of the
energy sources to the power yield as a ratio. This ranges from up to 1 (for
PAR), through wind energy (623) and hydropower (23 500) to proteinacecus
foods (1 to 4 million).
Exergy Exergy is defined as the maximum flux of short-wave energy that a system
can hold in the form of entropy-free chemical bonds, pending its eventual dissi-
pation as heat (Mejer and Jørgensen, 1979). Exergy represents the information
stored in the structure of the system and its ability to increase (Salomonsen,
1992). It may bolster the system against structural change when its mainte-
nance becomes more energetically costly than its return in energy harvesting
(see p. 376).
Ascendency In the context of accumulating ecosystems, this is the quantifiable
synthesis of growth, development and organisation of an ecosystem. System
growth is defined as the increase in total system throughput and development
as the increase in information of the network of flow structures (Ulanowicz,
1986). Thus, ascendency is a measure of the structure of an ecosystem based
upon the degree of organisation (information) and functioning (system activity)
(Kutsch et al., 2001).
Succession. Here, usage is restricted to autogenic assembly processes that result
in the substitution of species, usually in a recognisable series. The process is
partly the consequence of changes to the environmental conditions wrought by
the activities of earlier-establishing organisms, to the extent that they become
more amenable to the later establishing of individuals of other species and
less so the earlier species already established (see p. 359). Successions may
culminate in a dynamic steady state, known as climax, where one species
dominates overwhelmingly. The competitively best-fit species survives at the
304 COMMUNITY ASSEMBLY IN THE PLANKTON
expense of its rivals but there is little predictable about the sequence of species
participating in the succession itself. Succession is now regarded as no more
than a probabilistic sequence of species replacements in a weakly variable
environment.
Stability Stability is the tendency of the species composition of a community
not to vary over a given significant period of time (Pielou, 1974). Observable
stability carries different connotations according to context; it is better to refer
to individual classes of stability by their own names (Lampert and Sommer,
1993): constancy, resistance and resilience.
Constancy Only minor fluctuations in the number of species, individuals and aggre-
gate biomass occur. The absence of change does not distinguish whether the
environment is itself is also relatively constant, or whether the structure accom-
modates and survives forcing by external variations. Constancy is also used in
the context of cyclical consistency and the return to precisely similar community
structures (see p. 381).
Resistance Resistance is the term for the situation in which despite external forcing,
internal structure is preserved. This may be relative (weak forcing, resistant
structure) or the forcing does not alter the current environmental constraint
(see p. 376).
Resilience (elasticity) Resilience describes the situation where environmental forcing
causes a deviation in structure but the system returns (‘reverts’: Reynolds,
1980a) to its original condition. Resilience leads to constancy over a long period
(Lampert and Sommer, 1993).
Disturbance Disturbance is the situation in which the environmental forcing
results in a significant shift from the original structure which is not recovered in
the short term. Disturbance is a community response to the forcing. It should
not be applied to the forcing (see pp. 372–9).
particular parts of the sea or an inventory of past Phytoplankton of the North Pacific
surveys. Subtropical Gyre
From the time of the earliest comparative The deep-water oceanic provinces of the Pacific,
observations (e.g. those of Gran, 1912), system- Atlantic and Indian Oceans cover over half the
atic differences in the abundance and composi- surface of the planet. Remote from the conti-
tion of phytoplankton were recognised. At first, nental land masses, they are host to about one-
it was supposed that these were due to differ- quarter of global net primary production. Within
ential temperature and salinity preferences of these great basins, surface water movements con-
the algae. Gradually, however, floristic assem- stitute clockwise (in the northern hemisphere) or
blages became associated with the extent and anticlockwise (southern hemisphere) geostrophic
longevities of water masses in the major oceanic flows, or gyres (see Fig. 2.3). The North Pacific
basins and their circulations (cf. Fig 2.3), sub- Subtropical Gyre, occupying an elliptical area
ject to the modifying effects of adjacent shelf of some 20 million km2 located between lati-
areas and coasts, and of such local aberrations tudes 15◦ N and 35◦ N and between longitudes
as upwellings, coastal currents and frontal activ- 135◦ E and 135◦ W, is the largest of these cir-
ity. Some examples of the floristic distinctions culation features. It is also the Earth’s largest
among these zones are noted below, with com- contiguous biome (Karl et al., 2002). The circu-
ments on their environmental characters, con- lation substantially maintains a coherent mass
straints, fertility and variability. of water and separates it from adjacent habitats.
Even the permanent pycnocline, located at a tripos) normally make up only a small part of the
depth between 200 m and 1 km, isolates the sur- plankton.
face waters from the deep, nutrient-rich layers The variability revealed by the ALOHA obser-
beyond. As with the other major oceanic gyres, vations occurs on several timescales. Day-to-day
the severe nutrient deficiencies and low support- differences in primary production relate to dif-
ive capacities of the surface waters of the North ferences in PAR income, especially that available
Pacific have long been appreciated (TN < 3 µM, TP to the deep chlorophyll maximum. There are
< 0.3 µM, SRSi < 20 µM: Sverdrup et al., 1942). On broad seasonal variations in primary production
the other hand, the water has a high clarity (εmin rates (slightly higher in the northern-hemisphere
∼0.1 m−1 : Tyler and Smith, 1970, quoted by Kirk, summer) and biomass (slightly higher in the
1994). Its low planktic biomass and weak areal winter). However, these oscillations are also sub-
production have also been accepted (Doty, 1961; ject to larger fluctuations on supra-annual or
Beers et al., 1982; Hayward et al., 1983). The sup- aperiodic scales (chlorophyll range over the full
posed constancy of these conditions nurtured an study period, 13–36 mg chla m−2 ). Longer peri-
idea that the system had achieved the steady state ods of weak winds and enhanced stability, in
of a successional climax (Venrick, 1995). There 1989, were followed by blooms of the nitrogen-
is certainly little doubt that the North Pacific fixing Cyanobacterium Trichodesmium. During
Subtropical Gyre is a feature of great antiquity March–April 1997, nutrient upwelling, caused by
whose properties and boundaries have persisted a strong wind divergence, was followed by a
since the Pliocene period (10 Ma B.P: Karl et al., significant bloom of Rhizosolenia styliformis. Like
2002). Hemiaulus (Heinbokel, 1986) this diatom is also
Since the establishment in 1988 of a monthly able to supplement its supply of fixed nitro-
monitoring programme of the biomass and pro- gen which, in this instance, is provided by an
ductivity of the phytoplankton in North Pacific endosynbiotic Cyanobacterium, Richiella.
Subtropical Gyre, at a station near Hawaii called These fluctuations are caused proximally by
ALOHA, a much clearer picture of its community events that either increase the nutrient resource
dynamics has become available (Karl, 1999; Karl et or decrease the mixed depth. In turn, these
al., 2002). This confirms the low depth-integrated follow the supra-annual ‘El Niño’ oscillations
phytoplankton biomass (averaging 22.5 mg chla in wind forcing and gyre circulation brought
m−2 ; concentrations typically increase from < 0.2 about by high water temperatures in the west-
mg chla m−3 near the surface to a typical deep ern Pacific. As these events subside, vertical mix-
chlorophyll maximum at between 80 and 120 m ing in the gyre weakens, there is a decreased
depth of up to ∼0.8 mg chla m−3 ) and low pri- nutrient flux from depth and the upper waters
mary production rates (484 mg C m−2 d−1 . S.D. become more oligotrophic. Dominance reverts to
± 129). However, these data reveal more variabil- picocyanobacteria and nanophytoplankton and,
ity than might be expected of a fully equilibrated potentially, to nitrogen-fixing bacteria. Annual
steady state. carbon fixation settles back to ∼160 g C m−2
The floristic composition also proved to be sur- a−1 which is consumed mainly in turning over
prising. Whereas the earlier published accounts a microbial food web.
focused on the dinoflagellates and diatoms, This pattern comprises what are mani-
much the most abundant phototrophs and the festly the organisms best suited to the rar-
largest fraction of the standing biomass are efied resources. Significant sedimentary losses
prokaryotes, especially Prochlorococcus (some 50%) are obviated, except during episodes of diatom
and smaller amounts of Synechococcus. Lesser num- abundance (Karl, 1999). The point to emphasise
bers of eukaryote haptophyte (Umbellosphaera) and is that although the biomass remains generally
chrysophyte (Pelagomonas) nanoplankters (Letelier very resource-constrained, it is not in an unmov-
et al., 1993; Campbell et al., 1994) are present. The able steady state and the events to which it is
diatoms (species of Rhizosolenia, Hemiaulus) and susceptible influence the rate and direction of
dinoflagellates (Prorocentrum, Pyrocystis, Ceratium change in community composition and function.
Phytoplankton of other low-latitude gyres Pacific and Atlantic Oceans, whereas Antarctica
Few other modern ocean surveys match the com- is wholly surrounded by an almost continuous
prehensiveness of the ALOHA series but more lim- Southern Ocean which interfaces with the South
ited datasets suggest a common organisation. In Atlantic, the Indian and the southern Pacific.
the South Pacific Gyre, DiTullio et al. (2003) found In the Arctic Basin, seasonal dynamics are
picoplanktic Cyanobacteria to be major compo- supposed to be dominated by changes in day
nents of the assemblage, Prochlorococcus dominat- length and radiation intensity. There are few
ing over the scant biomass of Synechococcus and seasonal studies from which to gauge succes-
other eukaryotic plankters that included Phaeocys- sional changes (Smayda, 1980). In the winter,
tis and prasinophytes as significant constituents. underwater light availability is severely con-
Prochlorococcus is abundant in the low-biomass straining in both ice-covered and ice-free areas
region of the central Indian Ocean (between 5◦ N and phytoplankton biomass is low. While there
and 30◦ S and 55◦ and 100◦ E; < 0.1 mg chla m−3 ) is little evidence of much seasonal growth
where Trichodesmium is also common. Coccol- under permanent ice, lengthening days bring
ithophorid nanoplankton, diatoms and dinoflag- significant increases in standing crop in ice-
ellates contribute to modest seasonal biomass free areas (Smayda, 1980). Generally, diatoms are
increases during May and June, in the wake overwhelmingly dominant, with biomass being
of the monsoon period (various entries in Ray- shared among a relatively small number of main
mont, 1980). In the North Atlantic Tropical Gyre, species (Achnanthes taeniatum, Chaetoceros diadema,
picoplanktic communities supporting fully devel- Corethron criophilum, Skeletonema costatum, Coscin-
oped microbial food webs have been demon- odiscus subbulliens, Rhizosolenia styliformis and Rhi-
strated by Finenko et al. (2003). The dominant zosolenia hebetata). Phaeocystis sp. is one of the few
primary producers in the Sargasso Sea (part of non-diatoms that can be plentiful in summer;
the adjacent subtropical gyre) are Prochlorococcus, three species of Ceratium (C. longipes, C. fusus and
again being most abundant in a deep (∼75 m) C. arcticum), three of Peridinium (P. depressum, P.
maximum (Moore et al., 1998). Previous investi- oratum and P. pallidum) and the prasinophyte
gations of the seasonality of microphytoplank- Halosphaera are also noted.
ton seasonality of the Sargasso Sea (Riley, 1957; In intermediate zones, where the ice cover
Hulburt et al., 1960; Smayda, 1980) recognised may be broken but its melting proceeds through
that the limited abundance of spring diatoms much of the short summer, density differences
(Rhizosolenia, Chaetoceros spp.) gives way to a long contribute to a substantially reduced mixing
period of relative abundance of nanoplanktic depth. Within such polynia, there may be a char-
Umbellosphaera and other coccolithophorids, and acteristically strong seasonal development of phy-
to Trichodesmium and ceratians such as Ornithocer- toplankton. Green flagellates may appear under
cus during summer. the thinning ice pack, to be followed by such
diatoms as Achnanthes and Fragilariopsis species
Phytoplankton of high latitudes through the period of ice-break. These may per-
Towards the edges of the great oceanic provinces, sist until new ice is formed towards the end of
altered environmental conditions (mostly related summer (perhaps as little as 2–4 months later)
to the proximity to land masses: depth, light and, but are often joined by other diatoms (Chaetoceros,
especially, elevated nutrient availability) support Thalassionira spp.), Phaeocystis and Ceratium species,
alternative assemblages. In addition, the ten- together forming a distinctive species assemblage
dency towards the polar regions is to weaker (Smayda 1980).
and more seasonal thermal stratification. How-
ever, there are further contrasts between the cir- Phytoplankton of deep boreal waters
culatory patterns of the northern and southern The Atlantic to the north of the North Atlantic
hemispheres that relate to the distribution of Drift Current, together with its arms to the
land masses. The physical structure of the north- Labrador and Norwegian Seas (roughly, the open
ern high latitudes is separable between the Arc- ocean to the north of a line from 35◦ N, 75◦ W and
tic Polar Basin and the boreal reaches of the 45◦ N, 10◦ W) and the Pacific Ocean to the north
of the Kuroshio Current (roughly, from 35◦ N, nitzschioides becoming increasingly prominent
145◦ E to 55◦ N, 145◦ W) constitute the two main towards the late summer. Dinoflagellates are also
boreal oceanic provinces. The Continuous Plank- conspicuous in the summer months, especially
ton Recorder (CPR), devised originally by Sir Alis- Exuviaella spp., several Peridinium spp., ceratian
ter Hardy (1939) to investigate mesoscale patch- spp. (Ceratium longipes, C. fusus, C. furca and C.
iness, was used extensively and over a number tripos). Gymnodinioids are common throughout
of years to reveal microphytoplanktic structure the May to November vegetation period (synthe-
in the boreal North Atlantic Ocean (Colebrook et sis of Smayda, 1980, based on the observations
al., 1961; Robinson, 1961, 1965). These identified of Halldal, 1953).
the ceratians C. carriense, C. azoricum, C. hexacan- In the North Pacific, a mainly sub-Arctic cold-
thum and C. arcticum and the diatom Rhizosolenia water community occupies the boreal waters sep-
alata as being primarily oceanic species, quoted arated from the Central Pacific by the warm
in a sequence from warmer southerly water to Kuroshio Current (Marumo, in Raymont, 1980).
the colder northern masses. Many other species The biomass supported is contrasted with the
found were adjudged to have originated in adja- meagre populations of the gyres to the south.
cent shelf waters (e.g. Ceratium lineatum, Thalassio- Typically, diatoms (species of Chaetoceros, Corethron
thrix longissima, Nitzschia spp.), yet others to have and Fragilariopsis, Rhizosolenia hebetata, Thalas-
a more general occurrence (Thalassionema nitzschi- sionema nitzschioides, Thalassiosira nordenskioeldii)
oides, Rhizosolenia styliformis, R. hebetata). Phyto- dominated the spring–summer maxima. Sum-
plankton abundance increases with increasing mer dinoflagellates included Ceratium fusus.
day length, to a typical maximum of ∼2 mg
chla m−3 in May or June. Among many species Phytoplankton of the Southern Ocean
present, diatoms (Thalassiosira, Thalassionema and A discrete ‘Southern Ocean’ was distinguished
Rhizosolenia) generally dominate. In summer, Cer- from its contiguous oceans (Atlantic, Indian and
atium fusus, C. furca and C. tripos are relatively com- Pacific) after its effective isolation by a series
mon. of circumpolar fronts (see below) was first fully
Smayda (1980) reviewed Holmes’s (1956) recon- appreciated (Deacon, 1982). The key subtropi-
struction of algal periodicity in the Labrador Sea. cal frontal system stretches around the entire
Many species are common to the CPR listings Antarctic continent, generally between the lati-
for the open Atlantic, with species of Thalassiosira tudes 40 and 50◦ S, uninterrupted but for the pen-
and Thalassionema prominent, along with several etration of the Patagonian region of South Amer-
species derived from adjacent shelfs and coasts ica. The area enclosed represents almost 20% of
(notably species of Chaetoceros, Coscinodiscus, Fragi- the planetary ocean surface and it is now known
lariopsis) or from the Arctic (Ceratium arcticum and to play an important role in regulating the plan-
Peridinium depressum). Diatoms (especially Fragilar- etary climate (Boyd, 2002). Variations in the syn-
iopsis nana, Fragilaria oceanica, Rhizosolenia hebetata, thesis of the DMSP precursor (see Section 4.6.2),
Pseudonitzschia delicatissima) dominated the single the turnover of macronutrients and the passage
late spring/early summer maximum. Dinoflag- of photosynthetically fixed carbon to Antarctic
ellates, including Peridinium depressum, were rel- heterotrophs are all mediated by the phytoplank-
atively common in summer. Coccolithophorids, ton. Over geological time, the carbon exchanges
especially Emiliana huxleyi, are also numerous in of the Southern Ocean have impinged signifi-
the North Atlantic Ocean during summer, in cantly on planetary climate.
some years producing significant blooms (Balch Within the ocean, frontal systems separate
et al., 1996). approximately concentric inner water masses,
In the Norwegian Sea, the sequence of dom- each having distinct physical and chemical iden-
inance moves from diatoms (initially Chaetoceros tifiers. The seasonal advance and retreat of sea
spp., Fragilariopsis nana in the late spring), with ice cover adds to the physical complexity of
Chaetoceros convolutus, Corethron hystrix, Thalas- the Southern Ocean, defining permanently open-
siothrix, Rhizosolenia hebetata, R. styliformis, ocean and seasonally ice-covered zones (POOZ,
Pseudonitzschia delicatissima and Thalassionema SIZ). However, a general property of all the
substructures is the degree of environmental con- interface with the land masses, or with the con-
trol of phytoplankton that they exert. In short, tinental shelves of which they are part, or with
chlorophyll concentrations (generally <0.3 mg the margins of another oceanic province. Each
chla m−3 ) are rather lower than the perennial interface creates distinctive physical environ-
capacity of the macronutrients to support (up ments for phytoplankton growth and selection.
to 15 µM DIN, up to 2 µM BAP). The condi- Fronts at the mutual interface of two oceanic
tion, now referred to as HNLC (for high nitro- masses persist because differences in their tem-
gen, low chlorophyll: Chisholm and Morel, 1991), perature and salinity (and, hence, density) resist
is attributable to several constraints (the interac- intermixing. The most significant fronts sepa-
tion of daily irradiance, vertical attenuation and rate the permanently stratified tropical oceans
mixed-layer depth; silica exhaustion; grazer con- from the cold, well-mixed waters of the polar
trol) but experiments (such as IRONEX, SOIREE: seas, and which, roughly, are located at around
see Section 4.5.2) have shown iron deficiency to 40◦ N and 40◦ S. The polar convergences are well
be often the critical factor (for a full review, defined, especially in the Pacific, by sharp gradi-
see Boyd, 2002). With ambient TFe levels <10−9 ents of surface temperature that correspond to
M, the short ‘window of blooming opportunity’ the area in which the denser cold water is slid-
allowed by the austral summer sustains only ing under the lighter warm surface flow. A few
modest phytoplankton growth before further degrees to the equatorial side of the polar front is
recruitment becomes iron-limited. Even though a further, rather less abrupt subtropical conver-
the severity and timing of the limitation may gence, formed between the warmer central por-
vary among zones and the growth of diatoms tions of the gyre interface with the peripheral
depletes the stock of dissolved silicon (data of geostrophic flow.
Boyd et al., 1999), the ultimate controlling role of In both instances, the interfaces abound with
iron is quite general across the Southern Ocean. instabilities and a degree of intermixing. Phy-
The dominance of Antarctic waters by toplankton present in either water mass are
diatoms was detected in the earliest plankton sur- confronted with altered environmental condi-
veys (Gran, 1931; Hart, 1934). The major species tions, offering improved insolation to the polar
include Chaetoceros neglectus, C. atlanticus, Corethron assemblage and, perhaps, more nutrient to the
criophilum and species of Fragilariopsis, Nitzschia species in the subtropical water. In a transect
and Rhizosolenia (which include R. alata and R. across the South Pacific Ocean, following longi-
hebetata). In the seasonally ice-covered zones, tude 170◦ W from the Antarctic continent to the
Chaetoceros neglectus and Nitzschia closterium are Equator encountered its highest chlorophyll con-
abundant close to the receding ice edges, before centrations (>0.5 mg chla m−3 ) in the vicinity
giving place to the summer assemblage. The hap- of the polar and subtropical fronts (DiTullio et
tophyte Phaeocystis antarctica also blooms in some al., 2003). In the latter instance, there was a dis-
parts of the Southern Ocean. It benefits from tinct subsurface maximum at a depth of about
modest iron enrichment (as do other species) but 40 m. Between the fronts, the dominant phyto-
its growth does not run into silicon limitation plankters were coccolithophorids, chrysophytes
and it is also more efficient in adapting to vari- and Pelagomonas; prasinophytes, cryptophytes and
able light and carbon deployment (Arrigo et al., Phaeocystis also contributed to the standing crop.
1999). The abundance of the prymnesiophyte flag- Diatoms were scarce on this occasion (the silicon
ellate Pyramimonas is also sensitive to light and concentrations were <1 µM) but other authors
iron levels (M. van Leeuwe, quoted by Boyd, 2002). have found the Polar Front to be the main loca-
However, picoplanktic prokaryotes become relation of Antarctic diatom blooms (Smetacek et al.,
tively more scarce with increasing latitude (Det- 1997).
mer and Bathmann, 1997).
Upwellings
Oceanic fronts The major upwellings occur along continental
The margins of each of the major circulatory seaboards, where the circulation current and
provinces so far considered in this section can the prevailing wind have the same direction.
The Coriolis forces acting on the flow tends to the Benguela upwelling, where Chaetoceros species
drag surface water away from the coast (Ekman are similarly dominant and where primary pro-
transport), thus entraining deeper (usually from duction rates of 0.5–2.5 g C m−2 d−1 have been
below the pycnocline) to the ocean surface, where reported (Steemann Nielsen and Jensen, 1957).
it ‘upwells’ a short distance from the shore. This Off California, the most common phytoplank-
water is generally cold but relatively nutrient- ton during intense upwelling included Rhizosole-
rich and, entering the photic zone, becomes sup- nia stolterforthii, Skeletonema costatum and Lepto-
portive of high biological production and prized cylindricus danicus (Eppley, 1970). The recent time
fisheries. For example, the Canaries Current – the series of Romero et al. (2002) on the upwelling
south-flowing arm of the North Atlantic Subtrop- fluxes in the region of Cap Blanc, north-west
ical Gyre – coincides with the North-East Trade Africa, distinguished periods when character-
Wind along the coastline of Mauritania and Sene- istically oceanic diatoms (Nitzschia bicapitata,
gal. The matching structure in the North Pacific, Thalassionema nitzschioides, Fragilariopsis doliolus)
the California Current, is also responsible for were dominant from those when near-shore
San Francisco’s legendary fogs. Yet more strik- species were ascendant (including Cyclotella
ing are the Benguela and Peru Currents, respec- litoralis, Coscinodiscus and Actinocyclus spp.).
tively moving up the west coasts of Africa and During relaxations, production falls, diatom
South America, enhancing biological production dominance fades as cells are lost by sink-
in their wakes (to >400 g C m−2 a−1 ; see Behren- ing and decay, and smaller or motile algae
field et al., 2002) and especially where they divert assume relative importance. Coccolithophorids
from the coast lines (again respectively) near (Umbellosphaera) and small dinoflagellates (Lingulo-
Gabon and the Galápagos Islands. dinium polyedrum, Gymnodinium catenatum, both
The intensity of upwelling is not continu- notable as harmful bloom species) mark the
ous, fluctuating with seasonal variations in wind transition to calmer, nutrient-depleting condi-
strength and provenance. In the Indian Ocean, tions (Smayda, 2002). With falling fertility at
upwellings in the Arabian and Andaman Seas are the surface, chlorophyll maxima develop at
strongly seasonal, being related to the monsoonal depth and in which other coccolithophorids (e.g.,
episodes. The El Niño oscillations alternate sup- Florisphaera spp.: Raymont, 1980) and dinoflag-
pression and enhancement of the strength of ellates (notably Dinophysis spp.: Reguera et al.,
upwelling of the Peru Current. When coastal 1995) may become relatively abundant (Raymont,
winds weaken and the sea surface calms, the mix- 1980).
ing of the water also becomes less intense and
warm, nutrient-poor water persists at the surface. Shelf phytoplankton
This phenomenon is known as ‘post-upwelling What are recognised as the terrestrial land
relaxation’. masses are blocks of continental crust that, in
Both the upwellings and the relaxations cre- geological time, are moved apart or are coalesced
ate distinctive environmental conditions for phy- by the tectonics of the plates of oceanic crust.
toplankton development. Guillen et al. (1971) Whereas the active formative ridges and the deep
refer to the periods of high primary produc- subduction trenches power the changing posi-
tion in the coastal Peru Current (0.3–1.0 g C tions of the continental masses, the gaps between
m−2 d−1 ), nurtured by macronutrient levels them are water-filled. In fact they are the reposi-
shown by others to be up to 2 µM P and 20 tory of 97% of the planetary total. At the present
µM N and supporting phytoplankton biomass time and, subject to fluctuation of ±50 m due to
equivalent to the order of 2 mg chla m−3 . the changes in the volume stored as polar ice, for
Diatoms (especially Rhizosolenia delicatula, Thalas- most of the last 250 million years, the low-lying
siosira subtilis, Skeletonema costatum and Chaeto- continental perimeters have been inundated by
ceros debilis) dominated the flora, species of the sea, to a depth of ≤200 m. Collectively, these
other groups of organisms (dinoflagellates, coc- are the continental shelves. Grading at first gen-
colithophorids) making up only a small percent- tly away from the coastline, the shelf extends to
age of the biomass. Similar conditions obtain in the true continental edge, the abrupt, steep slope
to the ocean floor, >2 km (and locally up to 6 km) development of a May diatom bloom (typical
beneath the water surface. maximum 2–3 mg chla m−3 ), generally featuring
The width of the continental shelves varies Thalassionema nitzschioides, Rhizosolenia hebetata, R.
with location, literally from a few kilometres delicatula, Chaetoceros and Thalassiosira species. In
(as along the coast of central Chile, of north- the summer, dinoflagellates are relatively much
east Brasil and the south-eastern seaboard of Aus- more abundant, especially Ceratium fusus, C. furca
tralia) to the great expanses that characterise and C. tripos; Heterocapsa triquetra, Karenia mikimo-
the Sea of Okhotsk, the East China Sea, the Ara- toi and Dinophysis acuminata also feature. Rhizosole-
fura Sea, the Gulf of Maine, the Grand Banks nia alata and R. styliformis are most common in
of Newfoundland and the seaboards of north- autumn (various sources, compiled by Raymont,
west Europe (the Baltic Sea, the North Sea and 1980).
the English Channel). Though mutually very dif- The Sea of Okhotsk is generally partly or
ferent from each other in their oceanic inter- wholly ice covered in winter and the onset of the
faces, latitude, temperature and fluvial influence, spring bloom (of diatoms Chaetoceros spp., Rhizo-
they share common attributes of close mechani- solenia hebetata, Thallasionema nitzschioides, Thalas-
cal coupling to the adjacent littoral and/or ben- siosira nordenskioeldii and Leptocylindrus danicus) is
thic habitats through wind and tidal mixing. quite abrupt. The production and phosphorus
Even so, the dilution of underwater light by mix- dynamics of these episodes have been the sub-
ing in shallow areas is generally less extreme ject of recent investigations by Sorokin (2002) and
than in the oceanic areas beyond. Coastal shelf Sorokin and Sorokin (2002).
waters are, indeed, distinctive from the ocean
in supporting alternative planktic comunities Coastal and near-shore waters
under distinctly differing environmental condi- Near-shore shelf waters may be further distin-
tions (Smayda, 1980). guished by the potential to support greater lev-
Smayda (1980) developed his overview of phy- els of biomass, production and diversity, and
toplankton succession in open shelf water based with more variability of abundance and domi-
on well-studied examples from the Gulf of Maine nance. Shallower water, experiencing more rapid
and the North Sea. The phytoplankton of the Gulf interchange of resources with the bottom sedi-
of Maine, located between 41◦ and 44◦ N, is char- ment, together with the inflow of ‘new’ resources
acterised by the development, commencing in from the land, provide more growth opportuni-
March or April, of a spring diatom bloom, usu- ties and more support to accumulating biomass.
ally dominated by Thalassiosira nordenskioeldii with The phytoplankton that can be supported may
Porosira glacialis and Chaetoceros diadema. As water be regarded as a more productive sub-set of the
temperatures warm, other species of Chaetoceros adjacent shelf-water assemblage but some species
(notably C. debilis) become relatively more abun- may benefit or express advantage more than oth-
dant, before yet others (including C. compressum) ers and there may be additional species that are
take over dominance in early summer. Diversity rare or absent further off shore.
is further and variably increased by the rela- The relative logistic ease of sampling coastal
tive abundance of dinoflagellates (ceratians Cer- waters also makes for the assembly of more
atium longipes, C. tripos, C. fusus and peridinians detailed time series, although, as Smayda (1980)
Peridinium faroense, Heterocapsa triquetra, Scrippsiella reminded us, the sequence of species abundances
trochoidea) and coccolithophorids (especially Emil- are not necessarily successional. Rather, these
iana huxleyi). During the autumn, diatom domi- are temporal sequences of effects that may be
nance is restored by Rhizosolenia species (R. alata, under physical (wind, weather, offshore current)
R. styliformis, R. hebetata, a.o.) and Coscinodiscus control or, at best, the results of successional
species. events generated elsewhere. Smayda’s (1980) own
In the shelf waters around the British Isles, data from Naragansett Bay, Rhode Island, USA,
the influence of penetrating Atlantic water is reveal an initial spring flowering of the dominant
variable but it fails to suppress the indigenous shelf water diatom, Thalassiosira nordenskioeldii,
though this is under way at least a month ear- peridinian (especially Scrippsiella trochoidea), pro-
lier than it is offshore. Later common diatoms rocentroid and ceratian (Ceratium furca, C. fusus
included Skeletonema costatum, Asterionella japonica and C. tripos) dinoflagellates. As seas slacken and
and the large-celled Cerataulina pelagica. In the nutrients weaken, ‘Stage 2’ blends into ‘Stage
English Channel, there is a variable influence 3’, the genera Bacteriastrum, Corethron and large
of Atlantic water supporting dominant Thalas- species of Rhizosolenia (e.g. R. styliformis) become
sionema or Thalassiosira populations but it is gen- the main diatoms, while dinoflagellates of the
erally masked by indigenous developments of genera Ceratium, Dinophysis, Gymnodinium, and
diatoms and dinoflagellates (Chaetoceros compres- Lingulodinium may increase, along with coccol-
sus, Rhizosolenia delicatula, Heterocapsa triquetrum, ithophorids such as Emiliana. By high summer
Karenia mikimotoi, Prorocentrum balticum) in early and the onset of ‘Stage 4’, the ria is substantially
summer and of ceratians (Ceratium fusus, C. tri- stratified and nutrients in the surface waters
pos) in mid to late summer (Holligan and Har- are severely depleted. Only the large Rhizosolenia
bour, 1977). Typical maximum chlorophyll con- species (they may include R. calcar-avis) and Hemi-
centrations were in the range 3–4 mg chla m−3 . aulus hauckii persist in a plankton otherwise dom-
At the regular sampling station used by Harbour inated by Ceratium, Peridinium and Prorocentrum
and Holligan (depth ∼70 m), areal biomass gen- species. Alternatively, nitrogen-fixing Cyanobac-
erally varied between ∼20 mg chla m−2 in winter teria may appear at this time. It is noteworthy
and ∼150 mg m−2 during the April bloom period. that all these species have rather low surface-
However, a feature of these enriched near-shore to-volume ratios (sv−1 ≥0.3) and the populations
areas is the appearance in calmer weather of are mostly small (∼10 mL−1 ). In enriched coastal
the haptophyte Phaeocystis and occasionally abun- waters, prolonged stratification may instead pro-
dant growths of such nanoplankters as Carteria, ceed from ‘Stage 3’ in which the smaller ‘red-tide’
Dunaliella or Nannochloris. dinoflagellates (including Alexandrium tamarense)
One of the most significant and seminal persist and continue to grow into substantial
studies of near-shore marine phytoplankton was nuisance populations (see also Section 8.3.2).
maintained by Ramón Margalef over a num- Autumn cooling and renewed mixing restores the
ber of years on the Ria de Vigo in north-west plankton back to ‘Stage 1’.
Spain. He incorporated his findings into a devel- Margalef ’s descriptions provide a basis for
opment of a general explanation of the mecha- comparison with those of Smayda (1980) refer-
nisms of seasonal change in community struc- ring to the inshore waters of Norway’s fjord coast-
ture (Margalef, 1958, 1963, 1967). Although influ- line. Although some 25◦ –30◦ of latitude further
enced by later observations made in the Mediter- north and experiencing lower summer temper-
ranean Sea, Margalef recognised four distinct atures, the sequences of phytoplankton domi-
stages of the development. Early in the northern- nance have many similarities to those of Spain’s
hemisphere year, when coastal waters are still rias. In Ullsfjord (71◦ N), an April diatom bloom-
cool, well-mixed (he called them ‘turbulent’) and ing of Chaetoceros species, Fragilaria oceanica, Tha-
relatively charged with nutrients, small-celled lassiosira decipiens and T. hyalina is followed by
diatoms (Skeletonema costatum, Leptocylindrus dani- a July–August bloom featuring Chaetoceros debilis,
cus, Chaetoceros socialis, C. radicans, Rhizosolenia alata Pseudonitzschia delicatissima, Skeletonema costatum,
and R. delicatula) predominate, with small flagel- Thalassiosira nordenskioeldii, Leptocylindrus danicus
lates (Eutreptia, Platymonas, Rhodomonas a.o.), occur and Rhizosolenia alata. Dinoflgellates are also
during this ‘Stage 1’. All these algae have high present in summer (see below) but they are much
surface-to-volume ratios (sv−1 ≥1), are capable less abundant than diatoms. In Trondheimsfjord
of rapid growth at low temperatures and form (64◦ N), blooming starts in March, with Fragilar-
populations of between 102 and 103 cells mL−1 . iopsis cylindrus, Porosira glacialis, Skeletonema costa-
‘Stage 2’ tends to be dominated by a more mixed tum, Thalassiosira hyalina and several other species,
community of larger-celled diatoms Chaetoceros then continues through to May/June (with Chaeto-
species, Lauderia annulata, Eucampia cornuta) and ceros debilis, Leptocylindrus danicus, Pseudonitzschia
delicatissima, Skeletonema costatum and Thalas- large crops of this alga had been observed previ-
sionema nitzschioides prominent). Cerataulina pelag- ously and have been since but the 1988 event was
ica, Eucampia zodiacus and Rhizosolenia fragilissima startling in its magnitude. The general eutroph-
are summer species. Dinoflagellates, including ication of the Danish coastal waters was blamed
Exuviaella baltica, Heterocapsa triquetra and Scripps- but the proximal cause was the combination
iella trochoideum and ceratians, approximately in of a strong outflow from the Baltic with calm
the sequence Ceratium longipes, C. tripos, C. fusus, and sunny weather in the Kattegat, favouring
develop in summer but rather more strongly shallow stratification and the growth of oppor-
than they do further north. tunist motile algae – in this case Chrysochromulina
In the Oslofjord (59◦ N), Skeletonema is the (Edvardsen and Paasche, 1998).
most important diatom during the March Extending eastwards from the Kattegat and
bloom, with increasing representation by Thalas- the Øresund, the Baltic Sea is a substantially
siosira nordenskioeldii, various Chaetoceros species landlocked shallow sea. It is characterised by the
(C. debilis, C. socialis, then C. compressus), Rhizo- dilution of its salt content by an excess of pre-
solenia alata and Pseudonitzschia delicatissima. By cipitation over evaporation, and by some large
summer, Lauderia borealis, Cerataulina bergonii, river inflow discharges. Whilst the structure of
Cyclotella caspia and Chaetoceros species are promi- the plankton varies with the salinity gradient
nent, together with Phaeocystis pouchetti and a across the Baltic, the tendency of warm fresh
sequence of dinoflagellates (Scrippsiella, Hetero- water to float on the colder, salt water emphasises
capsa and Prorocentrum micans, and ceratians Cer- the vertical stability and minimises the horizon-
atium longipes, C. tripos and C. fusus). In the inner tal gradient in the summer. West of the constric-
fjord, many species of nanoplanktic flagellates tion between Denmark and Sweden, this effect is
are recorded, among which Micromonas, Eutrep- normally overwhelmed by tidal and wind mixing
tia, Cryptomonas, Rhodomonas, Ochromonas, Pseudo- but, as mentioned above, there have been excep-
pedinella and Chrysochromulina can be abundant. tions with dramatic results. Within the Baltic
Coccolithophorids (Emiliana, Calciopappus, Anthos- and its two arms, the Gulfs of Bothnia and Fin-
phaera) are also numerous in the outer fjord at land, the sequences of phytoplankton dominance
this time. In recent years, Alexandrium and other differ from those on the west side of Scandinavia.
red-tide species have tended to be abundant in The sea being usually ice-covered in winter, the
the locality. In the autumn, diatom dominance first growths of the year may be small flagellates
(Skeletonema, Leptocylindrus) is re-established. (e.g. Chlamydomonas, Cryptomonas) beneath the ice
A little to the south, in the Kattegat, the surface. After the break-up of the ice, diatoms
spring diatoms (Skeletonema costatum, Chaetoceros (including Achnanthes taeniata, Skeletonema costa-
compressus, Rhizosolenia delicatula, Rhizosolenia alata tum and Thalassiosira baltica) generally dominate
and Pseudonitzschia delicatissima) give place to but with developing column stability, essentially
Cerataulina and Leptocylindrus in the summer freshwater species of Oocystis, Monoraphidium a.o
months, together with variable amounts of Phaeo- (Edler, 1979; Wasmund, 1994; Samuelsson et al.,
cystis pouchetti, Heterocapsa triquetrum, Karenia miki- 2002) and picocyanobacteria (Kuuppo et al., 2003)
motoi, Prorocentrum minimum and Ceratium species. become established. However, it is the prominent
In the spring (May–June) of 1988, Chrysochromu- cyanobacterial flora (Anabaena lemmermannii, Aph-
lina polylepis, a thitherto minor component of anizomenon flos-aquae and Nodularia spumigenea are
the nanoplankton of Norwegian and Danish the most notable) that now most characterises
coastal water, produced a significant and harmful the Baltic Sea plankton. Blooms of toxic species
bloom. Although equivalent to ‘only’ (Edvardsen currently exercise the academics and responsible
and Paasche, 1998) about 10–20% of the diatom authorities alike (Kuosa et al., 1997). The autumn
chlorophyll, the concentration of 40–80 mg chla flora is dominated by large centric diatoms,
m−2 proved toxic to local fish populations as well Coscinodiscus and Actinocyclus (Edler, 1979).
as to a variety of molluscs, echinoderms ascidi- Strong stratification is a feature also of
ans and cnidarians (Dahl et al., 1989). Unusually the Black Sea. It has suffered intensive
eutrophication from its main influent rivers be about double this, but only during summer.
(Danube, Dnestr, Dnepr and Don) since the Areal concentrations >100 mg chla m−2 are sub-
1970s: it is relatively nutrient-rich and its deep stantially confined to continental shelf waters;
water is severly anoxic in summer (Aubrey et al., standing crops equivalent to >200 mg chla m−2
1996). The dominant species in spring include are restricted to enriched near-shore habitats and
diatoms (Chaetoceros curvisetus, Rhizosolenia calcar- coastal lagoons.
avis) but dinoflagellates (Prorocentrum, Heterocapsa Why is the biomass so relatively low in
triquetra, Scrippsiella trochoidea, Ceratium tripos, most places? Supposing the primary producers
Ceratium fusum a.o.) form a major part of the of the plankton to be everywhere sharing their
biomass, and to which Emiliana huxleyi can environments with heterotrophic bacteria and
sometimes make the largest contribution (Eker phagotrophic zooplankton, top–down depletion
et al., 1999; Eker-Develi and Kideys, 2003). During is a less likely generic explanation than is severe
the summer, diatoms become rare while coccol- bottom–up regulation by a poverty of nutrient
ithophorids and red-tide dinoflagellates become resources and by an inadequacy of photosyn-
dominant (Velikova et al., 1999). thetic energy, consequential upon deep mixing
Another weakly flushed, river-enriched and the erratic dilution of the harvestable pho-
coastal area of the Mediterranean is the north- ton flux. Only where an adequate supply of a full
ern Adriatic Sea. Its main phytoplankton species spectrum of essential nutrients coincides with
are diatoms (Chaetoceros, Rhizosolenia, Cyclotella high light income into a shallow, clear, mixed
and Nitzschia spp. and dinoflagellates Prorocen- layer is there a carrying capacity sufficient to sup-
trum and Protoperidinium: (Carlsson and Granéli, port potentially high producer mass. However,
1999). Finally, on the Tyrrhenian Sea (western) inadequacies in either restrict the carrying capac-
coast of Italy, Sarno et al. (1993) compared the ity, by imposing limitations on the ability of algae
phytoplankton of the Fusaro Lagoon with the to grow and divide.
adjacent waters of the Golfo di Napoli. Here, Is one of these more important than the
dinoflagellates (Prorocentrum micans) maintained a others? Nutrient poverty and light limitation
large winter population but this was replaced by place quite different impositions on algal growth,
a February–March diatom bloom, dominated by while differences in tolerances and adaptations
Skeletonema costatum, Chaetoceros socialis and other to survive extremes are instrumental in species
species and by Cyclotella caspia. Alexandrium and selection. The effects can be represented graph-
Dinophysis featured in the summer plankton as ically to demonstrate the interaction of these
did a number of prasinophytes (e.g. Pyramimonas factors, both in terms of habitats and of the
spp.) and euglenoids (Eutreptiella sp.). Chlorophyll attributes of the species for which they select.
concentrations varied up to a maximum of ∼70 We may start with Margalef’s (1978) ‘tentative’
mg chla m−3 at the surface, and to ∼50 mg m−3 plot to illustrate the sequence (he called it a
at a depth of 4 m. ‘succession’, which usage will be discussed in
Section 7.3.2) of phytoplankton dominance in
7.2.2 Species assemblage patterns relation to nutrients and stratification (he used
in the sea the term ‘turbulence’). A simplified version is
Several deductions emerge from the above excur- shown in Fig. 7.1. The original diagonal of his
sion around the world’s seas and their repre- plot tracked the four developmental stages in
sentative phytoplankton assemblages. One of the the Ria de Vigo, progressing from the Stage-1
most self-evident of these is just how rarefied diatoms of the still well-mixed, nutrient-rich con-
is the phytoplankton over much of the ocean. ditions of the early vegetative season, through
For much of the tropical ocean, the concentra- the Chaetoceros- and Rhizosolenia-dominated stages
tion of primary-producer chlorophyll is generally to the Stage-4 preponderance of dinoflagellates
much lower than 1 mg chla m−3 , and equiva- capable of exploiting the well-stratified and
lent to little more than 20 mg chla m−2 . In the resource-segregated water column to compensate
temperate ocean, maximal concentrations may the nutrient exhaustion of the surface water.
ssion
c ce
Su
Figure 7.1 Simplified version of Margalef’s diagram

summarising seasonal change in phytoplankton composition in
the sea as a direct function of weakening ‘turbulence’ and Figure 7.2 The ‘mandala’ (of Margalef et al., 1979),
diminishing free nutrients. Redrawn with permission from developed from Fig. 7.1, relating seasonal change to
Smayda and Reynolds (2001). environmental selection of life forms and life-history traits. L
and T are standard dimensionless units of length and time.
Redrawn with permission from Smayda and Reynolds (2001).
Margalef (1978) contended that the progression

moved from r-selected to K-selected species (see of life forms, rather than species, and a further
p. 209). The ‘tentative’ element was the widening trajectory selecting for the smaller, rounded ‘red-
of the representation to embrace more oceanic tide’ dinoflagellates.
diatoms (Thalassiosira, at the top right of the In an early attempt to apply Margalef’s (1978)
successional diagonal), highly adapted dinoflag- conceptual model to freshwater phytoplankton,
ellates (Ornithocercus at the bottom left) and to Reynolds (1980a) pointed to the frequent inci-
insert coccolithophorids about half-way along. dence of nutrient-rich, low-mixing conditions
Thus, the entire ocean flora was potentially expli- among shallow lakes. It was also recognised that
cable through the relationship between mixing these conditions generally promoted the rapid
and fertility. growth of small, exploitative organisms – in
This was a remarkable and stimulating con- fact, precisely those with the classical attributes
cept, relating evolutionary adaptations and sur- of r-strategists. Reynolds’ (1980a) view at that
vival strategies to habitat factors. It is flawed, time was that, against the two major axes of
in that to insert additional species from other Margalef ’s model (Fig. 7.1), true r → K succession
locations into a single r–K continuum breaks the would track more or less vertically. This means
understanding of the succession. A more serious that the diagonal really relates only to mixing
issue is the implication that nutrient availabil- intensity and consequential effects on nutrient
ity and mixing are mutually correlated, whereas redistribution of nutrients and that the distin-
they are independent variables. Margalef was guishing ability of the Thalassiosira–Chaetoceros
clearly aware of these difficulties, for his original diatoms that were considered to be exclusively
plot (Fig. 2 in Margalef, 1978) contains a reference r-selected by Margalef needed a new classifica-
to ‘red-tide’ dinoflagellates in the upper left-hand tion. Reynolds et al. (1983b) referred to them as
corner (‘high nutrients, low turbulence’), off the w-selected species, based on their morphologi-
main successional diagonal. He noted their occur- cal and physiological adaptations to maintain
rence as an aberration and a sort of ‘system ill- growth on the low average light incomes incum-
ness’. The issue was confronted again by Margalef bent upon deep column mixing, especially in
et al. (1979), whose ‘mandala’ representation (Fig. high-latitude winters. Limits to the refinement
7.2), based on a reorientation of Margalef’s first of the light-harvesting adaptations required for
plot (Fig. 7.1), accommodates a ‘main sequence’ effective operation under such conditions were
selection, selecting increasingly for the character-

istics of R-strategists, and which may well occur
in a sequence determined by r–K selectivity. The
relevant attributes of the algae favoured by the
environmental conditions are shown in Box 5.1
(on p. 212). The corresponding phytoplankton per-
formances and morphologies are also indicated
in Figs. 5.8 and 5.10.
Smayda and Reynolds (2001) used the same
axes to define a ‘habitat template’ (cf. South-
wood, 1977) for marine environments. The lay-
out, shown in Fig. 7.4, was conceptual, insofar as
the axes are not precisely quantified and merely
Figure 7.3 The ‘intaglio’ (of Smayda and Reynolds, 2001), indicative of the ranges of integrated light avail-
which allows selection of species within a wide ecological
ability and accessible nutrients. The superim-
space, according to their primary adaptive life-cycle strategies
(C, R, S), except where nutrients and light are both
posed boxes show the approximate positions of
continuously deficient (the ‘void’). Redrawn with permission specific pelagic habitats and their phytoplankton
from Smayda and Reynolds (2001). referred to in the preceding Section (7.2.1). The
broad diagonal is included as an approximate
border of habitat tenability and separation from
recognised, while the combination of low light the void areas. It also serves to maintain the anal-
and low nutrients was considered to be unten- ogy with Grime’s (1979, 2001) C–S–R triangular
able as suitable habitat for phytoplankton. configuration (Fig. 7.3).
There was, by now, a striking resemblance This representation allows us to reflect the
between the three viable habitat contingencies observation that the habitats able to support sub-
developed from Margalef’s first tentative plot and stantial phytoplankton biomass are those with
Grime’s (1979) (see Fig. 5.9 and Table 5.3) repre- the least enduring or least severe constraints
sentation of vegetational habitats and the C-, S- of energy and/or nutrient resources. Whereas
and R-life-history strategies that plants needed nutrient availability tails off in the downward
to adopt to live in them. Reynolds (1988a) pro- direction and harvestable light income dimin-
posed the basis of fitting phytoplankton, accord- ishes rightwards, the high-light, high-nutrient
ing to their morphologies and physiological sur- habitats in the top left hand corner are chiefly
vival adaptations, to the three viable pelagic habi- represented by near-shore habitats and coastal
tat combinations of mixing and resource gradi- lagoons, characterised by a potential for high
ents (reflecting, respectively, Grime’s ‘duration’ net production and a relatively high support-
and ‘productivity’ axes, marked in Fig. 5.9). This ive capacity for planktic biomass. These areas
can now be very simply summarised in the form are also relatively rich in the range of species
of a diagram (Fig. 7.3). In stable, well-insolated that they are observed to support. The strik-
columns, algal uptake is expected to deplete ing association of such waters with outbursts
nutrients, making the available resources more of nanoplanktic flagellates, of a variety of phy-
inaccessible and demanding specialist adapta- logenetic affinities (prasinophytes, chlorophytes,
tions of the phytoplankton for their retrieval. euglenophytes, cryptophytes, chrysophytes and
This sets the direction of true autogenic suc- small haptophytes) invokes a common adapta-
cession, moving inexorably from r-selected C- tion of small, nanoplanktic size and the poten-
strategists towards K-selected S-strategists. The tially rapid rates of growth conferred by high
income of harvestable light and its subjection to unit sv−1 ratios. These properties are shared by
the effects on entrainment of the variable mixed the Type-I gymnodinioid and Type-II small peri-
depth forms the horizontal axis: mixing events or dinian and prorocentroid dinoflagellates (classi-
continuous deep mixing cut across the autogenic fication of Smayda and Reynolds, 2001; Smayda,
Diminishing I*, increasing h m Figure 7.4 Schematic summary

of marine pelagic habitats, along a
Increasingly inaccessible nutrients
notional onshore–inshore transect

and separating deep-mixed and
I well-stratified systems of varied
nutrient deficiency. I∗ refers to the
F T
integral of irradiance received by
phytoplankters in mixed-water
S T
M
layers of variable thickness (hm ).
O
Smayda and Reynolds (2001).
H P
2002) that include Gyrodinium species, Katodinium, tolerance and self-regulation that characterise
Heterocapsa triquetra, Scrippsiella trochoidea and Pro- the Group-VII dinophysoids.
rocentrum species and which are apparently near- The lower left-hand apex of Fig. 7.5 covers the
cosmopolitan among coastal waters. The com- extreme resource-deficient environments of the
mon Type-III ceratians (Ceratium fusus, C. furca stratified tropical ocean. The major constraint is
and C. tripos) extend into deeper shelf waters, to gather from the very low concentrations of
where they become more abundant in columns essential nutrients, of which phosphorus and,
when they are at least weakly stratified, proba- especially, iron may be the most deficient (Karl,
bly to within 20–40 m of the surface. Resource 1999). Conforming to the encounter–sufficiency
segregation is likely but these larger and more relationship of Wolf-Gladrow and Riebesell (1997)
motile dinoflagellates are better adapted to alter- (see also Section 4.2.1), the most efficient pri-
nate between satisfying their energy and nutri- mary producers are of picoplanktic size. With
ent requirements. These distributions are sepa- a biomass capacity unsupportive of mesoplank-
rately shown on the unlabelled template repre- tic phagotrophy, the arguable selective advantage
sented in Fig. 7.5. in favour of a dominant SS-strategist picoplank-
The shallow mixed, moderately enriched habi- ton (see p. 211) is well supported by the vast
tats of fronts, coastal currents and upwellings are extent and monotony of the Prochlorococcus mono-
represented in the centre of Fig. 7.5. These are culture in tropical oceans. Respite comes in
able to support smaller nanoflagellates (includ- the form of mixing episodes, stimulating mod-
ing a wealth of coccolithophorids at lower lati- est growths of nanoplanktic coccolithophorids
tudes) as well as the group of distinctive small, and such specialised microplankters as nitrogen-
rounded dinoflagellates that include the harm- fixing Trichodesmium, the diatoms Rhizosolenia styli-
ful species Karenia mikimotoi, Lingulodinium poly- formis, R. calcar-avis and Hemiaulus hauckii and
edra and Pyrodinium bahamense respectively rep- the dinoflagellates Ornithocercus and Pyrocystis, all
resentative of Smayda’s Types IV, V and VI. The of which are incumbent upon hydraulic vari-
resource depletion of the upwelling relaxation ations in physical stability. These episodes are
zones requires the attributes of low-resource usually relatively short-lived, gradually reverting
Figure 7.5 Summary of marine

pelagic habitats (as shown in Figure
7.4), now populated by
functional–morphological categories
of phytoplankton (especially of
dinoflagellates), provisionally
identified by roman numerals: small,
rounded gymnodinioids (I) and
peridinians (II); migratory ceratians
(III); frontal-, upwelling- and
current-associated species (IV –VI);
heterotrophic dinophysoids (VII),
species of the ultraoligotrophic
oceans (VIII) and tropical DCM
species (IX); for further details, see
text. Figure redrawn with
permission from Smayda and
Reynolds (2001).
to the Prochlorococcus-dominated ambient steady in shallow and inshore areas, including several of
state (Karl et al., 2002). the large-celled, discoid species of centric diatom,
Diatoms are represented almost everywhere such as Cyclotella litoralis, C. caspia and species of
within the triangular space in Fig. 7.5. Although Actinocyclus, Cerataulina and Coscinodiscus, that are
their basic requirements for light and nutrients encountered also in lower-latitude upwellings.
are similar to those of all other phytoplank- Their performances are clearly favoured by rela-
ters, satisfaction of two diatom-specific specialist tively high nutrient levels and may depend upon
needs – a supply of skeletally progenic silicic acid high levels of insolation. The large group of
(SRSi) (see Section 4.7) and frequent or continu- diatoms whose ranges extend into deeper, but
ous entrainment in a surface mixed layer >1–3 still nutrient-rich offshore shelf areas – includ-
m in depth (or more if only a slow rate of growth ing species of Thalassiosira, Chaetoceros, Leptocylin-
can be sustained) (see Section 6.3.2) – is still possi- drus, Skeletonema and the slender-celled Rhizosole-
ble over all but the extreme left-hand side of the nia species – all show the attenuated antennal
template. Nevertheless, a reasonable first suppo- morphologies of R-strategists (sometimes exag-
sition is that planktic diatoms of seas and oceans gerated by chain formation). Many of these
should invoke the strongly R-strategist adapta- same species appear in the summer plank-
tions suited to passive entrainment in highly fluc- ton of the boreal oceans and polar seas. The
tuant, low-average-light environments. restricted diatom flora (Hemiaulus spp., broad-
Many of the oceanic and shelf species, indeed, celled Rhizosolenia styliformis and R. calcar-avis) tol-
comply with this anticipation. However, the great erant of warm, nutrient-poor but often well-
majority of these are found in coastal and near- insolated waters show little tendency towards
shore waters, where they are exposed to moder- superior light-harvesting but have special adap-
ately high nutrient levels and moderately high tations to contending with chronic nutrient lim-
light levels. Some of these are recorded mainly itation Indeed, these diatoms show characters of
more K-selected S-strategists and they may be per- structures of distinct pelagic ecosystems. Platt
haps considered as intermediate RS-strategists. and Sathyendranath (1999) visualised globally
Several other general deductions about the segregated provinces of the sea, distinguished by
composition of planktic communities generally their susceptibility to environmental forcing, the
and the functional role that they fulfil emerge primary production that each might sustain and
from the patterns identified. One relates to the the fates of their primary products. The species of
high species richness of the relatively benign phytoplankton at the hearts of these structures
environments that are not hostile to the majority are, often, both the architects of the processing
of species as a consequence of severe resource and and the best-fitted respondents to the prevailing
energy constraints. Extremes in either direction environmental constraints.
lead to the failure of all those species that lack
the adaptations to be able to tolerate the increas- 7.2.3 Species assemblage patterns in lakes
ingly exacting circumstances – species richness To review the composition of phytoplankton in a
is ‘squeezed out’ (Reynolds, 1993b). The toler- diverse range of inland waters – great lakes and
ant survivors are, by definition, well-adapted spe- small lakes, deep and shallow, saline and soft,
cialists and their presence in low-diversity com- acidic and calcareous, rivers, reservoirs, ponds –
munities constitutes a robust indicator of the in anything like the same way as was done for the
particular severe conditions. More, their pres- sea (Section 7.2.1) would be a daunting exercise,
ence will always help to identify the function for author and reader alike. Fortunately, there
of species clusters associated with slightly less is an easier course to be steered, although the
extreme circumstances (Dufrêne and Legendre, rules of navigation need some prior explanation.
1997). Because less exacting conditions are acces- Part of a personal quest to be able to define what
sible to many more species, more outcomes are algae live where and why (Reynolds, 1984a) has,
possible and, thus, they tend to lack positive over a period of 20 years, developed a tentative
species identifiers. and still-evolving phytoplankton flora (Reynolds
Secondly, species-poor, highly selected assem- et al., 2002). Cataloguing natural assemblages of
blages of species will dominate the behaviour of phytoplankton species generally reveals interest-
the community and control the fate of primary ing patterns. Not only are many species observed
products. For instance, assemblages dominated periodically in a given lake but their periodicity is
by diatoms are most liable to the dynamic con- also generally quite regular. Moreover, they often
trols set by sinking loss rates. Biomass is mainly co-occur with other species whose numbers fluc-
exported to depth, where it is processed by ben- tuate similarly and broadly simultaneously, as if
thic or bathypelagic consumers through spatially in response to the same seasonal or environmen-
large recycle mechanisms. Heavy grazing may tal drivers. Further, in part or in whole, the same
reduce the direct sedimentation of phytoplank- clusters of co-occurring species are recognised in
ters but, in part, substitute a flux of zooplankton other water bodies, despite mutual hydrological
cadavers, faecal pellets and an export of particu- isolation in many instances, and in what appear
late silica. Planktic primary products are more to similar kinds of water bodies but at remote dis-
likely to be accumulated in the pelagic if the tances. In between times and at many other loca-
algae are simultaneously small and ungrazed. tions, these species clusters are not represented.
Export is proportionately least when most of the They may well be replaced in abundance by quite
carbon is fixed by picophytoplankton and pro- different sets of co-occurring species but which,
cessed through a microbial food web (Legendre nevertheless, form equally distinctive, recurrent
and LeFevre, 1989). clusters.
Finally, it is the matching of processes to func- The pattern is scarcely obscure but it is diffi-
tional groups of phytoplankton species and, in cult either to describe or to explain. Before the
turn, to the overriding environmental circum- days of the sophisticated and readily available
stances biassing their selection, that leads to the statistical packages, the best-known techniques
elaboration and definition of macroscale spatial were those introduced by the European school of
phytosociologists to diagnose plant communities et al., 2003; Naselli-Flores and Barone, 2003;
and associations (Tüxen, 1955; Braun-Blanquet, Naselli-Flores et al., 2003). The scheme is still
1964). They would make a list of species in each evolving and two further algal groups have since
of a series of intuitively judged small areas of been proposed (Padisák et al., 2003c). A new con-
uniform vegetation, called relevés, scoring for the fusion is the fact that some species are correctly
relative area covered by each species. Listing the classifiable in more than one cluster, according
species in the same order made it easy to build up to their life histories (see p. 269).
frequency tables in which regularly co-occurring However, the scheme is not just about recog-
species are blocked together, while those that nising and giving labels to groups. The species
are avoided will appear in other blocks. These forming the particular groups have demonstra-
blocks, or associations, can be named and classi- bly similar morphologies, environmental sensi-
fied, just as if they were individual species. The tivities and tolerances, and they are not necessar-
task of explaining the ecologies of the compo- ily confined to one phylogenetic group (Reynolds,
nent species is arguably easier to progress if the 1984b, 1988a). They feature prominently the
vagaries of variable presence or relative impor- strategic adaptations that are required in the
tance of individual species is suborned to the habitats in which they are known to be capable
higher level of the species cluster. of good growth performances (Reynolds, 1987b,
Confronting accumulating records of species 1995a). These various aspects were summarised
counts in preserved samples collected weekly in tabular form in Reynolds et al. (2002). The
(sometimes more frequently) from each of several coda can be used to represent seasonal changes in
separate water bodies, I applied myself to the very dominance (Naselli-Flores et al., 2003), responses
tedious task of treating each count as a phyto- to eutrophication (Huszar et al., 2003) and the
sociological relevé and to diagnosing species that effects of non-seasonal physical forcing (Reynolds,
co-occurred frequently, rarely or not at all. Some 1993b).
weighting for larger species occurring in small The groupings themselves are tabulated in
numbers was the only modification needed to Table 7.1 for reference. Their properties are briefly
diagnose 14 such species-clusters that were ade- noted but these are also amplified in the context
quate to describe the entire seasonal periodic- of the compositional patterns observable in the
ity of the phytoplankton in five contrasted lakes freshwater systems exemplified.
in north-west England and five managed annual
sequences in the Blelham experimental enclo- Phytoplankton of large oligotrophic and
sures (Fig. 5.11). The clusters were not identified ultraoligotrophic lakes
beyond an alphanumeric label but the patterns We start with examples of the phytoplankton
and periodic sequences were conveniently ratio- assemblages that are encountered in some of
nalised in these terms (Reynolds, 1980a). the world’s larger lakes. According to Herden-
The original scheme has been much modified, dorf (1982), 19 of the inland waters currently
mainly through the addition of more alphanu- on planet Earth have areas greater than 10 000
meric groups to embrace algal assemblages in km2 and another 230 are greater than 500 km2 .
other types of water and in many other global Together, they contain about 90% of its inland
locations. Most of the new groupings have been surface water. To put these in a single cate-
delimited using statistical methods, which, inci- gory of ‘large lakes’ can be justified only in the
dentally, have been used to validate almost all present context of waters overwhelmingly domi-
the original ones. Some of these have been sub- nated by open-water, pelagic habitats. Here, the
divided or realigned slightly in arriving at the grouping will exclude examples that are ‘shal-
31 groups defined by Reynolds et al. (2002). Sev- low’. This term itself requires careful definition;
eral independent studies have been able to apply following Padisák and Reynolds (2003), shallow-
and amplify the scheme without undue contro- ness is only sometimes a self-evident absolute.
versy and, thus, help to confirm its utility (Kruk The statement that a lake is ‘relatively shal-
et al., 2002; Dokulil and Teubner, 2003; Leitão low’ is based upon the ratio between absolute
Table 7.1 Trait-separated functional groups of phytoplankton
Group Habitat Typical representatives Tolerances Sensitivities

A Clear, often well-mixed, Urosolenia, Cyclotella Nutrient deficiency pH rise
base-poor lakes comensis
B Vertically mixed, Aulacoseira subarctica, A. Light deficiency pH rise, Si
mesotrophic islandica depletion,
small–medium lakes stratification
C Mixed, eutrophic Asterionella formosa Light, C Si exhaustion,
small–medium lakes Aulacoseira ambigua, deficiencies stratification
Stephanodiscus rotula
D Shallow, enriched turbid Synedra acus, Nitzschia Flushing Nutrient depletion
waters, including spp., Stephanodiscus
rivers hantzschii
N Mesotrophic epilimnia Tabellaria, Cosmarium, Nutrient deficiency Stratification, pH
Staurodesmus rise
P Eutrophic epilimnia Fragilaria crotonensis, Mild light and C Stratification Si
Aulacoseira granulata, deficiency depletion
Closterium aciculare,
Staurastrum pingue
T Deep, well-mixed Geminella, Mougeotia, Light deficiency Nutrient deficiency
epilimnia Tribonema
S1 Turbid mixed layers Planktothrix agardhii, Highly Flushing
Limnothrix redekei, light-deficient
Pseudanabaena conditions
S2 Shallow, turbid mixed Spirulina, Arthrospira Light-deficient Flushing
layers conditions
SN Warm mixed layers Cylindrospermopsis, Light-, nitrogen- Flushing
Anabaena minutissima deficient
conditions
Z Deep, clear, mixed layers Synechococcus, Low nutrient Light deficiency,
prokaryote grazing
picoplankton
X3 Shallow, clear, mixed Koliella, Chrysococcus, Low base status Mixing, grazing
layers eukaryote
picoplankton
X2 Shallow, clear mixed Plagioselmis, Stratification Mixing,
layers in Chrysochromulina filter-feeding
meso-eutrophic lakes
X1 Shallow mixed layers in Chlorella, Ankyra, Stratification Nutrient deficiency,
enriched conditions Monoraphidium filter-feeding
Y Usually small, enriched Cryptomonas, Peridinium Low light Phagotrophs!
lakes lomnickii
E Usually small, Dinobryon, Mallomonas Low nutrients CO2 deficiency
oligotrophic, (Synura) (resort to
base-poor lakes or mixotrophy)
heterotrophic ponds
(cont.)
Table 7.1 (cont.)
Group Habitat Typical representatives Tolerances Sensitivities

F Clear epilimnia Colonial chlorophytes Low nutrients ?CO2 deficiency,
like Botryococcus, high turbidity
Pseudosphaerocystis,
Coenochloris, Oocystis
lacustris
G Short, nutrient-rich Eudorina, Volvox High light Nutrient deficiency
water columns
J Shallow, enriched lakes Pediastrum, Coelastrum, Settling into low
ponds and rivers Scenedesmus, light
Golenkinia
K Short, nutrient columns Aphanothece, Deep mixing
Aphanocapsa
H1 Dinitrogen-fixing Anabaena flos-aquae, Low nitrogen, low Mixing, poor light,
nostocaleans Aphanizomenon carbon low phosphorus
H2 Dinitrogen-fixing Anabaena lemmermanni, Low nitrogen Mixing, poor light
nostocaleans of larger Gloeotrichia echinulata
mesotrophic lakes
U Summer epilimnia Uroglena Low nutrients CO2 deficiency
LO Summer epilimnia in Peridinium willei, Segregated Prolonged or deep
mesotrophic lakes Woronichinia nutrients mixing
LM Summer epilimnia in Ceratium, Microcystis Very low C, Mixing, poor light
eutrophic lakes stratification
M Dielly mixed layers of Microcystis, High insolation Flushing, low total
small eutrophic, low Sphaerocavum light
latitude
R Metalimnia of Planktothrix rubescens, Low light, strong Instability
mesotrophic stratified P. mougeotii segregation
lakes
V Metalimnia of eutrophic Chromatium, Chlorobium Very low light, Instability
stratified lakes strong
segregation
W1 Small organic ponds Euglenoids, Synura, High BOD Grazing
Gonium
W2 Shallow mesotrophic Bottom-dwelling ? ?
lakes Trachelomonas (e.g. T.
volvocina)
Q Small humic lakes Gonyostomum High colour ?
Source: Updated from Reynolds et al. (2002).
depth and wind fetch and the ability of a lake physical recycling and the frequency access of
to maintain a density-differentiated stratifica- phytoplankton to resources accumulated by and
tion for some weeks or months on end (see discharged from the deeper sediments. For many
Fig. 2.19). The relevance of the distinction is years, the bathymetry of (and, hence, the vol-
the extent to which the plankton-bearing sur- ume of water stored in) large lakes remained less
face waters are affected by the internal rates of familiar than their respective areas. Several have
been added to the category of ‘deep lakes’, even confined to the stratified periods (Kozhov, 1963;
since Herdendorf’s (1990) listing. These include Kozhova, 1987; Kozhova and Izmest’eva, 1998;
Lago General Carrera/Buenos Aires, straddling Goldman and Jassby, 2001; Popovskaya, 2001).
the Chile/Argentina border, Danau Matano in The spring development, which takes place under
Indonesia and Lake Vostok, Antarctica. ice, is subject to sharp interannual variability. In
Setting aside those that are saline (relatively) high-production years, diatoms (espe-
(Kaspiyskoye More, Aralskoye More), the natural cially Aulacoseira baicalensis, A. islandica, Nitzschia
condition of the water in most of these large, acicularis, Synedra ulna var. danica and Stephano-
deep lakes is to be deficient in nutrient resources. discus binderanus), small dinoflagellates (includ-
They occupy large basins, fashioned either by ing Gymnodinium baicalense) and chrysophytes
tectonics or scraped out by glacial action, and (especially Dinobryon cylindricum) are prominent
are presently filled with water that is renewed (Popovskaya, 2001). Although several of these
only very slowly. Large lake volumes in relation algae are endemic, the assemblage corresponds
to catchment area also make for low support- mainly to Association-B diatoms, which have
ive capacities and, indeed, the phytoplankton a high sv−1 , and whose growth is tolerant of
they carry is typically dilute. Where known low temperature, poor insolation and low phos-
(or where approximated from published bio- phorus concentrations (Richardson et al., 2000),
volume estimates), average seasonal maxima with some representatives of the E- and Y-groups
of chlorophyll are <4 mg chla m−3 , although of flagellates. In low-production years, all algae
greater concentrations may be found locally are scarce (the interannual difference in pop-
(Reynolds et al., 2000). In many instances, the ulations of dominant Aulacoseira baicalensis is
paucity of phytoplankton may be determined between 100–200 cells mL−1 to 1 to 2 orders of
principally by energy limitation in deep, mixed magnitude fewer: (Popovskaya, 2001). The critical
layers. variable seems to be the extent of snow cover
Among the most systematically studied of on the ice: besides letting more light through,
these large lakes is Ozero Baykal. Formed in a gap snow-free ice allows more rapid heating of water
between two separating tectonic plates, Baykal is directly beneath the ice, which then sinks to the
also the deepest (1741 m), stores the greatest vol- point of isopycny, some 10–20 m below. This sets
ume (nearly 23 000 km3 ) and is probably the old- up a convective motion which resembles the epil-
est (∼20 Ma) of all the world’s lakes. Despite sig- imnion of a warm, ice-free lake (Rossolimo, 1957).
nificant industrialisation and settlement of the Indeed, this structure allows rather better inso-
catchment (especially around Irkutsk) and pol- lation of entrained diatoms than is possible in
lution of its two major inflows (Angara, Selenga the ice-free column, until surface heating allows
Rivers), the lake remains oligotrophic in charac- the lake to stratify directly (maximum surface
ter. Retention time is estimated to be 390 a. In temperatures may then reach 12–16 ◦ C: Nakano
the offshore areas, levels of soluble phosphorus et al., 2003). Then, algae tolerant of stratifica-
(SRP) and dissolved inorganic species of nitrogen tion, high insolation and low nitrogen and phos-
(DIN) are <15 µg P and <100 µg N l−1 (Kolpakova phorus concentrations (F-group colonial chloro-
et al., 1988; Goldman and Jassby (2001). The lake is phytes, including Botryococcus, and potentially
classically dimictic. Near-surface warming of the nitrogen-fixing Anabaeana lemmermannii of group
water surface in the summer induces thermal H2: Goldman and Jassby, 2001; Popovskaya, 2001)
stratification (July to September); in the winter, predominate. However, the main pelagic primary
the lake is ice-covered from January to late May. producers in summer are group-Z picoplanktic
At other times, the lake is subject to deep con- Cyanobacteria: in Baykal, populations of Syne-
vective mixing that is sufficiently intense to aer- chocystis limnetica may exceed 105 cells mL−1 , are
ate the profundal waters (Rossolimo, 1957; Vot- responsible for, perhaps, 80% of the pelagic pri-
intsev, 1992). Despite substantial year-to-year vari- mary production and support a well-developed
ations in the production and standing biomass microplanktic food web (Nakano et al., 2003). Like
of phytoplankton, it is plain from each of the Prochlorococcus in the tropical sea, they are able to
main overviews that increase of both is mainly function in clear, well-insolated water, turning
over fixed carbon to the microbial heterotrophs, of the surface), phytoplankton increases during
whilst nevertheless maintaining a weakly grazed spring to a maximum of ∼7 µg chla L−1 in
biomass. June. A second peak in August has about half
Lake Superior covers a larger area than Ozero this magnitude (Munawar and Munawar, 2001).
Baykal (82 100 against 31 500 km2 ) but its depth In Lake Michigan (area 57 750 km2 , mean depth
(maximum 407 m, mean 149 m) and volume 85 m, TP 5–8.5 µg P L−1 , DIN ≤260 µg N L−1 ,
(12 200 km3 ) are inferior (Herdendorf, 1982). Its stratified July–September to within 20–30 m of
basin is tectonic in origin but was significantly the surface), a similar diacmic pattern of abun-
scoured by ice during the last (and possibly pre- dance is observed, with maxima in June and
vious) glaciation. The present lake is little more August. The earlier is larger (maximum ∼11
than 15 ka in age (Gray et al., 1994). The lake area µg chla L−1 ) than the second. The dominant
is large relative to that of the drainage basin; species are similar in either case: diatoms of
currently, the hydraulic retention time is around the A-group (Cyclotella spp., notably C. bodan-
170 a. The input of nutrients has for long been ica, C. radiosa, C. glomerata; also Urosolenia erien-
very low and, except in the vicinity of industri- sis) and such B-group representatives as Aulaco-
alised cities like Duluth and Thunder Bay, con- seira islandica and Tabellaria flocculosa are rela-
centrations of TP (3.5–7.0 µg P L−1 ), SRP (≤3 µg P tively abundant throughout and dominate the
L−1 ) and DIN (≤300 µg N L−1 ) are typically dilute. earlier peak. Nanoplanktic flagellates of the X2
Annual TP loads are <0.1 g m−2 a−1 (Vollen- group (Chrysochomulina, Ochromonas) and E- and U-
weider et al., 1974). The phytoplankton biomass groups of microplanktic chrysophytes (Dinobryon
(supposed to average ∼1–1.5 µg chla L−1 ) is cor- spp., Mallomonas spp., Uroglena spp.) are relatively
respondingly modest and the water (Secchi-disk more abundant during the second peak, when
transparency 10–17 m: Gray et al., 1994). How- picoplanktic Chroococcoids (Z) are also at their
ever, for much of the time, low water temper- most numerous (Munawar and Munawar, 2001).
atures and weak insolation may provide more Comparison with the findings of Johnson (1975,
severe production constraints. Enhanced produc- 1994) shows that similar species assemblages
tion and elevated phytoplankton biomass in the make up the very sparse ‘peaks’ (probably <0.5 µg
open water are noted between July and October, chla L−1 ) of phytoplankton biomass encountered
when surface water temperatures are sufficiently at the beginning and towards the end of the ice-
differentiated for mixing to be restricted to the free period (July–November) in Great Bear Lake
upper 20–30 m, allowing weak summer stratifica- (area 31 150 km2 , mean depth 72 m, SRP ≤0.1 µg
tion (Munawar and Munawar, 1986, 2000, 2001). P L−1 ).
According to these synopses, diatoms (especially Since the late 1980s, the Laurentian Great
Cyclotella radiosa and Tabellaria fenestrata) develop Lakes have been experiencing the spread of a
at the start of this period, followed by nanoplank- Eurasian alien, the bivalve Dreissena polymorpha.
tic phytoflagellates of the genera Cryptomonas, Its motile larvae have taken great advantage of
Plagioselmis, Ochromonas and Chrysochromulina and the canal systems of Europe and the ballast tanks
microplanktic Uroglena. Picoplanktic chroococ- of ocean-going ships to spread from its native
coid Cyanobacteria become numerous (Munawar areas of the Caucasus to the Atlantic seaboard
and Fahnenstiel, 1982; Fahnenstiel et al., 1986) (by the mid twentieth century) and, eventually,
and, despite being a small part of the phyto- to the St Lawrence basin. The mollusc colonises
plankton biomass, contribute significantly to the almost any firm submerged surface where, multi-
annual carbon budget (∼50 C m−2 a−1 ). The plied by its numbers and explosive reproductive
sequence of phytoplankton can be summarised performances, its dense aggregations can gener-
A/B → X2/Y → U → Z. ate significant filtration capacities. Dreissena has
Superior’s neighbouring Great Lakes are sim- been particularly successful in Lake Erie and the
ilar in the modesty of phytoplankton they sup- Saginaw Bay area of Lake Huron where its inva-
port. In Lake Huron (area 59 500 km2 , mean depth sion has contributed to the reduction in phyto-
59 m, TP 5–7 µg P L−1 , DIN ≤300 µg N L−1 , plankton and TP content of the water (arguably
stratified July–September to within 20–30 m more so than newly imposed statutory pollution
controls) and improvements in clarity and macro- (Reynolds et al., 2002). Lake Ontario supports an
phyte growth (Nalepa and Fahnenstiel, 1995). abundant chroococcoid picoplankton in summer
Among the other large North American lakes, and whose dynamics and distribution were the
the spring phytoplankton generally includes subject of a detailed study by Caron et al. (1985).
a wider range of diatoms. Various studies Diatoms occur throughout the year in Lake
(Munawar and Munawar, 1981, 1986, 1996; Ontario but, during the early part of the year,
Duthie and Hart, 1987; Pollingher, 1990) reveal Stephanodiscus binderanus, especially, ‘blooms’ in
that C-group Asterionella formosa and Stephan- the near-shore areas of the lake, where popu-
odiscus binderanus are well represented in the lations become substantially isolated by a phe-
plankton of Great Slave Lake (area 27 200 km2 , nomenon known as ‘thermal barring’. This was
mean depth 58 m, TP: 3–8 µg P L−1 , ice-free described in detail by Munawar and Munawar
June–November, stratified to within ∼30 m of the (1975) but it is now known to be a common
surface July/August: details from Moore, 1980), feature in other large, cold-water lakes, includ-
Lake Erie (area 25 660 km2 mean depth 19 m, ing Michigan (Stoermer, 1968), Ladozhskoye and
TP 11–45 µg P L−1 ) and Lake Ontario (area 19 100 Onezhskoye (Raspopov, 1985; Petrova, 1986) and
km2 mean depth 86 m, TP 10–25 µg P L−1 , like Baykal (Shimarev et al., 1993; Likhoshway et al.,
Lake Erie, it is stratified to within ∼22 m of the 1996). In essence, vernal heating proceeds more
surface June–October). In the eastern and cen- rapidly in the shallow near-shore waters than in
tral basins of Lake Erie, phytoplankton biomass the open, offshore areas of the lake but, while
develops during June culminating in summer temperatures remain <4 ◦ C, the warmer water
(August) maxima in the order of up to 30 µg chla is retained within inshore circulations separated
L−1 and in which small dinoflagellates (especially from the open lake by pronounced frontal bound-
Gymnodinium helveticum, G. uberrimum and Glen- aries. The slightly higher temperatures and the
odinium spp.) and Cryptomonas species are com- rather higher average insolation experienced by
mon reprsentatives of a Y-type assemblage. There the algae thus retained promotes the early-season
is also a well-developed nanoplankton in which growth and recruitment of phytoplankton, domi-
X2-group Plagioselmis and Chrysochromulina are nated by group-C diatoms, group-Y flagellates and
joined by Chlorella, Monoraphidium and Tetraedron nanoplankton.
species, all X1-group, overtly C-strategist species The phytoplankton of the two northern great
whose growth requires higher half-saturation lakes of Eurasia shows closer affinites to that
concentrations of SRP (Section 5.4.4). Chroococ- of Baykal, Erie and Ontario than to that of the
coid picoplankters (Z) are also present in the sum- upper Laurentian Great Lakes. In Ladozhskoye
mer. In the shallower and most enriched western Ozero (area 18 140 km2 , mean depth 50 m, TP:
basin of Lake Erie, there is a substantial spring 13–40 µg P L−1 , ice-covered February–May, strat-
bloom dominated by Stephanodiscus binderanus ified July–August, but very variable stability), a
and other B- and C-group diatoms, abundant flag- spring diatom bloom of B- and C-group diatoms
ellates of group Y and Pediastrum and Scenedesmus, (including Aulacoseira subarctica, A. islandica, Asteri-
representing the J group of eutrophic chloro- onella formosa, Diatoma spp.), apparently nur-
coccaleans. The offshore phytoplankton of Lake tured in thermal-bar conditions (see above and
Ontario behaves similarly to that of eastern p. 89), spreads through the lake (chla gener-
Lake Erie, except that the summer plankton is ally 1–5 µg L−1 ). Asterionella continues to domi-
more biassed towards dominance by mucilage- nate the unstable open lake conditions, together
bound, non-motile, colonial chlorococcalean and with variable quantities of group-P Aulacoseira
tetrasporalean genera of green algae (group F) granulata and Fragilaria crotonensis. Often, signif-
such as Oocystis, Coenochloris and Pseudosphaerocys- icant quantities of filamentous Cyanobacteria
tis. Hutchinson (1967) called these ‘Oligotrophic (especially Planktothrix agardhii of group S1) and
chlorococcal plankters’; it is a widespread group xanthophytes (Tribonema of group T) are selected
(see below) that survives low SRP concentrations by their tolerance of low average insolation
but is apparently intolerant of poor insolation in mixed conditions (Raspopov, 1985). Under
calmer conditions, the Cyanobacteria Aphan- Phytoplankton of smaller temperate

izomenon and Woronichinia sometimes form sig- oligotrophic lakes
nificant blooms. I have no information to hand The ‘small-lake’ category as adopted here applies
concerning the nano- and picoplankton but to waters <10 km2 . Whilst most of the fresh water
the dominant calanoid zooplankton and well- is stored in ‘large’ lakes (>500 km2 ), most of the
developed ciliate microplankton are suggestive world’s lakes (∼8.4 × 106 in total: Meybeck, 1995)
of an active microbial food web. In Onezh- are ‘small’. A further 13 450 in the range 10–500
skoye Ozero (area 9 900 km2 , mean depth 28 m, km2 might be described as ‘medium-sized’. So
TP 5–10 µg P L−1 , ice-covered January–May, far as is known, a large proportion of these
thereafter, thermal bar formation separates the represents lakes in unyielding rocky basins and
inshore from the isothermal circulation of the forested catchments and the phytoplankton car-
central water mass; stratification develops from rying capacity is unambiguously constrained by
July to September, to within 30 m from the sur- the availability of nutrient resources, rather than
face), the spring assemblage is dominated by B- by low temperatures and low insolation. Their
group Aulacoseira and C-group Asterionella. The sea- smaller sizes facilitate a high frequency of sam-
sonal progression passes by way of Tabellaria to ple collection and a robust picture of seasonal
a sequence of oligotrophic–mesotrophic assem- change in production and biomass as well as
blages featuring Dinobryon (E) species, Coenochlo- species composition. In previous work (Reynolds,
ris (F) species, Woronichinia (LO ) and Planktothrix 1984a), I have cited Findenegg’s (1943) thorough
agardhii (S1) but the biomass at the summer max- two-year (1935–36) survey of the phytoplankton
imum scarcely exceeds 6 µg chla L−1 (Kauffman, in some 15 moderately alkaline lakes of Kärnten
1990). (Carinthia), Austria, to exemplify the genre (in
In Ozero Issyk-kul (area 6 240 km2 , mean particular, Millstätter See: area 13.3 km2 , mean
depth 279 m, TP 2–4 µg P L−1 , stratified depth 86.5 m, stratified May–November to within
April–December, to within ≥13 m of the surface: 7–10 m of the surface, TP, not reported but SRP
Shaboonin, 1982), relative phytoplankton abun- <1 µg P L−1 , DIN ≤300 µg N L−1 ). There is a sin-
dance is diacmic. The earlier (May) peak involves gle (monacmic) summer biomass ‘maximum’ in
group-A Cyclotella species and group-B Aulacoseira summer (chla ∼1.5 µg P L−1 (Fig 7.6), when the
species the later (October) peak is dominated A-group spring diatoms (Cyclotella comensis, C. glom-
by the Cyanobacterium Merismopedia and the erata) are supported by a mixture of other species
dinoflagellate Peridinium willei which both rep- that include self-regulating dinoflagelleates (Peri-
resent group LO . Species of Coenochloris, Oocys- dinium willei, some Ceratium hirundinella) and
tis, Gloeocapsa, Lyngbya, as well as an ‘abundant colonial Cyanobacteria, now known as Woroni-
nanoplankton’ also feature in this faintly saline chinia (representing group LO ), and such (N-group,
lake (Savvaitova and Petr, 1992). high-nutrient-affinity) desmids as Staurodesmus
In conclusion, the phytoplankton of large, species Findenegg (1943) also recorded species
high-latitude lakes involves elements character- of Coenochloris, Oocystis, Gymnodinium, Rhodomonas
ising either some of the ultraoligotrophic func- and other nanoplankters, as well as the non-gas-
tional groups, distinguished by their high affin- vacuolate, unicellular Cyanobacteria Chroococcus
ity for phosphorus (A, Z), or other substantially and Dactylococcopsis. Moreover, Findenegg (1943)
oligotrophic groups (B, X2, E, F, LO , U) toler- found similar assemblages in several of the other
ant of low nutrient concentrations (especially of lakes in the region: in Wörthersee, Uroglena and,
phosphorus). However, there is no clear evidence especially, Planktothrix rubescens, constituted met-
that these algae habitually fill the nutrient- alimnetic maxima. Findenegg’s explanation for
determined capacity, except when average inso- the seasonally distinctive assemblages and dis-
lation allows it. Excess of nutrient capacity over tributions invoked the interaction of light and
average light income favours the R-strategies of temperature preferences of participating algae.
the C, P, S and T groups of attenuated and fila- The four contingencies (cold-water, low-light
mentous ‘antennal’ algae. forms; cold-water, high-light forms; warm-water,
Figure 7.6 Annual cycles of

phytoplankton biomass (as
measured or approximated from
biovolume) in some temperate
lakes: (a) Millstättersee, 1935 (after
Findenegg, 1943); (b) Windermere,
North Basin, 1978 (Reynolds,
1984a); (c) Sjön Erken, 1957
(Nauwerck, 1963); (d) Crose Mere,
1973 (Reynolds, 1976a). Redrawn
high-light forms; warm-water, low-light forms) (F-group Coenochloris, Oocystis), then desmids
approximated to winter, spring, summer and (Staurastrum spp.), before Aulacoseira ambigua and
autumn plankton. other spring diatoms are entrained from the
Other small, low-nutrient lakes show simi- deepest layers.
lar seasonality, involving similar species of alga. In Finland, many small lakes are simul-
Hino et al. (1998) recently published the results taneously soft-watered, oligotrophic and often
of an investigation of a small, low-alkalinity lake strongly humic. The high latitude and short win-
in Hokkaido, Japan (Akan-Panke: area 2.8 km2 , ter days greatly constrain the ability of phyto-
mean depth 24 m, TP ≤10 µg P L−1 , DIN ≤300 plankton to sustain net photosynthetic gain.
µg N L−1 , stratified May–October to within 5 Any growth at these times must be sustained
m of the surface, ice-covered November–May). heterotrophically; few surviving cells even con-
This lake is classically dimictic but the overturn tain chlorophyll (Arvola and Kankaala, 1989).
periods are brief; the lake is usually stratified However, there are some species of Chlamy-
and its resources become strongly segregated. domonas, Chlorogonium, Peridinium and Gymno-
The water has a high clarity (εmin ∼0.15 m−1 , dinium that are able to maintain high population
Secchi-disk transparency 7–18 m). The spring densities just below the surface while ice cover
overturn (maximum, 2–3 µg chla L−1 ) is dom- persists (Arvola and Kankaala, 1989).
inated by diatoms (A-, B- and C- group species In Stechlinsee, another well-studied olig-
include Cyclotella radiosa, Asterionella formosa and otrophic system (area: 4.3 km2 , mean depth 22.8
Aulacoseira ambigua). As they settle out (taking m, TP ≤16 µg P L−1 , DIN ≤95 µg N L−1 , stratified
most of the bioavailable P and TP with them, May–October, to within 7–10 m of the surface,
the epilimnion is left with a sparse population ice-covered January–March: details from Gervais
of E-group chrysophytes (particularly Dinobryon et al., 1997), the phytoplankton biomass peaks in
cylindrica). By summer, the surface waters are spring, dominated by A-group Cyclotella species
substantially depleted of all but chroococcoid and picoplanktic Cyanobacteria (Padisák et al.,
picoplankters and chlorococcal nanoplankters 1998). Development starts in February or March
(together accounting for <1 µg chla L−1 ). Dom- and is terminated by the onset of the summer
inant microplankters (in which U-group Uroglena stratification and the sinking of the diatoms into
americana and LO -group Peridinium aciculiferum the hypolimnion. Exhaustion of the epilimnetic
and Merismopedia are prominent) are based in nutrient base confines autotrophic production
the metalimnetic region (at a depth of 10–25 m). to the metalimnion but, in this lake, it is the
With increasing mixed depth towards the end picoplanktic Cyanobacteria that dominate both
of the summer, there are first more green algae the biomass and the production (see also Gervais
et al., 1997). Originally identified as a Synechococcus zooplankton abundance but both seemed to be
species. the alga is now recognised as Cyanobium incidental to the continuing predominance of
(Padisák, 2003). Peridinium. The success of the alga relates to its
In the 20 years covered by the study of Berman efficient perennation and its ability to exploit its
et al. (1992), the phytoplankton periodicity of mobility to ‘scavenge’ (or ‘glean’) the water col-
Yam Kinneret (Sea of Galilee) showed great inter- umn of its nutrient reserves during the early stag-
annual similarity in the abundance, distribu- nation period. Depletion of both phosphorus and
tion and composition of the phytoplankton. This nitrogen to the limits of detection attests to the
warm monomictic rift-valley lake in the upper efficiency of this process. The post-bloom phyto-
Jordan Valley (area 168 km2 , mean depth 25.5 plankton is redolent of other very oligotrophic
m, stratified March–December) experiences a typ- pelagic systems where the primary products of
ically Mediterranean climate, with rainfall con- similar organisms are tightly coupled within
fined to the winter period (Serruya, 1978). It is microbial food loops. Such interannual variations
briefly isothermal with a minimum temperature in the size of the Peridinium crop as had been
of ∼13 ◦ C but, by April, is generally strongly strat- observed seem to be inverse to the fluctuations in
ified to within 10 m of the surface, as epilim- Aulacoseira production: years with larger diatom
netic temperatures reach 30 ◦ C. Its waters are maxima heralded smaller Peridinium maxima.
mildly alkaline and slightly saline. The maxi- Interestingly, variations to the stable sequence
mum bioavailability of both nitrogen and phos- have been observed in the 1990s. There have been
phorus is modest (≤200 µg N L−1 , ≤5 µg P L−1 ). one or two years when the excystment and spring
In each of the years from 1970 to 1989 inclusive, recruitment of Peridinium has been poor and
the phytoplankton developed in a characteristic green algae or Microcystis becoming briefly abun-
way. Starting with the autumnal circulation, the dant. The summer appearance of nitrogen-fixing
main components were unicellular Cyanobacte- Anabaena ovalisporum, in 1994 and in one or two
ria (Chroococcus) and such (X2-) nanoflagellates as subsequent years, is another departure from the
Plagioselmis and Chrysochromulina. With full over- stable pattern. The tempting explanation is that
turn and winter flooding, coenobial and filamen- the increase in phosphorus loading has triggered
tous plankton develop, including the apparently these events but careful analysis shows a trend of
warm stenothermic diatom, Aulacoseira granulata, decreasing winter concentrations in the lake and
of Group P. At this time, there would also be a lengthening of the period of thermal stratifica-
excystment of the large, self-regulating dinoflag- tion (Hambright et al., 1994). Berman and Shtein-
ellate now known as Peridinium gatunense and man (1998) also comment on the effect of a weak-
ascribed by Reynolds et al. (2002) to Group LM . ening of diffusivity in the water column working
Under conditions of reduced vertical mixing, this against the selective exclusivity in favour of Peri-
alga typically built to a stable maximum, often dinium at critical points in its annual cycle.
lasting from March to May, which represented The last example in this section is from Biwa-
the greatest annual biomass. Its termination, ko, which, on criteria set in this chapter, is
through encystment and settlement, generally really a ‘large lake’ (area 674 km2 , mean depth
coincided with high epilimnetic temperatures 41 m, TP ≤10 µg P L−1 , DIN ≤350 µg N L−1 ,
(between 27 and 30 ◦ C) and virtual exhaustion of stratified April–November to within 20 m of the
the epilimnetic reserve of nitrogen. The summer surface: Nakamura and Chang, 2001). Its phyto-
biomass remained relatively very low, comprising plankton was described in detail by Nakanishi
a sparse nanoplankton and numerous picoplank- (1984) but ongoing concerns about the water
ton (Malinsky-Rushansky et al., 1995). Metalim- quality and the appearance of Microcystis pop-
netic layers tended to be dominated more by ulations in the lake (which supplies drinking
group-V photobacteria (Chlorobium) than algae. water to 15 M people in Kyoto and Osaka) have
Berman et al. (1992) were concerned to inter- encouraged frequent monitoring of the plank-
pret this interannual stability against a trend ton. The winter plankton of the mixed water
of rising phosphorus loads and to fluctuating column is sparse and dominated by Aulacoseira
solida (Group B?) with increasing quantities of (Vicente and Miracle, 1988; Guerrero and Mas-
Asterionella formosa (group C) through spring and Castellà, 1995) and Planktothrix of the rubescens
a summer assemblage of Staurastrum dorsiden- group (R) (Findenegg, 1943; Zimmermann, 1969;
tiferum, Closterium aciculare and Aulacoseira gran- Bright and Walsby, 2000). Cryptomonas species
ulata (group P). Since 1977, Uroglena americana (group Y) also maintain stable layers in karstic
has ‘bloomed’ each spring (maximum 6–10 µg dolines (Pedrós-Alió et al., 1987; Vicente and Mir-
chla L−1 ) and, since 1983, small numbers of acle, 1988) and, at times, in larger lakes (Ichimura
(group H) Anabaena and (group LM ) Microcystis et al., 1968). Deep chlorophyll layers involving
are encountered in the water column. These motile chrysophytes (E, U: Pick et al., 1984; Bird
are observed to form striking lee-shore scums and Kalff, 1989), non-motile chlorococcals (X1:
in quiet weather (Nakamura and Chang, 2001; Gasol and Pedrós-Alió, 1991) and picoplankton
Ishikawa et al., 2003). Rod-like picoplanktic syne- (Z: Gervais et al., 1997) tend to be rather more
chococcoids (0.4–1.5 µm) become numerous in diffuse.
summer (reportedly, up to 106 cells mL−1 : Maeda,
1993). Close interval sampling after a typhoon Phytoplankton of sub-Arctic lakes
suggested that microplanktic dynamics respond Based on his earlier investigations in the arctic
much more to hydrodynamic variability than do and sub-Arctic regions of Sweden and Canada
picoplankton numbers (Frenette et al., 1996). (Holmgren, 1968; Schindler and Holmgren, 1971),
To conclude the section, the seasonal patterns and on a careful review of the literature, Stef-
in the phytoplankton of small nutrient-deficient fan Holmgren devised a systematic and long-term
lakes in temperate regions involve oligotrophic experimental study of the lakes in the Kuokkel
functional groups in the approximate sequence B district of northen Sweden (centred on 68◦ 25 N,
→ X2 → E and/or F → U or LO → N; picoplank- 18◦ 30 E). These lakes are small (0.01–0.03 km2 ),
tic group Z is often an important component. shallow (mean depth 1–6 m) and soft-watered,
In the more alkaline waters, E and U are sub- are ice-covered for up to nine months per year
stantially missing and more eutrophic (carbon- and, at times, experience 24-h nights (December–
concentrating) functional groups C, LM and P are January) and 24-h days (June–July). In the natu-
better represented than B, LO and N. ral condition, the lakes are oligotrophic (TP <9
A common feature of stratifying oligotrophic mg P m−2 , DIN <240 mg N m−2 ) but Holmgren’s
lakes is the tendency for phytoplankton to form enrichment experiments (+P, +N and +N+P)
deep chlorophyll layers. These are not just the raised levels selectively to up to 300 mg P m−2 and
consequence of sinking and formation depends DIN to up to 6 g N m−2 . Among the many inter-
upon algae being able to self-regulate their ver- esting findings he reported (Holmgren, 1983) is
tical position, either through their own motil- a proposed ecological classification of arctic–sub-
ity or by regulation of their density. Reynolds Arctic phytoplankton (Table 7.2). This recognised
(1992c) surmised that their maintenance is gener- the ubiquity of Chrysophyceae and the differing
ally conditional upon low diffusivity (they have to matches with other algae, signifying between-
be below the epilimnion) and adequate light pen- lake and seasonal variability. Biomass varied with
etration (they have to be within the range of net nutrient fertility, being well correlated to N; wind
photosynthetic gain or, at worst, balance. The sta- was a bigger influence on seasonality than either
tion also offers advantage over a position higher temperature or insolation. Greater water hard-
in the light gradient and, usually, it is access to nesses supported more diatoms (Urosolenia, Syne-
nutrients. The likely algal components are also dra) and cryptophytes. Increased nitrogen lev-
influenced by the trophic state and growth con- els promoted Uroglena and the small phagotro-
ditions prior to stratification. The size of the lake hic dinoflagellate Gymnodinium. Simultaneously
and relative remoteness of the metalimnion from elevated availabilities of higher phosphorus and
the surface also affect the physical tenability of nitrogen favoured Chromulina, Ochromonas and the
the structural layers. Among the ‘tightest’, plate- green alga Choricystis. In all fertilised systems
like layers are constructed by photobacteria (V) chlorococcal green alga developed.
Table 7.2 Phytoplankton assemblages in Arctic and sub-Arctic lakes, according to the scheme of
Holmgren (1983) but using the identifiers proposed by Reynolds et al. (2002)
Spring Summer Autumn

1. Chrysophyceae lakes X3, E, Y E
2. Chrysophyceae–diatom lakes X2, X3,E,Y A, (B?) A,Y, LO
3. Chrysophyceae–Cryptophyceae lakes X2, X3, E B, Y B,Y
4. Chrysophyceae–Dinophyceae lakes Y U, B, F, LO LO
Phytoplankton in selected regions: in terms of a few key dominant species and

medium-sized glacial lakes of the their phylogenetic groups. These patterns were
European Alps further distilled by Reynolds et al. (2002) who
Besides distinguishing patterns from a selection ascribed Sommer’s (1986) key species to their rel-
of particular lake types, it is also helpful to make evant trait-differentiated functional groups (see
comparisons of phytoplankton periodicity among also Table 7.3). In this way, (A-group) diatoms
regional series of water bodies. Here, between- of the Cyclotella bodanica – C. glomerata group
lake differences owe relatively more to edaphic dominated the sparse spring bloom in the olig-
distinctions among the individual lake basins otrophic Königsee (area 5.2 km2 , mean depth 98
(and their hydrology, hydrography and hydro- m, TP 5 µg P L−1 ); in Attersee (area 46 km2 ,
chemistry) than to the general commonality of mean depth 84 m, TP 5 µg P L−1 ), Tabellaria
location or of formation. Regional clustering is species were relatively more prominent. Aulaco-
exemplified well by the residual water bodies seira species and Asterionella were abundant in
in upland areas where recent glaciations have the mesotrophic lakes, such as Vierwaldstättersee
scoured out the typically linear basins of ribbon (area 114 km2 , mean depth 104 m, TP 20 µg P
lakes or finger lakes. As a generality, these lakes L−1 ) and Ammersee (area 48 km2 , mean depth
are usually oligotrophic or mesotrophic in char- 38 m, TP 55 µg P L−1 ), with increasing subdom-
acter. However, it is often also true that the indi- inance of (Y-group) cryptomonads but (D-group)
vidual lakes of a given cluster can be ordered Stephanodiscus in the richer waters of Bodensee
by productivity or mean biomass supported, or (area 500 km2 , mean depth 100 m, TP (then)
by typical composition and abundance of phyto- 100 µg P L−1 ) and Lac Léman (area 582 km2 ,
plankton. Such arrangements are usually illus- mean depth 153 m, TP (then) 80 µg P L−1 ). The
trative of the selective effects on the assembly onset of thermal stratification marked the end
of planktic communities within the respective of the spring diatom bloom and, because surviv-
regions. ing cryptomonads are always vulnerable to devel-
Sommer (1986) undertook such a comparison oping filter-feeding populations of Cladocera, a
of the phytoplankton of the deep lakes in the phase of high water clarity ensues. This ‘vacuum’
European Alps. He ably demonstrated a similar- is filled by a summer assemblage whose composi-
ity in the year-to-year behaviours in individual tion is particularly sensitive to physico-chemical
lakes as well as sequential seasonal stages (spring conditions. Continued bioavailability of phospho-
bloom, summer stratification, summer–autumn rus in the upper water column supports the
phase of mixed-layer deepening) common to all growth of CS-strategist algae such as Pandorina
of them. He suggested that the different phyto- morum (G-group of motile chlorophytes) and/or
plankton assemblages reflected between-lake dif- Anabaena species (H1-group of self-regulating and
ferences in trophic status and the availabilities potentially nitrogen-fixing Cyanobacteria). Even-
of limiting nutrients, and which were also sen- tual phosphorus depletion is generally marked
sitive to between-year variability in individual by increasingly motile (LM -group Ceratium) and
lakes. Sommer reported the periodic patterns mixotrophic (E-group Dinobryon) algae, and the
Table 7.3 Seasonality of dominant phytoplankton in nine lakes of the European Alps, according to
Sommer (1986) but rendered in terms of the trait-differentiated functional groups of Reynolds et al. (2002)
(see also Table 7.1)
Spring Clear Summer Summer Summer Late

bloom water P↑ Si↑ P↓ Si↑ P↓ Si↓ summer Autumn
Königsee A N/Y A
Attersee N Y B
Walensee N N/Y
Vierwaldstättersee B/Y N N H R
Lago Maggiore C/R Y P L/R H L/R
Ammersee C/R Y R/P/Y P L H R/P
Zürichsee C/D/Y Y Y/H P/Y R/P
Lac Léman C/D/Y Y Y/G P L H T/P
Bodensee C/D/Y Y Y/G P L/H/E T/P
development of deep chlorophyll maxima, dom- B grouping. Tabellaria (N) has a long temporal
inated by (R-group) Planktothrix rubescens. On the period (late autumn – spring) in Lago Maggiore
other hand, deeper mixing in summer may sup- (area 213 km2 , mean depth 176 m, TP ≤12 µg P
port the growth of summer diatoms of the N- L−1 ) and Fragilaria and Aulacoseira granulata are
(Tabellaria) and P- (Fragilaria, Aulacoseira granu- frequent throughout the year in Lago di Garda
lata) groups, so long as soluble reactive silicon (area 368 km2 , mean depth 133 m, TP ∼20 µg P
remains available, with the appropriate desmids L−1 ), Lago d’Iseo (area 62 km2 , mean depth 122
being more common if it does not. Indeed, nitro- m, TP ≤68 µg P L−1 ) and Lago di Como (area 146
gen limitation may develop in these instances, km2 , mean depth 154 m, TP ≤38 µg P L−1 ). P-
when H-group nitrogen-fixers (Anabaena, Apha- group desmids (notably Closterium aciculare) and T-
nizomenon) become active. Autumnal mixing in group filamentous forms (Mougeotia spp.) are also
deep lakes enhances the effect of shortening common in these three lakes. Cryptomonas species
days in imposing increasingly severe light limi- (Y group) are common throughout the vegetative
tation, favouring, in some instances, abundance period in all the lakes; X2-group nanoplankters
of (group-T) filamentous chlorophytes and, espe- (Chrysochromulina, Ochromonas, Plagioselmis) occur
cially, Mougeotia. in summer, and E-group Dinobryon, LM -group
The classification broadly holds for the mildly Ceratium–Gomphospaeria and stratifying R-group
alkaline (0.7–2.0 meq L−1 ), subalpine Italian lakes Planktothrix rubescens are prominent in all the
(including Lago di Garda, according to Salmaso lakes during summer. Uroglena species (group U)
(2000), and Lago Maggiore, featured in Table 7.3). occur in Lago di Lugano (area 28 km2 , mean
In a recent synthesis, Salmaso et al. (2003) anal- depth 167 m, TP ≤172 µg P L−1 ), Lago Maggiore
ysed the structure of the phytoplankton in five and Lago di Como.
of these lakes. They used different methods from
Sommer (1986) and, in consequence, subdivided Phytoplankton in selected regions: small
the vegetative season in a different way but the glacial lakes of the English Lake District
regional similarities are remarkable. The main Few lakes have been studied so intensively in
features are summarised in Table 7.4 by reference frequency or so extensively through time as
to the functional groups of Reynolds et al. (2002). those of the Lake District of north-west England.
Diatoms prominent in the early part of the year Time series, starting in some instances in the
in all five lakes include Aulacoseira islandica and 1930s, attesting to variable degrees of change
Asterionella formosa, representing a mesotrophic attributable to eutrophication (Lund, 1970, 1972;
Table 7.4 Summary of phytoplankton seasonality in five deep subalpine lakes, according to Salmaso et al.
(2003) but rendered in terms of the trait-differentiated functional-groups of Reynolds et al. (2002) (see also
Table 7.1)
Late winter to Late spring Early mid Late summer –

mid spring summer mid autumn
Lago di Garda, B, C, P, T, Y Y, P, X2 Y, P, R, T, E, H, Y, P, R, S
Lago d’Iseo X2
Lago di Como B, C, N, P, T, Y Y, P, L, U, X2 Y, P, R, T, E, U, Y, P, R
H, X2
Lago Maggiore, B, C, N, Y Y, P, L, U, X2 Y, P, R, E, U, H, Y, P, R
Lago di Lugano X2
Talling and Heaney, 1988; Kadiri and Reynolds, than silicon or carbon. Nanoplankton is sparse
1993), are now being analysed for sensitivity to (Gorham et al., 1974; Kadiri and Reynolds, 1993).
climatic variation (George, 2002). The lakes them- Picoplanktic Cyanobacteria survive the rarefied
selves are small, glacial ribbon lakes radiating resource availability in the summer epilimnion,
from a central dome of hard, metamorphic slates, where they may form a significant, if not domi-
excavating the valleys of a pre-existing radial nant, fraction of the biomass (Hawley and Whit-
drainage. The natural vegetation is temperate ton, 1991), while Peridinium willei and small num-
Quercus forest but this has been mostly cleared bers of Ceratium may ‘glean’ (sensu the usage
for pasture. The lakes themselves are univer- on p. 327) from deeper water. Desmids of the
sally soft-watered (alkalinity ≤0.4; most are <0.2 genera Cosmarium and Staurodesmus (group N)
meq L−1 ). Variation among their characteristics are also prominent among the otherwise sparse
is dominated by differences in morphometry, ori- microplankton.
entation, hydraulic flushing and catchment load- Derwent Water, Coniston Water and Hawes
ings of nutrients. They were first arranged in Water are classically mesotrophic. Besides the
a series of ascending ‘productivity’ by Pearsall same A-group diatoms, the spring phytoplankton
(1921), which template has been used to illumi- of these lakes may support rather larger quan-
nate comparisons of the solute concentrations, tities of Aulacoseira sub-Arctica, Cyclotella praeteris-
gross planktic photosynthesis, microbial activity sima (both considered typical of Group B), as
and invertebrate associations of the lakes (for fur- well as Asterionella formosa which also appears in
ther details, see Sutcliffe et al., 1982; Kadiri and Group C (Table 7.1). Nanoplanktic species (includ-
Reynolds, 1993). ing Plagioselmis, Chrysochromulina, Ochromonas, rep-
Between-lake differences in the phytoplank- resenting Group X2) are also quite numerous
ton have been investigated and reported by Lund in these lakes and may persist (and flourish)
(1957), Gorham et al. (1974) and Kadiri and for some time after the onset of stratifica-
Reynolds (1993). Based on extrapolations made tion. However, the dominant late-spring/early-
in these works, it is reasonably easy to deduce summer plankters in these lakes are typically
general summaries of phytoplankton periodicity non-motile colonial green algae (Coenochloris fot-
in these lakes (see Table 7.5). Among the olig- tii, Dictyosphaerium pulchellum, Pseudosphaerocystis
otrophic lakes, the supply of MRP rarely exceeds lacustris, Botryococcus braunii, Paulschulzia pseudo-
1 or 2 µg P L−1 and may remain below detection volvox, Radiococcus plantonicus are among the com-
limits for months on end. Chlorophyll concentra- monly observed species) of Group F and/or such
tions are also substantially <5 µg chla L−1 . Group- colonial chrysophytes as Dinobryon divergens, D.
A diatoms, such as Cyclotella comensis, C. radiosa sertularia, D. bavaricum, D. cylindrica, Mallomonas
and Urosolenia eriensis, exhaust phosphorus rather caudata and Synura uvella representing group
Table 7.5 Summary of phytoplankton seasonality in 19 stratifying lakes in the English Lake District,
according to Kadiri and Reynolds (1993), rendered in terms of the trait-differentiated functional-groups of
Reynolds et al. (2002) (see also Table 7.1)
TP, c.1991
Class Lake Area (km2 ) (µg P L−1 ) Phytoplankton
Larger oligotrophic Wast Water, 0.9–3.0 3–9 A → Z/LO → N
lakes Ennerdale Water,
Thirlmere,
Crummock Water,
Buttermere
Larger mesotrophic Derwent Water, 3.9–5.4 7–11 B(C) → X2/F/E →
lakes Hawes Water, X3/Z/LO /Y → N/R
Coniston Water
Well-flushed Brothers Water, Rydal 0.2–5.3 4–33 B or C → X1/X2 →
short-retention Water, Grasmere, Y (E, F, H2, P)
lakes Bassenthwaite Lake
Eutrophied larger lakes Windermere (North), 0.6–9.0 14–40 C(B)
Ullswater, →X1/X2/Y/G→
Windermere H/LM /S/T → P
(South), Esthwaite
Water, Lowes
Water
Enriched small lakes Loughrigg Tarn, ≤0.1 20–45 C/Y → X1/X2 /E/F
Blelham Tarn or H1 → LM → P/S
E. In summer, the biomass generally falls to the phytoplankton stimulated by late-summer

a phosphorus-depleted minimum during which and autumnal mixing. Phytoplankton biomass is
the nanoplankton diversifies with algae such as faintly diacmic in these lakes, the spring peak
Chrysococcus, Monochrysis, Pseudopedinella, Bicosoeca (in the order of 6–15 µg chla L−1 ) usually being
tubiformis and Koliella longiseta recruiting (from the larger. The level of MRP generally falls below
X3). Picoplankton may also be numerous, some of detection limits for between four (May–August)
which may well be Chroococcoid (Z) but some is and nine (March–November) months of the year.
eukaryotic (including Chlorella minutissima, which The summary notation shown in Table 7.5 for
Reynolds et al. (2002) placed in X3. These lakes ‘larger mesotrophic lakes’ also fits well to the pat-
have the potential to support metalimnetic max- terns among the greatest Scottish lochs (Bailey-
ima of (group-R) Planktothrix, although the num- Watts and Duncan, 1981).
bers of P. mougeotii produced are generally small. Several further lakes are, to varying extents,
Cyanobacteria are mainly represented by mod- enriched beyond the mesotrophic state. Since the
est growths of Anabaena solitaria and A. lemmer- mid-1960s, Windermere and Ullswater have been
mannii (of group H2) and by colonies of Woroni- subject to the discharge of effluents from sec-
chinia (formerly Gomphosphaeria) that, together ondary sewage-treatment works. Esthwaite Water
with dinoflagellates Peridinium willei, P. incon- was already a eutrophic lake before sewage treat-
spicuum and Ceratium species, represent group LO . ment affected its waters. Lowes Water has a
Tabellaria flocculosa and Cosmarium species (includ- low human population in its catchment and
ing C. abbreviatum and C. contractum) make up the reason for its enrichment is not resolved.
All these lakes can support significant num- South Basin has returned to being dominated by
bers of (H1-group) Anabaena flos-aquae, Aphan- P-group diatoms (especially Fragilaria crotonensis)
izomenon flos-aquae and, in the case of Esthwaite and desmids (Staurastrum pingue)
Water, LM -group Microcystis aeruginosa (together, Before 1992, peak chlorophyll concentrations
the bloom-forming Cyanobacteria that have done in the North Basin had come close to 20 µg
locally much to give eutrophication its bad chla L−1 , in the enriched South Basin, Tychonema
name). Each may also support an abundant populations had forced a summer maximum of
nanoplankton in late spring, that may include up to 45 µg chla L−1 and a distinctly diacmic
species of Ankyra, Chlorella, Chlamydomonas, Mono- annual pattern of phytoplankton abundance.
raphidium, Mallomonas akrokomos (of Group X1) Subsequently, phosphorus load reductions have
in addition to X2 representatives, at least until restored the earlier pattern of a spring maximum
grazed down by filter-feeders. followed by irregular, smaller summer peaks, all
The phytoplankton response to enrichment constrained within the supportive capacity of the
(and to the post-1992 restoration) of Windermere biologically available phosphorus. The illustra-
has been reviewed and summarised in Reynolds tion in Fig. 7.6 shows the periodicity of chloro-
and Irish (2000). Between 1965 and 1991, the phyll concentration in North Basin during 1978.
winter maximum of MRP rose from about 2 to Table 7.5 carries entries in respect of two
about 8 µg P L−1 in the larger North Basin and smaller Lake District lakes that carry effects
from 2.5 to almost 30 µg P L−1 in the South of recent changes in agricultural and domes-
Basin. The greater resource has permitted the tic P loadings. In Blelham Tarn, a long-standing
vernal diatom growth to escape the rate con- basic B → E/F → LM → N sequence has been
trol imposed by phosphorus and to ascend to the steadily altered by the diminution of the con-
silicon-determined maximum up to one month tributions of Aulacoseira subarctica, Dinobryon, Cer-
earlier (Reynolds, 1997b) (see also Section 5.5.2 atium and Tabellaria flocculosa in favour of Asteri-
and Fig. 5.13). More than that, from the early onella and Stephanodiscus minutulus (more strongly
1980s, vernal growth in the South Basin often C), several species of Anabaena (H1) species and
failed to exhaust the MRP in the lake, leav- a near year-round abundance of Planktothrix
ing resource to support enhanced early summer agardhii (S1).
production. Grazing by filter-feeders and carbon Finally, several of the lakes have extensive
dioxide depletion in the unbuffered water biasses catchment areas in relation to lake volume and,
the outcome in favour of the larger, efficient in this area of high annual aggregate precipita-
carbon-concentrating bloom-forming Cyanobac- tion (between 1.5 and 4 m annually), are liable
teria. Inroads into the DIN stocks has fur- to fairly frequent episodes of significant flush-
ther favoured (H1-group) Anabaena species, with, ing. In the case of Grasmere (mean retention
appropriately, high incidences of nitrogen-fixing time 24 d, range of instantaneous rates 5–2000
heterocysts. However, the most successful benefi- d), flushing has helped to dissipate the effects of
ciary of the eutrophication of the South Basin nutrient loads from sewage works commissioned
of Windermere has been the solitary, filamen- in 1969. Thitherto, the lake supported a simi-
tous, group-S1 oscillatorian Tychonema bourrellyi. lar assemblage to that of the contemporaneous
This particular species lacks gas vesicles and Blelham Tarn, save that the slow-growing algae
does slowly sink out to the bottom of the lake of the LM group were (and remain) poorly rep-
(thus exporting more oxygen-demanding reduced resented (Reynolds and Lund, 1988). The reason
carbon to depth where significant anoxia trig- for this is not that the algae have difficulty in
gered changes in the lake’s metabolism: Heaney growing against the instantaneous rates of flush-
et al., 1996). One mark of the success of the ing but that the autumn flooding is so effective
programme of tertiary treatment of Winder- in removing pre-encystment vegetative stocks. In
mere’s main sewage inputs has been the near- an effort to prevent massive algal growth in dry
elimination of Tychonema from the lake. The summer weather, the arrangements for effluent
late-summer phytoplankton in Windermere’s disposal were altered in 1982 so that the treated
liquor was piped to the hypolimnion directly. the oligotrophic tarns and pools in Grasmere’s
This admirable interim solution ‘locked’ the sum- catchment. With minimal grazing pressure, large
mer phosphorus load in store until the autum- populations of nanoplankton (over 105 cells mL−1
nal breakdown of stratification and the onset of in some instances: Reynolds and Lund, 1988, and
rapid hydraulic throughput, washing the phos- authors unpublished observations) develop in
phorus harmlessly from the lake. ‘Harmless’ to the physico-chemically favourable environment.
Grasmere, that is, for the phosphorus is moved These soon support correspondingly large pop-
through a second, short-retention lake (Rydal ulations of ciliates and rotifers. A quasi-stable
Water) to become part of the load to the relatively nanoplanktic biomass of about 0.5 mg C L−1
long-retention Windermere. ‘Harmlessness’ must and a similar biomass of microplanktic con-
also be judged in a temporal context: a chronic sumers make an unusual sight for a limnologist!
problem arising from infiltration by urban run- Such associations are, however, quite transient.
off during wet weather creates large volumes of A generation or two of Daphnia galeata recruit-
dilute sewage, which is impossible to store pend- ment is eventually capable of clearing the entire
ing any kind of treatment. It is normal practice nanoplanktic resource base from the water (and
in urban sewage works discharging to rivers to a lot of the ciliates too!).
allow incompletely treated effluent into storm One further point of interest that emerges
flows where the biological oxidation of resid- from Table 7.5 concerns the larger cryptomonads
ual organic carbon is completed naturally. In of group Y. Because the ubiquity of such com-
Grasmere, this residual carbon was being piped mon species of Cryptomonas as C. ovata, C. erosa
directly to, and collected in, the hypolimnion. and C. marssoni, there is a tendency to overlook
What happened, of course, was a shift in deep- their value in comparing assemblages. They also
water metabolism, hypolimetic anoxia and low share features of each of the three, primary C,
redox. As this is being written, a further upgrad- R and S strategies in being colonist, in adapt-
ing of the sewage-treatment works is in hand. It ing well to low insolation and in their ability
could be argued that the real solution (though to constitute deep monospecific plate-like lay-
expensive and disruptive) would be to re-sewer ers in the metalimnion of stratified lakes. How-
our towns so that foul- and surface-drainage are ever, their weakness is to be highly susceptible
kept completely separate, with only the former to grazing, by cladocerans, calanoids and cer-
needing to be submitted to treatment. tain rotifers (Reynolds, 1995a). In the English
In the meantime, the open water of Gras- Lake District, their numbers are broadly propor-
mere represents a highly variable environment tional to trophic state: they are sparse in the olig-
for phytoplankton. Flushing episodes, especially otrophic lakes (<10 mL−1 ); collectively, they may
during winter, are extremely effective at remov- achieve 10–100 mL−1 in the mesotrophic lakes
ing existing algal stocks from the water (ben- and 100–2000 mL−1 in the eutrophic examples.
thic propagules are substantially spared). They Moreover, the periods of their abundance occur
also deplete the crustacean zooplankton and it progressively earlier in the year with increasing
sometimes takes many months for a significant trophic state. This may be due to the interaction
feeding pressure to be recovered. What tends of growth potential and the nature and tempo-
to happen after a wet winter or spring is that ral phasing of phagotrophy: more nutrients nur-
the water column is repopulated from meagre ture the development of larger populations that
residual stocks of fast-growing algae (which cer- are detectable sooner (but are also exploited by
tainly include C and P groups of diatoms, X1- zooplankton earlier). An alternative view is that
group nanoplankters and, in Grasmere, Dinobryon the CR qualities of cryptomonad survival strate-
spp. which develop striking populations of many- gies are expressed more strongly than the SR
celled colonies). Typically, however, X2 and even traits with increasing trophic state. Among the
X3 nanoplankters are prominent first. Like other richer Lake District lakes, a small dinoflagellate,
rarities (chrysophytes, desmids), some of these formerly recorded as Glenodinium sp. (but now
are believed to be brought in by flood waters from recognised as Peridinium lomnickii) is common and
its dynamics coincide sufficiently closely with considered, Carrilaufquen Grande (area 16 km2 ,
those of the spring Cryptomonas that it has been mean depth 3 m, TP 298 µg P L−1 ) and Carri-
included in the spring Y-association (Table 7.1). laufquen Chica (area 5 km2 , mean depth 2 m,
TP 69 µg P L−1 ) were these algae dominant and
Phytoplankton in selected regions: glacial actively fixing nitrogen (Diaz et al., 2000).
lakes of Araucania Elsewhere, the lakes are steadfastly olig-
Thirty-six north Patagonian lakes, situated on otrophic or slightly mesotrophic in character.
either side of the Andean Cordillera between It has to be recognised that the character is
the latitudes 39◦ and 42◦ S, were the subject of a independent of the most remarkable feature of
careful survey carried out over 40 years ago by the region, which is the behaviour of west–east
Thomasson (1963). Though not as detailed as passing rain-bearing airstreams. Within a dis-
some of the other reports featured in this chap- tance of 50–70 km from the Pacific seaboard, the
ter, its inclusion is urged because the striking land rises to the crest of the granitic Andean
oligotrophy of the lakes in this region is gener- Cordillera (average altitude about 2000 m, with
ally regulated by nitrogen availability (Soto et al., peaks – several of which are volcanoes – of up
1994; Diaz and Pedrozo, 1996). Indeed, the molec- to 3800 m a.s.l.). Progressing a further 60–100
ular N : P ratios to be inferred from the data of km eastwards, the range falls to the level of
Diaz et al. (2000) on DIN and TP levels measured the Argentinian plateau, at an altitude of ∼1000
in samples from several lakes on the eastern m a.s.l. Precipitation in the mountains amounts
(Argentinian) side of the Andes range between to ∼4 m annually but tapers off abruptly to
0.4 and 1.0 (far below 16, reckoned to repre- barely 50 mm. The region is little affected by
sent parity). Yet more interesting is the fact that human settlement and a natural vegetation per-
maximum phytoplankton biomass observed in a sists over much of the area, in distinct bands
number of these lakes correlates well with nitro- corresponding to altitude and rainfall. The high-
gen concentration while it is even saturated by est rainforests are dominated by Fitzroya. These
MRP, excess of which remains measurable in lake give way to Australocedrus–Nothofagus woodlands.
water. From data in Diaz et al. (2000), it is clear Further to the east, this thins steadily, merg-
that in lakes where TP exceeded 11 µg P l−1 (most ing into Agrostis–Cortaderia grasslands (pampas)
values falling between 4 and 8 µg P l−1 ), the levels and, within 100 km from the Cordillera, to
of nitrate + ammonium nitrogen were relatively Festuca–Mulinum. This is surely one of the world’s
low. Almost everywhere, DIN levels were <30 µg most remarkable climatic ecotones. Our interest
N L−1 , generally, they were <14 µg N L−1 and, in is that most of the drainage to the lakes emanates
some cases, <3 µg N L−1 (i.e. in the range 0.2–1 from the mountains, flowing westwards in short
µM in which most phytoplankton are expected to rivers to the Pacific Ocean or eastwards into the
experience difficulties harvesting sufficient nitro- Rio Negro catchment that opens to the Atlantic.
gen to support further growth; see Section 5.4.4). The waters in the Araucanian lakes are thus
What is of particular interest is that the spare almost uniformly dilute in salts, low in alka-
phosphorus capacity seems not to be switched linity and weak in nutrients. The lake waters
to the support of nitrogen-fixing Cyanobacteria. are extremely clear (for photosynthetically active
Either the phosphorus is still too low (cf. Stewart wavelengths, εmin is between 0.12 to 0.2 m−1 ),
and Alexander, 1971), or there is a critical defi- except those charged with fine material emanat-
ciency in molybdenum, vanadium or iron (Rueter ing from glacial melt or from the dust of recent
and Petersen, 1987), or the energy thresholds for local volcanic eruptions.
nitrogen fixation are unsatisfied (Paerl, 1988) (see The phytoplankton assemblages represented
Section 4.4.3). Anabaena species are recorded in among these lakes also show a high degree of
several of the lakes and Anabaena solitaria (of the mutual similarity. According to Thomasson’s
‘mesotrophic’ H2 group) occurs widely among (1963) survey and the later information of Diaz et
the region’s lakes. However, only in the small al. (2000), the most ubiquitous species recorded
steppe lakes, on the eastern fringes of the region have been the diatoms Urosolenia eriensis (group
A) and Aulacoseira granulata (group P; Thomasson also noted the relative abundance of Aphanocapsa
also distinguished ‘Melosira hustedtii’), dinoflagel- species, with Staurastrum species and Aulacoseira
lates of the genus Gymnodinium and Peridinium in Lago Lacar (area 52 km2 ).
(variously including P. willei, P. bipes, P. volzii, It seems probable that, overall, the phy-
P. inconspicuum; presumed to be group LO ), the toplankton of the Araucanian lakes, with its
chrysophyte Dinobryon divergens (group E) and the basic A/P → E/LO → N or embellished A/P
desmid Staurodesmus triangularis (group N). No →X2/F/E/LO → N/P and A/P →E/LO /H2 →
information is to hand on the abundance and N/T sequences, conforms to the model of pat-
composition of the picoplankton but a presence tern for ultraoligotrophic to mesotrophic lakes.
as part of a developed microbial food web may In this context, it is interesting that ‘oligotrophic’
be inferred from Thomasson’s (1963) lists of is just as ‘oligotrophic’, regardless of whether
planktic ciliates and crustaceans (in which the nitrogen or phosphorus is the main constrain-
centropagid calanoid Boeckella gracilipes is a ing factor. The assemblages are, in reality, typi-
prominent component). In several lakes, such cal of low-alkalinity systems, although some sup-
as Llanquihue (area 851 km2 ), Ranco (408 km2 ) posedly more ‘eutrophic’ species, thought to be
and Todos Los Santos (181 km2 ) on the western poorly tolerant of low phosphorus conditions
side (Thomasson, 1963) and Traful (area 75 km2 , (not least Aulacoseira granulata, Fragilaria crotonen-
maximum depth >100 m, TP 8 µg P L−1 , DIN ∼3 sis and Anabaena spp.), are here able to function
µg N L−1 ) and Espejo (area 38 km2 , maxinum adequately.
depth >100 m, TP 8 µg P L−1 , DIN ∼11 µg N
L−1 ) to the east (Diaz et al., 2000), these are also Phytoplankton of small kataglacial lakes
the principal species. So far as it is possible to Wherever glaciers are, or have been in the
deduce from limited sampling frequencies, max- last 100 ka, active in eroding terrestrial land-
imum biomass occurs in a single summer ‘peak’, forms, there are usually to be found signif-
rarely achieving as much as 1 µg chla L−1 . An icant ‘downstream’ deposits of directly trans-
ubiquitous nanoplanktic (X2 group) component ported material, abandoned by the wasting
(Plagioselmis, Chrysochromulina) is weakly expressed glaciers (hence ‘kataglacial’). Associated forma-
in these ultraoligotrophic systems. Elsewhere, tions caused by ‘freeze-thaw’ cycles and solifluc-
the same species may achieve larger standing tion in the vicinity of ice fields are known
crops and where which the presence of other collectively as periglacial deposits. Depending
species may be more evident. In Lago Villarrica upon their age, morphometry and the contempo-
(area 851 km2 ), Thomasson (1963) noted a signifi- rary climatic conditions, these deposits may well
cant (F group) representation of Kirchneriella and enclose small lake basins. In the wake of the most
Dictyosphaerium species. The annual cycle in Lago recent glacial period that ended about 10.5 ka
Nauel Huapi (area 556 km2 , maximum depth ago (known as the Devensian in northern Europe,
464 m, mean depth 157 m, TP 11 µg P L−1 , DIN the Weichselian in the European Alps and the
∼10 µg N L−1 ) moves from a spring assemblage Wisconsinian in North America), vast terminal-
of Aulacoseira–Urosolenia–Dictyosphaerium, through moraine systems were deposited along the south-
Dinobryon–Dictyosphaerium–Urosolenia in sum- ern limits of the ice sheets. These abandoned
mer to a diatom (Aulacoseira, Urosolenia–Synedra ‘tidemarks’ of characteristically hummocky land-
ulna–Tabellaria)-dominated autumn plankton scapes are sometimes called moraine belts. These
(Thomasson, 1963). The (summer) chlorophyll kataglacial moraines, peppered with generally
maximum is still <2 µg chla L−1 (Diaz et al., small lakes, make up distinctive landscapes in the
2000). Several of the smaller lakes support an vicinity of Riding Mountain, Manitoba and across
array of desmids (Cosmarium, Staurastrum spp.), as northern Minnesota and Wisconsin. In Europe,
well as Fragilaria crotonensis (P) and Mougeotia (T). major moraine belts sweep through Jylland (Jut-
Anabaena solitaria (H2) also occurs quite widely; land), Holstein, Pomerania and Mazuria, and it
Thomasson recorded it as being dominant in is the morainic system striking north-eastwards
autumn in Lago Correntoso (area 26 km2 ); he across northern Russia that separates the Baltic
and Black Sea watersheds in that country. The texts show standing surface waters wherever the
largest moraines were formed during periods topographic contours dip below the level of the
of ice-front stagnation, when and where glacier local water table and, thus, merely its surface
recruitment and glacier melt were, for a time, manifestation. Hydrological studies of these drift
in approximate equilibrium. In front of them hollows confound this simplicity as annual lake
stand massive outwash plains of fluvioglacial levels fluctuate less than does the height of the
deposits. Behind them are the lands smoothed water table. Reynolds (1979b) proposed a model
by the advancing ice and plastered with com- in which lake basins were partially sealed by their
pacted, ground-down boulder clay, or till, which own deposits and that ground water entered by
may be only lightly and locally covered with ‘inspilling’ from a high water table. Lake water
later fluvioglacial deposits. Less expansive but ‘overflows’ normally leak away by seepage. When
wholly analogous structures abound in the low- the water table dropped, hydraulic exchanges
land outfalls of valley glaciers and ice sheets in were more or less confined to precipitation and
other parts of the world. Those with which I am evaporation from the lake surface. The fact that,
most familiar are located in the English north- on millennial timescales, basins have varied in
west midlands, especially the Wrexham–Bar Hill their trophic status according to the wetness
moraine (Reynolds, 1979b). of past climates, many becoming ‘terrestrialised’
The wane of the Pleistocene ice sheets into fens or peat bogs (Tallis, 1973), also demands
was never smooth but occurred in phases of a model of restricted basin permeability.
rapid retreat, alternating with phases of stag- Notwithstanding, much of the water supplied
nation or even readvance (West, 1977). Lesser to basins in drift has percolated through sub-
morainic features are widespread among areas soils and fluvioglacial deposits that are gener-
of kataglacial deposition. The variety of struc- ally finely divided, offering maximum contact
tures in drift material (which term embraces opportunities for the solution and leaching of
all glacial deposits) is increased by proximity to salts. In this way, the chemical composition of
solid geological features as well as in the for- lake water is strongly influenced by the local
mations themselves, such as kames (ice-deposited drifts. This does not mean that drift lakes are
mounds), eskers (englacial stream beds) and pin- necessarily rich in nutrients (many are not: Stech-
gos (periglacial frost heaves in drift outside the linsee is one that is quite nutrient-poor; see p.
ice fronts). Lake basins in drifts are just as varied 326) but many certainly are. Moderately to very
in the detail of their origins. Some are moraine- calcareous lakes are frequent among kataglacial
dammed gaps between drift hummocks and are series (Ca 0.4–4 meq L−1 , bicarbonate alkalinities
irregular or linear in outline; others are more ≤3.5 meq L−1 ), though they may be rare locally,
rounded with excentric underwater contours, depending upon the provenance of the drift. Sim-
corresponding to kettle-holes formed initially by ilarly, with nutrients, the dissolved silicon and
the melting of detatched blocks of ice. orthophosphate contents of unmodified ground-
However, lakes in kataglacial drifts do have water are variable from catchment to catchment
several generic and crucial common attributes. but may easily reach 7–8 mg Si L−1 and >300
The unconsolidated deposits in which they are µg P L−1 in some instances. Other major ions
formed are generally porous, so that present-day may be similarly enriched in particular systems.
precipitation percolates into the drift rather than Unless the topographical catchments have been
runs over the topographical surface. The water subject to considerable anthropogenic modifica-
collects in a zone of saturation above the basal tion, there is little scope for the enrichment of
till, which is relatively impervious. The ‘surface’ nitrogen, save through the same surface evapo-
of the saturated zone is called the phreatic sur- rative processes that affect all precipitation. This
face, or water table, and, again, represents a sort leads to another general attribute of morainic
of contemporary equilbrium between its recruit- and drift lakes, that the ratios of biologically
ment by percolation and the sluggish horizon- exploitable contents of nitrogen to phosphorus
tal permeation into regional catchments. Older and to silicon are lower or much lower than in
surface-fed water bodies. The capacity of the Transparency is reduced to 2–3 m. Silicon is gen-
nitrogen to support algal growth may be less erally heavily drawn down (from ∼2.5 to <0.1 mg
than that of phosphorus and algal growth rates Si L−1 ) while Monoraphidium, other green algae
in the field are, potentially, more likely to be lim- and the Y-group cryptophyte Chroomonas persist
ited by the external supply of nitrogen than by in the epilimnion (7–10 m in thickness), pend-
the phosphorus available. ing the exhaustion of DIN to <5 µg N L−1 .
The phytoplankton of a number of ground However, available phosphorus remains freely
water-fed drift lakes has been described in some available (SRP >150 µg P L−1 ) and, not surpris-
detail. Sjön Erken in Sweden (area 24 km2 , mean ingly, nitrogen-fixing H1-group Anabaena species
depth 9.0 m, TP 25–50 µg P L−1 ) has been observed (A. flos-aquae, A. planctonica, A. spiroides) dominate
regularly for many years. Rodhe et al. (1958) and through the summer. In autumn, Asterionella,
Nauwerck (1963) noted the April–May develop- Fragilaria and other diatoms dominate the declin-
ment of a spring ‘bloom’ (maximum ∼20 µg ing biomass. Epilimnetic pH in summer rises to
chla L−1 ) of flagellates (including Cryptomonas pH 8.8 or a little higher, with carbonate precipita-
spp., Plagioselmis, Chrysochromulina and Dinobryon tion. The X1 → C/Y → H1 → C/P sequence owes
divergens), starting under ice cover but soon much to the diminution of nitrogen and carbon
giving place to dominant diatoms (including levels during the vegetation season.
‘large’ Stephanodiscus rotula and ‘small’ Stephano- The drift lakes around Plön in North Ger-
discus hantzschii var. pusillus and Asterionella form- many also became classic sites in limnology,
osa). Green algae (including Eudorina and Pan- especially through the studies of August Thiene-
dorina spp.) and Cyanobacteria appeared in the mann. He distinguished their ‘Baltic’ biotic
early part of summer (Anabaena flos-aquae, Aphan- assemblages – of benthos as well as of plank-
izomenon flos-aquae and, especially, Gloeotrichia ech- ton – from those (‘Caledonian’ assemblages) of
inulata) to be replaced by Gomphosphaeria (now European mountain lakes. The former included
Woronichinia) and Ceratium hirundinella, building what we now refer to eutrophic diatoms (C,
to an August maximum of 30–35 µg chla L−1 D, P associations), cryptomonads, Cyanobacte-
(see also Fig. 7.6). With early autumnal mixing, ria and dinoflagellates. Sommer’s (1988c) exper-
Fragilaria and Staurastrum species became promi- imental investigation of Schöhsee (area 0.79
nent. Now, nearly 50 years later (D. Pierson, per- km2 , mean depth ∼9 m) demonstrated not just
sonal communication), the same floral elements the seasonal progression but, through the use
(C/D/X2/Y/E → G/H → LO → P) are encoun- of a series of enrichment bioassays, also the
tered but the dominance has changed slightly nutrients most severely constraining contempo-
in favour of Stephanodiscus hantzschii var. pusilla, raneous growth capacity. Vernal diatoms (Asteri-
nanoplanktic flagellates, summer Gloeotrichia and onella formosa, Synedra acus, Diatoma elongatum,
late-summer Asterionella in a sequence closer to Stephanodiscus rotula and S. minutula) were, in
(X2/D → H2 → LO → C). most instances Si-limited. The supportive capac-
Esrum Sø (area 17.3 km2 , mean depth 12 m) ity for Cyanobacteria (Anabaena flos-aquae) was
is a large kettle-hole in calcareous sandy moraine often P-limited, while summer dinoflagellate pop-
in Sjælland, Denmark which has, like Erken lake, ulations were sensitive to limiting availabili-
a long history as a focus for detailed studies ties of nitrogen (Ceratium) or phosphorus (Peri-
(recently reviewed, in part, by Jónasson, 2003). dinium cinctum, P. umbonatum, P. inconspicuum). The
Its phytoplankton, as described by Jónasson and growth of (group-F) colonial chlorophytes, (group-
Kristiansen (1967) develops during March under a E) chrysophytes (Dinobryon spp.), cryptomonads
thinning ice cover and dominated by nanoplank- (Cyptomonas ovata, Plagioselmis nanoplanctica) and
tic Ankistrodesmus falcatus, now known as Mono- other nanoplankters (Ankyra judayi, Chrysochromu-
raphidium contortum and ascribed by Reynolds et lina parva) was rarely constrained by nutrients
al. (2002) to group X1. After ice break, domi- but these would, perhaps, have experienced con-
nance of the April spring peak passes to diatoms trol through the carbon supply or through graz-
(predominantly B/C-group Asterionella formosa). ers or both. The summarised annual sequence
(C/D/Y → X1/X2/Y/E → H1/F → LM ) corre- may persist beyond spring (rarely resorting to
sponds to the drawdown of silicon, then nitrogen heterocyst production and nitrogen fixation). Cer-
and/or phosphorus, but always against a back- atium or Microcystis usually dominates the sum-
ground of frequent carbon dioxide deficiencies. mer biomass (though not, it seems, as a func-
In the small calcareous meres of the English tion of nutrient availability but of recruitment
north-west midlands, phosphorus availability is success: Reynolds and Bellinger, 1992). P-group
generally moderate to high, owing to a relative diatoms or even S1-group Planktothrix agardhii
abundance in the drift of minerals derived from have dominated in windy summers and, in
the underlying Triassic marls and evaporites. The one recent case, J-group Scenedesmus took over
oceanic climate to which these lakes are sub- a very shallow epilimnion during an unusually
ject ensures that they are generally ice-free in calm summer (Reynolds and Bellinger, 1992). J-
winter (warm monomictic lakes). In Crose Mere group Scenedesmus, Pediastrum and other chloro-
(area 0.15 km2 , mean depth 4.8 m, alkalinity coccalean algae are abundant in meres that have
3.2 meq L−1 , SRP ≤ 200 µg P L−1 ; DIN generally experienced recent enrichment from nitrogen-
≤2 mg N L−1 ), the phytoplankton often achieves rich agricultural fertiliser run-off while some
high biomass (150–250 µg chla L−1 in summer) low-alkalinity meres in sandy drift support more
in a distinct, diacmic seasonal pattern (see Fig. green algae of the F group (especially Botryococcus
7.6). The basic periodic sequence of the phyto- in Oak Mere: Reynolds, 1979b).
plankton (C/Y → G → H1 → LM → P) begins Finally, it may be added that the summary C
with a February–March maximum of Asterionella, → H1 → LM → P applies to Mazurian moraine
Stephanodiscus rotula and Cryptomonas ovata (which lakes such as Mikolajskie (Kajak et al., 1972) and
may, but more usually does not, exhaust the North American prairie lakes (Lin, 1972; Kling,
silicon). The onset of thermal stratification (in 1975).
late April–early May, to within 2–6 m of the sur-
face) leads to the rapid settlement of the diatoms Phytoplankton of large, low-latitude lakes
(see also Fig. 6.2), while surviving nanoplank- The phytoplankton of the great lakes of Africa
ters and cryptomonads succumb to intensifying differs markedly from that of the high-latitude
grazing rates. Eudorina unicocca, the next dom- examples considered above, but in ways that are
inant, mostly escapes grazing but depletes the more easily appreciated in the light of knowl-
DIN. By late June, nitrogen-fixing Anabaena circi- edge based on smaller lakes. The lakes themselves
nalis and/or Aphanizomenon flos-aquae are propon- differ in character, though most of the African
derent but Ceratium hirundinella, together with examples to be mentioned here are of tectonic
Microcystis aeruginosa, trawl into the metalimnion origin, being aligned in the opening and bifur-
for the nutrients to sustain the major biomass cating rift across East Africa. Its southern arm
peak of the year. Autumn mixing (or earlier sum- encloses Lake Malawi; a string of lakes traces the
mer storms) promotes renewed phases of diatom western section, including Lac Tanganyika, Lac
abundance (Fragilaria crotonensis, Aulacoseira granu- Kivu and, at the northern end, N’zigi (formerly
lata plus the desmid Closterium aciculare). For full Lake Albert and flushed by the upper reaches of
details, see Reynolds (1973c, 1976a). the White Nile). The eastern rift valley contains
There is some interannual variability about Turkana (formerly Lake Rudolf) as well as several
this sequence (see especially Reynolds and smaller lakes. In the plateau between the eastern
Reynolds, 1985) but a core C → H1 → LM → and western arms is a water-filled depression of
P is common throughout the deeper lakes of a quite different character – the gigantic saucer
the series (shallower lakes have less Ceratium and, that is Lake Victoria whose origin is not tectonic
sometimes, more Microcystis). In Rostherne Mere and, it is believed, quite recent.
(area 0.49 km2 , mean depth 13.4 m, SRP ≤ 300 The rift valley lakes are ancient (possibly up to
µg P L−1 , maximum DIN generally 1.5–2 mg N 20 Ma: Coulter, 1994), deep and, because of the
L−1 ), depth and turbidity suppress a spring bloom climate (‘endless summer’: Kilham and Kilham,
(unless stratification is delayed). H1 dominance 1990) at their respective latitudes, almost never
experience full convectional overturn. Malawi, populations (3–6 × 10−6 L−1 ) and abundant
Tanganyika, Kivu and Turkana are meromictic ciliates (especially Strombidium sp.) are indica-
lakes: each is vertically segregated into a peren- tive of an active microbial food web (Hecky
nially stagnant, lower monimolimnion and an and Fee, 1981; Hecky et al., 1991; Plisnier and
upper mixolimnion. The monimolimnion retains Coenens, 2001). As the windy season abates, algal
most of the nutrients in the system but is biomass increases at the northern end, with the
severely energy-limited; the mixolimnion is fre- development of Anabaena blooms, while pulsed
quently nutrient-deficient (Hecky et al., 1991). upwellings at either end deliver the nutrients
Even so, there is strong phytoplankton seasonal- to sustain the main phase of diatom growth
ity in these lakes which relates to variability in (Nitzschia spp. and Aulacoseira granulata are promi-
the depth of the mixolimnion through the year. nent), to maximal concentrations of ∼7 µg chla
Entrainment of deeper water during the period L−1 (Langenberg et al., 2002). However, as the
of increased mixing is extremely important to seiching weakens and the mixolimnetic nutrient
the recycling and reuse of some of the system’s base is dissipated in fish production and the sedi-
accumulated resources. Otherwise, new produc- mentary flux, it is green algae such as Coenochloris
tion in the pelagic relies on the supply of new and Oocystis that persist longest through February
resources (Hecky and Kling, 1981). to April (Hecky and Kling, 1981; Talling, 1986).
At a quoted maximum depth of 1471 m (Her- From the mixing to final relaxation and the next
dendorf, 1982), Lac Tanganyika is the second cycle, phytoplankton composition may be sum-
deepest lake on the planet (area 32 900 km2 , marised as P → F → Z (with the advantage to P
mean depth 574 m). Between May and Septem- perhaps alternating with one in favour of H with
ber, south-easterly winds are funnelled up the val- the seiche cycle).
ley and drive surface water (temperature ∼27 ◦ C) In Lake Malawi (area 22 490 km2 , mean depth
to the north, where the pycnocline may be 276 m), seasonal variations in the depth of the
depressed to a depth of ∼70 m (data of Plisnier mixolimnion range from ∼70 m in the cooler,
and Coenens, 2001). When the south-east winds windier part of the year (June to September),
stop, the tilted surface of the monimolimnion when its temperature falls to ≥24 ◦ C (barely more
continues to oscillate (or seiches) for several than 1◦ warmer than the monimolimnion), to
months until the stability is recovered (gener- <30 m during the hot, calmer months between
ally by February). Monimolimnetic water may October and March (temperature >27 ◦ C). The
be sheared off into the mixolimnion during the major nutrients occur at low concentrations,
windy months, while its simultaneous upwelling except at depth (>200 m, in the anoxic moni-
at the southern end (and also at the northern molimnion). Mixolimnetic concentrations are at
end during the period of oscillation) augments their highest during the mixed periods (SRP ∼5
the process. Between February and April, the µg L−1 , DIN ∼20 µg L−1 : data of Irvine et al., 2001)
only source of nutrients to mixolimnetic primary but these are soon drawn down after the winds
production is external. The mean mixolimnetic ease. Irvine et al. (2001) observed that planktic
nutrient concentrations (0–30 m) are higher (DIN chlorophyll concentrations are typically <2 µg
70 µg L−1 , SRP 50 µg L−1 ) in the period of wind chla L−1 with higher levels (≤7 µg chla L−1 )
shearing and upwelling than in the calm period coming during July. Diatoms are prevalent dur-
(DIN 50 µg L−1 , SRP 20 µg L−1 ), even though ing the windier months (Aulacoseira granulata,
this is also the wet season of enhanced lake Cyclostephanos sp.), with desmid species of Stauras-
inflows. trum and Closterium). Nitrogen-fixing Cyanobacte-
Phytoplankton biomass increases gradually ria Anabaena and Cylindrospermopsis species make
between May and August and progresses from up 50% of the planktic biomass in the months
south to north (Hecky and Kling, 1981), from of stable stratification (Allison et al., 1996; Irvine
the equivalent of <1 to ∼2 µg chla L−1 (data of et al., 2001). These authors give no specific infor-
Langenberg et al., 2002). Picoplanktic Cyanobac- mation on either pico- or nanophytoplankton,
teria contribute a large (>50%) proportion of the although the latter, at least, are likely to be rel-
particulate primary production, while bacterial atively most abundant during the early stages
of the intensification of the thermal stratifica- primary production and a doubling of the verti-
tion. Pending confirmation of this, the phyto- cal coefficient of light attenuation since Talling’s
plankton sequences are best summarised as (1965) observations in the 1960s (Mugidde, 1993).
P → H1 (SN ). The hypolimnion, which was previously aerated
The phytoplankton of Lac Kivu (area 2370 to the deepest sediment through most of the
km2 , mean depth 240 m, TP ≤55 µg P L−1 ) shows year, is now regularly anoxic. These may be
many compositional similarities to that of Lake responses to changes that started earlier than
Malawi. The stable part of the temperature gradi- the 1950s but they have most certainly acceler-
ent (and associated chemocline) begins at about ated since. There is an increasing nutrient load
60 m, which is sufficiently shallow to support originating from the activities (notably urbanisa-
a well-developed plate of photosynthetic bacteria tion and erosion consequential on intensifying
(Haberyan and Hecky, 1987). Eukaryote produc- agriculture) of a large (∼20 million) and increas-
tion in the mixolimnion is greatest following the ing (3–4% a−1 ) human population resident in the
periods of more active mixing, but nanoplank- lake’s catchment. Total phosphorus concentra-
ton and, later, the Cyanobacteria Microcystis tions in the lake water (now 45–72 µg P L−1 ) have
and Spirulina species feature with Anabaena and roughly doubled and the area-specific sedimenta-
Cylindrospermopsis in the calm period (Serruya and tion rates have gone up by a similar factor (Ram-
Pollingher, 1983). By way of summary, P → X → lal et al., 2001). Levels of DIN were, and remain,
H1/LM /SN /S2) is probably a fair representation. low: the nitrogen limitation of production recog-
In Lake Turkana (area 8660 km2 , mean depth 29 nised by Talling (1965) persists. Average silicon
m), where mixing involves a much larger propor- levels have fallen considerably. Coincidentally,
tion of the lake volume, phosphorus has accumu- the ecosystem has been catastrophically altered
lated to high levels (TP 1.8–2.4 mg P L−1 , [SRP]max by the introduction during this period of Nile
≤786 µg P L−1 : Hopson, 1982). The diatom assem- perch (Lates niloticus) and Nile tilapia (Oreochromis
blage conspicuously includes ‘large’ species of niloticus) which have expanded at the expense of
Surirella and Coscinodiscus; F-group colonial chloro- the lake’s endemic populations of haplochromine
phytes appear as stability increases in January cichlids (Ogutu-Ohwayu, 1990; Witte et al.,
but soon give dominant place to nitrogen-fixing 1992).
Anabaena species. Microcystis species also develop The phytoplankton of Lake Victoria in the
in some numbers (Liti et al., 1991). The sequence 1960s has been characterised by Talling (1965,
is partly captured in the summary notation P → 1986, 1987). Despite the great difference in basin
F → H1/LM . morphometry, Lake Victoria was subject to a pat-
In contrast to the rift-valley lakes, Lake Vic- tern of alternating dominance between diatoms
toria is young (possibly as little as 12 000 a) during the mixed periods (especially by what
and quite shallow (mean depth 39 m, max- are now referred to as Aulacoseira granulata and
imum depth 84 m) in relation to its area Cyclostephanos spp.) and nitrogen-fixers Anabaena
(68 800 km2 ). Like the rift-valley lakes, however, and Anabaenopsis species in the stratified phase.
it experiences seasonal alternation between a Also present were Closterium species and a variety
warm, wet (October–May) and a cool, dry season of chlorococcal algae associated with enriched
(June–September). The lake is wind-mixed to the shallow conditions (Scenedesmus, Pediastrum, Coelas-
bottom on frequent occasions during July and trum spp. of group J). Chlorophyll-a concentration
August. From the limnological information on in the open lake varied between 1.2 and 5.5 µg
the lake that has accumulated since systematic chla L−1 , though higher concentrations could be
information started to become available (Fish, found in the thermocline.
1952; Talling, 1965; Beadle, 1974), it is quite clear By the 1990s, Aulacoseira had declined to the
that changes within and beyond the lake dur- point of rarity and Anabaena, Anabaenopsis and
ing the last 40 years have had a profound influ- Aphanizomenon occurred only sporadically (Kling
ence upon the biota (Hecky, 1993; Kling et al., et al., 2001). Nitzschia acicularis had become the
2001). These include an eight- to tenfold increase dominant diatom, peaking between September
in the phytoplankton chlorophyll, a doubling of and November. This alga is typical of much
smaller, usually enriched and turbid, bodies of between November and April, to the bottom. In
water and some lowland rivers and ascribed by the intermediate quiescent episodes, microstrati-
Reynolds et al. (2002) to functional group D. fication develops to within a few metres of the
Though representing a larger biovolume than all lake surface. Although the absolute and rela-
the diatoms together in the 1960s, Nitzschia pro- tive abundances among them may vary among
duction in the Lake Victoria of the 1990s was episodes, the responding populations conform to
overshadowed by that of Cyanobacteria. Cylindros- a clear and general pattern. Aulacoseira is ususally
permopsis now dominated the nitrogen-fixer niche the dominant species among the diatoms that
in the stratified period, while solitary, filamen- increase with each mixing event. As the mix-
tous non-nitrogen-fixing species of Planktolyngbya ing weakens, cryptomonads briefly replace the
had become abundant in the mixed period. The settling diatoms but fall victim to cladoceran
former P/J → H1 alternation had been usurped filter-feeding. The clearing, stabilising epilimnion
by a D/S1 → SN sequence whose selection is is populated by various colonial green algae,
forced presumably by the requirement for supe- by Anabaena and then by migrating peridinoid
rior antennal properties. dinoflagellates. In chemical terms, the lake is
Brief reference should be made to one or more mesotrophic than eutrophic and deficient
two other, well-researched tropical systems out- in nitrogen rather than phosphorus. The sum-
side Africa which experience quite different pat- mary sequence P → Y → F → H1 → LM applies
terns of seasonal forcing. Lago Titicaca (area 8559 comfortably to the periodicity of species in many
km2 , mean depth 107 m, stratified October–July low-latitude lakes including that of another trop-
to within 40–70 m of the surface) supports a max- ical lake studied by Lewis (1986), Lago Valencia,
imum biomass during the early stages of mixing Venezuela.
(May–June, when Aulacoseira and other diatoms
are relatively abundant: Richerson et al., 1986; Phytoplankton in shallow lakes
Dejoux and Iltis, 1992). SRP levels are ≤23 µg In terms of number, most lakes are absolutely
chla L−1 , but DIN is low (<50 µg N L−1 : Vincent small (as defined on p. 325) and, commonly, ‘rel-
et al., 1984). Planktic nitrogen fixation is mod- atively shallow’ (as outlined on p. 320). The essen-
est, however, and nitrogen deficiency, together tial property of a shallow lake is that much
with light dilution, appears to most constrain or all of the bottom sediment surface is fre-
production. Diurnal heating and nocturnal cool- quently, if not continuously, contiguous with the
ing cause wide daily fluctuations in stability. Few open-water phase of the habitat (Padisák and
algae accommodate to this diel stratification but Reynolds, 2003). The consequences may be mani-
group-F colonial chlorophytes, together with self- fest in any of several ways. Subject to particle size,
regulating Peridinium species (LO ) are relatively depth and clarity, the bottom surfaces may sup-
tolerant and predominate during this part of the port epilithic algae or rooted macrophytic plants
year. (which, sensu lato, include large algae, mosses and
Finally, the apparent association of diatoms pteridophytes, as well as angiosperms). Finer sed-
and solitary filamentous cyanobacteria with iments are liable to entrainment by penetrat-
mixed water columns, and (for different reasons) ing turbulence; this may well increase turbid-
of rapid-growing flagellates, nitrogen-fixers and ity and light-scattering and, thus, impair light
self-regulating gleaners is well demonstrated in penetration. On the other hand, however, the
tropical Lake Lanao, the Philippines (area 357 simultaneous entrainment of interstitial water
km2 , mean depth 60 m). Lewis’ (1978a) penetrat- and biogenic detritus provides a mechanism for
ing analysis of phytoplankton periodicity in this the accelerated return of resources back to the
lake provides may insights into plankton ecology water column. From the point of view of ecosys-
elsewhere. The lake is subject year-round to fre- tem function, the pelagic systems of shallow
quent stormy episodes (10–15 a−1 ; the behaviour lakes differ from those of deep lakes and seas
is described as atelomictic) that mix the lake in not becoming isolated from a significant store
to depths >20 m, sometimes to >40 m and, of dead, decomposing biomass.
On this basis, some ‘large lakes’ (>500 km2 , 1993). One is a vegetated state with clear water;
at least when they are full of water) are also the other is a potentialy turbid, phytoplankton-
‘shallow’: these include the Lakes Winnipeg, dominated system. Moreover, many cases
Balkhash, Tchad, Eyre and Bangweolo (Reynolds together indicate that small lakes can switch
et al., 2000). Neither these, nor any of the greatly abruptly between these two stable states (Blindow
more numerous small lakes, support a character- et al., 1993: Scheffer, 1998), under the influence of
istic or recognisable ‘shallow water phytoplank- switches in the resource–consumer interaction.
ton’ – community assembly invokes other crit- Restricting the present discussion to the phy-
ical factors. Yet phytoplankton is important to toplankton composition, the structural influ-
perceptions of water quality in very many shal- ences in shallow waters fall into several cat-
low lakes (Scheffer, 1998), so it is valuable to egories. Phytoplankton may be present in
be able to extract some general principles. One small concentrations if bioavailable resources
of these is that, in a majority of truly ‘small’ in the water are modest (advantage to rooted
(<200 m across) and absolutely ‘shallow’ (≤5 m) macrophytes exploiting additional or alternative
water bodies have a theoretical light-supportive resources) or, if not, if macrophytic vegetation
capacity of ≤600 mg chla m−2 (say, 30 C m−2 ), harbours the food web that can exert ongo-
unless they are at high latitude or their waters ing controls on the development of phytoplank-
are heavily stained with humic substances. This ton. Phytoplankton can dominate in the open
provides plenty of scope for the intervention water above macrophytes rooted 1–1.5 m below
of other potential limiting factors. Interestingly, the water surface. In temperate waters, they can
Scheffer (1998) provided a plot of the average respond earlier and faster than macrophytes to
summer chlorophyll-a concentrations in each of lengthening days. If circumstances prevail where
88 shallow lakes in the Netherlands (mean of there is no adequate consumer control, phyto-
mean depths 2.1 m) against its average summer plankton have the potential to shade out rooted
TP concentration. Up to 0.3 mg P L−1 , the points benthos or, at least, to contain its distribution to
cluster beneath a slope of chla = 0.9 TP. At TP the shallowest margins.
concentrations >0.3 mg P L−1 , chlorophyll levels Some of the representative shallow-lake phy-
show a typical P-saturation response. An analo- toplankton assemblages may be noted. Unicellu-
gous dataset for the same lakes shows a similar lar nanoplanktic forms are collectively common,
behaviour with respect to TN, with the points benefiting from C-type invasive, fast-growth-
clustering beneath a slope of chla = 0.09 (TN – rate strategies. These often include nanoplank-
0.7), up to TN ∼4 mg N l−1 . However, a very tic chlorococcals (Chlorella, Monoraphidium spp.
large number of records apply in systems with of group X1), nanoplanktic flagellates (Chlamydo-
markedly lower TP and TN availabilities but, nev- monas, Plagioselmis, Chrysochromulina of group X2)
ertheless, show marked saturation of the chloro- and group-Y cryptomonads and small peri-
phyll actually supported. dinioids. A polyphyletic group-W1 association of
In many of these small lakes, there is a euglenoids (Phacus, Lepocinclis as well as the type
complex interaction with rooted, submerged genus, Euglena), chrysophycean (especially Synura
plants, which, in these shallow waters, are able spp.) and small volvocalean colonies (especially
to compete effectively in energy- and resource- Gonium) may be represented. On the basis of a
harvesting. The complexities arise through floristic comparison of small, shallow lakes in
the behavioural interactions among benthic Hungary, Padisák et al. (2003c) noted the frequent
macroinvertebrates, littoral cladoceran zooplank- co-occurrence of Phacotus spp. with this group
ton and their respective planktivorous and ben- and its dominance in calcareous waters; they pro-
thivorous consumers (including fish) (see Section posed a new group identity in YPH . Larger green
8.3.6). Suffice it to say here that aquatic primary colonies are represented by Volvox, Eudorina and
production and its heterotrophic consumers can Pandorina of group G. Colonial chrysophytes (espe-
strike quite different and alternative steady states cially Dinobryon spp.) are also abundant in small
in shallow lakes (Scheffer, 1989; Scheffer et al., water bodies, even quite calcareous ones, if there
is an good supply of carbon dioxide (e.g. from Lagerheimia, Closterium and Chlorella by one in
benthic decomposition of plant matter, includ- which Asterionella, Pediastrum, Anabaena and even
ing fallen leaves). Planktic diatoms are not gen- Microcystis formed successive populations. The
erally abundant in the small, shallow ponds save trigger was the introduction of base-rich water
for the small Stephanodiscus species, and perhaps from a trial bore that raised the ambient pH of
the smaller Synedra species (group D). Gyrosigma, the lake from 4.5 to 6.5 (Swale, 1968; Reynolds
Surirella and Melosira varians that are tychoplank- and Allen, 1968).
tic, as well as Aulacoseira species that are mero- Abundant populations (perhaps equivalent to
planktic (the distinction is actually very fine 100–600 µg chla L−1 ) of phytoplankton occur-
here), may be prominent in shallow lakes that are ring in small, shallow and continuously nutrient-
exposed to wind and wave action (including large rich, hypertrophic lakes and in which, through
lakes such as Balkhash and Tchad). Elsewhere, trophic imbalances, heavy grazing by zooplank-
B, C, N and P groups of diatoms and desmids ton is avoided, frequently comprise species of
are common, provided their suspension require- Scenedesmus, Coelastrum and Pediastrum (group
ments are fulfilled by the absolute water-column J), often with X1-group nanoplankters such
depth (see Sections 2.6.2, 6.3.2) in small lakes. as Chlorella, Ankyra and/or Monoraphidium. The
The diatoms remain sensitive to silicon exhaus- assemblage is well represented in the phyto-
tion and they may be replaced by other algae, plankton of habitats where high nutrient load-
including non-vacuolate, small-celled blue-green ing is equated with high hydraulic loads and
in their colonial phases (Aphanocapsa, Aphanoth- rapid flushing rates that discount against slow-
ece of group K) or by non-motile colonies of growing algae and most species of mesozooplank-
group-F green algae, such as Botryococcus. In shal- ton. These include hypertrophic rivers, some
low lake Śniardwy, Poland (area 110 km2 , mean flood-plain lakes flushed by river flow, natural
depth 5.9 m), Aulacoseira species alternate with ponds enriched with sewage and many artificial
group-J Scenedesmus and Pediastrum species (Kajak ponds constructed to bring about its oxidation
et al., 1972). During the 1970s, a similar assem- (Uhlmann, 1971; Boucher et al., 1984; Moss and
blage was prominent in Budworth Mere, England Balls, 1989; Stoyneva, 1994). The assemblage is
(Reynolds, 1979b). sometimes more prominent in the plankton of
The effects of nitrogen depletion in small, more substantial lakes, including Arresø, Den-
shallow lakes are no different from other mark (Olrik, 1981), and during the more stable
nutrient- and energy-rich habitats, where, of the alternating phases in the enriched Hamil-
accordingly, nitrogen-fixing (H1) Anabaena and ton Harbour, Lake Ontario (Haffner et al., 1980).
Anabaenopsis species and, in warm lakes, (SN ) At low latitudes, such enriched shallow sys-
Cylindrospermopsis may be promoted. Both are tems may succumb to monospecific (or, at least,
important components in Balaton, Europe’s monogeneric) steady-state blooms of Microcystis
largest shallow lake (area 593 km2 , mean depth (group M), where the alga’s ability to regulate
3.3 m: Padisák and Reynolds, 1998). The plank- its buoyancy helps it to avoid excessive light
ton of humic-stained Fennoscandian lakes is near the surface during diurnal stability and
frequently distinguished by the presence of chrys- to recover position after nocturnal mixing. The
ophytes, cryptomonads and the raphidophyte original demonstration of dominance through
Gonyostomum semen. Physiological studies on this this mechanism, by Ganf (1974b), in the tropi-
alga by Korneva (2001) seemed to Reynolds et al. cal Lake George, Uganda (area 250 km2 , mean
(2002) to justify its separation into group Q. The depth 2.5 m), has been repeated in subsequent
plankton of acidic small shallow lakes is biased studies of other low-latitude, hypertrophic shal-
towards chlorophyte (including desmid) and low lakes (Harding, 1997; Yunes et al., 1998). My
chrysophyte genera. An interesting observation own, unpublished observations of the large but
on the plankton of Oak Mere (area 0.2 km2 , mean shallow lake Tai Hu (2425 km2 , mean depth 2.1
depth 2.1 m) was the replacement of a plank- m) suggest that Microcystis similarly dominates
ton dominated successively by Ankistrodesmus, the plankton from quite early (April) in the year.
Finally, many shallow, hypertrophic lakes in low or is made shallow through density strati-
the temperate regions (especially those exposed fication. Such benign conditions are presumed
to frequent or continuous wind mixing), experi- to satisfy the minimum requirements of most
ence year-round dominance by Planktothrix agard- phytoplankton. However, the most widely promi-
hii and/or Limnothrix redekei or other slender, soli- nent trait-differentiated planktic groups achiev-
tary filamentous Cyanobacteria, such as Pseudan- ing large, dominant and persistent populations
abaena species. Phytoplankters of group S1 need under these conditions are epitomised by the
never normally experience nutrient limitation, groups LM , M and R and, perhaps to a lesser
and they are difficult for cladocerans to filter and extent, H1 and J (see also Naselli-Flores et al.,
for copepods to manipulate (although certain cil- 2003).
iates seem to have perfected a means of ingesting Decreasing day length and/or deeper mix-
them). They may dominate to the extent that they ing result in diminished opportunities for light-
exclude almost all other autotrophic phytoplank- harvesting, to the extent that the capacity
ton, to the limit of the energy-determined carry- for photosynthetic fixation of inorganic carbon
ing capacity set by their own highly efficient and becomes the constraint on the ability of phyto-
persistent light-harvesting antennae (Reynolds, plankton to function and survive (as discussed
1994b). Several examples from the literature were in Section 5.4.3). It was argued there that the
cited there in support, including the studies of more sensitive species are (literally) outcompeted
Berger (1984, 1989, on the polder lakes Dron- by those that are more robustly adaptable and
termeer, Wolderwijd and the pre-manipulated predisposed to light interception and harvest. The
Veluwemeer), of Whitton and Peat (1969, on St survey in the present chapter would confirm that
James’s Park, London) and of Gibson et al. (1971, late-summer mixing in temperate, mesotrophic
on Lough Neagh, Northern Ireland). In Kasum- and eutrophic lakes repeatedly selects for P-group
igaura (area 220 km2 , mean depth 4 m), Plankto- diatoms and desmids, and in some instances, for
thrix agardhii replaced Microcystis aeruginosa as one T-group Tribonema and Mougeotia and, especially,
low-diversity plankton gave way to another (Taka- S1-group Planktothrix agardhii. In highly enriched,
mura et al., 1992) and, apparently (Recknagel, shallow waters, dominance of self-shaded assem-
1997), as a consequence of an ongoing expansion blages by the latter may be near-perennial. Con-
of the resource capacity demanding the better- versely, the biomass recruited during the ‘spring
performing species under increasing light limi- outburst’ of phytoplankton in the mixed columns
tation. of mesotrophic and mildly eutrophic temperate
lakes is generally dominated by diatoms (B, C
7.2.4 Species assemblage patterns in lakes groups) and Y-group cryptomonads, gymnodini-
Certain consistent patterns emerge from the ans or small peridinians.
descriptive accounts making up Section 7.2.3, In many instances, improving light conditions
most of which begin with a recognition that phy- in lakes press the constraints towards the nutri-
toplankton production and the biomass that it is ent resources. Shortage of silicon is an obvious
possible to support are often constrained by the selective disadvantage to all diatom groups but
environmental conditions obtaining. A corollary it is not clear from the present review that sili-
to this statement is that high rates of produc- con deficiency becomes a limiting factor among
tion, sustaining and maintaining large planktic the most oligotrophic lakes (that is, other nutri-
crops, depend upon the alleviation of the typ- ents intervene first). The instances in which nitro-
ical constraining factors. Standing-crop concen- gen, theoretically or by interpretation, seems to
trations in excess of 40 µg chla L−1 are mainly limit the supportive capacity are more common
encountered in lowland and moraine lakes. All than the weight of literature might suggest, espe-
of these offer a substantial base of bioavail- cially at low latitudes and in well-leached catch-
able nutrients, especially phosphorus and nitro- ments. Nitrogen deficiency may be cited as a
gen. Moreover, autotrophs benefit from being selective factor operating in favour of the com-
entrained within a layer that is absolutely shal- mon occurrence of H1, H2 and, in warm-water
locations, SN groups of nitrogen-fixers. However, among smaller lakes that stratify to within 10
it must be emphasised again that their preva- m of the water surface and where accessible
lence is not confined to habitats that are low SRP concentrations persist in the metalimnion,
in nitrate or ammonium. Moreover, the argued the most successful species are the relatively
dependence of the ability to fix nitrogen on high motile gleaner species (of especially groups LO
energy inputs and an adequate reserve of phos- and LM ). Given water of sufficient clarity these
phorus and trace metals (see Section 4.4.3) is sup- may well be able to operate successfully simply
ported by the survey (see, especially, p. 335). In by remaining at depth. However, the deeper are
stratified lakes, conditions of low DIN and low the available resources, the more important is
SRP may be more amenable to motile ‘gleaners’, the light-harvesting ability and the less impor-
including those with known phagotrophic capa- tant is rapid motility. Ultimately, before energy
bilities. This explanation fits the frequent domi- and nutrient resources are finally uncoupled, the
nance of E, U and, especially, LO groups of algae. selective advantage may fall to superior, chro-
In water columns that are chronically deficient in matically adapted light-harvesting species with
nitrogen and phosphorus and, thus, in the ability a simultaneous ability to maintain and adjust
to develop even a detrital store of nutrients, the vertical station – and include cryptomonads (Y)
most tolerant survivors appear to be the Z-group and phycoerythin-rich Cyanobacteria, especially
picocyanobacteria. Planktothrix rubescens (R).
Phosphorus bioavailability, however, remains The surveyed cases support the contention
a major constraint upon the supportive capacity that particular adaptive traits distinguished
in a large number of stratifying and shallow lakes among the phytoplankton and the species in
and, hence, a key factor in defining the trophic which they are most strongly represented are
state. It has been recognised in earlier chapters better suited to particular sets of environmental
that plankton algae are generally very effective conditions. The more severe are the latter, the
in garnering their own phosphorus requirements greater the selective power that works in favour
from SRP concentrations as low as 10−7 M (∼3 of the most tolerant species. This is the comple-
µg P L−1 ) (see p. 158). While greater concentra- mentary deduction to that of Dufrêne and Leg-
tions than this remain accessible, the supportive endre (1997) regarding the reliability of indica-
capacity of BAP has not been reached. Superior tor species and what they may convey in terms
affinity for phosphorus in solution is unlikely of constraints acting upon community function.
to confer a competitive advantage to phytoplank- There is an obvious mutualism between, on the
ton except in the range 10−9 –10−7 M. In those one hand, evolutionary strategies and adapta-
instances where BAP is always or nearly always tions of species that permit them to tolerate
<10−7 M (∼3 µg P L−1 ), species with high uptake particular environmental conditions and, on the
affinities experience both the immediate com- other, the conspicuous occurrence of these same
petitive advantage of winning resources and the species in locations where the critical conditions
longer-term advantage from being able to main- obtain. In Table 7.6, the various trait-separated
tain larger inocula from one growth opportu- functional groups are listed in terms of their
nity to the next (see p. 203). The trait is shared reactivity to selective variables distinguishing
among species identified in Table 5.2 and which among habitats or, for a given habitat, among
figure in groups A, E, F, N, U, X3 and Z. They seasonally varying conditions.
are represented in lakes throughout the descrip- Of course, alternative approaches are avail-
tive series but they provide the prominent com- able to establishing the link between organismic
ponents of the planktic assemblages in the low- adaptations and indicative habitat preferences.
P oligotrophic lakes, large and small. They are For instance, it is commonly revealed through
also well represented among those stratifying analysis of the size distributions of the total-
mesotrophic waters during periods in which epil- ity of organisms forming the assemblage, inde-
imnetic BAP has been previously be depleted pendently of its taxonomic composition, based
to levels of ≥10−8 M (∼0.3 µg P L−1 ). However, instead upon their functional contributions
Table 7.6 Sensitivities to habitat properties of functional groups of phytoplankton
hm I∗ θ [P] [N] [Si] [CO2 ] f

Group <3 <1.5 <8 <10−7 <10−6 <10−5 <10−5 <0.4
A − ? + + + + − −
B − + + + − − − −
C − + + − − − ? −
D + + + − − − + −
N − − − + − +/− − ?
P − − − − − +/− + +
T − ? − +/− − + ? +
S1 + + + − − + + +
S2 + + − − − + + +
SN + + − − + + + +
Z + − + + + + ? −
X3 + − + + − + − −
X2 + − + ? − + ? −
X1 + − + − − + + −
Y + + + − − + ? −
E + + + + − + − −
F + − + + − + − −
G + − + − − + + +
J + ? + − − + ? −
K + ? − − − + + ?
H1, H2 + − − − + + + +
U + − ? + − + − +
LO + − − + − + − +
LM + − − − − + + +
M + − − − − − + +
R + + + − − + ? +
V + + + − − + ? −
W1 + + + − − + ? −
W2 ?
Q + + + ? ? + − −
Notes: Entries in table are to denote tolerance (+) or no positive benefit (−) of the environmental
condition set; ‘+/−’ is used to denote that some species in the association are tolerant; ‘?’ denotes
that tolerance suspected but not proven. Some representative genera or species only are listed.
Variables signified are: depth of surface mixed layer (hm , in m from surface); mean daily irradiance levels
experienced (I∗ , in mol photons m−2 d−1 ); water temperature (θ , in ◦ C); the concentration of soluble
reactive phosphorus ([P], in mol L−1 ); the concentration of dissolved inorganic nitrogen ([N], in mol
L−1 ); the concentration of soluble reactive silicon ([Si], in mol L−1 ); the concentration of dissolved
carbon dioxide ([CO2 ], in mol L−1 ); and the proportion of the water processed each day by rotiferan
and crustacean zooplankton (f )
Source: Updated from Reynolds (2000a).
Figure 7.7 Schematic summary

of freshwater pelagic habitats,
defined in terms of I∗∗ (an integral of
light income and its dilution through
the mixed layer) and K∗∗ (an integral
of nutrient accessibility), and where
I∗∗ = (I 0 ·Im )1/2 hm −1 and K∗∗ =
[K]/(1 + δ[K]); for further details
see text. Redrawn with permission
(Bailey-Watts, 1978; Gaedke and Straile, 1994b). this applies to photoautotrophic picoplankton as
Size-spectral analyses are particularly sensitive to much as to nanoplankton (Carrick and Schelske,
the structure of the pelagic food web and its 1997). High concentrations of picophytoplank-
responses to interannual enviromental variabil- ton dominated the algal assemblage of a small
ity and to fundamental alterations to the nutri- Antarctic pond, under the conditions of low
ent base (Gaedke, 1998). However, care is needed ambient temperatures and nutrient enrichment
in making ecological interpretations from algal through its use as a roost by elephant seals (Iza-
morphometry without allowing for its alterna- guirre et al., 2001).
tive indications. For instance, the many sepa- The truth is that smaller algae have physiolog-
rate studies on the abundance and production of ical and dynamic advantages over large ones. It is
picophytoplankton support the conclusion that the intervention of other factors (most especially,
these organisms fulfil the major contribution to the influence of grazing and the segregation of
the carbon dynamics of oligotrophic pelagic sys- the energy and resource bases) that alters the
tems (Stockner and Antia, 1986; Chisholm et al., structural balance in favour of larger or motile
1988; Agawin et al., 2000; Pick, 2000). In contrast, algae. This too impinges upon the spectral anal-
the greatest concentrations of large algae like ysis. In reality, the procedure detects functional
Ceratium and Microcystis are supposedly excluded aspects of the assemblage and the extent of its
from nutrient-poor systems (Wolf-Gladrow and successional maturity with respect to the driving
Riebesell, 1997). Thus, a size spectrum biassed variables and not necessarily what the limiting
towards a predominance of smaller or larger variables might be.
forms should reflect a lesser or greater produc- To summarise the compositional patterns of
tive capacity. The counter to this simplicity is freshwater phytoplankton assemblages, a habi-
that, indeed, organisms at the diminutive end tat template, analogous to the one for marine
of the size spectrum gain full advantage of their environments (Fig. 7.4), is proposed in Fig. 7.7.
high surface-to-volume ratio to acquire and con- Habitats are characterised on axes of nutrient
vert resources into biomass faster than larger resources and energy distribution, the interac-
organisms with less favourable surface-to-volume tion of which is suggested to drive the pri-
ratio lowers (Raven, 1998) (see also Section 5.3.1). mary selective criteria in lakes (variations in
Other things being equal, then, the greater is acidity/alkalinity are not addressed specifically).
the productive capacity, the more the size spec- Unlike Fig. 7.4, however, the axes are quanti-
trum should be biassed towards smaller rather fied, following Reynolds (1999a), in units that
than larger forms. Recent research confirms that were designed to embrace variabilities in both
Figure 7.8 (a) Habitat template for trait-separated photons m−3 d−1 . Using actual measurements
categories of freshwater phytoplankton (cf. Table 7.1), based (Ganf, 1974b) of average daily irradiance at
on limitation gradients of decreasing energy harvest and equatorial Lake George (average daily irradiance
accessible resources, as originally envisaged by Reynolds ∼2000 J cm−2 d−1 , or 20 MJ m−2 d−1 , PAR ∼9.4
(1987b). The superimposition of one shape on another MJ m−2 d−1 = 43 mol photon m−2 d−1 ) and verti-
implies that the algal category is more likely to succeed than
cal extinction coefficients through a 2.5-m water
those that it obscures. (b) As (a) but unlabelled; the shaded
column of up to 7.7 m−1 , I∗∗ falls as low as 1.45
triangles embrace approximately the floristic representation
in the named systems: 1, eutrophic pools in Shropshire; 2,
mmol photons m−3 d−1 Similarly, at the end of
Montezuma’s Well; 3, polder lakes such as Veluwemeer; 4, the summer in the Araucanian lake Nauel Huapi
Norfolk Broads; 5, Hamilton Harbour, Lake Ontario; 6, Crose (p. 336), even a low attenuation coefficient (ελ ≤
Mere; 7, Lough Neagh; 8, Volta Grande Reservoir; 9, Lagoa 0.2 m−1 ) does not prevent convectional mixing
Carioca; 10, Windermere, pre-1965; 11, Millstättersee. to 60 m from diluting the light availability to a
Redrawn with permission from Reynolds (1997a). similar level (I∗∗ ∼1.2 mmol photons m−3 d−1 ).
The vertical scale (K∗∗ ) is based upon mixed-
deep and shallow lakes. The scale of the hori- layer nutrient concentration, divided by a factor
zontal axis, I∗∗ , integrates the income of photo- based on the concentration gradient (1 + δ[K])
synthetic energy not just through time but its through the whole trophogenic layer. This index
dilution through the mixed layer. Whereas I∗ , distinguishes chronic deficiency of the critical
from Eq. (3.17), is calculated from the average nutrient ([K] low, δ[K] low, so K∗∗ = [K]/(1 + δ[K]) is
vertical absorptance of photosynthetically active also low) from the effects of near-surface deple-
solar radiation over the mixed depth (hm ) [as I∗ = tion (as δ[K] increases, so K∗∗ decreases. In terms
(I 0 · Im )1/2 , where Im = I 0 · hm exp(−ε λ ) is the of phosphorus, for example, the review lakes
residual irradiance flux reaching the base of the cover a range of measured availabilities from 0.01
mixed layer], I∗∗ is the average harvestable photon to 25 µM (0.3 to 750 µg P L−1 ). However, the scale
concentration. responds to the seasonal depletion of resources
Thus, in the surface mixed layer to the point of critical
I ∗∗ = (I 0 · I m ) /2 h −1
1
(7.1) deficiency.
m
Earlier versions of this template (especially in
If daily integrals of the photon flux are used, Reynolds 1987b, c; see also overview in Reynolds,
we can distinguish not just high from low light 2003a) have been used to accommodate the distri-
incomes but compare dilution of the income as a butions of most of the trait-separated functional
consequence of mixed layer depth and turbidity. groups. The shapes drawn in Fig. 7.8a are pro-
For example, a photon flux of 60 mol photons posed to represent the respective environmental
m−2 d−1 on a shallow water body (hm = H = 1 limits of each of the algal groups, with group-S
m) with a low coefficient of vertical light extinc- species extending furthest rightwards along the
tion (say ελ = 0.2 m−1 ), I∗∗ solves at 54.3 mol I∗∗ scale (shallow hypertrophic habitatats) and
the general track shown in Fig. 7.9. This pur-

ports to move through summer stratification,
nutrient depletion, autumnal mixing (less light,
re-enrichment and nutrient uptake during the
spring mixing). In theory, at least, the trajectory
inserted into each of the triangles superimposed
onto Fig. 7.8b should be capable of describing
the compositional changes that characterise the
annual plankton cycles in the water bodies rep-
resented. At least the pattern, if not the detail,
of phytoplankton structure is captured.
Figure 7.9 Idealised year-long timetrack of the selectivity 7.3 Assembly processes in the
trajectory imposed by seasonal variability in a temperate
system, in relation to the zones favoured by C, S and R
phytoplankton
primary strategists. Redrawn with permission from Smayda
and Reynolds (2001). The purpose of the present section is to explore
the active mechanisms that influence the vari-
group-E dinoflagellates furthest downwards in ations in the structure of the phytoplankton
the K∗∗ axis (depleting epilimnia of oligotrophic in natural communities, according to circum-
lakes). Less tolerant groups terminate closer to stances and dominant functional constraints.
the origin in the upper left-hand corner repre- The objective is to determine the extent of pre-
senting resource-rich, energy-rich habitats. Super- dictability of natural communities, with the
imposition of shapes, one upon another, implies processes quantified wherever possible. The key
superior group performances, with the fastest topics considered to be relevant to community
nutrient- and light-saturated growth rates of X1 assembly in the plankton – species richness and
species dominating the top left-hand corner. diversity, trait selection, succession and stabil-
This representation is illustrative but it helps ity, structural disturbance and organisational
to put into perspective the floristic descriptions resilience – are no different from those believed
of many kinds of lakes on to a single two- to impinge upon the community ecology of other
dimensional figure. The representation in Fig systems. However, the dynamics of pelagic sys-
7.8b is identical to that in Fig. 7.8a, but for the tems operate at such absolutely small timescales
stripping out of all the alphanumeric insertions. (when compared with those of terrestrial sys-
The numbered triangles that replace them are tems) that the outcomes are not only observable
really quite effective in circumscribing the plank- phenomena, as opposed to speculative extrapola-
tic flora of named lakes or lake series. These tions, but they are amenable to meaningful, con-
include shallow, enriched pools and well-flushed trolled experimental manipulation. This is the
pools (1, 2), the turbid, Plankothrix-dominated part of the book in which the community ecology
polder lakes (3), through to oligotrophic lakes of the phytoplankton can be championed as the
such as Millstättersee (11). model for community ecological processes every-
Collectively, the shapes used in Fig. 7.8 com- where. Some of the terms used require clarifica-
prise the familiar triangular layout, correspond- tion of usage; some working definitions are given
ing to the disposition of the primary algal in Box 7.1.
strategies (with C, R and S at the apices). Sea-
sonal changes in any individual system will, to 7.3.1 Assembly of nascent communities
a greater or lesser extent, be tracked through The first challenge of this discussion is to estab-
the triangle, in terms of variations in I∗∗ and lish whether the observable assemblages of dif-
imposed changes in resource accessibility, along ferent species of phytoplankton (and, for that
ASSEMBLY PROCESSES IN THE PHYTOPLANKTON 351
matter, cohabitant microbes, zooplankton and way that material cycling is internalised and, to
nekton) are merely fortuitous and random com- a large extent, closed’ (Ripl and Wolter, 2003, p.
binations of species that happen to be present, 300). The components are:
by chance and in sufficient concentrations to be
(1) Primary producers, which manufacture
encountered readily, or whether these are selec-
organic material and supply the energy
tively biassed, invoking either mutually interspe-
needed for all heterotrophic structures.
cific dependence or mutual tolerance. If the lat-
(2) Water, as a moderating feedback control of
ter, then the central question becomes ‘How are
production.
they put together?’
(3) A reservoir of organic detritus.
(4) Decomposers – bacteria and fungi – draw-
Minimal communities
ing energy from the oxidation of detrital
The word ‘community’ presupposes some sort
stores, which process also recycles nutrients
of functional differentiation, with populations
and minerals to primary producers; the store
of several species each doing different things
is exploited efficiently and with a low level
but whose activities, in total, achieve some
of losses.
linked reactivity that defines their combined out-
(5) The food web – that network of consumers
puts – the community function. Historically, ecol-
comprising higher and lower animals
ogists have experienced difficulties in defining
through whose activities much of the
the nature of communities. At one extreme,
original energetic input is finally dissipated.
Clements (1916) anticipated a supra-organismic
regulation of tightly bound organismic func- It need hardly be added that the remarks of Ripl
tions. At the other, Gleason (1917, 1927) coun- and Wolter (2003) were inspired by a view of ter-
tered that the species were present more or less restrial ecosystems. Yet, but for an alternative
by chance. Thus, species composition is funda- emphasis on the relevance of water and a ten-
mentally non-predictable beyond the influence of dency of open waters to store organic detritus
species-specific habitat preferences and the con- remotely (i.e. in the sediments), the model ade-
sequence of selective interspecific interactions, quately describes a self-sufficient pelagic ecosys-
including predation and competition. Current tem. Moreover, in either case, although all the
thinking is rather closer to the second view than biotic components are essential to the sustain-
the first but the interactions are recognised to able, integrated function of the whole structure,
be crucial and far from straightforward. A help- primary producers play a critical role in inter-
ful modern model is that of Ripl and Wolter ceding in the dissipative flux, in synthesising the
(2003), which envisages community function as organic base and, hence, in initiating the assem-
a series of successive organisational hierarchies, bly of the ecological unit. This will be taken as
beginning with the interdependence of molec- a sufficient justification for concentrating upon
ular transformations and working through to assembly processes among aquatic primary pro-
the physiological control of individual organisms ducers (and, especially, the phytoplankton) and
and to the beneficial coordination of the separate for deferring consideration of the assembly of the
activities of the various component organisms. heterotrophic and phagotrophic elements of the
They cited several illustrative examples, includ- pelagic community to Section 7.3.2.
ing the nitrogen-fixing symbionts of water plants Of special interest are the answers to the
(including diatoms) and the use of algal exudates questions, how many and which species of pri-
by bacteria. They proposed the DEU (for dissipa- mary producer will be present? The supposition
tive ecological unit) as the minimal interspeci- made throughout this book is that phytoplank-
fic assembly within which entities contribute ton species will grow wherever and whenever
to an organised, functioning structure with a they can, provided that (i) the supportive capac-
measurable and increasing thermodynamic effi- ity to satisfy their minimal requirements for
ciency. Thus, the DEU comprises five distinct com- their growth is in place and (ii), coincidentally,
ponents, capable of mutual coupling ‘in such a viable propagules are already present and able to
take advantage of the amenable conditions. Ear- upstream location (including the river itself) is
lier chapters have probed extensively the inter- likely to deliver an inoculum of phytoplankton
specific differences in environmental require- with a composition reflecting the upstream habi-
ments and tolerances but the uncertainties relat- tats that supplied it. Species that produce ben-
ing to propagule dispersion have, so far, been thic resting spores and propagules and which
avoided. were previously resident in the plankton of the
Both issues are relevant to the role of species flushed lake and which succeeded in recruit-
richness and diversity in assembling communi- ing a large number of overwintering propag-
ties. It has to be admitted, however, that the ules to the sediment prior to displacement of
large numbers of widely distributed species with the water mass by throughflow will also enjoy
generally rather small differences in their basic a biassed opportunity to recruit inocula to the
requirements do not offer clear prospects for expanding development opportunity. In spite of
separating critical processes. Thus, it helpful to this, the number of species colonising and, cer-
consider the governing principles as they are tainly, the relative numbers of individual organ-
expressed in the establishment phases of new, isms of each contributed is mainly a matter of
open and pristine habitats. This approach fol- chance (Talling, 1951). The larger the system and
lows the powerful ‘island biogeographic’ concept, the greater its spatio-temporal connectivities, the
developed by MacArthur and Wilson (1967) to greater is the likelihood of annually reproducible
describe the processes leading to the establish- patterns.
ment of a biotic ecosystem on a newly formed In large and small systems alike, however,
island, arising Surtsey-like from the ocean. The the one statement that seems true is that the
model translates well to the context of phyto- species that initially become abundant either
plankton development in a small, temperate or ‘arrive’ (with or without the benefit of a rest-
sub-polar, inland water body, which, in this case, ing inoculum) in strength, or grow rapidly,
may be considered as an aquatic island in a ter- or they do both. These traits are analogous
restrial sea. Seasonality confers an element of to those of the pioneering, invasive, island-
vernal ‘newness’ of such a habitat. If objection colonising species of MacArthur and Wilson
be made to the bias attributable to the pres- (1967), (r-)selected by their investment in short
ence of overwintering of propagules, then the life histories and prolific production of small, eas-
case of flood-plain lakes (varzeas) left isolated ily transmissible seeds. The corresponding phyto-
by falling rivers (Garcı́a de Emiliani, 1993) or plankters are manifestly the small-celled, quick-
of mining subsidence pools (e.g. Dumont, 1999) growing C-strategist species (Section 5.4.5) (Box
could be adopted. In the open sea, the seasonal 5.1). Freshwater examples come from the X1, X2
onset of thermal stratification in a water column and Y functional groups (Section 7.2.3; Table 7.1).
‘sterilised’ of phytoplankton during months of Indeed, some of the species involved (of such gen-
deep mixing also conveys the idea of open, avail- era as Chlorella, Chlamydomonas, Chlorococcum, Cocco-
able habitat, depauperate in indigenous plank- myxa, Diacanthos, Golenkinia, Micractinium, Mono-
tic algae. Even highly flushed, river-fed small raphidium, Treubaria, Westella) are also variously
lakes (like Grasmere, English Lake District), may encountered in small or temporary aquatic habi-
become so depleted of phytoplankton during wet tats, such as rain puddles of a few days’ age,
periods that, effectively, they become amenable, bird baths, rinsing water left in open contain-
open habitat for the next phytoplankters to arrive ers and bottles, and in the phytotelmatic habi-
there (Reynolds and Lund, 1988). tats of water retained in the foliage of epiphytic
plants such as bromeliads. The only consistently
The pioneer element of nascent communities plausible means of dispersal for propagules of
In the truly novel planktic habitat, the first these algae is through the air. The extent of
colonists have to arrive, de novo, from some spore production and the tolerance of dessica-
other external location. In other, quasi-open habi- tion is not well known, while the viability of
tats, fluvial transport of propagules from an cells in aerosols needs further investigation. The
point that needs to be emphasised is that small, Each was previously unrecorded in the lake and,
potentially planktic, C-strategist algae are not supposedly, recruited from the nearest sea, some
just highly mobile between mutually isolated, 1500 km distant!
potentially viable habitats but that, collectively, The implication of this ready colonisation
they are sufficiently abundant to have an evi- by invasive species is that, for some species at
dently high probability of early colonisation of least, the dispersal channels are effective and
new planktic habitats. well exploited. Not all species are equally mobile:
The observation is relevant to one of those colonial Cyanobacteria such as Microcystis are rel-
enduring puzzles of phytoplankton ecology, still atively slow to establish in newly enriched lakes,
lacking formal solution, which is just how appar- even though the habitat might seems amenable
ently conspecific phytoplankters move among to its growth (Sas, 1989). A number of species have
hydrologically isolated lakes and achieve near- very limited distributions and seem endemic to
global, cosmopolitan distributions. Many con- particular regions or even localities (Tyler, 1996).
tributory investigations, for instance, those of However, water bodies throughout most of the
Maguire (1963), Atkinson (1980) and Kristiansen world, not just new ones, must be constantly
(1996), reveal something of the variety and vari- assailed by the propagules of invasive species.
ability of dispersal routes for planktic organisms, This process helps to maintain their apparent
especially aquatic insects and water birds. Rather cosmopolitan and pandemic distributions. These
less than the much-speculated ‘duck’s foot’, guts species turn up everywhere and, if the opportu-
and faeces and avian feathers retain propagules nity of habitat suitability and resource availabil-
and (often) live vegetative algal cells, providing ity is there, populations may establish wherever
important avenues for the potential transfer of they land.
planktic inocula from one water body to another. These deductions apply to many bacteria and
This may apply over relatively modest distances, other microorganisms and, to a lesser extent, to
before desiccation or digestion destroys the via- mesozooplankters (Dumont, 1999). They apply at
bility of the potential inocula. Recent investiga- an approximately comparable level among the
tions by Okamura and her co-workers (see espe- microzooplanktic and nanozooplanktic ciliates,
cially Bilton et al., 2001; Freeland et al., 2001) studied extensively by Finlay (2002). Besides coin-
have related the spread of genetically separa- ciding with the size ranges and body masses
ble clones of sessile bryozoans (small aquatic of dispersive phytoplankton, ciliates also occur
filter-feeders that produce distinctive propagules locally in abundance and enjoy world-wide ubiq-
called statoblasts) to the migratory paths of wild- uity, and, as Finlay’s experiments amply demon-
fowl. The work makes it clear that avian move- strate, common ecotypic species quickly estab-
ments do greatly influence the genomic compo- lish in contrived new environments open to their
sition and species richness of island freshwater colonisation from the atmosphere. Ubiquity is
communities. dependent upon mobility of propagules, which
Even marine phytoplankton is readily trans- is, in turn, a function of propagule size. Then
portable via other agents. The recent diminution their relative transmissibilities are subject to the
in the area and standing volume of the Aralskoye influence of distance and physical barriers. Moun-
More, as a result of disastrous mismanagement of tain ranges and oceans may constrain the dis-
its hydrology over the past few decades (see e.g. tributions of aquatic invertebrates and insects.
Aladin et al., 1993) has been accompanied by a Fish experience similar difficulties, even on a
rise in its salinity. In the 1990s, by the time that catchment-to-catchment basis. Aquatic mammals
the salinity had reached 28–30 mg L−1 , productiv- spread to places that they can walk or swim to.
ity had collapsed and all freshwater and brackish Finlay’s (2002) hypothetical separation of protist
species of phytoplankton had been eliminated. species that are ubiquists from those that have
According to Koroleva (1993), the depleted plank- biogeographies on the basis of their sizes (criti-
tic flora included marine species of Exuviella, Pro- cal range, ∼1 mm) appears to hold for all pelagic
rocentrum, Actinocyclus, Chaetoceros and Cyclotella. organisms.
Species richness of nascent communities number of niches is a reasonable assumption.

Planktic colonisation of new habitat or the Fitted to species lists of macroscopic flora and
reopening of existing habitat through the sea- fauna of existing islands, Eq. (7.3) has been found
sonal relaxation of physical exclusions may be to account for up to 70% of the variability in
expressed in quantitative terms as the filling of species number. The slope, z, is reciprocal to
previously unoccupied niches, or the opportuni- the immigration rates and, in these cases, usu-
ties available for the exploitation of the aggre- ally solves at between 0.2 and 0.35. Similar val-
gate pool of resources. The more that the niches ues have been obtained when the solution has
become occupied, the slower is the process of been applied to the occurrences of phytophagous
filling. MacArthur and Wilson (1967) formulated insects on their food-plant ‘islands’, though val-
a differential equation (7.2) to describe colonisa- ues are conspicuously influenced by the abun-
tion, thus: dance (thus mutual proximity and accessibility)
of hosts (Janzen, 1968; Lawton and Schröder,
1/n(t)dn/dt = û(ñ − n) (7.2) 1977).
Smaller values of z (0.1–0.2) represent
where n(t) is the number of exploitable niches in
enhanced opportunities for immigration. Inter-
the location at a time t, ñ is the number reached
phyletic differences in immigration rates to given
at t = ∞ and û is the rate at which they are filled.
locations are also apparent. For instance, birds
Practical solution of the equation is not possi-
have lower z values than land snails. Finlay et al.
ble without independent quantification of the
(1998) have worked out that the slope z for fresh-
number of potential niches (ñ) and experimental
water planktic ciliates is yet smaller again (z =
observations on the arrival of occupant species
0.043) as a consequence of their rapid rates of
(û). Occupancy is asymptotic to ñ but its value is
dispersal.
difficult to predict from observation, especially
If a low value of z in Eq. (7.3) reciprocates
in the early stages of colonisation, when the
a high initial dispersive value of û in Eq. (7.2),
resource availability (S) is likely to be large in rela-
the potential richness of planktic species may
tion to the exploitative demand (D). Thus, what
approach the maximum (ñ) that the habitat-
may be, effectively, a single broad niche may be
size constraint will allow. Certainly, in those
simultaneously and non-competitively exploited
instances where investigations of species struc-
by several species, which together act as a multi-
ture has been sufficiently competent and exhaus-
species guild, or functional group, of species
tive, the species richness of individual lake sys-
(Tokeshi, 1997). However, it is readily predictable
tems is demonstrably large. Dumont and Segers
that the number of distingushable and viable
(1996) estimated the likely maximum species
niches is likely to be a function of the size of the
richness of cladocerans to be ∼50 and of rotifers
habitat (a large lake offers more colonist oppor-
to be ≤270. Representation by a significant pro-
tunities than, say, a pool in a bromeliad). This
portion of the extrapolated global total of ∼3000
is back-extrapolable from another prediction of
species of free-living ciliates is probable (Finlay et
island biogeography, pointed out by MacArthur
al., 1998). Compared to a supposed potential of
and Wilson (1967) and supported empirically in
4000 to 5000 taxa, 435 named species of fresh-
several subsequent investigations, that there will
water phytoplankton in tropical Lake Lanao were
be a positive correlation between the species rich-
noted by Lewis (1978). Consecutive daily sam-
ness of an island and its size. The number of
ples of the phytoplankton of shallow Lake Bal-
species (nsp ) to be found in a defined area, A, con-
aton accumulated a list of 417 separate taxa
forms to the relationship
in 211 days (only ∼300 were considered to be
nsp = k A z (7.3) truly planktic: Padisák, 1992). The accumulated
species list of phytoplankton encountered in Win-
where k and z are characteristics of particular dermere (Reynolds and Irish, 2000) includes 146
categories of organisms and invasiveness. The named taxa. In samples of phytoplankton from
existence of at least an approximately equivalent the upper 100 m of the central North Pacific
over a 12-year period, Venrick (1990) recorded 245 organic carbon, which is powered ultimately by
species. short-wavelength, photosynthetically active solar
energy. The role of the phototrophic primary pro-
7.3.2 Properties of accumulating ducers is to intercept, harvest and invest energy
communities in the production of high-energy organic carbon
Following the establishment (or re-establishment) compounds. This process is continuously subject
of an assemblage of colonist primary producers, to the laws of thermodynamics. The second of
many subsequent events are set in train. Fur- these determines that the direction of flow is
ther species of primary producers may arrive and from the concentrated to the diffuse and that
respond positively to the favourable growth con- short-wavelength energy is dispersed irrecover-
ditions. Total biomass increases and individuals ably at longer wavelengths (i.e. as heat). Short-
begin to impinge on each other’s environments, wave electromagnetic solar radiation falling on
inevitably setting up interspecific interactions. a lifeless planet causes molecular excitation and
Moreover, the accumulation of producer biomass, dispersion (a rise in the surface temperature) but
detritus and organic waste soon constitutes food most is back-reflected to space at longer wave-
and substrate for the heterotrophs (phagotrophic lengths. This dissipation of heat is irreversible.
herbivores, microbial decomposers) and the open- It follows that the effect of a pulse of energy is
ing of primary product as a resource to consumer to reduce order and increase randomness, or, in
components of the DEU (Ripl and Wolter, 2002); thermodynamic terms, to raise the entropy of the
the functional community is born! recipient structures.
Several aspects of this development command Against the inevitability of entropic dissi-
our attention. Intercepting an increasing por- pation, the ability of growing communities to
tion of the solar flux, the accumulating sys- assemble structure, order and complexity seems,
tem involves the deployment of greater levels at first sight, to be in contravention of the laws
of power and resources. Sooner or later, con- of the universe (Prigogine, 1955). In truth, ecosys-
straints in the supply of one or the other feed tems are an integral part of the thermodynamic
back, invoking adaptive reactions in the pat- system, doing little more than to reroute and reg-
tern of community assembly. In turn, there ulate the velocity of the dissipative flux. A reason-
are consequences for species composition, rich- able analogy is that of a waterwheel located on
ness, dominance and diversity, and key species a stream (Reynolds, 2001a). Some of the kinetic
thus favoured characterise the overall commu- energy of flow is drawn to drive the wheel
nity function. These developments are some- (the energy-harvesting primary producers), whose
times referred to collectively as community ‘self- motion may be exploited as a source of alterna-
organisation’ (or autopoesis: see, for instance, Pahl- tive mechanical or electrical power (the ecosys-
Wostl, 1995). They are, indeed, contingent upon tem). However, the overall direction of the flow is
behaviours and mutual interactions at the basic unaltered.
levels of community assembly (the predilections Moreover, the proportion of the available
and adaptabilities of the individual participating power that is realised in assembled biomass is
species) and the aggregate of which is the struc- really quite small. Chlorophyll harvests no more
ture and future of the emerging community. The than ∼6% of the radiative flux to which it is
process of increasing system throughput, infor- exposed (at sub-saturating levels); barely 1.5–2%
mation and complexity of network flows is also of the available energy is captured in carbon fix-
known as ascendency (Ulanowicz, 1986) (see also ation (Section 3.5.3). Much the largest propor-
Box 7.1). tion (≤95%) of the solar flux is consumed in
non-biological (but often biologically important)
Ascendency: power and exergy processes, such as heating the water and driv-
These concepts are not difficult to grasp if the ing evaporation and cooling. The investment of
appropriate analogies are used. Biological sys- the photosynthetically active radiation actually
tems are organised about the managed flow of absorbed into plant biomass, typically ≥0.08 mol
the community, the DEU or the ecosystem, these

Figure 7.10 Ecosystems and the dissipative energy flux. In
an abiotic world (a) all the incoming short-wave radiation, Qs ,
structures dissipate energy, just as surely as the
is reflected or reradiated at longer wavelengths (i.e. as heat). abiotic environment, and, ultimately, they do so
This entropic disssipation is noted as dS and, in (a), is external to the point of balanced exchanges.
to living organisms (hence dSe ). The intervention of The apparent affront to the second thermody-
photoautotrophic organisms intercepts and reroutes a namic law is that the inward flux may not be dis-
fraction of the solar flux to drive the assembly of biogenic sipated at once but, instead, be retained within
material. Much of this eventually shed in respiration and molecular bonds. While these persist, they block
decomposition (as dSi ), in conformity with thermodynamic
entropy. In effect, the biological system is siphon-
law, but only after the captured energy has been invested in
ing off a part of the energy flux into a loop of
the assembly and accumulation of biomass. This flux is
equivalent to system exergy. A positive exergy flux enables
reduced carbon, whence its eventual release is
ecosystems to build, perhaps to the point where all the entire regulated and delayed. This entropy-free poten-
solar flux is dissipated as dSi . B, biomass. Redrawn with tial has been variously referred to as negen-
permission from Reynolds (1997a). tropy (for negative entropy) or exergy (Mejer and
Jørgensen, 1979). Equally, it has been conceptu-
C (mol photon)−1 , or 0.2 MJ (g C)−1 , again at alised in different ways, in terms of informa-
sub-saturating fluxes (Section 3.2.3), implies that tion stored (Salomonsen, 1992) or the size of the
the power requirement to assemble producer accumulated gene pool (Jørgensen et al., 1995;
biomass is around 20 MJ (g C)−1 . There is a further Jørgensen, 1999).
energetic cost to assembling and maintaining Here, the preference is to adhere to expres-
the structure, which is met from the controlled sion in units of energetic fluxes (cf. Nielsen, 1992).
metabolic re-oxidation (R) of part of the accumu- Then, one of the preconditions for community
lated carbon. When producer biomass becomes ascendency is that the rate of energy harvest-
part of the biomass of its consumer, some 40–90% ing and deployment in organic carbon synthe-
of the power investment may be dissipated as sis shall exceed the aggregate of the require-
heat. This is repeated at each successive trophic ment of internal maintenance and its dissipa-
level. Alternatively, the cadavers and waste prod- tive losses. The cartoon in Fig. 7.10 illustrates the
ucts may be exploited by decomposers for the increasing biomass of an ascendent community
last molecules of organic carbon and the last ves- through time, from bare ground to high forest.
tiges of the power investment. Whether we take From wholly dissipating the flux of short-wave
radiation (Qs ) as heat (units, W m−2 or J m−2

d−1 ), the intervention of biological systems is to
reroute part of the dissipative flux (dS) into the
building (P) and maintenance (R) of biomass (B).
That part of the abiotic flux that is now passed
through organisms is now distinguished as (dSi ),
leaving a reduced portion (dSe ) of the original abi-
otic flux external to the organisms. For B to accu-
mulate, the sum (dSi ) + (dSe ) has to be smaller
than the original flux, dS. As biomass is accu-
mulated, R increases absolutely. Although more
energy is siphoned off, absolutely more (and,
eventually a bigger proportion) is dissipated as
heat (dSi ). The achievable steady state is when the
biotic component harvests the entire energy flux
(dSe → 0) and consumes it exclusively in its own
maintenance (R = dSi → dS). No further increase
is then possible: the community has achieved its
highest state (Fig. 7.10c).
Figure 7.11 Graphical representation, based on Fig. 3.18,
Data are not available to quantify all the rele-
of the relationships sketched in Fig. 7.10. The vertical axis
vant fluxes. However, the principles can be illus-
represents the incoming short-wave radiation. The curve
trated by adopting the light-harvesting proper- represents the proportion that is ‘siphoned off’ into primary
ties and phototrophic growth of the alga Chlorella production, thereby (temporarily) reducing the external
(from Fig. 3.18) to scale the theoretical ability dissipative flux (hence, –dSe ). The efficiency reduces with
(and diminishing efficiency) of the producers to increasing mutual interference of the producer
harvest and allocate the maximum energetic flux. light-harvesting but whose the maintenance costs (dSi ) are
The revised plot, in Fig. 7.11, shows how low, supposed to be proportional to the biomass. The energy
pioneer levels of biomass harvest more energy difference between the curves (−dSe + dSi ) is available to
support growth, reproduction and the ascendency of the
than they expend and that the resultant posi-
system. The difference between the harvested energy and the
tive exergy flux can be allocated to increasing
internally dissipated flux of a system is equvalent to its exergy
the energy-intercepting biomass. The size of the flux. For more information see text. Redrawn with permission
waterwheel paddles has been increased! Main- from Reynolds (1997a).
taining more biomass carries a higher cost (it
is shown as a linear function of biomass in Fig.
7.11) but, as more light-harvesting centres that are relatively most capable of overcoming the
are placed in the light field begin to shade each effects of small inoculum sizes. Once again, r-
other out, the harvesting return per unit invest- selected C-strategist species can be seen to hold a
ment declines asymptotically. The limiting con- fitness advantage in aspiring to dominance.
dition comes when the cost of maintenance can
no longer be balanced by the harvested income Ascendency: environmental constraints
and the exergy flux falls to zero. Some changes in community metabolism and
Long before that condition is reached, ascen- resource partitioning through a phase of biomass
dant accumulation is led by the recruitment accumulation in the Blelham enclosures (Section
and expansion of species that contribute most 5.5.1; Fig. 5.11) are illustrated in Fig. 7.12. The
strongly to the aggregate biomass. These are not large panels summarise the variations in the
necessarily the fastest growing (cf. Fig. 5.19) but, main contributing species to the phytoplanktic
in the early stages of accumulation, the greatest biomass in the enclosure (Fig. 7.12b, as biomass
rates of expansion are likely to come from the carbon) and their rates of population change
fastest-growing of the available pool and which (Fig. 7.12c: shading separates the reconstructed
Figure 7.12 Production and dynamics of phytoplankton recruitment in a Blelham Enclosure (A, 1983), during a period (a) of
weak mixing (vide zm ) and following high transparency (zeu ) and artificial fertilisation with nitrogen and phosphorus. Changes in
the biomass of prominent species of phytoplankton are shown in (b) on a common scale of carbon concentration. The rates of
net population change (plus or, below the horizontal line, minus) are shown in (c), relative to reconstructed growth rates, after
correcting net rates for grazing losses. Changes in the net daily photosynthetic rates, measured directly (P) and as the rate of cell
recruitment by growth, before grazing and sinking (Pn ), each specific to the upper 5 m of water, are shown in (d) (curves fitted by
eye). Changes in the standing biomass (B) are shown, and similarly summarised in (e). Productivity trends (P/B, and especially,
Pn /B), are shown in (f). Changes in the concentrations of SRP and DIN (= nitrate + ammonium nitrogen) (g) reflect uptake by the
phytoplankton. Changes in the stock of phosphorus (TP) in the water column and its apportionment between soluble (SRP) and
particulate (PP) components are shown in (h) together with the aggregate of phosphorus collected in sediment traps. Original data
of Reynolds et al. (1985), as reworked and presented in Reynolds (1988c) and redrawn with permission from Reynolds (1997a).
growth rate from net rate of positive or nega- irradiance, zeu ). At the end of June 1983, the onset
tive change), in relation to day-to-day variations of a phase of very stable thermal stratification
in the depth of the mixed layer (zm ) and the depth coincided with the rapid depletion in the dom-
of the conventionally defined photic zone (to 1% inant populations of Planktothrix (sinking and
penetration of the immediate subsurface visible photooxidation) and Cryptomonas (to grazers) and
an attendant increase in transparency. The enclo- balance or exceed the rates of its recruitment,
sure was then well fertilised with solutions of but shows well the increasing interference of the
nitrate, phosphate and silicate. The responses of a major constraining forces on energy-harvesting
now-diminished phytoplankton to this relatively and resource-gathering with the process of
sharp increase in carrying capacity was followed recruiting new biomass. These effects carry other
over the next two months, after which cooler interesting hallmarks of autogenic change in
and windier conditions obtained. The changes to developing communities. The change in compo-
the community during July and August are pre- sition, first in favour of strongly r-selected C-
sumed to have been substantially autogenic (self- strategist species, such as Ankyra, towards slower-
imposed), working within the limits of the con- growing, strongly K-selected, biomass-conserving
straints of the carrying capacity of the light and S-strategist species such as Microcystis, is reflected
nutrient resources. in the changing size spectrum of the popula-
At the metabolic level, gross volume-specific tion (Fig. 7.13a). During early ascendency (take
primary production (P, in mg C L−1 d−1 ) as indi- the window for 15 July), the major part of the
cated by conventional radiocarbon fixation rates biomass resides in the smallest organisms: vir-
(data of Reynolds et al., 1985) and phytoplank- tually all are individually smaller than 104 µm3 ,
ton biomass (B, as mg C L−1 ) increased over the and over 50% of them are each less than 102 µm3 ).
two-month period (Fig. 7.12d,e). However, produc- Towards the end of the accumulation phase (9
tion net of respiratory losses (Pn ) and productiv- August and persisting until October), the major
ity (sensu production normalised to biomass, P/B fraction is made up of units each >105 µm3 .
and, especially, Pn /B) decreased (Fig. 7.12 d, f). The Moreover, the distribution of biomass among
increase in biomass took place at the expense of individual species also declines, with an increas-
the nutrients in solution (Fig. 7.12g). Removal of ing proportion of the primary productive poten-
dissolved phosphorus, ostensibly into phytoplank- tial of the community residing in the assembled
ton, was effectively complete by early August, biomass of a decreasing number of species (Fig.
while the general decline in the total phospho- 7.13b). The tendency is towards 1, as the best-
rus fraction is explained by the sedimentary flux adapted species in the pool progressively outper-
of planktic cadavers (mostly of the animals that forms, to their eventual exclusion, all the oth-
had grazed on the phytoplankton). ers competing for the same, diminished resource.
At the compositional level, the dominance of This is an issue not of species richness but of
the phytoplankton changed twice – the leading species diversity, to which the discussion will
contender for the opening resource spectrum was return (see p. 364).
Ankyra, whose increase in biomass was initially
supported by rapid net specific rates of increase Succession
of ∼0.86 d−1 (reconstructed cell replication rates, Distinctive patterns in the developing meta-
r ∼1.1 d−1 ). These were not sustained and the bolism and community organisation of major
population was reduced by zooplankton feeding, biological systems, manifest as an ordered
especially by filter-feeding Daphnia. The second sequence (or succession) of substitutions of
dominant was the non-motile colonial chloro- species, have long been recognised by plant
phyte Coenochloris, whose net rate of increase botanists and geographers (Margalef, 1997). Early-
(≤0.43 d−1 ) was less sensitive to grazing but twentieth-century plant ecologists were espoused
nevertheless weakened as zeu diminished in rela- to the idea of an internally controlled process,
tion to zm . Microcystis increased steadily through culminating in a final, climactic stage of max-
July and early August (≤0.24 d−1 ), achieving dom- imal persistent biomass. Moreover, some well-
inance in place of Coenochloris. Its biomass per- characterised successions (for instance, from bare
sisted, even though there was almost no net soil to high forest: or the hydroseral stages of
recruitment from phytoplankton growth. vegetation development at the edge of a lake,
The example illustrates not just the increas- passing from swamp to marsh or fen through
ing burden of accumulating biomass, as loss rates to Quercus forest: Tansley, 1939) have sharpened
Figure 7.13 (a) The changing size distribution of changes wrought by the activities of individual
phytoplankton in Blelham Enclosure A, traced through the organisms, to the extent that they alter the envi-
allocation of the total live biomass among units (cells, ronment so that it becomes more amenable to
coenobia or colonies) in the size categories shown. Note the the establishment of individuals of other species,
predominance of small cells (mostly of Ankyra) on 15 July, are, in broad terms, essential and predictable
during the early part of the sequence depicted in Fig. 7.12,
aspects of succession (Tansley, 1939). ‘Succes-
and the predominance of larger algae later in the year. (b) The
sion’ should be reserved to refer to these wholly
changing apportionment of biomass among species, moving
from being invested in the first two to four co-dominants to
autogenic responses of the community (see also
being almost wholly invested in the first. Original data of Reynolds, 1984b).
Reynolds (1988c) and redrawn with permission from Successional changes, contingent upon auto-
Reynolds (1997a). genic drivers, are the principal manifestation
of ascendency in developing ecosystems (Odum,
a resolve to find deterministic explanations. This 1969). We may accept ‘autogenic succession’ and
quest has had to reject some well-intentioned but its attributes, as circumscribed by Odum (Table
erroneously rigid statements of successional gov- 7.7), as being symptomatic of ascendency. Sev-
ernance. At the other extreme, understanding of eral of the stated attributes plainly apply to the
succession has not been helped by the use of this events depicted in Fig. 7.12. The ‘causes’ of suc-
term by students of the pelagic to refer to all cession are the reactivity and interactions of the
temporal changes in the species composition and pool of the species to the environmental oppor-
the abundance and relative dominance of the tunities presented. It is these that may eventually
plankton (Smayda, 1980). The fact that succession lead to structures that, subject to the turnover of
‘conforms to no plan’ and is mostly concerned resources, achieve a steady state of energetic bal-
with the relative probabilities of certain possi- ance, to which the notion of successional climax
ble emergent outcomes (Reynolds, 1997a), might is fully applicable.
overcome the need for a precise explanation of
its mechanics. Filtration and community assembly
However, some aspects of the concept of suc- Where strict successional selection fails to
cession are too valuable to reject. Change in com- explain the mechanisms underpinning repro-
munity structure, in the plankton as in systems ducible sequences, then we need to invoke
involving higher plants and animals, has a num- another model. The opposite view from posi-
ber of quite separate drivers. These may owe to tive selection invokes the premise (see p. 351)
allogenic, non-biological, physical forcing mecha- that most species can grow anywhere they arrive
nisms (earthquake, fires, storms and floods) that at, provided the habitat is suitable and can
destroy structure or modify the environment in sustain their growth and survival needs. Then,
favour of tolerant species. On the other hand, the habitat that is, or becomes, unsuitable will
Table 7.7 Attributes of early and maturing stages of autogenic succession
Ecosystem attribute Early stage Maturing stage

Community energetics
1. Gross production/community respiration (P/R) (P/R) >1 (P/R) → 1
2. Gross production/biomass (P/B) high low
3. Biomass supported/unit energy income (B/Q) low high
4. Net production yield (Pn ) increasing low
5. Food chains linear web-like
Community structure
6. Total organic matter small arge
7. Inorganic nutrients extrabiotic intrabiotic
8. Species diversity rising high, possibly falling
9. Species equitability decreasing low
10. Biochemical diversity low high
11. Structural diversity poorly organised well organised
Life histories
12. Niche specialisation broad narrow
13. Organismic size generally small generally large
14. Life cycles short, simple long, complex
Nutrient cycling
15. Mineral cycling open substantially closed
16. Exchanges among organisms and environment rapid slow
17. Role of detritus in nutrient unimportant important
Selection
18. Growth selection r K
19. Production for quantity for quality
Community homeostasis
20. Internal symbiosis undeveloped developed
21. Nutrient conservation poor good
22. Response to external forcing resilient resistant
23. Entropy high low
24. Information low high
Source: Based on Odum (1969).
simply select against the persistence of intoler- Then the organism can survive to the limits of its
ant species. In this way, environments act as a own adaptive capabilities. These may be superior
sort of filter, separating off the ill-adapted species or inferior to those of the next species and these
from those whose traits and adaptations allow may well determine which of them can continue
them to pass to survival. For instance, the poten- to function under persistently low oxygen con-
tial of aquatic habitats to support aerobes is centrations. Pelagic environments pose many and
constrained by a limited, finite supply of oxy- more subtle conditions and constraints to their
gen. To be able to live in water, organisms are exploitation by aquatic organisms and the puta-
required to be functionally or physically adapted tive systems to which their presence contributes.
to ensure that diffusion gradients are adequate These may well operate simultaneously, as filters
to the needs that their size and activity dictate. of varying coarseness, to favour or disfavour the
Table 7.8 The rules of community assembly in the phytoplankton, proposed by the International
Association of Phytoplankton Taxonomy and Ecology (see Reynolds et al., 2000b) with minor modifications
(1) Provided suitable inocula are available, planktic algae will grow wherever and whenever they can
and to their best potential under the conditions obtaining.
(2) Then, of those present, the species initially likely to become dominant are those likely to sustain
the fastest net rates of biomass increase and/or to be recruited from the largest inocula (‘seed
banks’).
(3) The largest autochthonous inocula (seed banks) are furnished by species that have been abundant
at the location in the recent past.
(4) Environments may select, preferentially and with varying levels of intensity, for certain specific
organismic attributes or traits.
(5) Species with preferred attributes are likely to build bigger populations than those that lack them
and, where appropriate, to found larger inocula to carry forward.
(6) Plankton assemblages become biassed in their species composition by the conditions typically
obtaining in the host water body.
(7) The species most frequently present in specific environments share common suitable attributes.
(8) The outcome of assembly processes may be subject to the food web and to other interspecific
interactions.
(9) Of those species present (and quite independently of the initial conditions), the ultimate dominant
is likely to be the one most advantaged by its adaptive traits.
(10) Assembly is always subject to the overriding effects of environmental variability and the resetting
of the assembly processes.
relative abundances of each of the given species. of others (see for, instance, the papers of Belyea
The most decisive of these criteria becomes the and Lancaster, 1999; Brown, 1999; Straškraba
finest of the filters operating, selecting for the et al., 1999). Here again, the value of the short
fittest of the species available (Reynolds, 2001a). timescales of community processes in the pelagic
In this way, the fortuitousness of the species have permitted the substantial verification of the
composition of the various functional compo- role of trait selection in the development of
nents of a community and the stochastic nature pelagic communities (Rojo and Alvares Cobelas,
of its early dominance is squared against the 2000). A definitive set of rules guiding the assem-
increasing predictability of the outcome of auto- bly of communities in the plankton has been pro-
genic succession. The mechanisms have yet to be posed by the International Association of Phyto-
fully elaborated. My use of the word ‘filtration’ plankton Taxonomy and Ecology (see Reynolds et
(Reynolds 2001a) was inspired by the reference al., 2000b) (see also Table 7.8). These repeat some
of Lampert and Sommer (1993) to ‘environmen- statements offered earlier in this chapter.
tal sieving’. However, the crucial role of specific It has already been established (Chapters 5
organismic traits in, as it were, determining the and Sections 7.2.2 and 7.2.4) that the principal
filterability of species that may then participate correlatives of community composition are clas-
prominently in community function is provided sifiable as resource constraints and energy con-
by the seminal work of Weiher and Keddy (1995; straints (e.g. Fig. 7.8). It is now a simple exer-
see also Weiher et al., 1998). Their formulation of cise to suggest the assembly mechanisms under-
the guiding principles and rules governing com- pinning the distribution of trait-distinguished
munity assembly has attracted the concurrence functional groups against these constraints. The
(shown by hatching) of most planktic species

present. Thus, they furnish an ascendent oppor-
tunity, unconstrained save by their own exploita-
tive capabilities. Interspecific competition for the
opportunity is weak but, in accord with Rules
1–3 (Table 7.8), the species that become most
prominent will be the ones generating the high-
est exergy and investing in the fastest relative
rates of growth. Working downwards to Fig. 7.14b,
reduction of the base of the potentially critical
nutrient to growth-limiting levels (Section 5.4.4)
demands traits of high uptake affinity for the
limiting nutrient, or the ability to exploit other
less-accessible resources, which may be shared
by relatively few of the contesting species. Once
nutrient is exhausted from the upper column,
the only exploitable resources are located deep in
the light gradient (Fig 7.14c), requiring yet greater
Figure 7.14 Idealised environments, defined in terms of specialist organismic traits to be able to use them
downwelling solar irradiance (I) and the capacity-limiting and, therefore, fewer species able to compete for
bioavailable resource (K) in relation to mixed depth
them (Rules 5, 6 and 9).
(represented by the temperature gradient (θ ). The potential
Working rightwards in Fig. 7.14, the light-
biomass and its distribution is shown by shading. In the top
left diagram (a), resources and, at the top of the water
harvesting opportunities are diminished by weak-
column, growth is saturated at the maximum potential of the ening stratification and increased mixing depth.
producers present. Working downwards, many water Light-harvesting opportunities are diminished in
columns correspond to (b), wherein resources are so scarce consequence, although they are extended and
or depleted that they constrain the productive potential, renedered more uniform through the entire
possibly as far as to the metalimnion (c). Working rightwards, mixed layer (Fig. 7.14d). Ascendent biomass also
deeper mixing (d) relative to diminishing light penetration (f) impinges increasingly upon light penetration:
sets the major functional constraint. Most systems probably
increased depth of mixing and raised turbidity
fluctuate around a condition close to (e) but the inference of
levels impose greater adaptive efficiency of the
deep mixing and nutrient deficiency (g) as being untenable to
primary producers is clear. Based on an original figure of light-harvesting apparatus and enhanced ability
Reynolds (1987b) and redrawn with permission from still to cream off the remaining exergy genera-
Reynolds (1997a). tion into biomass accumulation (Fig. 7.14f).
It has to be said at once that these are ide-
alised states, none of which obtains constantly.
representation of pelagic habitats shown in Fig. Real systems show variability in both directions
7.14 is now well travelled and a little dated but it and may hover around conditions represented by
serves to illustrate how temporal habitat changes Fig. 7.14e However, the inference of simultane-
in light and nutrients intensify species filtration ous segregation of the nutrient base from the
of species traits. The individual subfigures (a–g) energy-harvesting opportunity is supports only a
making up Fig. 7.14 represent vertical profiles of desertified habitat (Fig. 7.14g), corresponding to
light availability (I) and nutrient concentration temperate oceans in winter (Fig. 1.8).
(K), in relation to temperature (θ , included in this The arrangement in Fig. 7.14 also picks up the
instance only as a correlative of structure and ver- layout of the C–S–R triangle, serving to amplify
tical mixing). In the first instance (Fig. 7.14a, in both the conditions and (for fresh waters) the
the top left-hand corner), nutrients and, at least notional ranges of some of the trait-distinguished
near the surface, light, are shown to be capable of functional groups (cf. Figs. 7.8, 7.9). For systems
saturating the immediate growth requirements that are chronically resource-deficient (K∗∗ always
Figure 7.15 The impact on ecosystem energetics of

chronic resource deficiency and low energy income, based on Species richness, diversity and evenness in
their representation in Fig. 7.11. The basic layout is shown as assembling communities
being constrained by axes delimiting low or variable Recalling the representation of seasonal varia-
resources or energy income, which, respectively reduce the tions in environmental selectivity for a given
exergy to the smaller shapes in (b) and (c). The preferential individual water body (Fig. 7.9), the C–S–R trian-
trend towards S or R dominance is predicted. Redrawn with gle can be used to match the successional tra-
jectories of favoured species (or at least their
respective trait-selected groups) to the extent and
low) or, equally, in which harvestable light (I∗∗ ) is direction of environmental variation. Some well-
habitually diluted by mixing or by turbidity, the documented pathways are plotted in this mode
selective bias is strongly pressed against C strate- in Fig. 7.16, relating compositional changes to
gies and towards those of either S or R species. the drift in nutrient and light availability con-
Moreover, the bias of inocula serves as a positive sequential upon community ascendency. This is
feedback to influence the floristic composition of not prescriptive but purely an example of the
the respective habitats. kinds of self-imposed ‘decision points’ that may
The link to the assembly of communities in steer the assembling community towards par-
these habitats is also simply represented through ticular outcomes. Starting at a point X, repre-
reference to Fig. 7.12. If we take the segment senting initially non-limiting nutrients and light,
enclosed by the curves representing potential the assembling community is more likely than
harvest over cost as the potential for ascendent not to be founded upon the primary production
investment and subject it to persistent but vari- of relatively fast-growing, r-selected nanoplank-
able constraints in either energy or resource, tic species of one or other X algal associations.
two new shapes may be derived income (Fig. Early dominance of these organisms is not bound
7.15a). Chronic resource limitation of supportable to collapse: two possibilities for perpetuation are
biomass lessens the likelihood of frequent limi- included in Fig. 7.16, where dilution by nutrient-
tation of growth by harvestable light (Fig. 7.15b). rich throughflow (‘FLUSH’) and avoidance of con-
High exergy can be maintained by species that sumers (‘NO GRAZE’) permit the endurance of the
cope with resource limitation (and are highly sen- status quo, provided nutrients and light remain
sitive to altered loadings of critical nutrient). For freely available. Modest, delayed or selective graz-
systems that are energy-limited, the biomass is ing in water columns sufficiently deep to be
unconstrained by nutrient whereas it can rise to subject to self-shading may operate in favour
the level of the limit of the exergy harvest by the of self-regulating cryptomonads or (eventually)
most efficient users of the available light (Fig. of efficient energy harvesters such as attenu-
7.15c). The respective selective biasses in these ate diatoms and filamentous Cyanobacteria. The
extremes favour S- and R-strategist species. greater becomes the light constraint, the more
Figure 7.16 Some pathways in

the development of phytoplankton
composition, with some key
‘decision nodes’. Starting at X, the
predicted initial establishment of
r-selected, invasive C-strategists may
be perpetuated by flushing or may
otherwise persist if there is no
grazing. Early imposition of grazing
may force the dominance of
sequences of larger and, eventually,
more versatile phytoplankters
(Eudorina → Ceratium); otherwise,
diminution in light availability pushes
dominance towards energy-efficient
R-species. If resources are not
replaced, or are chronically
deficient, sequences move vertically
towards specialist resource-gleaners
and, possibly, to the dominance of
picophytoplankton. The enclosure
within a triangle ties these temporal
patterns to the habitat template.
Reynolds (1997a).
intense is the competition for the available pho- pressure in favour of the late successional par-
tons, the tighter is the filter and the less is the ticipants and against the weaker competitors for
diversity of still-functioning species. The model the critical resource. The system moves towards
reconstructions of increasing light constraints on a steady state, because its rate of building and
the comparative growth rates of Chlorella, Asteri- the cost of its maintenance are balanced and,
onella and Planktothrix also amply predict this out- for so long as the constraints remain, no bet-
come. ter competitor is available to contribute new
Alternatively, the early onset of heavy grazing exergy. The community has, in fact, achieved
is likely to work against self-shading and directly its potential, analogous to its successional cli-
in favour of larger algae. Comparative growth max (Reynolds, 1993b, 1997a; Naselli-Flores et al.,
rates, inoculum size and the relative intensity 2003). The outcome is not altogether random:
of developing constraints in the resource sup- the overwhelming dominance of the best com-
ply (phosphorus, nitrogen and carbon flux) each petitor for the most severe constraint may be
influence the compositional trends among these anticipated, in part, from a knowledge of the
algae (whether as a consequence of high affinity properties of the habitat (deep or shallow, hard-
for phosphorus, the metabolic flexibility to fix or soft-watered, phosphorus-rich or phosphorus-
nitrogen or to resort to mixotrophy, mobility to poor). However, the outcome is fashioned entirely
glean or scavenge the resources available or the by the assembly process. Community develop-
ability to enhance carbon uptake). The greater ment pushes the selective pathway towards one
becomes the particular resource constraint, the or other of the selective apices (S or R in Fig. 7.9),
more important are the specific means of its where the growth of only a few specialist-adapted
retrieval. Again, the filter tightens and the num- species remains possible. It is as if the richness
ber of (increasingly S-strategist) species still able of species potentially able to contribute to the
to function diminishes. In this way, the develop- late stages of community assembly is increasingly
ing interactions increasingly exert the selective ‘squeezed out’ as the apices are approached. The
patterns analysed in Section 7.2 and, especially, have acknowledged, will be extremely rare. Thus
among some of the mesotrophic and eutrophic the value of s becomes partly a function of the
lakes (Section 7.2.4) provided several examples of diligence of search and it is likely to be influ-
quasi-steady states of extended dominance by one enced by the taxonomic competence of the asses-
(or a very small number of) persistent species. sor. Students of phytoplankton elect to adopt a
These are, moreover, considered to be typical for cut-off (generally, though quite arbitrarily, that
the particular conditions in particular groups they will include only those species each con-
of lakes. These include the sequences culminat- tributing >0.5% of B) that immediately weights
ing in long phases of dinoflagellate dominance the diversity indices they quote against the rare
of nutrient-depleted epilimnia (Peridinium in Kin- species in their original samples. This being so,
neret, Ceratium in Rostherne Mere); the forma- the valid measure that should be sought is the
tion of deep chlorophyll maxima in stable metal- equitability (or evenness) of the species represen-
imnia (as Planktothrix rubescens in Zürichsee); the tation (Es ), which approximates to:
association of dominant populations of Anabaena
spp. or Cylindrospermopsis with hydrographically E s = H /Hmax

(7.5)
stable periods in N-deficient tropical lakes (see
p. 344); and the near-perennial dominance of Scores for equitability are rarely quoted, partly
highly enriched, shallow polder lakes by Plankto- because the difficulties of estimating s persist.
thrix agardhii and/or Limnothrix redekei (pp. 345, Thus, the calculated diversity indices (H ) that are
349). usually quoted provide only a proxy estimate of
There may be little doubt about the domin- the course of temporal changes or between-site
ance but competitive exclusion of less-fit species comparisons of community organisation. There
by those more adaptable is rarely so complete have been other, less well-used approaches to
that small numbers of excluded species or (espe- capturing mathematically the drift in species
cially) their propagules do not persist within the equitability. For instance, Margalef’s (1958) index
system for many years. Here, they constitute a of diversity, Hs , simply relates the number of
sort of system ‘memory’, whence they may, fol- species (s) to the total number of individu-
lowing some future environmental modification, als (N). Applied to a fixed size of sample (vol-
have the opportunity to reassert their abundance ume of water), the diligence-of-search criterion
(Padisák, 1992). Species richness is thus a less sen- is cut off in an arguably less arbitrary way. How-
sitive barometer of the structural organisation ever, it has a transparent sensitivity to increased
of communities through succession than is the biomass and the number of species that are
relative distribution of the biomass among the common. The units are bits (of information) per
species present. individual:
Ecologists express the diversity and equitabil-
ity (or evenness) of species assemblages using Hs = (s − 1)/ loge N (7.6)
terms borrowed from Shannon’s (1948) theory
of information. Pielou (1975) proposed that the Margalef’s index is used to construct the quan-
most useful estimate of biotic diversity (H ) tified analogy to the changes in species even-
comes from the specialised form of the function ness through the summer accumulation phase
(from Shannon and Weaver, 1949): in a Blelham enclosure (Section 5.5.1) (Fig. 5.11)
and the progressive onset of Microcystis domi-
H = −bi /B log2 (bi /B ) (7.4)
nance through exclusion, shown in Fig. 7.17.
where B is the total biomass, and bi is the biomass The observed changes in contributory specific
of the ith of the s species present. H increases biomass (shown in the upper part of the figure)
with the number (richness) of species. Thus, we and the rates of community change (σ s ), calcu-
should expect the biotic diversity to be a contin- lated over 3- to 4-day intervals are included for

uous function of richness, such that Hmax = log2 comparison. The latter is calculated, according to
s. The majority of the those species will, as we Equation (7.7), based on the original derivation of
in a steady state of overwhelming dominance,

reflected by a very low rate of community change
(σ s < 0.03 d−1 ), any residual evenness is lost.
In spite of the implicit shortcomings of the
Shannon–Weaver index, it continues to be much
the more commonly used. Moreover, experience
indicates that the application of the cut-off
reveals very well the change in individual species
contributions to biomass through the course of
succession. For instance, it is a well-attested prin-
ciple that, in a fully operational ecosystem, ‘most’
(≤95%) of the biomass at each functional level
(primary producer, herbivore, decomposer, etc.) is
accounted in the first six to eight species (Hildrew
and Townsend, 1987). In my own career expe-
rience of counting upwards of 6000 individual
samples of phytoplankton from lakes, reservoirs
and rivers, I have never encountered one in which
more than 20 species were simultaneously repre-
sented at >1 cell mL−1 , or by more than 60 species
exceeding 10 L−1 . This is despite the fact that, in
some cases, the full list of recorded species at the
various given locations has exceeded 150. In the
many cases where cell counts were transformed
to biomass, there has not been one instance in
which the eight most abundant species made up
<95% of the total standing mass of the phyto-
plankton crop. Indeed, the 95% threshold is fre-
quently surpassed by just one to three of the most
Figure 7.17 Periodicity of dominant phytoplankton in abundant species. So it is that, even within a vari-
Blelham Enclosure A in 1982. From changes in the relative abilty range of the one to eight species that will
abundances of each species shown in (a), the rates of typically comprise ≥95% of the mass of phyto-
community change (σ s ) and community diversity (Hs ) are plankton, the truncated Shannon–Weaver index
calculated and plotted in (b). Redrawn with permission from is perfectly sensitive to variations in the compo-
Reynolds (1997a).
sitional diversity. Examples of its use are cited in
the following section.
a ‘succession rate index’ of Jassby and Goldman
(1974b): 7.3.3 Species diversity and disequilibrium
σs = {[bi (t1 )/B (t1 )] − [bi (t2 )/B (t2 )]}/(t2 − t1 ) (7.7)
in natural communities
High diversity is a feature of many biological sys-
where B is the total biomass and bi is the biomass tems and it is generally considered to be impor-
of the ith species on each of two occasions (t1 and tant for their healthy functioning (Reynolds and
t2 ), so that σ s is averaged over the period that sep- Elliott, 2002). Whether this is always or only
arates them (units, d−1 ). The plot in Fig. 7.17 con- sometimes true, high species richness, equitabil-
firms the increase in the number of participating ity and genetic representation are believed to be
species in the early phases (Hs : 1–2 bits per indi- under threat and, therefore, worthy of conserva-
vidual) and that diversity is kept up while species tion (Lawton, 1997). It follows that the ‘protec-
composition varies considerably (σ s : 0.1–0.3 tion of biodiversity’ must be a good thing, even
d−1 ). Once Microcystis becomes abundant and if there has been an incomplete understanding
Figure 7.18 Time courses of

Margalef’s diversity index in the
mesotrophic North Basin of
Windermere, the eutrophic Crose
Mere and the well-fertilised Blelham
Enclosure B during 1978. Redrawn
of just what this means, much less any clear well-fertilised Blelham Enclosure (Fig. 7.18) was
strategy for bringing it about. Biodiversity (origi- used by Reynolds (1984a) to illustrate the then
nally BioDiversity, coined as a shorthand for Bio- general belief that lake systems maintaining high
logical Diversity: Wilson and Petr, 1986) has no nutrient availabilities support poorer species
precise definition and it has no units of mea- diversities than oligotrophic systems (Margalef,
surement. As a synonym for ‘species richness’, it 1958, 1964; Reynolds, 1978c). This point now
can assume fairly precise terms but it is either requires further discussion (see Section 7.3.3
reliant on a sophisticated and continuously accu- below) but Fig. 7.18 presently serves to show that
mulating knowledge base of how many species the low values observed in the Blelham Enclosure
there are. At the same time, this understanding B in 1978 (Hs 0.24–1.75 bits individual−1 ) had not
requires an area-based focus. For instance, there been dissimilar from those noted in the Enclo-
is (presumably) a finite number of species on sure A in 1982 (Fig. 7.17). The differences between
the planet, each with an instantaneously finite the measures for the eutrophic Crose Mere (Hs
number of individuals; elimination of these is 2.00–4.15) and the mesotrophic North Basin of
to extinguish the species. On a regional or local Windermere (Hs 1.78–6.44 bits individual−1 ) are
basis, the extinction of species may be reversed generally small but for the wide departures in
by subsequent invasions of new individuals of the summer months. These may be compared
the same species from remote populations. It also with Margalef’s (1958) own assessments of
seems important to establish the mechanisms diversity in samples taken from the Ria de Vigo,
maintaining high local species diversity, again through the summer progression from diatom to
through reference to (apparently) cosmopolitan dinoflagellate dominance (Hs 0.8–5.4) (see p. 311
species whose short-generation organisms render and Fig. 7.19). Recent studies of phytoplankton
the timescales of exclusion and dispersal conve- diversity among a broad range of European lakes
niently measurable. quote seasonal fluctuations in Shannon diversity
Phytoplankton assemblages in natural lakes (truncated H values) in the range 1–3.5 in strat-
and coastal waters often comprise several main ifying lakes (Leitão et al., 2003; Salmaso, 2003),
species simultaneously and these contribute to or slightly higher (2–5) in shallow lakes (Padisák,
relatively higher measures of ecological diver- 1993; Mischke and Nixdorf, 2003).
sity, as measured by the Margalef or trun- Hutchinson (1961) famously drew attention
cated Shannon–Weaver indices, than is suggested to the paradox that high diversity among phy-
by the plot of Hs in Fig. 7.17 for an experi- toplankton species competing for essentially the
mental enclosure. The plot comparing diversity same limiting resources in an apparently homo-
indices from Windermere, Crose Mere and a geneous environment is counter-intuitive and
Figure 7.19 Variations in the

index of diversity in a number of
samples of phytoplankton collected
from surface waters of the Ria de
Vigo in 1955 plotted against the
successional stages diagnosed by
Margalef (1958). Redrawn with
permission from Fogg (1975).
against Hardin’s (1960) principle of competitive to grow and increase specific biomass under
exclusion (see also Box 7.1). Since that time, tolerable environmental conditions but which
there have been numerous attempts, often with show a net loss of biomass under conditions as
compelling experimental evidence, to argue they become intolerable. Just the simultaneous
for a satisfactory explanation. These broadly occurrence of species whose numbers are increas-
divide among three categories. Following Paine ing while those of another are in net decline
(1966), species coexistence is promoted primarily is sufficient to maintain species richness and
through food-web interactions. Following Tilman local species diversity throughout the period that
(1977) and others, coexistence reflects simulta- the two sets of responses can be detectable. For
neous, niche differentiation of the surviving this reason, low diversity (H , Hs ) must always
species. Disciples of Connell (1978) attribute a be accompanied by low rates of compositional
rich biota to temporal and simultaneous spatial change (σ s ). Equally, the maintenance of high
variability of the environment. These explana- diversity must be viewed within the context of
tions are not mutually exclusive and, indeed, time constraints set by changing rates of popula-
they may be summative. However, the impor- tion recruitment and attrition.
tance of the topic requires us to examine the The time-frame of ecologically relevant
mechanics of these concepts in more detail. responses to an infinitely variable environment
is supposed to be bounded (at the lower end)
Disequilibrium by the generation time of the shortest-lived
Hutchinson’s (1961) own explanation of the para- species and (at the upper) by the time taken to
dox lay in the error of the first assumption reach a competitively excluded climactic state
of environmental homogeneity and steady-state (Reynolds, 1988c, 1993b). Smaller-scale variabil-
conditions. In all the examples considered in ity may demand reactivity at the biochemical
this chapter, species composition is shown to and physiological levels. Provided that the cell
change conspicuously through time. In every can maintain its cycle of growth and mitotic
instance, this is shown to be driven by a bal- division, the variability is smoothed to ecolog-
ance of dynamic responses of planktic species ical constancy. Environmental changes outside
that have ‘arrived’ from elsewhere, that are able the period of ascendency to climactic steady state
that may destroy biomass and precipitate a new eventually averaging 700–1000 cells mL−1 . These
potential climax – such as inundation of vegeta- maxima, observed in late August or early Septem-
tion by a marine incursion or the desertification ber, arise almost exclusively, through serial divi-
of forest to scrub – have intrinsic interest but tell sions, from a base inoculum of excysting overwin-
us little about successional maturation. tering propagules of between 0.1 and 1 cells mL−1 ,
Against the many scales of physical variabil- recruited to the water column in February. The
ity in aquatic environments (Sections 2.1, 2.2.2, 11–14 doublings required to achieve this occupy
2.3.2, 2.3.3), the timescales of phytoplankton pop- around 200 days, although the fastest rate of
ulation ecology are readily discerned. The maxi- increase is generally attained during July (∼0.15
mum growth rates of the weedy and more inva- d−1 , coinciding with the warmest water, longest
sive C-strategist phytoplankters – exemplified by days and nutrients still not limiting). The obser-
Chlorella and Ankyra – have standardised specific vations are comparable with those for Esthwaite
replication rates (r 20 ) of >1.8 d−1 in culture Water of Heaney et al. (1981) and the growth rates
(Table 5.1). Verified field growth rates of 1.1 d−1 observed in attaining smaller maxima in the Blel-
(Table 5.4) bring the potential time of biomass ham enclosures (Lund and Reynolds, 1982) (see
doubling to matter of several hours (certainly to also Fig. 5.18) are also comparable.
<1 d under ideal field conditions). Of course, Several summer maxima in the Blelham
progress may well be impeded by the interven- enclosures, similarly achieving concentrations
tion of poor average light conditions, nutrient equivalent to >600 mg chla m−2 (>30 C m−2 ),
exhaustion or grazer responses and be truncated have been overwhelmingly dominated by Micro-
far short of a climactic steady state. Physical fac- cystis populations of 300 000–400 000 cells mL−1 .
tors are also important but rapid flushing rates Again the initial recruitment has come from
may benefit fast-growing organisms over slower small colonies that entered the plankton between
species and their consumers. However, an ade- April and June (see Section 5.4.6). Sustained expo-
quately discretionary rate of dilution leaves the nential growth during the summer (rn between
net rate of population increase only just positive 0.15 and 0.24 d−1 , slowing towards the end)
and thus the attainment of the potential pro- would raise an invading the standing population
vided by growth is much protracted. The physi- from the equivalent of 100–200 cells L−1 to its
cal conditions suspend the advance of the suc- maximum through some eight or nine genera-
cession, maintaining it in a sort of plagioclimax tions in about 5–8 weeks of growth. Prior to that,
(Tansley, 1939). Elsewhere, selection through the three or four divisions of the overwintering cells
foraging preferences of the zooplankters present contributing the invasive colonies whilst still
for particular foods or particular particle sizes, on the sediment perhaps needs to be included
constraints in the availability and accessibility of (Preston et al., 1980; Reynolds et al., 1981).
resources and light all affect the relative success It was these considerations that allowed
of the tolerant species in moving towards dom- Reynolds (1993b) to propose that the period
inance. A phase of apparently enhanced coexis- required for a phytoplankton succession to move
tence and (thus) higher measurable diversity is from initiation to a competitively excluded cli-
passed before the strongest competitor eventu- max is equivalent to some 12–16 generations.
ally emerges as the dominant species of a low- Given favourable conditions and the resource
diversity climax. capacity to sustain it, the entire process might
We have considered examples of Ceratium, occupy 35–60 days.
Microcystis and Planktothrix moving to this stage However, it is also plain that, in the plankton
of community development. In Crose Mere (Fig. as on land, such low-diversity climactic, steady
7.6), Ceratium has several times been observed states are exceptional occurrences. That diversity
(Reynolds 1973c, 1976a) to build up to a sta- is normally kept high is seen to be largely a
ble (low-σ s ), low-diversity (Hs ∼2.0) maximum of consequence of processes that prevent, or delay,
between 150 and 250 µg chla L−1 (∼900 mg chla perhaps indefinitely, the progress to the poten-
m−2 , ∼45 mg C m−2 ), comprising concentrations tial steady state. Communities remain, in fact, in
various states of ecological disequilibrium, hav- d−1 ) increases while alga B (r ∼0.1 d−1 ) decreases
ing a persistent tendency to move towards a given when both are subjected to a filter-feeding con-
(though possibly changing) climactic outcome sumer Z (F = 0.15 d−1 ). Food selection increases
but in which progress is seriously impeded or the benefit to the ungrazed alga C (r = 0.1 d−1 ).
stalled. In many instances, diversity and species The possibilities increase greatly if depth distri-
richness are quite unrelated to trophic state and butions of food and feeder are considered; alga
productive capacity (Dodson et al., 2000). B may now increase faster than A if B swims
to the surface while A fails to avoid the down-
Internal (biotic) mechanisms of coexistence ward migrations of Z. Then the effects are again
The activities of heterotrophs (principally, but multiplied if consumers X and Y feed on larger
not exclusively, phagotrophs) within the same or smaller foods than Z or show particular pref-
system interact with the population dynamics erences for a given alga. Finally, the growth
of phytoplankton in a variety of ways. Easi- dynamics of algae and feeders are themselves
est to understand is the concept of zooplank- fluctuating. It is not difficult to recognise the
ton feeding indiscriminately on phytoplankton. complexities of phytoplankton interactions with
By removing individual phytoplankters from the phagotrophs or how their fluctuating fortunes
pool, even light grazing surely delays succes- contribute to the maintenance of diversity to the
sional progress. It is theoretically possible that extent that Paine (1966) indicated.
individual food organisms might be removed Phagotrophic grazers are not the only organ-
at the same rate at which they are recruited, isms that influence the structure of the planktic
which would ensure zero instantaneous rates assemblage from within (i.e. autogenically). Para-
of advance. However, such balances are rarely sites and pathogens are generally highly specific
struck (well-nourished zooplankters increase in in the algae they affect and it is frequently the
size and recruit further generations of indi- most common (or recently most common) that
viduals). The more common outcome is that attract epidemics (see Section 6.5). The progres-
zooplanktic animals (at least, the filter-feeders; sive intervention of chytrid fungi was responsi-
see Sections 6.4.2, 6.4.3) draw down their foods ble for breaking the years of summer dominance
to concentrations at which their own popula- of Ceratium in Esthwaite Water, and for accord-
tion growth is impaired. The intervention of a ingly raising the subsequent diversity of the phy-
third trophic component with a relatively slow- toplankton (Heaney et al., 1988) (see also Section
changing impact intensity (such as planktivorous 6.5.2).
fish or invertebrate species) is usually necessary
to impart a simultaneous top–down regulation Resource competition as a mechanism of
on zooplankton consumption (Reynolds, 1994c). coexistence
The interactions among the consumers of the The theory of resource-based competition devel-
microbial food web (nanoflagellates, ciliates) with oped by Tilman and co-workers (Tilman, 1977;
the producers (picoalgae, bacteria) are presumed Tilman et al., 1982) (see also p. 197) proposed an
to place mutual constraints on the dynamics of elegant explanation of species coexistence and
each other in a way that helps to ensure the assemblage diversity, invoking internal responses
simultaneous survival of each of the components (specific resource uptake and growth rates) to
(Riemann and Christoffersen, 1993; Weisse, 2003). external drivers (resource availability). The exper-
Zooplankton grazing on phytoplankton is gen- iments of Tilman and Kilham (1976) showed how
erally more selective in its effect. This selectiv- it was possible for two species of diatom to grow
ity has several dynamic components that need simultaneously (and, hence, to coexist) for as long
to be distinguished. Just the difference in the as one of them remained limited by the sup-
replication rates of two phytoplankton species ply of phosphorus and the other by the supply
may determine the relative success of one over of silicon. They reasoned that while they were
the other whilst both are simultaneously grazed not in direct competition for limiting resource,
by the same consumer species. Alga A (r ∼0.2 their coexistence would endure. Moreover, if two
species may coexist while each is independently limiting resources are thus able to maintain
resource-limited, the further corollary is deduced larger inocula. On the other hand, the effective
that there may be as many coexisting, non- exhaustion of a limiting nutrient, by species of
competing species as there are simultaneously all affinities, means that the most likely outcome
limiting factors (cf. Petersen, 1975). The high- of algal activity is that the uptake capacity for
est species diversities, accordingly, might be the limiting resource is insufficient to support
expected in environments of general resource further population increases. Ultimately growth,
deficiency and, hence, the greatest likelihood of then cell division (r ) is brought to a halt, and
simultaneous rate limitation. population change (rn ) falls to zero or is negative.
There is no shortage of experimental infor- Competition may certainly be keen but the rates
mation to support the differentiation of interspe- of succession and exclusion are reduced to very
cific growth rates along resource-ratio gradients. low values. It follows that species richness and
Rhee and Gotham (1980) demonstrated analogous diversity endure because the mechanisms that
performance differences among various common work against them are themselves severely con-
species of phytoplankton with respect to the ini- strained. The low-diversity steady state is simply
tial relative availability of nitrogen to phospho- postponed.
rus, while Bulgakov and Levich (1999) cited exper- It must be noted that this logic does not inval-
imental evidence in support of the widely held idate resource competition as a direct mecha-
supposition that a low N : P ratio favours the dom- nism for sustaining coexistence, provided that
inance of Cyanobacteria. Sommer (1988c, 1989) the turnover of the limiting resources, between
has argued for competitive interactions among organisms and the labile pool, is demonstrably
algae for diminishing levels of Si, N and P favour dynamic. Possibly the clearest evidence for this
the seasonality of assemblages in kataglacial is the widespread co-dominance of ultraoligo-
lakes (see p. 338). The important role of carbon trophic plankton assemblages, both in the open
and its photosynthetic fixation is brought into ocean and in lakes, by nutrient-deficient photo-
the appreciation through the ratios of light to autotrophic picoplankton and carbon-deficient
nutrient and C : P availabilites (Sterner et al., 1997) bacterioplankton. The carbon (especially) is
(see p. 199). rapidly exchanged through a labile DOC/DIC pool
At first sight, simultaneous multiple limita- whereas carbon and nutrients are turned over
tion is an attractive basis for explaining coex- by consumers of the microbial web. There is a
istence. On the other hand, it is losing favour thus a near-continuous, differentiating compe-
except in qualified circumstances. This is partly tition among the components and it maintains
because simulation models of the dynamics of an ambient, diverse steady state. In these terms,
multiple species limitations have not borne out the independent introduction of the erstwhile
the prediction of stable coexistence but, rather, limiting nutrient is bound to help the photoau-
have pointed to instability and chaotic outcomes totrophs rather more than the chemoautotrophic
when more than two species compete for more bacteria, which will experience continuing severe
than two limiting resources (Huisman and Weiss- or intensifying carbon deficiencies.
ing, 2001). However, some of the experimental
conditons and assumptions do not necessarily Disturbance as a mechanism of coexistence
obtain in natural environments. As has been The third category of processes that pro-
pointed out previously, resources either saturate mote interspecific coexistence is rather gener-
the sustainable growth rate of phytoplankton or ally referred to as disturbance. Disturbances are
they fall to levels that interfere with the abil- imposed by external (allogenic) forcing factors
ity of just about all species to maintain a finite and are recognised by their effects in variously
rate of growth (Section 5.5.4). Persistent low lev- delaying, arresting or diverting successional
els of nutrients favour species with relatively sequences from achieving their low-diversity,
high uptake affinities over species having weaker steady-state climaxes. Commonly cited types of
affinities and to which the same resources are terrestrial disturbance include geophysical events
unavailable. Species with higher affinity for (landslides, lava inundations) and abnormal
weather conditions (hurricanes, floods, droughts: hypothesis (IDH). In essence, it anticipates

Sousa, 1984). In the pelagic, the best-known dis- that:
turbance events are associated with hydraulic
(1) In the absence of any disturbance, compet-
or hydrographic disruptions of the water col-
itive exclusion will eventually reduce the
umn (flushing, strong mixing) or with nutrient
number of surviving species to minimal lev-
pulsing (Sommer et al., 1993; Sommer, 1995).
els.
For example, significant wind-mixing events dur-
(2) At high frequencies of intense disturbance,
ing summer in a small, stratifying, eutrophic
only pioneer species are ever likely to re-
lake (Crose Mere) were several times observed to
establish themselves in the wake of each
cut across the usual G → H → LM sequence
disturbance.
of dominant algae by restoring growth con-
(3) If such disturbances are of intermediate fre-
ditions suitable for diatoms (Reynolds, 1973a,
quency or intensity, there are more and
1976a). The species that grew had been repre-
longer opportunities for community ascen-
sented in the vernal bloom (Association C; see
dency and a greater variety of species will
Table 7.1) but others were confined to such sum-
establish in the wake of each disturbance
mer growths (notably Aulacoseira granulata, since
but these will rarely be allowed to mature
ascribed to P). However, in the post-disturbance
to competitively excluded steady states.
relaxation, a tendency for the G and H elements
to recapitulate their post-stratification sequences The obvious consequence is that a peak of diver-
was noted. In the highly disturbed early sum- sity should be found at intermediate frequencies
mer of 1972, diatoms, G-chlorophytes and H- and intensities of disturbance.
Cyanobacteria were maintained as co-dominant This is an elegant theory: its development is
species (Reynolds and Reynolds, 1985). Later logical and its statements are intuitively correct.
on, artificial mixings applied to the Blelham It has long remained a hypothesis for it is dif-
Enclosures imitated these effects experimen- ficult, in most ecosystems, to collect appropri-
tally (Reynolds et al., 1984). The imposed delay ate data to support or confound it. Planktic sys-
to the successional maturation, whilst retain- tems provide an exception to this statement,
ing extant mid-successional populations and where successions from the establishment of pio-
periodically resurrecting pioneer species, helps neer communities through to their total eclipse
to bolster the diversity of the phytoplankton by steady-state, competitively excluded climaxes
represented. can be completed potentially in under 60 days
In this way, the impact of disturbance (see p. 370). It may be deduced, incidentally, that
on diversity is related to the intensity and, the equivalent of 12–16 generations required for
especially, the frequency of the forcing events potentially climactic tree species (with genera-
(Polishchuk, 1999). However, the scales of impact tion times of 200–800 years) to achieve anything
and timing have to be judged against the gen- closely resembling a competitively excluded cli-
eration times and succession rates of the main max will occupy 2.5–100 ka. This represents is
species involved (Reynolds, 1993b). Without the a substantial fraction of an interglacial period.
context of successional processes, understanding Given the reconstructed climatic variability over
of the complex mechanisms underpinning the even the last ten thousand years (10 ka) since the
diversity–disturbance relationship has been last great glaciation, the history of the chang-
rather slow in development and not without ing dominant vegetation of the northern land
controversy (Wilkinson, 1999). Even now, the masses (e.g., Pennington, 1969) may be legiti-
mechanisms are widely misunderstood, despite mately comparable to the periodicity of phyto-
the fact that most of the relevant theory was plankton in a small temperate lake during a sin-
in place in the mid-1970s (Grime, 1973; Fox and gle year (Reynolds, 1990, 1993b).
Connell, 1979). The work cited most frequently is As has also been pointed out on many previ-
that of Connell (1978) whose strictures on species ous occasions, phytoplankton successions are not
diversity in rain forests and coral reefs led to him merely observable but are amenable to practical
to propose his memorable intermediate disturbance experimentation. IDH has been invoked in the
explanation of periodic sequences of phytoplank- Juhász-Nagy, 1993 is most apposite), the litera-
ton in natural lakes and reservoirs (among oth- ture carries examples of instances when expe-
ers, Haffner et al., 1980; Reynolds, 1980a; Trimbee riences differed from expectation. For instance,
and Harris, 1983; Viner and Kemp, 1983; Ashton, the fact that an episode of summer storms and
1985; Gaedke and Sommer, 1986; Olrik, 1994). floods failed to break the dominance of Aphani-
Reynolds’ (1980a) usage of the terms ‘shift’ zomenon in the plankton of a small Danish lake,
and ‘reversion’ to refer, respectively, to the dis- while earlier, relatively trivial, weather events
ruption of the succession by cooling/mixing had triggered upheavals of species composition
events associated with passing weather systems, was quite clearly contrary to the anticipation
and to its subsequent recapitulation, now seems of Jacobsen and Simonsen (1993). Other authors
too mechanistic (too Clementsian? cf. p. 351); have noted that the resistance of communities
abandonment of these terms was recommended to disruption by external forcing is acquired pro-
(Reynolds, 1997a). Nevertheless, the experimen- gressively with increased successional complexity
tal manipulations of succession and disturbance (Eloranta, 1993; Moustaka-Gouni, 1993; see also
imposed by Reynolds et al. (1983b, 1984) attest to Barbiero et al., 1999). Sommer (1993) has shown
the importance of steady physical conditions in that in neighbouring lakes of similar chemistry
allowing autogenic processes to structure com- but differing morphometry, disturbance events
munities and to dominate assembly processes. were more extensive and effective in increasing
Equally, they show how variations in the physical diversity in the deeper example. The responses of
conditions may terminate one autogenic phase the plankton differed among three small lakes of
and initiate the next. contrasted trophic state and compared by Holz-
The physiological responses of the phyto- mann (1993), being weakest in the most olig-
plankton to an altered physical environment are otrophic of them. In reviewing each of these con-
immediate but take several hours or more to feed tributions, Reynolds et al. (1993b) concluded that
through to altered population recruitment rates. there is no consistent or predictable relationship
Accelerated rates of community change (σ s ) and between diversity and external physical forcing.
higher instantaneous diversity indices (H , Hs ) For terrestrial ecologists, the development of a
are evident within days. From the cited Blelham consistent theory of disturbance has been just as
enclosure studies, Reynolds (1988c) deduced that, troublesome (e.g. Wilkinson, 1999). There is little
depending upon the relative dominance of the difficulty in understanding the intervention of
disfavoured species, σ s reaches a maximum after catastrophes, from fires and storms to volcanic
two to four generations of the favoured species eruptions and lava flows, in being able to arrest,
have been recruited to the plankton. The diver- not to say to obliterate, the autopoetic develop-
sity index (Hs ) follows closely. In real time, this ment of terrestrial vegetation and the reopen-
might be equivalent to 4–16 days after the stim- ing of the land surface to colonising propagules.
ulus is applied. This deduction agreed substan- Neither is there a problem in recognising that
tially with the findings of Trimbee and Harris stochastic, smaller-scale forcings create just the
(1983) and Gaedke and Sommer (1986). There is generally fluctuating environment that prevents
now general agreement that diversity is highest the extinction of opportunist strategists by persis-
early in the disturbance response, around the sec- tent resource gleaners (see, for instance, Pickett et
ond to third generation, or within 5–15 days of al., 1989). Indeed, the simultaneous existence of
the imposition of the stimulus (Reynolds et al., sources of invasive species actually necessitates a
1993b). continuity of disturbances and a continuum of
patches in differing stages of successional matu-
IDH and diversity ration among which invasive species may migrate
Despite the satisfyingly simple elegance of (Reynolds, 2001a). It is also self-evident that, were
the concept of disturbance-induced community this not the case, opportunism (r-selection) would
restructuring and of disturbance-impeded suc- cease to have any viability as an adaptive survival
cessional progression (the searching critique of strategy. This view of patch dynamics is explicit
in Connell’s (1978) hypothesis and is implicit in only must it be a signal that can feed through
Hutchinson’s (1961) suggested explanation of the to species-specific growth rates but the potential
plankton paradox, when he referred to ‘contem- respondents have to be moved from their contem-
poraneous disequilibrium’. A further corollary is poraneous limitation. Put simply, if the phyto-
that, as all species have evolved to contend with plankton is experiencing growth-rate limitation
disturbed environments, they are, in a sense, through an effective lack of (say) iron or phospho-
matched to recurrence of perturbating events rus, it is difficult to imagine how exposure to 1 to
(Paine et al., 1998). Thus, we may note again that 2 days of strong winds and a simultaneous deep-
dispersal constraints are as important to com- ening of the surface mixed layer from <3 to ∼7
munity assembly as is the inevitability of self- m will overcome that. The forcing does not dis-
organisation and the stochasticity of external disturb the status quo ante. Yet this was the extent of
turbances. forcing applied to Crose Mere needed to disturb
The problems experienced by terrestrial ecol- the summer succession, to stimulate a significant
ogists relate mainly to inconsistencies of pattern, bloom of diatoms and to raise the planktic diver-
especially relating to the intensity and frequency sity (Reynolds, 1976a) (and see p. 373).
of disturbances considered to be effective in influ- The experience of this observation from the
encing diversity, and which of these might be field should always prompt the question: is the
the more critical (Romme et al., 1998; Turner et potential of the forcing sufficient to overcome
al., 1998). There are difficulties in the scaling of the existing structure? At the present time, this
models based on field observations and differ- is not easy to answer from any quantified stand-
ing inputs can raise or suppress the response of point, without some common units. The poten-
diversity to disturbance (Huston, 1999; Mackey tial of comparing external mechanical forcing
and Currie, 2000; Hastwell and Huston, (2001). with the current cushion of exergy flux (in
In another modelling approach, Kondoh (2001) MJ m−2 d−1 ) has been explored theoretically in
purported to show the important interaction Reynolds (1997b). An accumulating phytoplank-
between system productivity and distubance in ton was challenged by wind-mixing events of
influencing the competitive outcome of multi- known kinetic energy. From the starting assump-
species dynamics on species richness. In essence, tions about temperature (set at 20 ◦ C), the possi-
a relationship between diversity and disturbance ble income of solar energy (≤26.7 MJ m−2 d−1 ;
is confirmed as being unimodal but the pro- say, 12.6 mJ m−2 d−1 harvestable energy), and
ductivity level that maximises diversity itself the mechanical energy required to disperse this
increases with increasing disturbance. uniformly through 1 m (65 J m−2 d−1 ; equiva-
In a way, this seems to match the deduc- lent to a wind of 3.4 m s−1 ), wind forcing was
tions in respect to lake productivity (Holzmann, increased selectively with a view to balancing the
1993; Reynolds et al., 1993b): that productive or incoming flux. By itself, mixing energy offers lit-
nutrient-rich systems are more susceptible to dis- tle challenge to the harvestable energy flux: even
turbance than nutrient-poor ones. This, indeed, with a tenfold increase in wind speed (34 m s−1 ),
would make a satisfying conclusion. However, it the kinetic energy flux is still only 6.5 kJ m−2
offers no real explanation for its mechanisms in d−1 . However, the simultaneous deepening of the
nature and it also perpetuates one of the most mixed layer dilutes greatly the harvestable energy
persistent shortcomings in applications of the available to entrained phytoplankters (see p. 138).
IDH. Taking the second point first, every observ- Even a doubling of wind speed is sufficient to
able disturbance is a response reaction, an out- increase the depth of mixing of the modelled
come of internal dynamics of the species inoc- water by the cube, that is, to 8 m. Using various
ula present. It is to be differentiated from the interpolations of existing phytoplankton biomass
forcing applied, which may, or may not, evoke a (1–100 mg chla m−2 ), its assumed light absorp-
response at the level of population recruitment. tion, εa = 0.01 m2 (mg chla)−1 , and of the back-
Only then is a disturbance precipitated. This is ground extinction of the water (εw = 0.2 m−1 ),
where the intensity of forcing is critical – not the levels of energy harvestable by phytoplankton
Figure 7.20 Energy exchanges and exergy accumulation

Figure 7.21 Representation, against the
under a varying capacity. The sigmoid line traces the potential
energy-accumulation model (Fig. 7.11) of the responses of
accumulation (cf. Fig. 5.20, 7.12e) but this experiences
biomass to capacity oscillations, including readjustment (a
setbacks every time that the sustainable biomass exceeds the
disturbance) when the accumulated mass is left unsustainable.
capacity of the oscillating energy income. On each occasion,
Redrawn with permission from Reynolds (2002b).
there has to be an adjustment (losses of structure, mass and
exergy) which is manifest as system disturbance. So long as
energy or resources continue to fluctuate, ecosystems go on
alternating between expansion into available capacity and by wind mixing, variability is such that it is
readjusting to contraction Redrawn with permission from indeed possible that the energy that is actually
Reynolds (1997a). harvestable may, on occasions fall below the min-
imum maintenance requirement of the accumu-
lated phytoplankton biomass. Against the plot in
could be reduced to <6 MJ m−2 d−1 . When the Fig. 7.11, this is equivalent to saying there is insuf-
exclusive effect of increased mixing due to the ficient harvestable energy to maintain a positive
doubling of mechanical stress is compounded by exergy flux. In short, the system is unsustainable.
changes day length, cloud cover and lowered radi- So long as the external condition persists, the
ation intensity, the harvestable energy is quite requirement to restructure (lose biomass, place
liable to reduction to low values (often <2 MJ biomass in a lower energy resting state; switch
m−2 d−1 ). to more light-efficient species) becomes likely to
Thus, mechanical disturbance arguably car- invoke a response that might be perceived as a
ries entropic penalties of a magnitude sufficient disturbance. The energetic representation is used
to disturb community development in the plank- as the basis of Fig. 7.21, in which the variabil-
ton. Without more empirical data, however, it ity in the potentially sustainable biomass (against
remains a conjecture. Nevertheless, the conjec- variable energy income and mechanical dilution)
ture yields a further, simple conceptual model is tracked as a function of the actual biomass.
of the link between forcing and the manifes- We can readily appreciate that forcing variabil-
tation of diversity-raising disturbances. In Fig. ity absorbed within the harvesting capacity and
7.20, a hypothetical plot tracks through time the leaving a positive exergy flux is fully sustainable
biomass potential of an ascendent phytoplank- and does not constitute a disturbance. Should
ton (and, hence, of its finite energetic require- the forcing result in an energy income that is
ments of maintenance) and the simultaneous insufficient to balance maintenance, there is no
supportive capacity of the harvestable energy exergy buffer left. Eventual reaction is inevitable,
flux. With daily fluctuations in the incoming although the scale of disturbance may yet depend
solar flux and of daily fluctuations in its dilution upon the scale of exergy shortfall (intensity), its
Figure 7.22 Arithmetical means

of (a) species number, (b)
equitability and (c) Shannon diversity
in periodic phytoplankton showing
disturbance responses as a function
of their frequencies. Most of the
data refer to events in Balaton (B) in
the summers of 1976–78, 1980 and
1982, before (B82bb) and during
(B82b) the enduring
Cylindrospermopsis bloom that year.
Other points are for (N)
Neusiedlersee, an experimental
pond (e) and its control (c). Note
that all three plots peak at
disturbance frequencies of 3–7 days.
Original data of Padisák (1993) and
redrawn with permission from
Reynolds (1997a).
duration and the physiological resistance of the judgement. The evidence from Padisák’s (1993)
accumulated biomass to withstand the energetic collected sequences of phytoplankton in various
disequilibrium. The track in Fig. 7.21 shows the Hungarian lakes (see Fig. 7.22) shows species rich-
necessary scaling down of biomass costs and the ness, equitability and Shannon diversity each to
reversion to an earlier ascendent stage. Inciden- rise steeply from daily disturbances up to separa-
tally, it also implies a low incidence of distur- tion intervals of 3–7 days. Thereafter, each index
bance at low biomass (as does Fig. 7.21), including decayed gradually back to low average levels at
those instances of severe nutrient constraints (cf. infrequent disturbances (separated by >30 days).
Fig. 7.15b). The other valid way of expressing frequency, as
With a clearer picture of what constitutes a a number of disturbances per year, was used by
disturbance, we may now return to the central Elliott et al. (2001a) as a basis for applying simu-
issue of diversity being a function of the fre- lated forcing events on periodic sequences mod-
quency of disturbance. From Connell himself to elled using PROTECH (see Section 5.5.5). Exter-
many subsequent authors elaborating on the IDH nal forcing to a controlled physical environment
(e.g. Wilkinson, 1999), the relationship between was devised (actually an instantaneous, complete
diversity and frequency of disturbance is repre- mixing of a hitherto stratified 15-m water col-
sented as a smooth parabola, generally referred umn, mixed to only 5 m) while all other vari-
to as ‘the humpback curve’. It successfully con- ables (day length, incident radiation, tempera-
veys the idea of low diversity occurring at zero ture and saturating nutrient availability) were
and at very high frequencies of disturbance and held constant. Grazing of algae was excluded.
of higher diversities at intermediate frequencies Eight species of phytoplankton having contrasted
(p. 373). However, there is no basis for assum- life forms and survival adaptations were seeded
ing a continuous relationship with frequency, on day 1 and their dynamic changes were then
even if scaled in terms of respondent gener- simulated over a period of 1 year. Model runs
ations. Neither do most terrestrial sources of were subjected to applied forcings, with vary-
data provide much information to form any real ing frequency and duration, and the population
Chlorella): it achieved immediate dominance and

the virtual exclusion of all rivals (i.e. low diver-
sity) within 50 days. Inserting two or three 7-
day forcings during the year (Fig. 7.23a) produced
the equivalent number of clear disturbances,
in which a brief maximum of species B (hav-
ing the relevant properties of R-strategist Asteri-
onella) was supported. On reversion to the ambi-
ent state, species A (‘Chlorella’) duly reasserted its
dominance and species B (‘Asterionella’) dropped
back. Increasing the forcing frequency to five per
year (Fig. 7.23b) has the effect of maintaining
a more varied plankton, with ‘Asterionella’ alter-
nating in dominance with ‘Chlorella’ and other
species being represented more strongly. At fre-
quencies of 25–30 forcings per year, the forced
and alternating quiescent periods become quite
similar in length (Fig. 7.23c) and are sufficient to
allow ‘Asterionella’ to dominate ‘Chlorella’ through-
out while other species remain in contention.
Over a broad range of disturbance frequencies
(6–30 a−1 ), the average Shannon diversity (H ) of
the simulated eight-species assemblage is about
1.8 of a possible 3.0 bits mm−3 (evenness ∼0.6).
Increasing the frequency yet further, a sharp
alteration in behaviour was found to occur
between 31 and 32 a−1 (i.e. quiescent periods
are each <5 d (Fig. 7.23d). The system changed
to persistent dominance by a third species, C
(having the relevant properties of R-strategist but
more K-selected Planktothrix agardhii), eventually
to the exclusion of all the other species. Its
pre-eminence was enhanced at still higher fre-
quencies up to the point of continuous mixing
Figure 7.23 PROTECH-simulated changes in the
to 15 m.
populations of phytoplankton species to a standard physical
forcings (complete and instantaneous vertical mixing of an
Changing the forcing to 5- or to 3-day peri-
otherwise stratified 15-m water column for 7 days) applied at ods produced analogous sequences of results,
various frequencies. Periods of imposed forcing are shown by although correspondingly higher frequencies
the short horizontal bars. The ‘species’ correspond to were needed to avoid the replacement of ‘Plankto-
Chlorella (A), Plagioselmis (B), Asterionella (C) and Planktothrix thrix’ during the ‘intermediate quiescence’. The
(D). Note the precipitous change to Planktothrix dominance latter term had been introduced previously by
when 32 rather than 31 disturbances are applied. Redrawn Chorus and Schlag (1993) in the context of inter-
with permission from Elliott et al. (2001a).
ruptions (disturbances) to the success of Plankto-
thrix in rich, turbid lakes that are normally sub-
responses were also monitored for 1 year. Under ject to persistent mixing.
the ambient, unforced conditions, the commu- In all the series, surpassing the critical scale
nity became quickly overwhelmed by the fastest of forcing frequency secured an abrupt decline
growing life form (species A in Fig. 7.23, actually in Shannon diversity (to H < 0.8). Plotting diver-
having the relevant properties of a C-strategist sity as a function of forcing frequency (Fig. 7.24a)
Figure 7.25 Species diversity plotted as a function of the

frequency of externally forced disturbances, as the smooth
‘humpback’ usually supposed in many previous considerations
of the IDH, and as the ‘cliff-shaped curve’ revealed by
PROTECH modelling. Redrawn with permission from
Reynolds and Elliott (2002).
Figure 7.24 Annual mean Shannon diversity of

PROTECH-simulated phytoplankton assemblages, in terms of contributing to high diversity but several. As
(a) frequency of 7-day forcing events, and of (b) the time Hutchinson himself understood well, the para-
intervals separating forcing events of 3-, 5- and 7-day duration. dox lies wholly in the initial assumptions. Except
Redrawn with permission from Reynolds and Elliott (2002). under very well mixed conditions in a small
(<10 km2 ) and relatively shallow lake basin, there
is no single, homogeneous, isotropic environ-
shows the expected low diversity at both very low
ment. Even there, dynamic water movements are
and very high frequencies of alternation. Plotted
so variable that there is almost no opportunity
in terms of the interval between forcing events
for any kind of physical steady state to estab-
(Fig. 7.24b), diversity first increases steeply as qui-
lish. There is a virtual simultaneity of adjacent
escent periods lengthen, achieves a maximum in
patches, each experiencing differing conditions,
the range 10–50 days and then falls away with
yet which will themselves soon alter again, per-
ongoing physical stability. The modelling plainly
haps critically and perhaps in less time than it
confirms the prejudices about intermediate dis-
takes an alga to complete a generation. Only in
turbance and its critical times. What is also of
the case of the most severe and ongoing depriva-
general interest is the deduction of Elliott et
tion of a particular nutrient, or of the harvestable
al. (2001a) that the relationship of diversity to
energy input sufficient to satisfy the mainte-
forcing frequency is not the smooth humpback
nance requirements of most producer biomass is
curve that is popularly represented but one with
the effect of this physical environmental variabil-
a more cliff-like cut-off at critically high distur-
ity overridden. Moreover, because algae are dif-
bance frequencies (Fig. 7.25).
ferentiated by their adaptations to function and
to continue to grow in differing combinations of
Explaining the plankton paradox: where does various environmental constraints, they show dif-
the diversity come from? fering dynamic responses and response times to
Over 40 years of research since Hutchinson (1961) environmental variability, sufficient indeed for
posed the paradox of the phytoplankton have habitats to maintain several species in simulta-
succeeded in revealing many more verifications neous and alternating contention. And, because
of the diversity of natural phytoplankton assem- local species exclusion, whether as a direct result
blages, whereas its explanations have emerged of unsuitable prevailing conditions or through
only gradually. For there is not one mechanism one species being outperformed by others when
the prevailing conditions are suitable, requires Though strongly apparent in the phytoplank-
a lapse of time equivalent to several algal gen- ton, the dynamic nature of ecological diver-
erations, local extinction is also correspondingly sity with respect to habitats or global popula-
rare. Finally (and perhaps crucially), most locations had not always be appreciated or empha-
tions preserve an ecological memory in the infor- sised by ecologists generally. More recently, ter-
mation banks represented by the overall species restrial ecologists have embraced the importance
richness. Like the handful of conspicuous domi- of the larger-scale processes, within the separate
nants, their numbers also wax and wane through subdiscipline ‘macroecology’ (Brown and Mau-
time; local diversity is enriched by the poten- rer, 1989). Of particular interest are the rela-
tial of what is usually the majority of rare local tionships among what is referred to as the α-
species to be able to appear in one of the niches diversity (that recruited locally and equivalent to
that events conspire to make available. the phytoplankton species list of a single, hydro-
It is also evident that, in the event of relatively logically isolated small lake) and to the richness
more extreme environmental forcing events, of the regional pool, whence local exclusions
resulting in larger-scale, catastrophic reductions might potentially be replenished (γ-diversity; say
in biomass (e.g. flushing, storm disruption), the that of the same geographical zone, or part of
recovery of a functional phytoplankton (albeit the ocean). Of considerable concern to the con-
‘disturbed’ as defined herein) is fully possible. servation of terrestrial biodiversity, especially of
In addition to a certain resistance of late suc- macrovegetation and vertebrate fauna, is the risk
cessional stages to destruction by mixing events to both local and regional diversity of habitat
(that is, species-specific biomass is robust in with- fragmentation into smaller units (e.g. Lawton,
standing – and recovering from – conditions 2000). Just as macroecological systems are sepa-
that, for a time, are arguably unable to satisfy rable on the basis of their habitat (β-) diversity,
its maintenance needs), the species composition they too tend either to Type-I behaviour, in which
is also demonstrably resilient to forcing events. local species richness is fluid and proportional
By this, we mean that the respondent commu- to regional richness or to Type-II local richness
nity is based upon new growth of a similar being constrained above a certain proportion of
selection of the same species that were present the regional richness.
before the destructive forcing, so that the post- Being mutually separated, most freshwater
disturbance species composition may come to systems have the character of isolated islands
resemble the pre-disturbance one, albeit in new and, so, are expected to conform to the con-
proportions. This is not surprising in itself, being straints of island biogeography (MacArthur and
attributable in many instances to the recruit- Wilson, 1967). Moreover, the obvious connec-
ment of inocula biased by the survivor vegeta- tivities within predominantly linear structures
tive biomass or by the germination of conspe- are essentially gravitational. Thus, the broad-
cific local propagules (the ‘seed bank’) formed est recruitment pathways for freshwater biota
in previous extant phases (Reynolds, 2002c). How- are provided by downstream transport, whereas
ever, the further source of specific biomass that mobility between catchments relates to organis-
is important to re-establishing pelagic assem- mic size and efficiency of dispersal. As is we have
blages is the arrival of species-specific inocula seen (Eq. 7.3), microalgae, like bacteria and micro-
from adjacent and more distant patches, exploit- zooplankton, move very freely between catch-
ing effectively the channels of dispersion iden- ments and are highly cosmopolitan among suit-
tified above (see p. 353). Post-disturbance phyto- able habitats over most of the world. Given
plankton assemblages retain an ongoing capacity that detailed lists of species of phytoplankton
to yield just those compositional surprises that recorded in individual well-studied sites may
make the study of phytoplankton ecology so fas- comprise between 100 and 450 species (see p. 354)
cinating. This is also the main mechanism of the and supposing that the regional species richness
apparently cosmopolitan and pandemic distribu- lies within the order of magnitude 500–5000,
tions of most phytoplankton. then habitat suitability probably distinguishes
the 100 or so commonly encountered species ecology permit the application of observations
in the local assemblage. Of the latter, fewer to the resolution of outstanding theories, predic-
than eight species account for most of the local tions and models derived in ecological investiga-
biomass at any one time. These also account for tions of other systems.
most of the measured local (α-) diversity.
The implied generality of these relationships 7.3.4 Stability and fidelity in phytoplankton
is perhaps indicative of the intuitive Type-I phy- communities
toplankton diversity, in which local species rich- After the preceding discussion of variability and
ness represents a sub-set of the regional pool that disequilibrium in the plankton and the great
has somehow satisfied the twin constraints of distance away from steady state at which most
having ‘passed the filters’ (Keddy, 1992) of dis- planktic communities are poised, any explo-
persal and habitat suitability. Beyond that, the ration on the topic of their stability may seem
actual mix of species and the relative propor- vaguely fatuous. On the other hand, many exam-
tions in which they appear seems stochastic and ples of patterns in species composition and tem-
unpredictable (Rojo et al., 2000). On the other poral change in phytoplankton assemblages in
hand, planktic assemblages frequently support a Section 7.2, both in marine and freshwater sys-
high information content (H 3–5 bits mm−3 ). tems, were shown to apply widely among similar
This might equally imply that species interac- types of water body. Many examples of broadly
tions contributed only weakly to a Type-I diver- repeatable annual sequences of species abun-
sity. dance were cited. They evidently carry a rela-
To be set against this are the many cases tively low coefficient of interannual variation
of low-diversity, near-unialgal structures wherein (CV), calculable as the ratio of variance to mean
similar species or life forms are consistently biomass. Moreover, taken at the widest geograph-
and predictably selected in geographically quite ical scales, individual species are typically either
remote locations: we think of the common domi- common or generally rather rare.
nance of Microcystis or Planktothrix or Ceratium. The Given a predictable level of constancy about
organisation of species assemblages around per- the relative global abundances of phytoplankton
sistent gradients of nutrients, light and redox is species, this high fidelity of cyclical behaviour
similar from the equator to the poles (Reynolds, is symptomatic of the kind of long-term stabil-
1988c). Regional variations in the participating ity that was once supposed to characterise all
species are evident but the functional specialisms developed ecosystems. It used to be believed,
of the dominants are wholly analogous. The uni- for example, that the perceived stability of for-
formly low α-diversity that is achieved is indepen- est ecosystems was a consequence of their com-
dent of the regional richness and, thus, is quite plexity (Elton, 1958; Hutchinson, 1959, quoted
unmistakably indicative of Type-II behaviour. by Southwood, 1996) and food-web interactions
Poised between Type-I and Type-II mecha- (Noy-Meir, 1975). These assumptions about sys-
nisms, the overall deduction about biodiversity tem stability are now seriously questioned (May,
in the plankton is that it is maintained through 1973; Lawton, 2000). On theoretical grounds, May
the combination of variable forces – environ- (1973) showed that, far from increasing stability,
mental oscillations, disturbances and recovery complex interactive linkages increase the oppor-
from catastrophic setbacks – backed by a pow- tunities of chaotic behaviour and, thus, should
erful dispersive mobility of organisms among decrease stability. Appropriate modelling of phy-
spatio-temporal patches. The same mechanisms toplankton species composition concurs with this
resist the extinction of the majority of rarer sub- argument (Huisman and Weissing, 2001; Huis-
dominant species. The conclusion concurs with man et al., 2001). Yet it is perfectly evident and
the theoretical predictions of Huston (1979) and generally accepted that complex ecosystems con-
the models of Levin (2000), both based on the form to the ideal of low-CV stability, oscillat-
behaviour detected among terrestrial ecosystems. ing about one or another characteristic stable
Once again, the small timescales of plankton condition. May (1977) memorably analogised the
behaviour to a ball in a depression on an uneven cycle has remained the prerogative of just one or
surface, gently rocked by external forcing. The two species. Any of the others of the same func-
movements of the ball are focused on a stable tional group is presumably able to fulfil the role
location position; however, it remains liable to of the dominant, should the incumbent species
a more violent, chaotic event that may dislodge underperform for any reason. Significant inter-
the system, perhaps irreversibly, into an adjacent annual differences in species composition are
hollow, representing an alternative stable state. prompted by periodic habitat variations, such as
As May (1973) had deduced, this means that temperature structure and wind-mixing, forced
ecological stability depends upon the stability by El Niño activity (Karl, 1999, 2002).
of the physical habitat and that the determin- The observation is relevant to the question,
ing interactive linkages are not at all random raised at the outset of Section 7.3.3, about the
but are a selected set. Viewed from the stand- importance of high diversity to the function-
point of phytoplankton ecology, the first of these ing of complex ecosystems. The early debate lay
conditions is explicit (see p. 317). The notion between two extremes. The ‘rivet theory’ (Ehrlich
that, where physical constraints allow, increas- and Ehrlich, 1981) supposed that because every
ing species richness helps to stabilise aggregate species plays a part in the ecosystem, like the riv-
community responses is also well established. ets in an aircraft, each one that is lost impairs
Cottingham et al. (2001) recently reviewed sev- performance, eventually to a point where it can
eral studies that measured species richness and no longer fly. At the other end of the spectrum,
variability. They concluded that the evidence it was recognised that, provided each of its essen-
for tight coupling of these properties is not as tial DEU functions is fulfilled (see p. 351), it is per-
unequivocal as they had supposed previously. fectly possible for species-poor systems to func-
Methodological difficulties of sampling and scal- tion adequately. Thus, most species tend to be
ing in the original studies may have contributed functionally redundant within their habitats (B.
to their findings. However, they were able to ver- H. Walker, 1992; Lawton and Brown, 1993). Nowa-
ify that, while fluctuations in the populations of days, a more consensual understanding pervades,
individual species vary independently of species that the contributions of individual species to
richness, or may actually increase with greater given systems are generally unequal. Some, the
species richness, the total biomass frequently so-called ‘keystone species’ (Paine, 1980), will ful-
aggregates around a stable level. In other words, fil the role of major repositories of organic car-
community variability decreases with increased bon or they may play a disproportionately bigger
species richness, despite possibly increased vari- role in energy transfer. Some species may play
ability in the contributions of individual species. a structural role in the sense that they modify
This concurs with analyses of zooplankton data (or ‘engineer’: Jones et al., 1994) the environment
subject to planktivory. Ives et al. (1999, (2002) pro- to the benefit of other exploiters (forest trees
posed that, subject to the condition of low envi- and the microhabitats and trophic niches they
ronmental variability, increasing the number of furnish come immediately to mind). This still
species in a community decreases the coefficient leaves a large number of species that are func-
of variation of the summed species concentra- tionally superfluous or, at best, mere ecosystem
tions (aggregate biomass). passengers.
With reference to the interannual variabil- The results of some ingenious field experi-
ity of the supposedly most stable phytoplankton ments, involving terrestrial herb communities,
structures (those of the North Pacific Subtropical help us to resolve a general view of the func-
Gyre; see p. 304), we are now able to appreciate tional role of species richness and local species
the disproportion of the long-term species rich- biodiversity. Wilsey and Potvin (2000) reduced
ness. Just 21 of the 245 species recorded over 12 the numbers of dominant plants from old-
years by Venrick (1990) have together contributed field communities, though without reducing the
90% of the aggregate biomass, while the long- overall species richness. They found that aggre-
term dominance of any given stage of the annual gate biomass increased in proportion to evenness,
independently of which species had previously network linkages (for further discussion, see
been dominant. In the experiments of Wardle Loreau et al., 2001).
et al. (1999), plants of the most aggressive of
the grass species present, Lolium perenne, were 7.3.5 Structure and dynamics of
removed altogether from a perennial meadow phytoplankton community assembly
in New Zealand. There followed an increase in In order to weave together the multiplicity of
biomass of the species remaining, while rich- threads gathered in this exploration of the pro-
ness was increased by germlings of invading cesses governing the assembly of phytoplankton
species. For a time, at least, a broadly similar communities, some conclusions now need to be
function was maintained in the grassland. When aligned.
they removed all the plants, of course, there
were immediate repercussions in other ecosys- (1) Development of phytoplankton can take
tem components, most notably among the nema- place, subject to the assembly rules proposed
tode consumers and their predators. Wardle et in Table 7.8, within the constraints of the
al. (1999) deduced that the deliberate exclusion environmental carrying capacity. The latter
of the dominating (high-exergy) species promotes may be set by the nutrient resources avail-
the next fittest of the species available within the able or the daily harvestable light income, as
same functional group to assume the same func- regulated under water through the interac-
tional role in the modified ecosystem. tion of mixed depth and the vertical coef-
I am not aware of a comparable experiment ficient of light extinction. While capacity
involving phytoplankton, save as an alternative remains unfilled, there can be recruitment of
to macrophytes and periphyton, as considered biomass through growth (Rule 1, Table 7.8). At
in Chapter 8. However, the PROTECH model (see such times, there need be no competition for
Section 5.5.5) is eminently suited to the simula- the light and resources available. The oppor-
tion of this kind of manipulation. Elliott et al. tunity is exploitable by all species present
(1999a, 2001b) compared the rate of development, and whose minimal requirements are satis-
the maximum biomass attained and the diversity fied. Of these, the species with the highest
of phytoplankton communities serially stripped exergy and fastest sustainable net rates of
of the best-performing species. In each instance, increase growth should benefit most (Rule 2,
attainment of capacity was noticeably delayed Table 7.8). Spare capacity is thus beneficial
but evenness among the remaining species was to net productivity (Pn /B) and to increased
increased. species richness. Equally, if biomass is equal
The essential contribution of otherwise appar- to, or exceeds, the current capacity, recruit-
ently functional redundant species to commu- ment is weak. Survivorship is influenced in
nity and ecosystem function seems to lie in their favour of species having greater affinity for
potential to assume primacy when the perfor- the capacity-limiting factor and/or superior
mance of the existing dominants is impaired adaptations to withstand the adverse con-
for any reason, either internal (due to interac- ditions. Competition for the limiting factor
tion with other species) or external (imposed works against species richness but competi-
changes in filtration by environmental variables). tive exclusion is slow if the development of
A resilient ecosystem is characterised by a net- even the superior contenders is constrained.
work of energy-flow linkages, whose individ- (2) The locally available species (richness)
ual connections may strengthen or weaken in depends, in part, upon the local ‘seed bank’
response to fluctuations elsewhere. Interventions of propagules (Rule 3, Table 7.8). However,
affecting the performance of key species and, the high rates of transmissibility among
in consequence, those to which they are troph- phytoplankton (low z value in Eq. 7.3) make
ically linked, are compensated to a great extent for abundant immigration opportunities
by the enhanced performances of hitherto sub- and high rates of invasion or reinvasion
dominant species and the strengthening of their from remote sites. Relative importance of
local or invasive recruitment varies among through succession. Given sufficient time and
phytoplankton, mainly with relative size and with constant probabilities, the possible cli-
r–K selectivity. mactic endpoints of succession are few in
(3) Species richness may vary through the number and predictable for the region, its cli-
dynamics of local recruitment and local mate, catchment and limnological character-
extinction. There is evidence of long sys- istics.
tem ‘memories’, with vestigial populations re- (6) Species diversity in accumulating communi-
enacting the seasonal cycles of growth and ties is likely to be relatively higher than in
attrition evident in previous days of abun- maturing successions. Species diversity is also
dance. Although some prominent species greater where there is structural diversity of
return to dominance year after year for long the habitat (such as in a stably stratified lake).
periods, there may be a slow change in Species richness in accumulating communi-
the species representation and richness (Ven- ties is not generally related to productivity
rick, 1990). This turnover may be slow on or trophic state. According to Dodson et al.
interannual scales but it may vary over cen- (2000), species richness shows a flattish but
turies (a fact exploited by palaeolimnologists unimodal relation to area-specific primary
reconstructing habitat changes at the millen- production, peaking in the range 30–300 C
nial and greater scales: Smol, 1992; Smol et m−2 a−1 (which range covers moderately olig-
al., 2001; O’Sullivan, 2003). Species richness otrophic to moderately eutrophic systems).
and/or dominance is liable to abrupt alter- (7) Autogenic maturation of communities is
ation in the wake of critical environmental exposed to disruption by allogenic physical or
change (eutrophication, acidification). chemical forcing. This the communities may
(4) Planktic systems are continuously liable to survive (resistance) or recover from (resilience)
invasion by species that are not already or the community is restructured through a
present or abundant. Such invaders do not forced replacement of biomass and dominant
become prominent unless their adaptations species by others more suited to the new con-
are better suited to the local environment ditions. The response is disturbance. Commu-
than are those of existing species (a recent nity disturbances imposed at a frequency of
example is the spread of Cylindrospermopsis two to four algal generation times maintain
raciborskii, as catalogued by Padisák (1997)) the highest mix of available species and thus
and achieve a higher exergy, with faster net high levels of diversity. By virtue of the near
recruitment rates, than existing dominants. prevention of competitive exclusion, distur-
(5) While habitat variability remains slight, com- bances of this frequency constitute a major
munities develop towards a steady state, gen- driver of continuing high diversity and rich-
erally dominated by one of a small selec- ness levels (p. 377). However, the mobility of
tion of suitably adapted species (Rules 4–7, species and their ability to re-establish from
Table 7.8). This autogenic development has other locations may be critical to this process.
long been known (as succession); the properties The mechanisms resist the extinction of the
of maturing successions are also well known majority of rarer subdominant species.
(Table 7.7). The long-standing desire to explain (8) Annually recurrent cycles of seasonal algal
successional sequences of species composi- dominance suggest a high level of inter-
tion must defer to the evidence that there annual constancy in pelagic habitats. Simi-
is no continuous sequence of species, rather lar environmental conditions, roughly reca-
a probability of certain outcomes relating to pitulated each year, renew similar filtration
the fitness and adaptations of the species pool effects upon a species pool, which in conse-
available (Rules 8 and 9, Table 7.8). Succes- quence becomes biassed in favour of species
sion is a cycle of probabilities of replacement that have grown well in recent seasons.
of dominant species. Diversity (though not The development of ‘commonness’ among
necessarily overall species richness) declines regional species pools also influences the
SUMMARY 385
similarity of seasonal pattern among region- the physical environment in regulating both
ally similar types of aquatic systems, consid- the abundance and composition of the phyto-
ered in Sections 7.2.1 and 7.2.3. plankton. These provide the axes of the gener-
alised habitat templates, developed by Smayda
and Reynolds (2001), to which the main assem-
7.4 Summary blages of marine phytoplankton are consistently
aligned.
Patterns in the abundance and composition of Species assemblage patterns in lakes respond
natural phytoplankton assemblages in the sea to analogous drivers, with some strong coher-
and in lakes are sought in this chapter. In ences among major floristic components with
the sea, characteristic floristic associations with regional climates, lake morphometries and
the extent, longevity and supportive capacity catchment-derived nutrient loads. Among the
of the water in the major oceanic circulations world’s largest lakes, pelagic environments
are demonstrated. Environmental distinctions resemble the open ocean in the deficiency of
from adjacent shelf and coastal areas and from nutrients and the importance of mixing, and
localised upwellings, coastal currents and fronts they show a similar community organisation
are shown frequently to offer greater carrying based upon picophytoplankton and microbial
capacities, higher levels of biomass and different processing of fixed carbon. However, at all lat-
planktic associations. itudes, density stratification (including under
The tropical gyres of the Pacific, Indian ice) is a prerequisite of significant communal
and Atlantic Oceans are profoundly oligotrophic, growth and assembly. Among moderate to small
nutrient deficient and permanently stratified lakes, phytoplankton abundance and composi-
beyond a depth of ∼200 m. Phytoplanktic tion is demonstrably constrained by nutrient
biomass is severely constrained, often <20 mg supplies and by the underwater light climate.
chla m−2 . For long periods, the dominant Both are instrumental in setting limits on the
primary producers are picoplanktic Cyanobac- carrying capacity and in influencing the adap-
teria, the fixed carbon being cycled mostly tive requirements of the favoured phytoplank-
through the microbial food web. Changes in ton species. Seasonal periodicities, broadly mov-
near-surface stratification, wrought by weather ing from (w-selected) diatom abundance in mixed
events or longer-term climatic fluctuations, stim- columns to colonist (r-selected) nanoplankters
ulate episodes of recruitment of nitrogen-fixing and then to increasingly specialist (K-selected)
cyanobacteria (Trichodesmium) and diatoms (Hemi- microplankters, reveal consistent patterns. How-
aulus) or deep-migrating dinoflagellates. ever, differences in maximum biomass and in
In contrast, primary production in the high dominant functional types usually reflect differ-
latitudes is constrained by strong seasonal fluc- ences in resource richness. Phosphorus-deficient
tuations in light income, temperature and phys- oligotrophic lakes support low biomass com-
ical mixing. A small number of diatom species prising distinctive diatoms (Cyclotella bodanica
are overwhelmingly dominant, but dinoflagel- group, Urosolenia), desmids (Staurodesmus, Cosmar-
lates and the haptophyte Phaeocystis are some- ium), chrysophytes (Dinobryon) and dinoflagel-
times abundant. In boreal and mid-latitudinal lates (Peridinium). Picophytoplankton may make
waters, diatom dominance alternates seasonally up a substantial proportion of the biomass.
with nanoplankter and dinoflagellate abundance. In eutrophic (P-rich) lakes, other diatoms, nos-
In shelf and coastal waters, supportive capacity is tocalean Cyanobacteria and self-regulating Cer-
generally much higher than in the open ocean atium species may have eventually to be able
but the more abundant phytoplankton shows to contend with shortages in the rates of
strong seasonal periodicity. supply of carbon and nitrogen, while high
Margalef’s (1958, 1963, 1967, 1978) deep turbidity may force light constraints favour-
insight has contributed greatly to a broad under- ing filamentous cyanobacteria, diatoms and
standing of the separate roles of nutrients and xanthophytes.
These behaviours help to explain the fit of ultimately depend upon their specific evolution-
groups of species (trait-selected functional types) ary attributes and adaptive traits and upon their
to analogous templates defined by axes scaled appropriateness under the environmental condi-
in limiting resources and harvestable underwa- tions obtaining. Conversely, it is deduced that
ter light. These templates are suggested to rep- the species compostion that is typical in a par-
resent the filterability of species by pelagic habi- ticular water body, or in a particular type of
tat conditions: where light and nutrients are not water body, is biassed by the conditions typi-
constraining, many species are potentially able cally obtaining. Moreover, other species that are
to grow, given that suitable inocula are present. frequently present in the same specific environ-
Successful species will have evolved rapid-growth, ments, as a result of simultaneous recruitment,
exploitative life-history (C) strategies that are share common traits and are thus allied to the
preferentially (r-) selected under these conditions. same trait-selected functional groups.
Moving down the gradients of resource and of Phytoplankton communities can continue to
availability is analogised to ever constricting fil- develop so long as nutrients and harvestable light
tration of (R- or S-) specialisms that will be are available to sustain it. Once either capacity
increasingly K-selected by the ever more exacting is exceeded, however, recruitment rates weaken.
constraint. Adaptive traits representing greater affinity for
Analysing the mechanisms behind the observ- the limiting factor, or greater flexibility in access-
able patterns, the near ubiquity of common phy- ing or overcoming the deficiency assume pre-
toplanton species is deduced to rely on efficient mier importance. Species with the appropriate
and highly effective dispersal mechanisms. These traits to withstand the deficiency survive while
are comparable to those of bacteria, some pro- populations of more generalist species stagnate
tists and the micropropagules of sedentary inver- or regress. Competition for the limiting factor
tebrates. The ability of phytoplankters to reach works against species richness. Given sufficient
suitable habitat soon after it becomes available time, competitive exclusion leaves only the supe-
means that their ecological behaviour is readily rior contender.
describable in the terms of island biogeography This mechanism underpins the long-
(MacArthur and Wilson, 1967). recognised process of succession. The equally
The assembly of biomass by colonist species long-standing desire to explain successional
(ascendency) is subject to the flux of energy and sequences of species composition are found to be
carbon, and the size of the base of resources unhelpful. Autogenic maturation of communi-
that is available to sustain a supportable biomass. ties is exposed to disruption by allogenic physical
The standing mass and abundance of phyto- or chemical forcing. Community structure may
plankton is also influenced by the fate of pri- resist or recover in the wake of such episodes, or
mary product – to metabolism, or to the loss the community is restructured through a forced
of intact cells by sedimentation or consumption replacement of biomass and dominant species
by phagotrophic or parasitic heterotrophy. There by others more suited to the new conditions.
remains a strong element of fortuity about phy- The response is disturbance. Such disturbances
toplankton composition, what has arrived, how imposed at a frequency of two to four algal
well it can function there (or how well it is generation times contribute to the endurance of
allowed to do so). However, accumulating pop- high levels of diversity.
ulations interact, with dynamic consequences. The diversity of phytoplankton, for so long
Assembly becomes increasingly contingent upon considered paradoxical, is found to be main-
the sum of accumulating behaviours, conform- tained by a combination of variable forces – envi-
ing to patterns (‘rules’) summarised in Table 7.8. ronmental oscillations (habitat instability), more
Of those present, the species initially likely to severe disturbances and recovery from catas-
become dominant are those likely to sustain the trophic forcing – backed by the powerful disper-
fastest net rates of biomass increase. These will sive mobility of organisms.
Chapter 8
Phytoplankton ecology and aquatic

ecosystems: mechanisms and management
aging, flight, reproduction) that contribute to the

8.1 Introduction survival and genomic preservation of the con-
sumer species in question. In thermodynamic
The purpose of this chapter is to assess the terms, the food web serves to dissipate as heat
role of phytoplankton in the pelagic ecosystems that part of the solar energy flux that was pho-
and other aquatic habitats. The earliest suppo- tosynthetically incorporated into chemical bonds
sitions to the effect that phytoplankton is the (see p. 355).
‘grass’ of aquatic food chains and that the pro-
duction of the ultimate beneficiaries (fish, birds
and mammals) is linked to primary productiv- 8.2.1 Fate of primary product in the
ity are reviewed in the context of carbon dynam- open pelagic
ics and energy flow. The outcome has a bearing Based upon this simple premise, the performance
upon the long-standing problem of phytoplank- of pelagic systems can be quantified in units of
ton overabundance and related quality issues in energy dissipated (or of organic carbon reduced
enriched systems, its alleged role in detracting and re-oxidised) per unit area per unit time, by
from ecosystem health and the approaches to its a biomass also quantified in terms of its organic
control. carbon content (or its energetic equivalent) and
The chapter begins with an overview of the partitioned according to function (primary pro-
energetics and flow of primary product through ducer, herbivore, carnivore, decomposer). For the
pelagic ecosystems, especially seeking a reap- open sea and the open water of large, deep lakes,
praisal of the relationship between biomass and the scales of the input components and the rates
production. at which they are (or can be) processed have been
established in the preceding chapters. In Sec-
tion 3.5.1, the attainable net productive yield of
8.2 Material transfers and energy pelagic photosynthesis, across a wide range of fer-
tilities, was suggested to be typically in the range
flow in pelagic systems 500–600 mg C m−2 d−1 (the thickness of the
photic layer compensates for differences in con-
One of the essential components of ecological sys- centration of photosynthetic organisms). Extrap-
tems is the network of consumers that exploit olated annual aggregates (in the order of 100–200
the investment of primary producers in reduced g C m−2 a−1 ) agree well with the generalised find-
organic carbon compounds. Some of these are re- ings of oceanographers for the open ocean, as
invested in consumer biomass but much of the well as those deduced from satellite-based remote
food intake is oxidised for the controlled release sensing (Section 3.5.2). On the other hand, they
of the stored energy in support of activities (for- clearly underestimate the cumulative production
388 PHYTOPLANKTON ECOLOGY AND AQUATIC ECOSYSTEMS
in many small to medium lakes, as well as in product may be exported to the sediments (but
what were referred to as ‘oceanic hotspots’. Even generally ≤100 g C m−2 a−1 ) (Jónasson, 1978),
so, from an investment of PAR of some 40 kJ to be processed there by benthic food webs.
(g Corg )−1 , the synthesis of even this amount of By subtraction, the potential yield of primary
photosynthate (4–8 MJ PAR m−2 a−1 ) is acknowl- product available to pelagic microzooplankton
edged to be just a tiny fraction of the annual or to herbivorous mesozooplankton can scarcely
harvestable PAR flux (less than 1% of the solar exceed 80% of the annual primary production
energy flux) (Section 3.5.3). So far as its role in (say, 30–150 g C m−2 a−1 ) (Callieri et al., 2002).
supplying the conventional energy requirements Even these consumer pools turn over much of the
of the dependent food chain is concerned, biolog- carbon that they harvest in a matter of days; the
ical turnover represents only a small proportion proportion that is available to larger metazoans
of the dissipative flux. (10–25%; now rather less than 40 g C m−2 a−1 , or
At the same time, it is evident from the <1.6 MJ m−2 a−1 ) (Legendre and Michaud, 1999)
modest year-to-year changes in the photosyn- is yet smaller. The balance is, of course, respired
thetic biomass in the open sea that there is lit- to carbon dioxide.
tle net accumulation of primary-producer mass
or carbon. Taking chlorophyll levels of 20–40 mg
chla m−2 to be typical for the trophogenic zone of 8.2.2 Food and recruitment of consumers
the tropical ocean, phytoplankton standing crops in relation to primary production
remain steadfastly constrained, probably in the It is nevertheless worth emphasising again that
range 1–5 g cell C m−2 (0.04–0.20 MJ m−2 ). Phy- the flow of energy and materials to higher
toplankton concentrations in oligotrophic, high- trophic levels in the pelagic is a function of car-
latitude lakes may be lower still (in some cases, bon turnover rate and not of biomass per se.
≤0.5 g C m−2 ) (Section 3.5.1). These obdurately However, it is also relevant to remind ourselves
low levels of producer biomass (B), like the pro- that the general uniformity of the pelagic out-
duction (P) it yields, are deemed indicative of the puts (90 ± 60 g C m−2 a−1 ) is of a similar mag-
‘unproductive’ nature of the nutrient-poor sys- nitude to the rate of invasion of the water col-
tems. Yet, as pointed out (Section 3.5.1), a P/B umn by atmospheric carbon dioxide (see Section
yield of not less than 100 g C from not more 3.4.1). The fastest rates envisaged (up to 310 mg
than 5 g cell carbon represents an extremely high C m−2 d−1 , or ≤110 g C m−2 a−1 ) depend upon
biomass-specific productivity! winds that blow more strongly and more con-
Low areal biomass and low cumulated pro- tinuously than is generally the case. They also
duction are traditionally attributed to resource require a steep diffusion gradient from air to
poverty. Without a simultaneously available fund water, caused by severe depletion of CO2 in solu-
of other elements, including nitrogen, phospho- tion. With due allowance for turnover of the car-
rus, iron and other traces especially, new biomass bon by the cycling between photosynthesis and
and new cells cannot be built and recruited. respiration, it is feasible that the capacity of CO2 -
In the ocean and, to a large extent, in deep, limited pelagic primary production to sustain
oligotrophic lakes, very little of the photosyn- new secondary production can be up to about
thetically fixed carbon is deployed in new algal 40 g C m−2 a−1 but scarcely much more.
biomass: as much as 97% may be vented from The exploitation of even this meagre supply of
the active cells (Reynolds et al., 1985). Some of particulate organic carbon by mesoplanktic con-
this is respired to carbon dioxide but a propor- sumers depends upon successful foraging. This,
tion is released as DOC (see Sections 3.2.3, 3.5.4). in turn, depends upon adequate encounter rates
This is processed by bacteria, whose own respira- of suitable foods by consumers, which is often a
tion may account for 20–50% of the carbon thus function of food concentration. The range of con-
assimilated (Legendre and Rivkin, 2000b; Bidanda centrations of suitable food organisms providing
and Cotner, 2002). In smaller oligotrophic lakes, the encounter rates representing the minimum
a significant proportion of the original primary and the saturating levels for diaptomid copepods
MATERIAL TRANSFERS AND ENERGY FLOW IN PELAGIC SYSTEMS 389
is suggested to be 5–80 mg C m−3 (Section 6.4.3) 2000b). The amount of food required to cover the
(say, ≥5 g C m−2 in a 100-m layer). This is sustain- maintenance of body mass is about one-third the
able on a productive turnover of 100 g C m−2 satiating ration at 18 ◦ C, rising to about 50% at
a−1 , in the sense that it could yield up to 40 temperatures below 10 ◦ C.
g C m−2 a−1 and 1.6 MJ m−2 a−1 to consumers. From these data, we may compare the approx-
Supposing this was invested entirely in the pro- imate ranges between the minimum and maxi-
duction of diaptomids, a potential turnover of mum food requirements of trout at about 18 ◦ C
some 10 × 106 1-mm animals m−2 a−1 (or 2 ×106 (2–7 kJ d−1 for an 11-g fish, 25–75 kJ d−1 for a 250-
2-mm animals m−2 a−1 ) may be projected from g fish), and at temperatures below 10 ◦ C (0.4–1.5
the data in Box 6.2 (p. 281). Depending upon kJ d−1 for an 11-g fish, 5–15 kJ d−1 for a 250-g fish).
the depth of the mixed layer and the gen- Were these energy requirements to be derived
eration times of the copepods, the sustain- exclusively from copepods, a slightly lower intake
able recruitment is in the order of 100–1000 is required (Diaptomus yields ∼23.9 kJ g−1 dry
animals l−1 a−1 . weight) (Cummins and Wuychek, 1971). However,
What kind of resource is this to planktivorous to consume a minimum of 0.08 g d−1 of cope-
fish? The foods, feeding habits and bioenergetic pod (for an 11-g fish at 18 ◦ C), or up to 3.1 g d−1
requirements of several commercially important (for a 250-g fish), requires very efficient foraging.
species are relatively well studied. The models Referring to the data in Box 8, between 2500 to
of Kitchell et al. (1974, 1977) and Post (1990) 100000 2-mm calanoids would need to be cap-
have been widely applied in fish management. tured and eaten each day or between 13000 and
The work of Elliott (1975a, b) showed particu- 500000 smaller (1-mm) animals. Given potential
larly well how the maximum rate of growth of copepod recruitment rates equivalent to 10 ×106
brown trout (Salmo trutta) varies with tempera- 1-mm animals m−2 a−1 (say 27.4 ×103 m−2 d−1 ),
ture and with the quantity of food (Gammarus) the planktivorous 250-g trout would have to har-
consumed, within the range between total satia- vest the entire daily production under 18 m2 of
tion and the minimum needed to balance basal surface (i.e. up to 1800 m3 of water) in order to
metabolism. These have also been the subject of fulfil its maximum growth rate. Just to maintain
sophisticated modelling (Elliott and Hurley, 1998, its body mass, it would have to forage the entire
1999) that simulates the earlier observations of copepod production from under 6 m2 .
performance of trout feeding upon invertebrates Though plainly approximations, they serve
(Elliott and Hurley, 2000b) and reflects differ- to show two things about the scale of demand
ences when other food sources are offered (Elliott of planktivory. First, that truly pelagic sys-
and Hurley, 2000a). Information based upon only tems, exchanging carbon only with the atmo-
one species of fish is not assumed to apply to all sphere, are bound to be oligotrophic and capa-
others but the data serve to establish some impor- ble of sustaining only very low densities of
tant deductions about pelagic foraging. The max- fish (averaging, perhaps, in the order of 1–10
imum growth rate (and, hence, the maximum g fresh weight m−2 ). Second, that the ability
satiating food intake) of all sizes of trout inves- of large pelagic consumers to harvest planktic
tigated increased roughly fivefold between 5 ◦ C production really does require specialist forag-
and its maximum at around 18 ◦ C, before tapering adaptations. Indeed, pelagic feeding herring-
ing off quickly at temperatures >20 ◦ C. Growth- like fish (clupeoids), including shads, freshwa-
saturating food intake was absolutely greatest in ter salmonids, coregonids (whitefish and ciscos)
the largest fish examined (250 g), being equiva- and tropical stolothrissids, are all characteris-
lent to ∼4.5 g Gammarus d−1 , or ∼75 kJ d−1 or tically strong swimmers, capable of covering
∼1.9 g C d−1 . Weight-for-weight, however, small vast distances in the near-surface layer. All have
fish eat more, 11-g fish consuming up to 400 mg large gill rakers that confer enhanced capabilities
Gammarus d−1 , or about 7 kJ d−1 . In either case, for straining small particles from several cubic
the efficiency of conversion of ingested carbon metres of water passed over the gills each day.
to trout biomass is 30–35% (Elliott and Hurley, To adult fish of almost all other species, pelagic
planktivory is normally a nutritionally unreward- Section 3.4.1 and cited work of Cole et al. (1994)
ing means of foraging. and Richey et al. (2002).
To be a net source of carbon dioxide (out-
gassing) also indicates that the lake is oxidis-
8.2.3 Carbon subsidies in lakes and the sea ing more organic carbon than was fixed in
In coastal and shelf waters, in estuaries and situ. The short-retention lake must be a repos-
embayments and in small to medium-sized lakes itory of organic carbon from the catchment,
(according to the dimensional ranges proposed delivered in the inflow. Particulate organic mat-
in Section 7.2.3: inland waters of less than 500 ter (POM, of various sizes), derived from ter-
km2 in area), cumulative net primary production restrial ecosystems, especially including anthro-
often exceeds the 100–200 g C m−2 a−1 , some- pogenic interventions (agriculture, industries,
times considerably so. Values as high as 800–900 settlements) is transported in abundance by
g C m−2 a−1 have been reported (see Section 3.5.1). inflowing streams and rivers. Yielding slowly to
Primary production at these levels is possible in microbial decomposition in the receiving water
one or both of two ways. One arises through a (or, more likely, on is bottom sediments), this
much-accelerated metabolic turnover of carbon material releases its component carbon dioxide
synthesis and respiration; the other is the result in solution. Quantification of these contribu-
of the system receiving additional carbon from tions is still in relative infancy. External POM,
external sources. Such subsidies come in various derived mainly from the terrestrial plant pro-
forms but their provenance is largely from terres- duction of open moorland, was shown to be an
trial sources. They are tangible in the sense that important food source in the oligotrophic, base-
they support finite sedimentary fluxes, known poor water of Loch Ness, Scotland, UK. Analysing
to amount to 100–300 g C m−2 a−1 in partic- seasonal variations in the stable-isotope composi-
ular eutrophic lakes (Jónasson, 1978). Inflowing tion, Grey et al. (2001) determined that the crus-
rivers are generally well equilibrated so far as tacean zooplankton, dominated by Eudiaptomus
their dissolved carbon dioxide content is con- gracilis, derives some 40% of body carbon through
cerned (typically 0.5–1.0 mg CO2 L−1 , or ∼0.15–0.3 ingestion of allochthonous particles; only the
g C m−3 ) (see Section 3.4.1). Delivered to the sea late-summer population of filter-feeding Daphnia
or to the inland water in solution, it is also seems to derive most of its nutrition through
immediately available for photosynthetic uptake. autochthonous pelagic producers. External POC
Maberly’s (1996) investigation of the carbon spe- sources are likely to be relatively less important
ciation in Esthwaite Water, a small (1.0 km2 ), in large lakes. Ozero Baykal is functionally olig-
eutrophic, soft-water lake in the English Lake Dis- otrophic, where at least 90% of the annual car-
trict (Section 3.4.1), showed that the lake received bon and oxygen exchanges take place in the open
much more of its inorganic carbon in the inflow pelagic. However, external POC sources should
streams than directly from the atmosphere. This not be discounted. Even in a lake as large as
is despite phytoplankton-driven summer episodes Michigan, stream and groundwater inputs con-
of high pH and CO2 depletion when the atmo- tribute as much as 20% of the usable carbon sup-
sphere became the principal source. For most ply to the ecosystem (Bidanda and Cotner, 2002).
of the year, pH is near neutral and the concen- Interestingly, the largest component of terrestrial
tration of free CO2 in the lake is close to 0.12 organic carbon contributed to lakes and seas is
mol m−3 (1.4 g C m−3 , or up to seven times the often dissolved humic matter (DHM) (see Section
atmosphere-equilibrated concentration). Far from 3.5.4) and, at first sight, the most likely subsidy
relying on CO2 invasion from the atmosphere, to bacterial metabolism and the microbial food
the net flux is in the opposite direction, the web. Although some DHM is amenable to bacte-
lake venting CO2 . This relative importance of the rial degradation, the metabolic yield, in terms
inflows in supplying inorganic carbon is likely to of both energy and liberated carbon dioxide, is
be general among lakes wherever the inflows dis- generally slight (Tranvik, 1998), and production
place the lake volume in less than a year (see also in oceanic bacteria remains more constrained by
the turnover of carbon than nutrients (Kirchman, that shallowness also assists the concentration of
1990). areal production and of the founding resources,
Frequent recycling of inorganic carbon should grants to shallow systems the opportunity to
be relatively neutral in the carbon profit-and-loss recycle raw materials efficiently and nearly
accounting. When the same atoms of carbon are continuously.
first incorporated into carbohydrates and then Shallow lakes (and the shallow margins of
released in algal, bacterial or microplanktic res- deeper ones) also offer potential habitat to macro-
piration, the net yield is tiny in comparison to phytic vegetation, which, once established, alters
the amount of carbon fixed. When this happens the turnover, internal stores and the direction
several times in a year, the aggregate of fixa- and dynamics of carbon pathways. Rooted lit-
tion (say 100 g C m−2 ) may still have contributed toral plants fix carbon into carbohydrates but,
little in the way of either producer output or unlike phytoplankton, they generally have the
export (perhaps <20 g C m−2 ). The balance (≤80 ability to retain and store carbohydrate poly-
g C m−2 ) would have been fixed, respired and mers within their tissue, rather than venting
refixed several times over. In this case, the 20 g C it back to the water. Through their rooting sys-
may have reached the bodies of large pelagic tems, they have potential access to nutrients nec-
animals or have been consigned to the basal essary for protein synthesis that are not avail-
sediments. able to phytoplankton. Relatively slower cycles
Much of the carbon fixed in biomass at the of growth, maturation, death and decomposition
bottom end of the trophic web is actually still also contribute to the long retention of resources
quite labile, with about two-thirds of it being in biogenic products. Macrophytes may also pro-
restored to inorganic components in less than a vide substratum for epiphytic algae as well as
year (Jewell and McCarty, 1971) (see Section 6.6). It habitat and nutritional refuge for a wide vari-
is interesting that, in the open sea, much of that ety of benthic invertebrates – herbivores, detri-
process takes place in the upper 200 m, before tivores and their predators. Littoral and sublit-
the material has passed into the more permanent toral benthos also offer far more attractive for-
depths, whence its subsequent return may be aging opportunities to most fish, mainly because
very delayed indeed. In many small to medium- potential prey organisms are both larger and dis-
sized lakes, however, planktic cadavers may reach tributed more nearly in two dimensions rather
the bottom of the lake within 200 m of the sur- than three. The reward of 4 g of Gammarus or
face. Here, they are substantially protected from Asellus is considerably more attainable to a 250-g
turbulent re-entrainment and they stay more or trout than gathering the same mass of zoo-
less where they settle. Though persistently low plankton. Offshore sediments, supplied in part
temperatures and a lack of oxidant in the inter- by autochthonous organic carbon from the phy-
stitial water (and perhaps the bottom part of the toplankton, also nourish detritivore-based chains
adjacent water column) may considerably slow of invertebrate consumers. These too mediate an
the rate of decomposition, benthic detritivory, alternative producer–fish trophic link, passing by
bacterial decomposers and bacterivorous proto- way of benthic macroinvertebrates, and avoid-
zoa nevertheless continue to mineralise sedimen- ing the tenuous bridge of a diffusely dispersed
tary organic material and release inorganic car- zooplankton.
bon. The return of carbon and other inorganic Empirical comparisons of the biomass and
elements to the trophogenic layers is extremely production of successive trophic levels in
slow when compared to the open sea. However, lakes of various depths and trophic states
in shallow lakes, where the bottom sediments have been used to determine the strength of
are in direct contact with the trophogenic layer phytoplankton–zooplankton linkages. Many fac-
and are liable to frequent entrainment by pene- tors are involved but energy-transfer efficiencies
trating turbulent eddies (Padisák and Reynolds, are found to be generally greater in deep lakes
2003), recycling of sedimentary materials back to than in shallow lakes, where the alternative path-
the water column is greatly facilitated. The fact ways are more likely to be available (Lacroix et al.,
1999). Where higher consumers are involved, ments from 2 m3 and the 11-g animal could be
modern techniques invoking the determination fully sated by clearing 200 L d−1 .
of the ratios of stable isotopes in the body mass Under these circumstances, feeding on zoo-
and the foods of predators reveal much about plankton once again becomes a reasonably
the principal pathways exploited. In a recent rewarding foraging option. The progressive
overview, Vadeboncoeur et al. (2002) were able to switching between benthivory and planktivory
confirm the relative importance of zoobenthos in the feeding of young roach (Rutilus rutilus)
in the diets of a wide variety of adult fish from in response to the abundance of planktic crus-
lakes in the north temperate zone. In their bioen- taceans, as revealed by Townsend et al. (1986), has
ergetic analysis of the fish production in the been highlighted earlier (see p. 278). The thresh-
large tropical Lac Tanganyika, Sarvala et al. (2002) old of ∼1 g m−3 zooplankton carbon is not a fixed
showed that the non-clupeid species were sub- one but the behaviour helps to explain the coex-
stantially supported other than by planktivory. istence of moderate concentrations of crustacean
Copepods were the principal food only of the zooplankton and potential fish predators.
dominant Stolothrissa tanganicae; alternative foods It is worth emphasising again that, quite
also featured in the diet of another clupeid, Lim- apart from the habits and preferences of the
nothrissa miodon. adults, the quasi-planktic 0+ hatchlings and juve-
Although benthic pathways probably provide niles of a wide range of non-pelagic fish species
the dominant link in the production of most feed almost exclusively in the plankton (Mills and
species of fish in many small and medium- Forney, 1983; Cryer et al., 1986). Seasonal peaks
sized waters, zooplankton may still be readily of phytoplankton production and zooplankton
exploited where and when it is abundant. High recruitment, especially the vernal pulse in tem-
concentrations of microzooplankton are neces- perate lakes and marine shelf waters, provide
sarily sustained by correspondingly high avail- the major feeding opportunity to young-of-the-
abilities of appropriate particulate organic car- year fish. Generally, they are recruited in large
bon, which include fine detritus, bacteria, micro- numbers to the pelagic, often coinciding with
zooplankters and, especially, planktic algae. Algal the maximum recruitment phases of zooplank-
abundance requires not just the subsidised car- ton. Survivorship is generally poor but individual
bon fluxes but the additional biomass supportive growth is potentially rapid. In temperate lakes,
capacity provided by an ample nutrient supply. the rates of recruitment and mortality experi-
Concentrations of ingestible algae (nanoplank- enced by the young-of-the-year are closely cou-
ters and small microplankters, generally <25 µm) pled to the seasonality of planktivore activity
(see p. 267) substantially greater than 100 mg and the optimal exploitation of the zooplankton
C m−3 are adequate to support growing popula- resource (Mills and Forney, 1983; Scheffer et al.,
tions of filter-feeding daphniids. These can grow 2000).
quickly and recruit further generational cohorts, At the same time, planktivory inevitably
in a matter of days rather than months. With- depresses the numbers and impacts upon the
out predation, daphniids can go on to achieve size distribution and recruitment rates of the
aggregate filtration rates that may well exhaust zooplanktic prey. On the basis of quantities con-
the food supply altogether. This level of filtra- sidered in this section, sustainable planktivory
tion capacity may be developed by 20–30 large might yield between 10 and 100 g C m−2 a−1
(∼2-mm) or 200 small (∼1-mm) Daphnia L−1 to consumers, depending upon the quality of
(Kasprzak et al., 1999) (see also p. 267). Multiplied zooplankton nutrition. Planktivory also has an
by the respective carbon contents (Box 6.2), the upper limit of sustainability (in the sense of not
equivalent biomass of filter-feeders is about 1000 exhausting the food supply): this may be encoun-
µg C L−1 (1 g C m−3 or ∼40 kJ m−3 of potential tered in carbon-rich ponds at fish densities equiv-
energy to a consumer). Now weighed against the alent to 30 g m−2 fresh weight (Gliwicz and Preis,
requirements of Elliott’s (1975a) trout, the 250- 1977). Among planktic systems generally, direct
g fish could expect to harvest its daily require- planktivory is likely to sustain mean biomass in
the order 1–10 g C m−2 . Larger consumer masses

are indicative of subsidies to the carbon produced
in the pelagic.
8.2.4 The relationship between energy

flow and structure in the plankton
What emerges is that phytoplankton provides
almost the unique capacity to supply carbon to
aquatic food webs in large, truly pelagic sys-
tems of oceans and large deep lakes. They are
unable to support a large biomass (as a conse-
quence of resource poverty) and they support no
greater net areal biomass production than the
rate of turnover of dissolved inorganic carbon
will allow. Terrestrial subsidy to the carbon may
relieve somewhat the constraints upon carbon Figure 8.1 Carbon- and energy-flow constraints on the
deployment: tangible production and biomass structuring of emergent pelagic communities. Accepting that
recruitment yields can be higher but are no the amount and distribution of native carbon sources vary
over several orders of magnitude, phytoplankton composition
longer reliant on phytoplankton production. The
varies with carbon dynamics, while the concentration of food
smaller and the shallower is the water body, the
particles determines the type and productivity of the
greater is the probability that the ecosystem ener- zooplankton and, in turn, the resource and its relative
getics will be powered by the adjacent hydrologi- attractiveness to fish. Shaded areas represent the transition
cal catchment, the weaker is their dependence but is generally close to a carbon availability of ∼0.01 mmol
upon phytoplankton production and the more L−1 in each instance. Redrawn with permission from
integrated are littoral and benthic pathways into Reynolds (2001a).
the food webs.
The predominant species and their organisa-
tional structures in the pelagic are thus closely level of CO2 solution are exclusively the conse-
coupled to the carbon dynamics of the entire sys- quence of withdrawal by photosynthetic organ-
tem: oligotrophic and eutrophic systems are dis- isms at a rate exceeding replenishment and indi-
tinguished less by nutrients than by the carbon cate the imminence of limitation of photosyn-
fluxes and the types of organisms most suited thetic rate by the carbon dioxide flux. This sit-
to their mediation. A provisional guide to the uation may be a rare occurrence in resource-
apparent thresholds separating these structural deficient, oligotrophic, ‘soft-water’ systems that
provinces was proposed by Reynolds (2001a) and are chronically unable to support high levels of
is reproduced in Fig. 8.1. The various trophic biomass. It would be yet more rare in ‘hard-water’
indicators are shown against a logarithmically- systems typified by a bicarbonate-enriched aug-
scaled spectrum of carbon availability (scaled in mented total carbon capacity. However, in either
mmol C L−1 ; 1 mmol L−1 is equivalent to 12 case, a more ample supply of nutrient resources
g C m−3 ) to register critical biological bound- not only sustains higher levels of biomass but
aries. Thus, the first row represents the avail- places greater demands on the capacity to sup-
ability of total CO2 : concentrations higher than ply carbon dioxide. Thus, enrichment ultimately
the air-equilibrium level of CO2 gas in water selects against phytoplankters with a low affinity
are due to the reserve of dissolved CO2 main- for carbon dioxide (like freshwater chrysophytes)
tained by internal recycling and external ter- and in favour of species (such as the group H1,
restrial subsidy, as well as to the store of dis- LM and M Cyanobacteria) (see Tables 7.1 and 7.2).
solved bicarbonate ions present. Instances where That the carbon capacity underpins the
CO2 concentration falls below the air-equilibrium nature and abundance of the POC availability to
planktic consumers also prejudices the selection ‘habitat template’ to accommodate the favoured
of zooplankton. Low POC demands the sophisti- life-history traits and ecological strategies of ter-
cated location and flexible capture adaptations restrial communities. Reynolds (2003a) reviewed
of calanoids; mechanised filter feeding is shown a series of attempts to develop an analogous
in Fig. 8.1 as being realistically sustainable only template for freshwater communities in a way
above 0.008 mmol L−1 (0.10 g C m−3 ). The thresh- that partitioned quantitatively the distinguish-
old of zooplankton abundance for planktivory to ing thresholds (or controlling nodes).
be attractive to adults of non-specialised species The outcome is presented as Fig. 8.2. Against
of freshwater is inserted at about the same level. logarithmically scaled axes representing the
Zooplankton populations at any lower concentra- resource constraints upon supportable biomass
tions can only be of interest to specialised plank- and the processing constraints upon the rates
ton feeders and send non-specialists inshore, to of its assembly, the space is divided up accord-
forage for benthos. ing to the most likely limitations regulating
There have been other important attempts the system. The resource axis is the easier
to relate species structures or, at least, the to explain, with the upper limit of biomass
dominant functional traits of ‘keystone’ species, depending upon the lowest relative bioavailabil-
to given habitats. Setting zoogeographical con- ity of nutrient (corresponding stoichiometrically
straints to one side, habitat characteristics pro- scaled axes for bioavailable N and P are inserted).
vide the most relevant filter of regional (or γ-) Low resource availability is thus the key to habi-
diversity (see Section 7.3.3). This has been demon- tats in which the elaboration of biomass usually
strated strikingly in the species structure of local confronts ‘nutrient-limiting’ conditions. The pro-
fish assemblages in different parts of contiguous cessing axis is constructed on the basis of photo-
rivers and to be strongly predictable from their synthetic assimilation rates and their various
adaptive traits (Lamouroux et al., 1999). At the dependences upon the fluxes of phota and inor-
level of primary producers, the direction and fate ganic carbon. Towards its right-hand side, assem-
of carbon is integral to the opportunities for its bly rates of primary producers and their depen-
use and the adaptive traits that are advantageous. dent heterotrophs and phagotrophs are light-
The conceptual explanations originally proposed dependent and critically constrained by mod-
by Legendre and LeFevre (1989) and based upon est harvestable PAR fluxes (‘photon flux-limiting’
hydrodynamical singularities have been refined conditions). Moving leftwards, photosynthetic
progressively to a model of regulation by food- assimilation rates become light-saturated but the
web ‘control nodes’ (Legendre and Rivkin, 2002a). carbon-flux capacity is increasingly strained to
In essence, the size and the trophic structure of the limits of the invasion rate of carbon diox-
the marine plankton direct the flow of carbon ide from the atmosphere and the various catch-
through particular, optimised channels (cycling ment sources of carbon. With increasing exter-
in the web, export to the benthos or bathypelagic, nal sources of organic carbon, rates of processing
etc.). The ‘deciding’ species structures are those moves from being less dependent on the carbon
whose functional traits best fulfil the opportuni- supply and more so upon the supply of oxidant
ties of the habitat. needed to make it available.
The idea that the habitat is the best predic- These axes describe the space available to
tor of the most-favoured species traits, survival ecosystem components according to their pri-
strategies and, thus, community structures has, mary adaptations and so define a template for
of course, been around for some years (Grime, community structures (Fig. 8.3). The axes read-
1977; Southwood, 1977; Keddy, 1992). Put in the ily accommodate the typical distributions of the
simplest terms, different communities comprise trait-separated functional groups of freshwater
species having certain suites of common charac- phytoplankton (Table 7.1) among waters of given
ters and their prevalence is related to particular habitat characteristics (see for instance Fig. 7.8)
features of the habitat in which the communities favouring particular morphological and physio-
occur. Significantly, Southwood (1977) proposed a logical adaptations (Section 5.4.5). In this way, the
ANTHROPOGENIC CHANGE IN PELAGIC ENVIRONMENTS 395
Processing fluxes
d/
/
/
Figure 8.2 Log/log representation of aquatic habitats in

and the speed of processing opportunities. Poten-
terms of resource-supportable biomass (which, it is assumed tial plaktivores may also be plotted, habitual
can be regulated by the stoichiometrically least available pelagic-feeding fish are distinguished from typ-
nutrient; nitrogen and phosphorus axes are included as ical benthivores in sediment-retaining habitats
examples) and processing fluxes, which may be set by the (the psammophilic cyprinids) and those from less
PAR income or the rate of carbon delivery or by the flux of silted habitats (lithophilic fish). The template is
oxidative flux required to process the organic carbon also tentatively advanced to plot the distributions
delivered. The area thus defined can be subdivided, as shown,
of macrophytes (Fig. 8.3c) and benthos (Fig. 8.3d).
according to the characteristic of the habitat thus defined
(nutrient-limiting, carbon-limiting, etc.). Figure combines
features of figures presented in Reynolds (2002b, 2003a).
8.3 Anthropogenic change in
pelagic environments
species characteristically dominating strongly
nutrient-constrained systems (representatives of The habitat template of pelagic systems, invok-
the Z, X3, A, E, F associations), those dominating resource and processing constraints, provides
ing turbid or light-deficient environments (C, D, a useful bridge to understanding temporal vari-
P and especially R and S species) and those cop- ation in stability and its consequences. We have
ing with carbon-flux challenges in high-nutrient, already considered how seasonal changes in habi-
high-light environments (X1, H, L) are separated tat conditions alter the selection among stocks
(Fig. 8.3a). The template similarly distinguishes of primarily C-, S-, or R-strategist phytoplankters
(Fig. 8.3b) among planktic phagotrophs: the daph- (Section 5.5.4; see Fig. 5.21), essentially as the
niid, moinid and diaptomid traits signify dif- coordinates on the axes defining relative mixing
ferent interplays in the availability of resources and nutrient are altered during the year. In the
Figure 8.3 Tentative use of the resource/processing tern of seasonal periodicity in the development
habitat template to summarise the distributions of familiar and community assembly in the phytoplankton
aquatic assemblages. (a) Phytoplankton shown by describes a series of loops tracking approximately
trait-segregated functional groups (summarised in Table 7.2); similar courses through the habitat template.
(b) representative functional groups of zooplankton and fish; The sketch in Fig. 8.4 shows a notional series
(c) freshwater macrophytes; (d) freshwater benthos. CPOM, of such loops that, basically, follow the selec-
coarse particulate organic matter; FPOM, fine particulate
tion pathway ‘a’ in Fig. 5.21. There are modest
organic matter. The figure follows ideas proposed in Reynolds
interannual variations that may affect the num-
(2003b).
bers, relative abundance and occasional domi-
nance changes of the phytoplankton in succes-
sive years. These loops could be said to be gov-
model thus developed, variations in the nutrient erned by the various behavioural contols, act-
availability were substantially the consequence of ing as Lorenzian ‘attractors’ (see, for example,
biogenic activity but the physical conditions were Gleick, 1988). These defy randomness and are
set, at least in part, by the external physical envi- variously susceptible to circumstantial weighting
ronment. We have also extended this concept in of the dynamics of the individual species and,
the context of predicted variations of unexpected thus, the communal outcome. Significant depar-
magnitude or of variations at unexpected tempo- tures from this track, forced by extreme combi-
ral intervals, sufficient to effect significant struc- nations of resource and processing opportunities
tural disturbances (Section 7.3.3). Over a number (‘strange attractors’), then conform to convention-
of years, interannual adherence to a basic pat- ally defined chaotic responses (Gleick, 1988, a. o.).
8.3.1 Eutrophication and enhanced

phosphorus loading
Defining the issues
The problem of eutrophication (most especially
of lakes) has been known for many years and, as a
topic, has been discussed and reviewed in numer-
ous previous publications. Here, the reader is
simply referred to other works (Ryding and Rast
(1989), is excellent) charting the history and the
fascinating early attempts to explain the quite
Figure 8.4 Sketch to show several annual sequences of the alarming increases in the plankton biomass that
phytoplankton in a eutrophic lake, tracked as in Fig. 5.21, to were observed in a number of prime European
show its Lorenzian attraction and the occasional chaotic,
and North American lakes during the middle
strange-attracted departure.
part of the twentieth century. The background
to the present section is that it mainly con-
Many of the interannual differences in cerns the raising of supportive capacity of oth-
seasonal dynamics of phytoplankton, from erwise clear-water lakes as a direct result of
oligotrophic oceans (Karl et al., 2002) to eutrophic increased loadings of hitherto limiting nutri-
lakes (Reynolds, 2002a), can be related to ents, arising from anthropogenic activities in the
chaotic displacement of the environmental hydrological catchment. The problem is not con-
attractors. It results in significant (though gen- fined to lakes as analogous enrichment of coastal
erally rather temporary) variation in the driv- seas has enhanced their fertility several-fold sev-
ing constraints, sufficient to evoke a series of eral with respect to the pre-agricultural period
population responses. These summate to com- (Howarth et al., 1995). In both lakes and shelf
munity responses, which, even if not precisely waters, eutrophication has led to alterations in
explicable, may still be tracked in the habitat the species structure of the plankton and to what
template. However, there may well be interan- are generally regarded as damaging changes to
nual differences that are more insidious and aquatic ecosystems.
more systematic, that occur in response to rel- Cultural eutrophication was defined by the
atively slow progressive environmental in mor- Organisation for Economic Co-Operation and
phometry, climate or in the anthropogenic influ- Development (1982) as the nutrient enrichment
ences to which they are subject. Prior to mod- of waters resulting in the stimulation of an array
ern concerns about accelerated climate change of symptomatic changes, among which are the
as a consequence of human intervention in the increased production of algae and macrophytes,
global carbon cycle (see Section 3.5.2), the first that are injurious to water quality, are undesir-
two named sources were principally the concern able and interfere with other water uses. This
of palaeoecologists, investigating the effects of definition distinguished the process from natu-
postglacial sedimentary infill from catchments ral eutrophication that many people associated
subject to changing erosion rates and vegeta- with lake ageing, recognising that the undesir-
tion cover. Environmental changes and system able changes were, in effect, merely some sort of
responses over a few decades and within the acceleration. There are such things as naturally
span of observation of individual laboratories eutrophic lakes (especially among kataglacial
and, in some cases, of individual scientists may outwash lakes) (see p. 336) but most of the
be tracked from extant data collections. The palaeological evidence points to the opposite,
most familiar of these changes have been the that as natural catchments mature and support
effects of eutrophication through anthropogeni- closed forest ecosystems, the nutrients released
cally enhanced nutrient supplies and increasing decrease rather than increase. Sediment deposi-
acidification of precipitation. tion diminishes the water depth and eventually
concentrates the nutrient load in a shallower are not optimised to give the best opportunity for
depth of water, contributing positively to the the assimilation of nitrogen in proteins and plant
apparent fertility of the water column. In this growth. Even so, loadings of nitrogen to lakes
context, eutrophication (or its reverse, ‘olig- and coastal waters from agricultural catchments
otrophication’) should be seen as no more than have probably been boosted two- to fivefold in the
a contemporaneous reaction to contemporaneous last 50 years (cf. Lund, 1972). The use of phos-
external nutrient loadings, which will exercise a phatic fertilisers has also increased over the same
regulatory role so long as the nutrient is critical period, during which phosphorus levels in rivers,
to the supportive capacity. lakes and seas have increased by a similar or
It is also well known that the nutrient that still greater factor. However, the main phospho-
is generally considered to act as the limiting rus pathway is probably quite different. Plants
control is BAP (see Section 4.3.1). Actually, dis- ‘compete’ with soil chemistry for the phosphorus
solved sources of assimilable nitrogen are just as applied, much becoming immobilised in the soil
likely to be in short supply, relative to demand. by clay minerals, iron and aluminium hydrox-
However, owing to the advantageous trait of ides and through co-precipitation with carbon-
certain cyanobacteria to be able to fix atmo- ates (Cooke, 1976). Drainage from well-managed
spheric nitrogen (Section 4.4.3), nitrogen-fixing pastures and cropfields generally tends to have a
species may still be able to grow, often to lim- relatively low soluble-P content (there are plenty
its imposed by the available phosphorus. This that are otherwise; see, for example, Haygarth
leaves phosphorus still the principal regulatory (1997), although it may carry a substantial PP
factor, even in nitrogen-deficient lakes (Schindler, load as eroded soil particles but which remains
1977). Deficiencies of micronutrients (Mo, Va, Fe) ‘scarcely bioavailable’ (cf. Table 4.1). On the other
may prevent much nitrogen fixation (Rueter and hand, much phosphorus remains in the soil or
Petersen, 1987) (Section 4.4.3). Availability of iron is incorporated into the biomass that is actually
constrains the plankton-supportive capacity of cropped by consumers.
large parts of the ocean (Section 4.5.2) but, with As a metabolite of consuming animals (includ-
the exception of those instances where plankton ing humans), excess phosphorus is eliminated
is simultaneously nitrogen-deficient and fixation- in solution (i.e. excreted as MRP). In the nat-
constrained, phytoplankton in lakes is more ural way of things, this is most likely to be
likely to be limited by phosphorus than any other returned to soils and to further chemical immo-
nutrient. bilisation. Changing cultural standards, driven by
Anthropogenic enhancement of phosphorus burgeoning human numbers, urbanisation and
comes from several sources. The clearing of for- public health concerns, insist that most of that
est, the promotion of agriculture and the use of ‘waste product’ is intercepted and subjected to
inorganic fertiliser have been trade marks of a secondary biological treatment (in which much
green revolution of food production on an indus- of the organic content is re-oxidised and re-
trial scale. The ascendency of a human popula- mineralised). On average, adult humans excrete
tion of over 6 000 000 000 would have been quite some 1.6 g P d−1 (0.58 kg P a−1 ) (Morse et al., 1993);
unsustainable without this. Agriculture rejuve- with modern standards of secondary sewage
nates ecosystem- and biomass-specific productiv- treatment, between 70% and 100% (Källqvist and
ity but also opens them to resource exchange Berge, 1990) of this is returned to the aquatic
(Ripl and Wolter, 2003). Often, deficiencies of environment in a readily bioavailable form.
nitrogen have to be overcome in order to realise Historically augmented by soluble phosphates
productive potential and, world-wide, applica- emanating from the hydrolysis of detergents
tions of inorganic nitrogenous fertilisers repre- based on sodium tripolyphosphate (STPP; see
sent a major component of agricultural practice. Clesceri and Lee, 1965) and boosted by other
Their high water solubility makes then vulnera- household wastes, domestic sewage has undoubt-
ble to leaching in drainage water, if applications edly enhanced P-loads to rivers and pelagic
systems in general. Reynolds and Davies (2001) related mean summer chlorophyll a concentra-
estimated human contributions to have con- tions to winter–spring concentrations of TP (Dil-
tributed up to 1.6 kg P individual−1 a−1 in the lon and Rigler, 1974; Oglesby and Schaffner,
recent past. Certain manufacturing industries 1975) or maximum concentrations to vernal BAP
also generate significant quantities of dissolved (Lund, 1978; Reynolds, 1978c) or, eventually, the
phosphorus. However, in both its coincidence mean annual chlorophyll concentration to the
with the implementation of widespread treat- annual P-loading factor (Vollenweider, 1975, 1976;
ment and the statistical allocation of contribu- Rast and Lee, 1978).
tions to the P budgets of contrasted river basins All these formulations yielded similar slopes
(e.g. Caraco, 1995; Brunner and Lampert, 1997), (Reynolds, 1984a): the greater the phosphorus
effluents from secondary sewage treatment are availability the greater is the phytoplankton
most strongly implicated as the leading proximal biomass likely to be supported. However, it is the
source of the phosphorus complicit in eutroph- Vollenweider–OECD approach which has been
ication (Reynolds and Davies, 2001). Lest this most frequently applied as it incorporates the fur-
is taken as an accusation, it must be recalled ther useful empirical step of relating in-lake avail-
that modern agriculture and the socio-economic ability to the nutrient loading from the catch-
structures that it sustains are founded upon the ment. The full formulation models the relation-
enhanced distribution, chemical mobility and ship between average chlorophyll biomass and
biotic dissipation of (e.g. Moroccan) phosphate the amount of phosphorus supplied to the lake,
and (e.g. Chilean) nitrate. Whether these ele- adjusted to availability through corrections for
ments are removed to water directly or through water depth and hydraulic retention. In its final
a complex anthropogenic food chain is really form (Vollenweider and Kerekes, 1980), the regres-
academic. What matters is that these pathways sion of the annual mean concentration of chloro-
are understood, that their impacts are quanti- phyll a ([chla]a , in mg m−3 ) is a direct function
fied, and that they inform rational strategies for of an averaged, ‘steady-state’ index of phosphorus
improving their future management. availability (P , also in mg m−3 ):
log [chla]a = 0.91 log[P ] − 0.435 (8.1)

Aquatic impacts of phosphorus loads to lakes
The Organisation for Economic Co-Operation and The derivation of P is interesting and is
Development (OECD) sponsored a series of co- extremely valuable when the management of P
ordinated studies of the relationship between loads becomes an issue. It begins with the dynam-
phytoplankton biomass (analogised to chloro- ical relationship between the total phosphorus
phyll a concentration) and phosphorus availabil- concentration in the lake water ([P], in mg m−3 )
ity in lakes (TP was preferred to MRP, which, if and the rate of supply, less the amounts lost in
limitation means anything, is rather scarce). The the outflow and to the sediments:
first report (Vollenweider, 1968) uncovered the
d [P]/dt = (qi [P]i )/V − w P [P] − σP [P] (8.2)
essence of a mathematical relationship between
in-lake concentrations of chlorophyll a and TP where qi is the inflow rate and [P]i is the phospho-
(although, in fairness, Sakamoto (1966) and rus concentration of the ith inflow stream and V
Sawyer (1966) had each previously recognised is the volume of the lake; σ P is the phosphorus
this). The final report (Vollenweider and Kerekes, sedimentation rate and w P is the rate of loss of
1980) provided definitive statistical fits of mean phosphorus in the outflow. At steady state,
annual chlorophyll a concentration against the
delivery of TP from the catchment (the ‘load- [P ] = {L (P)/z}/{w P + σP } (8.3)
ing’), corrected for the effects of depth and
hydraulic replacement. In between, a flurry of where L(P) is the aggregate areal rate of phospho-
other equations was published that (variously) rus loading (in mg m−2 of lake area per year)
and z represents mean depth. The most diffi-

cult term to estimate without detailed measure-
ment is the phosphorus sedimentation rate; Vol-
lenweider (1976) proposed an empirical solution
in which:
√
σP ≈ (z/qs )/tq (8.4)
where tq is the hydraulic residence time in

years, calculated as lake volume (V) divided by
the annual sum of the inflow rates ( qi ), and
where qs is the hydraulic loading rate (in m
a−1 ), approximately equivalent to z/ tq . Rewriting
Eq. (8.4),
√ √
σP ≈ [z/(z/tq )]/tq = (1/tq )/tq (8.5)
Supposing that wP = 1/tq , Eq. (8.4) may now be Figure 8.5 The Vollenweider–OECD relationship between
rewritten average chlorophyll a concentrations in lakes and the
phosphorus availability inferred from loading characteristics,
√
[P ] = {L (P)/z}/{1/tq + (1/tq )/tq } (8.6) with 99% confidence limits. The equation (8.8, in the text) of
the fitted line relates chlorophyll to mean in-lake phosphorus
Multiplying out by tq gives concentration (log [chla]a = 0.91 log P – 0.435, where P
√ is derived as P = {L(P) / qs } / {1 +(z / qs )} (see text and
tq [P ] = {L (P)/z}/{1 + [z/(z/tq )]}
Eq. 8.7). Redrawn with permission from Reynolds (1992a).
and cancelling
√
[P ] = {L (P)/qs }/{1 + (tq )}
is, the log–log format obscures a wide variation
or
in real average chlorophyll content for a given
√
[P ] = {L (P)/qs }/{1 + (z/qs )} (8.7) derivation of [P ]. For instance, a literal extrapo-
lation of Fig. 8.5 is that the average chlorophyll
The log–log relationship of chlorophyll con- supported in a lake having an availability index
centration to Vollenweider’s index of phospho- of 100 mg P m−3 may be predicted, with 95%
rus availability (P ) is plotted in Fig. 8.5. confidence, to be between 10 and 53 mg chla
The fitted regression is thus the definitive m−3 . On the other hand, the Vollenweider–OECD
‘Vollenweider–OECD model’; restated, model is a powerful, empirical statement of the
log[chla]a = 0.91 log[{L (P)/qs }/ long-range performance of large and medium-
√ sized lakes subject to varied differing phosphorus
{1 + (z/qs )}] − 0.435 (8.8)
loadings.
Great store has been placed on this formulation. Reducing the relationship back yet more
Some authors have apparently been perplexed towards a semi-quantitative word model, the
by sites which do not concur with the regression Organisation for Economic Co-Operation and
(such outliers are evident in Fig. 8.5) and have Development (1982) proposed approximate phos-
sought amendments to the equation in respect of phorus availabilities to coincide with the trophic-
other influences, including the effects of grazers state, metabolic-dependent categories of lakes
(e.g. Prairie et al., 1989). Were serious criticisms in use since the days of Naumann (1919). The
to be made, they might be directed at the origi- slightly modified scheme of Reynolds (2003c),
nal and inadvertent bias towards deep temperate incorporating the continuing prevalence of phos-
lakes. Had more shallow, heavily loaded, P-cycling phorus limitation is shown in Table 8.1.
or well-flushed systems been considered, a less Such deductions help to inform practical
satisfying result would have been obtained. As it approaches and strategies for the manipulation
Table 8.1 Proposed criteria for classifying lakes on the basis of trophic state and
metabolic constraints, based upon OECD (1982) and Reynolds et al. (1998)
[TP] or [P ] Months where MRP Average [chla]a

Category (mg P m−3 ) <3 mg P m−3 (mg m−3 )
Ultraoligotrophic <3 Always <2
Oligotrophic 3–10 9–12 per year 0.7–4.5
Mesotrophic 10–35 4–9 per year 2–24
Eutrophic 35–100 <4 months per year 3–53
Hypertrophic >100 Never >10
a
For predicting maximum chlorophyll yield from the bioavailable phosphorus, see
Section 4.3.4 and Eqs. (4.14) and (4.15).
of phosphorus loads. These are reviewed later chemicals, they had already achieved notori-
(see Section 8.3.3). Mention may also be made ety for their tendency to form surface scums.
of lake-specific derivations of the metabolic con- Accumulated by downwind drift, these paint-like
sequences of varied external phosphorus loads. shore-line swathes of stranded algae are con-
Supposing that the portion of the bioavailable sidered to be, at best, unsightly. As they die
phosphorus load that is actually retained gen- and decompose, they release pigment into the
erates a stoichiometrically equivalent mass of water and become evil-smelling. Thus, ‘leprous
organic carbon and that this is oxidised by a sto- and fetid’ (Sinker, 1962), they discourage con-
ichiometrically equivalent quantity of hypolim- tact with, enjoyment of and fishing from, the
netic oxygen, Reynolds and Irish (2000) success- waters so affected. In point of fact, rather few
fully simulated the differing impacts of the pro- species of Cyanobacteria do this: those belonging
gressive eutrophication and its subsequent rever- to the ‘bloom-forming’, gas-vacuolate genera that
sal on the hypolimnetic oxygen content of the either occur in coenobial or colonial structures
North and South Basins of Windemere. (such as Microcystis, Woronichinia and Gloeotrichia;
trait-separated functional groups L, M) or in fila-
ments that aggregate in clumps or flakes (as in
8.3.2 Blue-green algae and red tides
many Anabaena, Anabaenopsis and Aphanizomenon
In both lakes and shelf waters, progressive
species belonging to the H groups). Other gas-
eutrophication has resulted not just in the
vacuolate, filamentous forms that remain soli-
maintenance of a higher average phytoplankton
tary (Planktothrix, Limnothrix, some Anabaena and
biomass but also to shifts in species composition
Cylindrospermopsis of the S associations) can be
that apparently favour Cyanobacteria (in lakes)
abundant but rarely scum. Generally, small-celled
and small dinoflagellates (in certain seas). Nei-
species, including picoplanktic Cyanobacteria, do
ther change is straightforward. In either case,
not have gas vacuoles and, thus, have no means
the change brings consequential effects upon
of constituting surface scums at all.
environmental quality and both, by coincidence,
While such blooms are certainly not new
carry implications for human health. Beyond
(see Griffiths (1939) for a review), even entering
that the direct causes of the problems are quite
local folklore at locations where they have been
distinct and are referred to separately.
long recognised (Reynolds and Walsby, 1975),
reports of scums (or blue-green algal blooms)
Cyanobacterial ‘blooms’ became much more numerous since the mid
Long before it was confirmed that certain species twentieth Century and appeared, often spectac-
of Cyanobacteria (or blue-green algae) were ularly, in lakes where they had previously been
capable of producing several extremely toxic virtually unknown. At first, their appearance
was attributed to sudden, unexplained bursts of temperate Cyanobacteria would be confronted

rapid growth and replication (Mackenthun et al., by post-vernal phosphorus exhaustion. The his-
1968). In fact, the algae grow only relatively tory of eutrophication in Windermere (as told by
slowly. They accumulate at the surface only dur- Reynolds and Irish, 2000) illustrates very well the
ing quiet weather (minimal vertical mixing) and interannual stability of the silicon-constrained
only as a consequence of their failure to regu- diatom crops despite year-on-year increases on
late the buoyancy imparted by the gas vacuoles. annual P loadings. The record reveals the onset of
From being well dispersed through the water col- late-spring P surpluses and the advent of signifi-
umn, buoyant algae are disentrained to the sur- cant summer cyanobacterial blooms. It shows the
face when the wind drops. Becoming thus ‘tele- demise of both in the wake of reduced phospho-
scoped’ (Reynolds, 1971) from depth to the sur- rus loadings after 1992. Sas’ (1989) review of suc-
face and further concentrated along lee shores cessful lake restoration schemes in Europe iden-
by light winds, the scums give a greatly exag- tified a reduction in cyanobacterial crops in each
gerated impression of abundance. Nevertheless, of several lakes, only after BAP levels had been
the increasing incidence of blooms prejudicial reduced during the spring period to levels too
to water quality correlates well with increas- low to support significant blue-green recruitment
ingly enriched conditions, suggesting a power- during the early summer.
ful causal link between increased abundance
of bloom-forming Cyanobacteria with increased Cyanobacterial toxicity
phosphorus availability (Gorham et al., 1974). Problems with cyanobacterial scums have been
Indeed, such blue-green algal blooms have per- compounded by the recent ‘discovery’ (really it
haps done most to give eutrophication its bad was a confirmation of a long-held suspicion) of
name. their severe toxicity to humans. Again, the issue
Interestingly, bloom-forming Cyanobacteria does not arise as a direct consequence of eutroph-
need no more phosphorus to support growth ication but, on the other hand, the occurrence
or to saturate their growth-rate requirements of toxic organisms in health-threatening concen-
than other common eukaryotic algae (Reynolds, trations is dependent upon an enriched resource
1984a) (see also Section 4.3.2 and Table 4.1). Nei- base. Cyanobacteria are not the only group
ther does phosphorus availability guarantee blue- of phytoplankters to produce toxic metabolites:
green algal abundance: they are relatively intoler- besides the ‘red-tide’ dinoflagellates (see p. 407
ant of acidity and low insolation (Table 7.2). They below), several kinds of marine and brackish hap-
are extremely tolerant of high pH and may domi- tophytes (including Prymnesium and Chrysochro-
nate where algal production generally strains the mulina) produce substances toxic to fish and other
carbon supply (Shapiro, 1990). They commonly vertebrates. However, it is the self-harvesting in
do well in mildly P-enriched calcareous lakes scums that so magnifies the potential harm that
(Reynolds and Petersen, 2000). cyanobacterial toxins might pose to humans.
Also relevant is the fact that the growth of the Many instances of sickness and death of live-
‘large’ algal units that characterise the bloom- stock and, occasionally, humans after consum-
forming blue-green algae is especially sensitive ing water containing Cyanobacteria had been
to lower temperatures (see Section 5.3.2 and Fig. reported over a number of years, especially from
5.3). At least so far as the well-studied temper- warm, low-latitude regions. It had been supposed
ate lakes of North America and Europe are con- by many hydrobiologists working remotely that
cerned, it became apparent that the offending these were symptomatic of putrescence and bac-
bloom-forming species are excluded in the early terial pathogens. The possibility that the algae
year by low temperatures. These are, of course, themselves could be harmful appears not to have
no bar to the successful early growth of diatoms been taken very seriously, prior to the pioneer-
and nanoplankters. By the time that the physical ing investigation by a small but active group of
conditions are adequate for their growth, many scientists led by P. R. Gorham (see e.g. Gorham
et al., 1964). The work was continued by W. W. become dominant and its rivals have already
Carmichael and his colleagues (see Carmichael been competitively excluded. It is possible that
et al., 1985). Many of the present perspectives are the toxicity is fortuitous and the compounds
due to the efforts of G. A. Codd and his colleagues are the by-product of some unknown homeo-
at Dundee (for a recent overview, see Codd, static step in the ageing of the cyanobacterial
1995). The eventual interest of the World Health cells. Toxicity of Microcystis was long suspected
Organisation promoted the useful handbook by to be a symptom of stress in natural popula-
Bartram and Chorus (1999). tions and which was reversible by introducing
It is now well appreciated that Cyanobacteria colonies to nutrient-replete media under nor-
are capable of producing at least three classes mal illumination and temperature (Carmichael,
of toxins. The acutely hepatotoxic microcystins 1986). More recently, Orr and Jones (1998) pro-
attack the digestive tract of consumers, causing vided convincing evidence that the production
acute pneumonia-like symptoms and sickness in of microcystin in nitrogen-limited Microcystis cul-
humans. More than 60 structural variants have tures is proportional to the cell-replication rates.
been detected in cells or cell-free extracts from a Field evidence is inconsistent, at best suggest-
range of cyanobacterial species, not just of Micro- ing that toxin production is sporadic. Jähnichen
cystis (K. Sivonen and G. Jones, in Bartram and et al. (2001), experimenting with Microcystis har-
Chorus, 1999). It is believed that they act by block- vested from Bautzen Reservoir, Germany, showed
ing phosphorylation; weight for weight, the toxic- that microcystin was synthesised only during the
ity of microcystins is comparable to that of curare phase of exponential increase and only after the
and cobra venom. The neurotoxic anatoxins are external pH exceeded 8.4, free CO2 was virtually
also acute poisons. Chemically, they resemble exhausted and photosynthesis was drawing on
the dinoflagellate saxitoxins, and are produced bicarbonate. The suggestion that microcystin pro-
principally by the nostocalean genera. The third duction has some connection with the extreme
group of toxic compounds are lipopolysaccha- affinity of cyanobacteria for carbon uptake (see
rides: these are the least well characterised but Section 3.4.2) is an attractive proposition. It
possibly the most insidious of the three, being is also far from incompatible with the earlier
associated with sub-lethal skin irritations as a findings.
consequence of contact with affected water or Toxicity per unit of Cyanobacterial mass
with cumulative chronic effects of frequent expo- varies with the species of Cyanobacteria present,
sure. the potential of the resources to support their
The benefit to the organisms of producing growth and the availability of DIC. However, the
such poisonous chemicals is still an unexplained extent to which physical processes may have fur-
paradox. It is scarcely a necessary defence against ther concentrated the cyanobacterial mass mag-
herbivorous crustacean or ciliate consumers and, nifies the risk of human exposure to a toxic dose.
besides, to kill potential grazers is as protective As has been stated, the microcystins are them-
to other species of the phytoplankton as to the selves extremely toxic: the lethal intraperitoneal
Cyanobacterium and most of these, it will be lethal dose to mice of the common microcystin-
recalled, grow rather faster. There would be more LR is 1.25 µg/25 g (Rinehart et al., 1994), or 50
point to producing toxins against the other algae. µg kg−1 . It is not assumed that all mammals
There is some evidence for the suppression of are equally sensitive but, supposing it were simi-
growth of common algae, either in media from lar, the generally lethal oral dose would be some
which the cyanobacteria have been removed or 10- to 100-fold greater, i.e. in the order 0.5 to 5
in fresh medium, spiked with extract from spent mg kg−1 . Swallowing 30–300 mg of microcystin
Microcystis cultures (Reynolds et al., 1981). would probably be sufficient to kill a 60-kg adult.
What is interesting about this is that the However, the mass of microcystin produced by
toxin production seems most prolific at the time individual cyanobacterial cells is measurable in
of population climax, after the organism has femtogram quantities. Lyck and Christoffersen
(2003) have recently measured ∼85 ± 44 fg micro- reportedly harvested for food by natives of Chad:
cystin per Microcystis cell in field populations, Pirie, 1969). I am not aware of a scientific study
somewhat less than the 278 ± 115 fg cell−1 that that confirms my impression that most animals
Christoffersen (1996) had measured in laboratory avoid drinking water tainted by Cyanobacteria
cultures. Against the dry mass of the cell (aver- but there is certainly evidence that domestic live-
age 32 pg; Table 1.3), microcystin accounts for stock may be reluctant to drink until overcome
perhaps 1–1.5% of the typical dry mass (note, a by thirst (Reynolds, 1980b). It is noticeable that
similar amount to its chlorophyll complement). fish and Daphnia also avoid water sullied by scum.
Supposing the highest measured toxin content, All this makes the behaviour of domestic dogs –
then the smallest number of Microcystis cells that which seem uniquely attracted to wallow and roll
would have to be ingested in order to deliver in shoreline scums – quite difficult to explain.
a toxic dose is close to 100 thousand million The attraction proves fatal when the animals
(1011 cells) and, in all probability, ten times that start to lick their coats.
amount. When compared to the dispersed cell Pets apart, there is a need to be concerned
populations attained by natural Microcystis popu- about sub-lethal or chronic exposure to the
lations noted in nature (my record is ∼360 000 toxic Cyanobacteria, not least through drinking
cells mL−1 for a near-monospecific population in water purified from reservoir storages in which
one of the Blelham enclosures and equivalent to planktic Cyanobacteria may be abundant. Treat-
∼120 mg chla m−3 ), it is clear that a toxic dose ment processes that remove planktic cells by
is equivalent to scarcely less than 28 L of lake filtration, flocculation or dissolved-air scaveng-
water. ing are effective in removing intracellular toxin
One would be entitled to conclude that the but engineers need to be aware of treatments
risk of drowning in the water far exceeds that that induce cell lysis and secretion of toxin into
of poisoning by its suspended contents, but the water. Cyanobacterial toxins are effectively
for the phenomenon of surface-scum forma- removed from the water passed through gran-
tion. Through the abrupt flotation of buoyant ular activated carbon. Reservoir managers are
colonies to the water surface, the concentration also often able to select from several draw-off
of Microcystis colonies hitherto dispersed through options, in order to avoid the intake of Cyanobac-
a depth of ∼5 m is quickly compacted (‘tele- teria. Current guidelines from the World Health
scoped’: Reynolds, 1971) into a layer no thicker Organisation suggest 1 µg microcystin L−1 as the
than 5 mm (i.e. a 1000-fold concentration). The upper safe limit (see Bartram and Chorus, 1999).
scum is further concentrated by subsequent drift If reservoir populations approached the toxicity
to a lee shore, where the population from (say) a level observed by Christoffersen (1996), it has to
1-km fetch could be reasonably aggregated into be admitted that many reservoirs supply water
a shoreline scum of 1 m or so in width. Now, supporting Cyanobacteria rather more numerous
the microcystins are perhaps concentrated by a than the equivalent of 1 µg chla L−1 .
factor in the order of 106 , capable of supply- For recreational waters, the hazard posed to
ing the toxic dose within some 28 µL of lake swimmers, sailors and anglers alike remains the
water! ingestion of scum. In addition to providing peri-
The salutory deduction is that the shore- odic warnings, site managers usually seek a com-
line scums are rather more threatening than promise between banning public access to water
the dispersed populations. It is interesting that that they know may contain extant blue-green
our species has coexisted with bloom-forming algae and allowing activities to continue until
Cyanobacteria for some millions of years, mostly there is a significant risk of toxic algae aggre-
oblivious to the latent hazard. On the other hand, gating along the shore. This equation is not just
the scums are so uninviting that they invoke about how much alga is dispersed in the water
the life-saving instinct of disgust: people are gen- but whether, and by how much, it is likely to
erally dissuaded from consuming or contacting self-concentrate in areas of public access. Using
this water by revulsion (although Spirulina is a reverse calculation of what was needed to
generate a toxic dose in 28 L of lake water, quently dominate shallow, turbid lakes (Schef-
Reynolds (1998c) started with what seemed like fer et al., 1997) and that nitrogen-fixing Nos-
a reasonable estimate of a volume swallowed, tocales may dominate eutrophic lakes in defi-
how many toxic cells that that might contain and ance of low levels of combined nitrogen, pro-
how much that factor of concentration could be vided certain other conditions are satisfied (Sec-
sustained by flotation and leeward drifting. The tion 4.4.3). Cyanobacterial abundance is, like that
outcome for buoyant Microcystis suggested that a of most other forms, correlated to TP and TN
lake population equivalent to 5 µg chla L−1 is a (e.g. Downing et al., 2001), not least because their
level warning of significant risk and is a trigger biomass will account for much of the TP and
for careful monitoring for downwind scums in TN that is present. In temperate lakes, growth
calm weather. Most Anabaena, Aphanizomenon and of bloom-forming Cyanobacteria is slow in win-
Woronichinia species. float up only half as rapidly, ter and they may fail to grow at all if the
permitting a doubling of the concentration trig- competitors (especially vernal diatoms) clear the
gering the warning level. Planktothrix and Limno- water of dissolved phosphorus before the light
thrix float so relatively slowly that concentra- and temperature thresholds are past. Species that
tions equivalent to 100 µg chla L−1 may be toler- overwinter on the benthos experience critical
ated before warning levels are triggered. The dis- difficulties of seasonal recruitment (Reynolds
persed cell concentrations of named Cyanobac- et al., 1981; Reynolds and Bellinger, 1992; Bragin-
teria equivalent to these warning thresholds are skiy and Sirenko, 2000; Brunberg and Blomqvist,
set out in Table 8.2. 2003). If filter-feeders are abundant before the
new colonies are released into the water col-
Control of cyanobacteria umn, then grazing can be highly effective in
Interest persists in being able to eliminate and/or stemming recruitment to the plankton (Ferguson
exclude Cyanobacteria from managed water bod- et al., 1982; Reynolds, 1998c).
ies or, at least, to keep their numbers at back- By deduction, effective controls against
ground levels. Unfortunately, there is no simple cyanobacterial dominance are few. Low concen-
or universal means to attack Cyanobacteria per trations of available phosphorus are beneficial
se which is not likely to be destructive of all in temperate lakes (in the sense of not support-
other biota, desirable or otherwise. Deep-rooted ing bloom-forming Cyanobacteria) but not so in
suppositions about the nutrient requirements tropical lakes, especially if they stratify and have
of ‘the Cyanobacteria’ and about their suscep- long retention times (see Section 7.2.3). Of gen-
tibility to grazing rather ignore the very wide eral importance, however, is the poor tolerance
diversification evident among the group of mor- by bloom-forming genera, especially Anabaena,
phology, physiology and life history. Even when Aphanizomenon and Microcystis, of entrainment in
authors have focused on just the bloom-forming mixed layers that take the algae well beyond
genera, they have tended to seek mechanistic the conventional photic depth and thus dilute
explanations emphasising the importance of cer- and fragment their exposure to the light field.
tain correlative factors, such as nutrient ratios This sensitivity was detected in the early liter-
and biological interactions, in governing popu- ature reviews (Reynolds and Walsby, 1975), was
lation dynamics (see e.g. Levich, 1996; Bulgakov verified in the mixing experiments of Reynolds
and Levich, 1999; Elser, 1999; Smith and Ben- et al. (1983b, 1984) and was a demonstrable con-
nett, 1999). Other analyses suggest that cyanobac- sequence of the application of destratification
terial dominance is a more fortuitous outcome techniques in London’s Thames Valley Reservoirs
of interacting factors that include perennation, (Ridley, 1970; Steel, 1975). Since then artificial
weather effects, column mixing and carbon affin- mixing techniques have been used widely in the
ity (Reynolds, 1987a, 1989a, 1999b; Dokulil and protection of water quality in reservoirs (Steel
Teubner, 2000; Downing et al., 2001). The main and Duncan, 1999; Kirke, 2000); the reduction
generalisations that are possible seem to be in cyanobacterial mass has been generally wel-
that the filamentous group-S Oscillatoriales fre- comed, even where this was not the primary
Table 8.2 Approximate biomass equivalents of potentially toxic Cyanobacteria attaining a (‘warning’) level
that could deliver a lethal oral dose to a human adult; the derivations assume that intracellular toxin content
to be the highest reported at time of writing, that the organisms are at their most buoyant and that their
horizontal aggregation is subject to surface winds of 3.5 m s−1 . Developed from Reynolds (1998c).
Cells (or Suggested

mm) warning Number of cells
‘Average’ mL−1 level (or mm) mL−1
Cell volume ‘Average’ cell chla to 1 µg (µg chla equivalent to
Speciesa (µm3 )b cell C (pg) (pg) chla L−1 L−1 warning level
Microcystis 30 (65) 100 15 0.3 2 000–4 000 5 10 000–20 000
aeruginosa
Woronichinia 5(40)75 10 0.2 4 000–6 000 10 40 000–60 000
naegeliana
Aphanizomenon 5(12)20 3 0.06 15 000– 10 150 000–180 000
flos-aquae 18 000
Anabaena 70 (100) 130 22 0.45 1 500– 10 15 000–30 000
circinalis/ 3 000
flos-aquae/
spiroides
Anabaena 33(47)113 11 0.22 3 000–6 000 10 30 000–60 000
lemmer-
mannii
Anabaena 270(400)520 90 1.8 400–700 50 20 000–40 000
solitaria
Planktothrix 28 000 10 500 210 4–5.5 100 400–550
mougeotii∗ (46 600)
(1 mm) 71 000
Planktothrix 12 000 (24 5 500 110 8–11 100 800–1 100
agardhii∗ 000) 28000
(1 mm)
Limnothrix 1800 (3140) 700 14 60–80 100 6 000–8 000
redekei∗ 7500
(1 mm)
Pseudanabaena 780 (1220) 275 5.5 160–200 100 16 000–20 000
limnetica∗ 1800
(1 mm)
a
Asterisks indicate where instead of cell volume, the volume of a l-mm filament is given.
b
Bold figures indicate a typical middle-of-the-range value.
Source: Developed from Reynolds (1998c).
objective. The mechanical aspects of artificial of stream was cleared of most of its detached
mixing are reviewed in a later section (8.3.5). algae downstream of an abandoned straw bale,
Mention may be made here of the use of the idea quickly spread that straw added to
barley straw as a defence against Cyanobacteria. ponds and small reservoirs would provide effec-
From a chance observation in which a length tive protection from cyanobacterial growth. The
implicit confidence in the technique seemed mis- sition and local oxygen depletion. ‘Brown tides’
placed, as there was no information upon the of the picoplanktic Aureococcus and ‘green tides’
mechanism and the instances of successful treat- of Phaeocystis in its microplanktic colonial stages
ments represented a very small proportion of may cause considerable distress among fisher-
the attempts. By degrees, however, it became men. It has become fashionable to lump all these
clear that the effective agent was a phenolic organisms together as ‘harmful algae’; certainly,
substance, presumably produced by the barley they are the subjects of several recent reviews and
plant (as in other plants, including bark-bearing compendia (see e.g. Anderson et al., 1998; Reguera
woody plants) as a live defence against fungi et al., 1998).
and microorganisms. It is released from the However, the particular issue of the red-
straw as it ages and rots and the substance will tide dinoflagellates remains an intriguing one. A
inhibit growth in the laboratory of Cyanobac- dozen or so genera are known to produce low-
teria and other algae at quite low concentra- molecular-weight saxitoxins which are acutely
tions (Newman and Barrett, 1993). Suitably aged, neurotoxic to birds, fish and humans. Being endo-
broken up and dispersed, barley straw has been toxins (that is, they are not released into the
shown to be reliably effective in preventing the water by live cells), they are apparently far more
growth of nuisance blue-green algae (Everall and injurious to larger organisms than to planktic
Lees, 1996; Barrett et al., 1999). Although there assembalges. One well-known pathway leading to
is inevitably some imprecision in what makes an human poisoning is through the consumption of
effective dose and algae show some variation in filter-feeding shellfish taken from areas recently
susceptibility to barley toxins, the use of straw affected by red-tide organisms.
should be recognised as a legitimate and effective The development of significant concentra-
defence against Cyanobacteria, capable of induc- tions of relevant dinoflagellates is, in part,
ing species-specific algal mortalities and altering attributable to high local nutrient inputs. Over
dominance in mixed assemblages (Brownlee et al., the last three decades or so, red-tide events and
2003). The comment is still inspired, however, their constituent organisms have been becoming
that the use of a toxic agent to protect against more frequent in the enriched coastal and shelf
another toxic agent is faintly ironic, although, as waters adjacent to major urban centres of the
already admitted, there may be other reasons for world (north-west Europe’s Atlantic seaboard, the
seeking to eradicate Cyanobacteria from ponds St Lawrence, the Gulfs of Maine and of Mexico,
and reservoirs. off Baja California, the north-eastern seaboard
of Argentina, around Japan and Korea, and Aus-
Red tides tralia and New Zealand). However, abundance is
Several phyletic groups of marine phytoplankton compounded locally by the ability of the algae to
have toxic representatives. Besides the dinoflag- self-regulate in slack water and to become aggre-
ellates mentioned at the start of Section 8.3.2, gated by weak water movements. Thus, the devel-
they include a handful of haptophyte genera opment of red tides is substantially consequent
(mainly especially the Prymnesiales) that pro- upon the containment of populations by near-
duce substances causing osmoregulatory failure surface stratification of enriched waters, or in
and death in fish. Both Prymnesium parvum and shallow water columns (Smayda, 2002).
Chrysochromulina polylepis have been implicated in It seems that the natural habitats and prob-
fish kills around the North Sea coast, the alga able sources of many of the more troublesome
having first built unusually large local popula- of the red-tide species are associated with the
tions in each instance (see, for instance, Edvard- exploitation of rapid nutrient renewal in frontal
sen and Paasche, 1998). In addition, non-toxic zones and post-upwelling relaxations, where
species have been implicated in fish kills as reduced mixing and relative resource abundance
a consequence of local algal abundance, near- normally coincide (see Figs. 7.3, 7.4 and Section
simultaneous death, followed by rapid decompo- 7.2.2). Alexandrium tamarense and Karenia mikimotoi
8.3.3 Controlling eutrophication by

phosphorus load reduction
The worth of the Vollenweider–OECD regression
lies in the elegant empirical statement it makes
about the generalised behaviour of the lakes
included in the original dataset upon which it is
based. It makes no statement about any individ-
ual lake, beyond the qualitative deduction that
if it becomes more enriched, it may well sup-
Figure 8.6 (a) Selection trajectory in the sea, as proposed port more phytoplankton chlorophyll on average.
by Smayda and Reynolds (2001), accepting that more of the Still less does it confirm that reducing current
potential resource/processing interaction is inhospitable to phosphorus loads will result in the support of
phytoplankton. The trajectory now closely resembles more modest algal populations. Yet many man-
Margalef’s ‘main sequence’ (see Fig. 7.2). (b) By analogy, the
agers have been deceived by the power of the
‘red-tide sequence’ (here styled HAB sequence) runs parallel
regression to employ it as a driver and measure
to the main sequence but more deeply into environments
made more tenable by greater nutrients and reduced depth of of schemes intended to reverse eutrophication
mixing, representing the involvement of bloom-forming taxa and restore the lake to something supposed to be
in the active community. Redrawn with permission from closer to ‘pristine’ conditions. This latter, rather
Smayda and Reynolds (2001). nebulously used, term usually refers to a state
that preceded the agricultural–industrial anthro-
pogenic impacts and not to the conditions at the
lake’s birth. Even so, the Vollenweider–OECD rela-
are typical members of the frontal flora in mid- tionship is not a management tool, nor was it
dle latitudes, while Lingulodinium polyedrum and ever intended as such. It may be seen as a guide
Gymnodinium catenatum are notable toxic species to the determination of nutrient loads and to
of upwelling relaxations. The highly toxic Pyro- the prospects for the benefits to be gained from
dinium bahamense and Karenia brevis are adapted to their reduction. It is ‘not a slope, up or down
entrainment and dispersal in offshore currents. which a given lake will progress during a period
The links among size and morphology (life of artificial enrichment or deliberate restoration’
form), physiology and ecology that have been (Reynolds, 1992a, p. 5).
noted at frequent intervals in this book were On the other hand, some restoration schemes,
explored famously by Margalef (1978), and not involving the alleviation of anthropogenic phos-
least in the context of an early exposition of the phorus loads, either by diversion or tertiary treat-
causes of red tides (Margalef et al., 1979). In their ment of the sewage inflows, have been spec-
mandala (Fig. 7.2), the ‘red-tide sequence’ was tacularly successful. One of the best-known and
represented as a sort of system ‘sickness’, strik- most-cited examples concerns the abrupt reduc-
ing parallel to the successional ‘main sequence’, tions in the average phytoplankton biomass in
in an area of greater nutrient availability. In Lake Washington, in the north-west USA, fol-
the context of the C–S–R triangle, preferred by lowing the diversion of all sewage discharges
Smayda and Reynolds (2001), increased nutrient to the lake. Once the diversions began to take
enrichment of coastal waters permits a higher effect, from the mid-1960s, the phytoplankton
arc than the main sequence to be traced (Fig. biomass, for long dominated year-round by Plank-
8.6). This is more favourable to the Type-IV, Type-V tothrix agardhii, was quickly reduced to below
or Type-VI species associations than to the glean- its 1933 level (the first year for which quanti-
ing, K-selected S-strategists normally culminating taive records had been kept) (Edmondson, 1970,
Margalef’s (1978) main sequence. However, Fig. 8.6 1972). By 1976, the alga had effectively disap-
gives little new insight into why these species peared, leaving microplankters dominating the
should be toxic or what benefit obtains. This reduced phytoplankton biomass, and average
explanation is still sought. Secchi-disk transparency cleared from <2 m to
Figure 8.7 Stages in the recovery of three European lakes the editorship of H. Sas (1989), showed substantial
after reduction in external phosphorus loading. Heavy lines differences in their responses to reduced external
link the changing observed annual average chlorophyll a P loads, most especially between lakes judged to
concentrations against the phosphorus availability and are be ‘deep’ and those considered as ‘shallow’. There
superimposed upon the slope of the Vollenweider–OECD were also systematic differences in the onset of
regression. The bars to the right of each point represent the the biological response to the diminished deliv-
difference between TP (total phosphorus) and PP (the
ery of phosphorus to the lake (Sas’ Subsystem 1)
particulate fraction). Note that the difference, approximately
equivalent to unused dissolved P, is virtually exhausted before
which was consistently mediated through the
there is a response in the mean chlorophyll concentration. bioavailability to algal production (Subsystem 2).
Redrawn with permission from Reynolds (1992a). Three cases are illustrated in Fig. 8.7.
The first concerns the Wahnbach Talsperre,
a mainstem reservoir on the River Sieg that sup-
>4 m. Further consequential changes included plies water to Bonn and Köln in Germany. During
the rise of Daphnia species and their replace- the early 1970s, unacceptably large, year-round
ment of Diaphanosoma and Leptodiatomus as domi- crops of Planktothrix rubescens were proving expen-
nant zooplankters, with further benefits to water sive to treat for potability. The morphology of the
clarity (Edmondson and Litt, 1982). Other asso- valley in the inflow region lent itself to the con-
ciated changes in the structure of the plank- struction of a small pre-reservoir and a treatment
tic food web have been consistent with the lake plant that would remove biogenic debris, partic-
recovering an essentially mesotrophic condition ulate matter and dissolved phosphorus from the
(Edmondson and Lehman, 1981). inflow, so that the water in the reservoir was
The Lake Washington case has been an endur- rendered much less supportive of phytoplankton
ing example of what can be achieved when the growth. This approach to tackling the problem of
eutrophication is still relatively mild (Lorenzen, diffuse phosphorus loads was novel, costly and,
1974). Schemes elsewhere have not necessarily so far, scarcely imitated, but the savings in water
been quite so successful, either taking long peri- treatment and disinfection for distribution have
ods to take effect or, in some cases, having still repaid the investment. However, there was some
to show improvements (Marsden, 1989). The need nervousness in the early days following commis-
to understand these apparently quite different sioning, for although the dominance of Plankto-
sensitivities of lakes to altered phosphorus loads thrix was eroded, diatom blooms became more
provided the inspiration for the analysis carried prominent in the phytoplankton and these were
out by Instituut voor Milieu- en Systeem-Analyse still disappointingly expensive to treat (Clasen,
(IMSA) in Amsterdam. The study, published under 1979). Eventually though, as Subsystem 2 in-lake
Figure 8.8 (a) Revised explanation of lake responses to halved between 1975 and 1979 but with lit-
phosphorus-load reductions, including changes in
tle effect either upon the average phytoplank-
phytoplankton composition. Reading from right to left
ton biomass or its species composition. A steep
approximates the effects of eutrophication. (b) Hypothetical
P–chl relationships for individual lakes. The fitting of an decline in the phytoplankton mass occurred in
averaged regression probably underestimates the slope of chl the early 1980s, once the available phosphorus
a on P in those lakes where it is genuinely a limiting constraint had been lowered to a point below the demand
on biomass-carrying capacity Redrawn with permission from of previous maximum crops, MRP had become
Reynolds (1992a). a crop-limiting resource. Again, the slope of
response was close to 1 : 1 and steeper than Vol-
lenweider mean. In Schlachtensee (Fig. 8.7c), the
phosphorus availability slowly adjusted to the lag between the sharp reduction in phosphorus
sharp depletion in subsystem 1 phosphorus deliv- loading between 1981 and 1984 (when bioavail-
ery, so annual average chlorophyll concentrations ability was also reduced eight- to tenfold) and
also declined. By 1985, the reservoir had achieved any noticeable impact upon average phytoplank-
an oligotrophic status (Fig. 8.7a). Of particular ton crops was plainly related to the prior diminu-
interest is that the line fitted to the year-on-year tion of the large cushion of unused MRP. Once
averages is rather steeper than the Vollenwei- again, the eventual, near-linear biomass response
der OECD regression. It is in fact almost linear, depended upon the prior imposition of a demon-
at 1 : 1, reflecting the fact that phosphorus had strable capacity limitation by the bioavailable
become the capacity-controlling resource and, so, phosphorus. Without that, biomass is insensitive
truly limiting to phytoplankton crops supported. to the actual amounts of phosphorus supplied to
In Veluwemeer (Fig. 8.7b), one of the polder its growth medium.
lakes bordering the reclaimed Ijssel Meer, prob- The behaviours represented in Fig. 8.7 are
lems of excessive Planktothrix agardhii growth generic. The general case, proposed in Sas (1989)
were addressed by interception of the main point and reproduced in Fig. 8.8a, has received wide
sources of phosphorus and by progressive flush- acceptance. Reducing phosphorus loads have lit-
ing with a more dilute water source (Hosper, tle effect on biomass while unused PP capacity
1984). In-lake TP concentrations, then about ten saturates the requirements of the biomass suf-
times those in the Wahnbach Talsperre, were fering some other constraint. Movement from
stage 1 to a P-led biomass reduction (stage 3) gate of the MRP load and which is broadly pre-
and a switch, perhaps, away from relative abun- dictable (Eq. 8.3, Section 8.3.1 above). Its trans-
dance of Cyanobacteria to (e.g.) more benign fer to biomass alters the MRP concentration in
chlorophyte–chrysophyte associations (stage 4) the water but – theoretically – without alter-
depends upon the reduction in P availability to ation to the TP content of the water and with-
a point when it imposes the main constraint out affecting its overall bioavailability (though
on biomass carrying capacity. The critical quan- it is temporarily restricted to those organisms
tities are unique to each individual lake, how- that took it up). The subsequent fate of the par-
ever, so the coordinates of the response curve are ticulate P (and the organismic P in particular)
lake-specific. It is not difficult to recognise that does very much affect its future bioavailability
the Vollenweider–OECD regression is actually fit- and to whom. Whether the first beneficiary is
ted to what are ultimately the biomass response eaten and its phosphorus incorporated in the
curves to P loads for the individual lakes (notion- biomass of the consumer, or its cadavers are bac-
ally shown in Fig. 8.8b). Then, the more lakes terised and the phosphorus is recycled to the
that are featured in which the phytoplankton water, the effect of the intervention of the food
biomass is not severely or continuously P-limited, web is twofold. Either phosphorus ends up in the
then the flatter would be the fitted slope. The biomass of large organisms (macrophytes, fish,
Vollenweider–OECD coefficient of 0.91 reflects a birds, mammals), whence it is possibly exported
higher representation of P-deficient oligotrophic from the lake, or it contributes to a flux of bio-
lakes in the original dataset than (say) among the genic deposits. Typically, these products range
(mainly) North American lakes analysed by Rast from plant necromass (wood, rhizomes, leaves,
and Lee (1978) to yield a coefficient of 0.76. consigned to slow, microaerophilous decay) to
In the context of controlling Cyanobacteria, the rain of fine planktogenic detritus that set-
the analysis of site restorations considered by tles down towards the bottom of the lake. On
Sas (1989) unexpectedly revealed another com- geochemical scales, such sedimentary phospho-
mon behavioral feature. Before any noticeable rus may eventually be liberated but, in ecolog-
biomass or compositional response to reduced ical terms, the possibilities for its release and
in-lake P availability had occurred, several of reuse are strikingly habitat-dependent. Biogenic
the lakes with large or dominant component of organic materials reaching the bottom of the
bloom-forming Cyanobacteria showed a consid- water column contain still-reduced carbon and
erable increase in clarity (actually, Secchi-disk a variety of other co-associated elements (includ-
transparency). This observation was explained ing phosphorus). These become the resource of
by a deeper average vertical distribution of the benthic food webs in which detritivorous and
Cyanobacteria and other self-regulating species. decomposer organisms assemble and maintain
The expectation that average buoyancy in gas- their respective biomass and perhaps sustain the
vacuolate Cyanobacteria should respond to low- assembly of benthic consumers (including fish).
ered ambient nutrient concentrations (see p. Collectively, the potential effect of the benthic
206), thereby permitting the algae to scour for food web is progressively to oxidise the carbon
resources more deeply in the water column, is and to liberate the other elements which steadily
strongly upheld by these observations. Sas (1989) become in stoichiometric excess.
included this behavioural response in the generic The influence of habitat on this process oper-
curve (Fig. 8.8a) as a discrete stage 2. ates first through depth. As has been pointed
It remains to explore the between-site differ- out earlier (Section 8.2.3), much of the pelagic
ences in the Subsystem 1 – Subsystem 2 linkage sedimentary output in the sea is oxidised and
and the difficulty, or otherwise, of depriving phy- its more labile associated nutrients (including
toplankton of sufficient bioavailable phosphorus nitrates and phosphate) are liberated within 200
for them to continue to grow and replicate. The m of the surface. Nutrient cycling is not nec-
potential unexploited BAP concentration has a essarily impaired in lakes substantially <200 m
direct, steady-state relationship with the aggre- in depth, although the largely diffusive return
pathways, even for the most soluble nitrogenous through its combination with excess sulphide
components, operate more slowly than in the ions (which are liberated by sulphate reduction
open turbulence of the water column. The supply under slightly lower redox levels than required
of oxidant can also become problematic, insofar for oxidation of the ferrous ion) and preciptated
as annual net organic carbon fluxes of >100 mg as FeS, then the potential for chemical scaveng-
C m−2 carry an oxidative demand that may well ing of phosphate by ferric iron during the next
challenge the capacity of the ‘deep sediments’ oxidation is proportionately weaker. Accordingly,
(in this conext, anywhere between about 5 and more phosphate will be available to biotic recyc-
200 m from the surface) to satisfy, especially in ling.
periods of density stratification. Indeed, beneath What emerges is that in a majority of lakes
an oxidised microzone, the sediments and their of less than 200 m in depth, the bottom sedi-
interstitial liquor is anaerobic and strongly reduc- ments normally function as a phosphorus sink.
ing. Carbon oxidation and nutrient release are Phosphorus atoms take a single, albeit devious,
correspondingly slow, ecological function being trip from hydrological catchment to permanent
truly ‘oxidant limited’ (also as discussed in Sec- lake sediment. Tomorrow’s production effectively
tion 8.2.3 and represented in Fig. 8.2). With still depends upon tomorrow’s (or even today’s) deliv-
greater contributions of organic carbon from the ery of external phosphorus. Were this not so, nei-
production of eutrophic lakes, the oxygen deficit ther the Vollenweider model nor the restoration
may extend well into the hypolimnetic water methods that rest upon its central philosophy
above the bottom sediment, where the rate of can work. It is the mechanism by which trunca-
its biological reuse and organic decomposition is tion of the external phosphorus load to the water
equally depressed. column results directly lowered phytoplankton-
Against the background of differential oxida- supportive capacity. The new steady-state can
tive turbulent entrainment rates, the recycling be struck, usually within two to three reten-
of phosphorus becomes similarly varied (see e.g. tion times (Ryding and Rast, 1989; Reynolds
Reynolds and Davies, 2001). Orthophosphate ions and Irish, 2000). The limiting condition to this
released from decomposing biomass into the statement is the saturation of the phosphorus-
interstitial water of oxidised superficial sedi- binding capacity of the sediment iron, either by
ments may travel but a short distance before excessive orthophosphate recruitment or by the
being exchanged preferentially for hydroxyls on onset of high alkalinities (caused, for instance,
the surface of iron flocs and clay minerals in the by episodes of high productivity, carbon with-
sediment (Reynolds and Davies, 2001). Here, they drawal and base generation). With eutrophica-
are effectively immobilised, pending the onset of tion comes the prospect of accelerated recruit-
strongly reducing conditions (in which the fer- ment to sediments, possible saturation of the P-
ric ion Fe3+ ion is reduced, biotically and abi- binding capacity and its complete suppression
otically, to the soluble ferrous Fe2+ ion; see Sec- by anoxic reducing conditions. Even then, the
tion 4.3.1) or increased alkalinity, in which excess deeper sediments do not give up orthophosphate
hydroxyls now displace immobilised phosphate to the water column very readily and little is nec-
ions. Though potentially bioavailable in this form essarily returned to the water column for biotic
(Golterman et al., 1969), phosphate thus liberated reuse and recycling.
into the interstitial of anaerobic sediments (or
equally, into the anaerobic hypolimnetic water if 8.3.4 The internal load problem
present) is likely to be once again scavenged and As noted by Sas (1989), the further contribu-
immobilised by iron reprecipitating on exposure tory process which so delays the restoration of
to oxygen. This is important, because the act of eutrophied shallow lakes is that the phosphorus-
re-solution of iron-bound phosphate is not ‘recy- rich interstitial water is much more readily re-
cled’ to the biota if it is just re-precipitated in entrained than it ever is from deep sediments
the non-bioavailable form. If, in the interim, the protected by adjacent boundary layers. Struc-
mass of reduced iron is diminished, for instance tural disruption of the shallow deposits, mostly
through their physical penetration by wind- or municipal investment in P reduction had failed
wave-driven turbulence currents, may well result to ameliorate the eutrophication of the shal-
in the discharge of interstitial water and its low lake Trummen, in Sweden, and nothing less
solutes into the main water column. Applied to than the dredging and removal of the veneer
fine superficial sediments, turbulent shear, dissi- of P-saturated sediment was needed in order to
pating energy at rates in the range, E = 10−6 to restore an acceptable quality to the lake (Björk,
10−5 m2 s−3 (see Table 2.2 and Section 2.3.4), is 1972, 1988). This is an expensive, disruptive and
capable of penetrating and resuspending mate- problematic technique, not least because of the
rial from a thickness 30–40 mm (Nixon, 1988). high water content and fluidity of the dredged or
The process can be abetted through the activities air-lifted material. In spite of its potential fertil-
of burrowing animals and foraging fish (‘bioturb- ity as a soil conditioner, the material is not eas-
ation’; see, for instance, Davis et al., 1975; Petr, ily stored and de-watered except on level ground.
1977). The kinetics of ‘phosphorus release’ from Not surprisingly, lake restoration based upon sed-
the sediments to the water have been the sub- iment removal has been little practised.
ject of several major overviews (including Kamp- There is, however, considerable interest in
Nielsen, 1975; Baccini, 1985; Boström et al., 1988). a recent methodological development involving
Ultimately, however, it is the relative scarcity the application of clay minerals to shallow water
of significant chemical binding capacity in the bodies prone to free cycling of phosphorus (Dou-
water that permits the phosphorus thus released glas et al., 1998). These imitate the P-binding prop-
to become once again substantially bioavailable erties of metal oxides and hydroxides but with-
to planktic producers. out causing the toxicity problems that direct dos-
The further deduction that can be made is ing of (say) iron or alum salts might cause to
that, whereas deep lakes are likely to respond to water bodies. The binding performance of mod-
(i.e. be sensitive to) managed reductions in nutri- ified clay minerals based on lanthanum is espe-
ent loads, the prospect for successful restora- cially effective, removing phosphorus from solu-
tion of eutrophied small shallow lakes is prob- tion and firmly immobilising it in the particulate
ably much poorer after they have once been sedimentary fraction.
enriched significantly. The greater likelihood is As the techniques for treating and restoring
that historic phosphorus loads will, on ecologi- lakes through nutrient reduction become more
cal scales, thereafter be recycled indefinitely (van varied in concept and suitability, it becomes
der Molen and Boers, 1994). A case in point is that increasingly important to managers to seek guid-
of Søbygaard, a highly eutrophied shallow lake in ance in selecting the most effective approach and
Denmark, where a sixfold reduction in the exter- to predict its prospects of success. Such knowl-
nal phosphorus load of ∼30 g P m−2 a−1 had not edge might also assist the prioritisation of invest-
succeeded, even after 12 years, in reducing MRP ment among competing schemes of restoration.
levels in the lake to anywhere near a point where What is required is a simple basis for distin-
they might limit phytoplankton biomass capac- guishing among the sites where, to use the ter-
ity in the lake (Søndergaard et al., 1993; Jeppe- minology of Carpenter et al. (1999), eutrophica-
sen et al., 1998). By analogy, efforts to reduce tion is either readily reversible (response immedi-
external phosphorus loads on large, shallow Lake ate and in proportion to the change in P load-
Okeechobee, Florida, USA, have so far failed to ing), hysteretic (requiring profound reductions in
bring about restorative changes in the ecology P input over a protracted time period) or irre-
of the lake, which continues to function on the versible (recovery impossible by reducing P inputs
support lent by internal P cycles (Havens et al., alone). As indicated in the IMSA study (see Sas,
1996). 1989, above), the most sensitive direct measure
The available data make plain that mas- of the sensitivity of the biomass to change is the
sive P recycling continues to be maintained amount of bioavailable P capacity in Subsystem
quite independently of present external loadings. 2 that remains outside the biomass. This ‘cush-
In another much-cited instance, a considerable ion’ then represents the hysteretic resistance to
change that must be all but exhausted before Plainly, the lower the product, the more the
biomass is made dependent upon altered loads. lake is expected to respond to altered P loads.
Sensitivity is then reduced to the latitude of Reynolds (2003c) also proposed a further factor
TP–MRP transformation and the extent to which to cover the importance of good water quality
internal recycling of MRP is suppressed by sedi- at the particular site (internationally important
ment binding. amenity, drinking-water source, local wildlife
Following on from an earlier attempt reserve, etc.). It was urged that the total product
(Reynolds et al., 1998) to categorise the rela- should not determine prioritisation for correc-
tive dependence of biomass to each of a num- tive treatment but rather prompt further inves-
ber of environmental variables, Reynolds (2003c) tigation of the importance of overcoming high
used the inferred phosphorus thresholds to factors; nevertheless, it was noted that low over-
inform a sensitivity model. This involved scoring all scores (≤6) would refer to high-quality sites
against each of a number of categories (trophic whose trophic state should be ruthlesly defended,
state, hydraulic retention, bicarbonate availabil- whereas scores up to 32 are indicative of sites
ity and the extent of shallow sediments likely to that should respond hysteretically to altered
exchange phosphorus). The lower the score value, nutrient loads. Sites with higher scores may err
the greater is the sensitivity to change. beyond the bounds of a likely reversal of eutroph-
The first of these took the five trophic divi- ication.
sions in Table 8.1. Where small amounts of The UK Environment Agency now uses a
phosphorus are available, dependency is high scheme, based upon this approach, to assist the
and, accordingly, small changes in P loading implementation of its Eutrophication Strategy.
will invoke significant responses; thus ultraolig-
otrophic lakes score 1, hypertrophic 5. Flush- 8.3.5 Physical methods to control
ing lowers the responsiveness to nutrients; thus, phytoplankton abundance
lakes were considered to be sensitive (1) to P-load Alternative practical approaches to the control of
variation if the hydraulic retention time (tq ) was phytoplankton biomass yield, the rate of capacity
>30 d, not at all so (3) if tq <3 d and slightly attainment and, in many instances, its species
so (2) when 3 d < tq <30 d. Bicarbonate alkalin- composition have invoked the artificial enhance-
ity and the elevated frequency of opportunities ment of the physical processing constraints.
for phosphorus exchanges at high pH also affect Methods for extending the period of full arti-
sensitivity and this is reflected in the grading of ficial mixing in lakes deep enough to stratify,
alkalinity. Lakes are sensitive (1) where bicarbon- either by artificial destratification or by prevent-
ate alkalinity <0.4 meq L−1 , not at all sensitive ing the onset of stratification at all. The princi-
(3) to alkalinities >2 meq L−1 , and slightly sen- ple that is invoked is the one illustrated in Fig.
sitive (2) when 0.4 < alkalinity < 2.0 meq L−1 . 3.19, where the greater is the depth of the layer
Similarly, recognising the propensity for shallow- in which algae are entrained, then the greater is
water recycling, water bodies are considered to be the dilution of the light-determined supportive
sensitive (1) to changes in external load if <15% capacity. It is not just that the algae (say, at an
of the surface covered sediments <5 m deep but areal concentration of 80 mg chla m−2 ) may be
insensitive (3) if >50% of the lake is shallower diluted from a high near-surface concentration
than 5 m; lakes where >15% but <50% is under (80 mg chla m−3 , if confined to the top metre) to
5 m deep are considered to be slightly sensi- being spread uniformly through the top 80 m, at
tive (2). Multiplication of the four factors yields a concentration of 1 mg chla m−3 , but the plot
a product between 1 (a deep, ultraoligotrophic says this is the maximum concentration that an
lake with a chronically low phosphorus con- 80- m mixed layer could possibly support, as cal-
tent, low alkalinity and long retention) and 135 culated according to Eq. (3.25) in Section 3.5.3
(for a shallow hypertrophic river fed pool, sub- and assuming the most favourable conditions of
ject to rapid flushing by nutrient rich calcreous insolation (I0 ) and background absorption (εw +
flow). εp ). Mixed through only 40 m, the maximum
supportable concentration is closer to 20 mg chla This also requires energy expenditure and, thus,
m−3 (i.e. 800 mg chla m−2 ). Mixed through 10 an appropriate cost–benefit analysis; the benefits
m, the maximum supportable concentration is may include the direct effects of aeration, as well
about 160 mg chla m−3 (or 1600 mg chla m−2 ). as mixing. Yet another method is to force the
Using this logic in reverse, increasing the depth inflow through angled nozzles that direct the jets
of mixing ameliorates the concentration of algae of water to maximise their homogenising effect
in the reservoir water to be treated for potabil- (Toms, 1987).
ity, especially if the water is at all coloured or Caution is urged, however, against an assump-
carries a significant non-algal turbidity. The sev- tion that lack of stratification is necessarily
eral Thames Valley storage reservoirs that serve indicative of being well mixed. My experience
London have either been subjected to artificial from several reservoirs where pumps or bubble
mixing (Ridley, 1970; Steel, 1972, 1975) or they plumes are in operation agrees with the view
have been commissioned since that time with the that temperature gradients are generally weak,
capability built in at the design stage (Steel and or non-existent in the immediate environs of the
Duncan, 1999). The benefits, in terms of filtration device. Further away, phytoplankton may still be
efficiency and savings on the cost of treatment, disentrained and the self-regulating species may
have been substantial and worthwhile and the perhaps engineer slightly better growth condi-
nuisance Cyanobacteria are scarcely represented tions for themselves. It may be sufficient just to
any longer in the planktic flora of the reser- keep stratification weak, so that the work of wind
voirs in operation (Toms, 1987; Steel and Duncan, and cooling can more easily mix the water col-
1999). umn from time to time. Modelling algal growth
The methods for overcoming and prevent- in reservoirs with PROTECH (Section 5.5.5; work
ing thermal stratification and for extending the still in process of publication at the time of
period of non-stratification have been in use this book going to press) has strongly suggested
for many years and are well established (Irwin to me that many such reservoirs are often less
et al., 1969; Dunst et al., 1974; Tolland, 1977; well mixed than supposed. On the other hand,
Simmons, 1998; Kirke, 2000). The most success- other reservoirs without any other source of arti-
ful of these function as lift pumps, comprising ficial mixing also remain very weakly stratified,
a vertical cylinder with either a paddle or a at those times when water is drawn off for treat-
injected bubble stream to draw cold deep water ment from near the bottom of the water column.
to within a short distance from the surface where Unless it is unsatisfactory on chemical or biolog-
the flow diffuses laterally. Another device, the ical grounds, I believe it to be preferable to draw
Helixor® , uses an Archimedean screw to achieve water from depth as it yields water of reason-
a similar effect. By setting up entraining vor- able quality but (literally) undermines the ten-
tices, these devices move rather more water than dency to stratify. The power-generating plant sit-
flows through the cylinders. Moreover, they work uated at the dam of Brasil’s Volta Grande Reser-
reasonably efficiently so long as water is dis- voir draws its flow from the base of the reser-
charged from the pipe below the surface of the voir and, in complete contrast to others in the
water body: lifting water above the surface level cascade, fails to stratify (Reynolds, 1987b). A 70-
requires a rather greater pumping effort. Most m, near-isothermal water column in a tropical
use electrical power to drive the pumps (the alter- impoundment is an unforgettable surprise to a
native is wind power) and their operation car- limnologist!
ries significant costs, which have to be balanced The phytoplankton in this reservoir also dif-
against the savings to be made on treatment and fered from that of others in the chain. The
disinfection. unstable water column of Volta Grande was
Another popular method of weakening strati- dominated by the diatom Aulacoseira and the
fication is to use bubble plumes: punctured hose desmid Staurastrum, together constituting an
or tubing is laid out on the reservoir floor and assemblage classifiable as Association P. This is
compressed air is forced out of the perforations. one of the groups of mainly R-strategist groups
whose natural occurrence is in well-mixed water nique of directly applying powdered copper sul-
columns and whose ascendency can be stim- phate to reservoirs as an algicide. The salt dis-
ulated in lakes and reservoirs by episodes of solves easily in the water, where a strength of 1
artificial or experimental mixing (Lund, 1971; mg L−1 of the penthydrate is sufficient to kill
Reynolds et al., 1983b, 1984; Reynolds, 1986b). most species of phytoplankton and – inciden-
Restratification may revert the flora to species tally – more than enough to administer a lethal
either less dependent upon entrainment for sus- dose to crustacean zooplankters (my own unpub-
pension or more dependent upon disentrain- lished observations). The effectiveness of the
ment to have any chance of harvesting suffi- treatment against algae seemed to vary, appar-
cient light (i.e. C-strategist nanoplankton and ently in relation to the degree of mucilage invest-
motile microplankton, including the larger, self- ment (Coesel, 1994; see also Box 6.1, p. 273). Most
regulating S-strategist species). The possibility of of the copper is eventually precipitated and sinks
alternating conditions, before any particular set to the sediment. Therein resides the problem,
of adaptations led one group of species to become for the copper remains sensitive to changes in
overabundant, was specifically investigated in the acidity and redox potential. The salutary case of
study of Reynolds et al. (1984) and through the the gross copper poisoning of the Fairmont Lakes
population dynamics of selected algal respon- Reservoir, Minnesota, USA, following years of fre-
dents (Reynolds, 1983a, 1983b, 1984c). This led quent dosing to control algal growths (Hanson
Reynolds to propose an approach to manag- and Stefan, 1984) both confirmed the worst fears
ing reservoirs by mixing them intermittently, about the potential threats of long-term copper
in order to prevent any group becoming too applications and doubtless influenced the ban-
numerous. Reynolds et al. (1984) conceded that ning of its use in many countries.
it was probably too cumbersome to apply safely,
in the sense of maintaining close control over 8.3.6 The control of phytoplankton by
phytoplankton biomass. Generally (and always biological manipulation
provided that the water column is sufficiently As an alternative to being able to control ade-
deep and/or the water sufficiently turbid), per- quately the biomass-supportive capacity, the idea
sistent mixing to control light-determined carry- of regulating its allocation among desirable and
ing capacity is the simpler and more efficient undesirable biota has had a long-standing appeal
option. However, a potential drawback of doing in lake management. The term ‘biomanipula-
this, which was of concern to Lund (1971, 1975; tion’, coined originally by Shapiro et al. (1975),
Lund and Reynolds, 1982), is that prolonged mix- has become increasingly restricted to refer to the
ing would lead surely and ultimately to domi- control of undesired organisms by the deliberate
nance by such organisms from functional group adjustment of the abundance of the next trophic
S1. Once established, prevalence of Planktothrix level of consumers. The underpinning logic is
agardhii, Limnothrix redekei or Pseudanabaena lim- the same one that distinguishes ‘bottom–up’
netica is hard to overcome (Reynolds, 1989b). That from top–down processes (see pp. 287--8). Thus,
this has not happened in the case of the routinely not surprisingly, it has been just as controver-
destratified Thames Valley reservoirs seems to be sial and contested just as vehemently (see, for
a function of their short retention times (usu- instance, Carpenter and Kitchell, 1992; De Melo
ally <20 days) (Toms, 1987; Reynolds, 1993b). The et al., 1992). However, also like the ‘bottom–up
combination of a hydrological displacement time vs. top–down’ debate, the issues ceased to be
of just 10–30 days and a fairly intensive mixing so contentious, once the critical roles of vari-
regime is perhaps decisive in the continued ability and site-specific factors affecting resource
exclusion of Planktothrix. turnover became more clearly understood. There
Mention of the treatment or prevention of is, for instance, a much wider acceptance
algal growth in drinking water reservoirs gives an of the importance of the basic carbon path-
appropriate opportunity to refer briefly to once- ways in determining the optimum functional
common practice but now much discredited tech- adaptations and, so, the identity of the most
efficient traits (discussion in Sections 8.2.3 and trophogenic layer of the water column and its
8.2.4). Interestingly, the assumption of direct periodic exposure to the shear stress of the suface
and manipulable linkages between phytoplank- mixed layer. This property influences both the
ton, zooplankton, planktivorous fish and pisci- cycling of resources and the interactions between
vores is now argued to provide an acceptable benthic and planktic consumers. However, the
model only for truly pelagic systems. To apply remarks apply broadly to lakes in which >50%
the deductive reasoning that says (e.g.) the selec- of the area is shallower than 5 m (as in Section
tive removal of piscivores distorts the food web by 8.3.4).
sparing planktivorous fish to prey more heavily The supportive capacity of shallow lakes may
on zooplankton and promote a larger phytoplank- be regulated by hydraulic throughput, turbidity
ton biomass, requires effective management over or colour but, naturally, the supply and chem-
very substantial areas of the pelagic. Similarly, ical binding of phosphorus are frequently such
to approach the problem of phytoplankton over- to regulate the accumulation of biomass. Modest
abundance fuelled by extra nutrients simply by depth and high perimeter to volume ratio, how-
reducing planktivory or increasing piscivory in ever, leave the small water body quite vulnera-
a small lake is rather to ignore the role of the ble to drainage and the import of solutes and
terrestrial subsidies and benthic feedbacks in particulates arising from land-use change, loss of
undermining the system reliance upon phyto- vegetation, increased particulate loss from cul-
plankton photosynthesis. Such retrospective con- tivated soil, as well as to incursion by any of a
siderations aid the appreciation that the more large number of organic pollutants. With the pos-
successful attempts to devise sustainable bioma- sible exception of chronically nutrient-deficient
nipulative schemes for controlling phytoplank- small lakes, many are susceptible to quite rapid
ton and attaining an aesthetically pleasing water ecological change – between clarity and high
clarity have been most successful in small, shal- algal turbidity and between supporting exten-
low lakes and ponds and least so in larger, deeper sive macrophytic vegetation, both submerged and
lakes (Reynolds, 1994c). It is nevertheless neces- emergent. These fluctuations among periodically
sary to qualify this statement again by saying steady states have been studied for some time
that ‘if to biomanipulate is simply to distort the (Scheffer, 1989; Blindow et al., 1993; Scheffer et al.,
food chain simply attain a more beneficial con- 1993; Carpenter, 2001). Some of these transitions
dition, then there is no theoretical objection to have now been empirically characterised and the
biomanipulating an area the size of the Atlantic outcomes are helpful in informing strategies for
Ocean’ (Reynolds, 1997a). If, however, the intent is securing the successful biomanipulative manage-
an alternative, stable and self-sustaining system ment of shallow lakes. The following subsections
but running on, broadly, the existing resource recount these.
and energy inputs, then attention has to be
given to the means of substituting alternative sys- Nutrients, phytoplankton and grazers
tem components that process those inputs along There is no fundamental difference between
quite different pathways. These alternatives are large and small lakes in the supportive capacity
confined to small or shallow systems. Biomanip- of plant nutrients so the relationship between
ulation of the functional importance of phyto- the potential yield of phytoplankton chlorophyll
plankton is a practical and sustainable corrective and the supportive capacity of bioavailable phos-
to the symptoms of eutrophication only in such phorus (as suggested in Eq. 4.15, Section 4.3.4) is
water bodies. independent of lake size. Scheffer’s (1998) analy-
In the context of the present chapter, the sis of the average summer chlorophyll yields of
adjective ‘shallow’ is used more in its func- 88 shallow Dutch lakes, showing a slope close to
tional sense than within any absolute constraint 1 µg chla (µg TP)−1 and saturating at about 300
(Padisák and Reynolds, 2003). The most impor- µg TP L−1 is in accord with expectation. Such con-
tant property is the frequent or continuous concentrations of chlorophyll a are unexceptional
tact of most of the bottom sediment with the in nutrient-rich shallow water, where they may
typically characterise near-unialgal populations to kept in check from its own potential growth
of group-J or group-X1 chlorophytes or group-S1 and recruitment of filtration capacity (say 0.19
Planktotricheta (see Section 7.2.3). d−1 ), fish feeding needs to remove about 0.1 mg
Whereas the latter may dominate perenni- zooplankton C L−1 d−1 , but with rather narrow
ally, the green algae may be subject to abrupt margins of variation. If we take the model of
collapses, usually as a consequence of intensive Elliott and Hurley (2000a; see Section 8.2.2) and
grazing by filter-feeding zooplankton, or, sub- estimate a requirement of 40 mg zooplankton C
sequently, to recovery to large concentrations. per gram fresh-weight of fish per day, then the
The dynamics of these fluctuations are best required stability would be struck when the fish
known from systematic observations on com- biomass is equivalent to 1 g per 400 L or 2.5
mercially managed fishponds (Hrbácek et al., g fresh weight m−3 . No allowance is made in
1961; Hillbricht-Ilkowska and Weglenska, 1978; these approximations for variable temperature,
Korı́nek et al., 1987) but many have been imitated day-to-day differences in phytoplankton growth
in controlled experiments (Shapiro and Wright, potential or for the divorced timescale of changes
1984; Moss, 1990, 1992; Carpenter et al., 2001). in fish biomass, through progressive growth and
Unchecked by consumers, phytoplankton can increasing appetites.
grow to the limits of the supportive capacity Again, there is little encouragement here for
(which may be set by light rather than by phos- realising a reliable and stable managed state. In
phorus). Direct herbivorous consumers include one of the few really clear cases of an engineered
rotifers and crustaceans and, at low latitudes, cer- stability among three trophic levels, Kořı́nek
tain species of fish (Fernando, 1980; Nilssen, 1984). et al. (1987) observed a persistent standing popu-
Among temperate lakes, filter-feeding daphniids lation of cryptomonads, of 2–4 µg chla L−1 in
prove to be well capable of developing biomass a carp pond. The check was exercised by an
(equivalent to ∼1 mg C L−1 ) (Section 8.2.3) with apparently steady population of Daphnia pulicaria,
an aggregate filtration capacity (approaching numbering about 100 L−1 . Carp were present
1 L L−1 d−1 ) that will overhaul the rate of algal but at an unspecifed ‘low level’. On restocking
growth and rapidly clear the water of all fine with fish in the summer, there was, predictably,
particulates. Without access to filterable foods, an abrupt decrease in Daphnia, followed by a
the Daphnia quickly starve and there is mass rapid increase in chlorophyll a concentration,
mortality among the smaller instars especially to >100 µg chla L−1 .
(Fig. 6.4). Filtration collapses, and the next phase
of algal dominance can commence.
This see-saw between alga and grazer mass is The role of macrophytes
scarcely a management formula but the fluctu- This last example supports the conjectures about
ations may well be damped in the presence of stability but it also reinforces the intuition that
significant densities of young cyprinid fish. The heavy fish stocking is conducive to the over-
expectation is that fish consumption of zooplank- abundance of phytoplankton in small lakes. On
ton will reduce the exploitation of the phyto- the other hand, the lurching, cyclical alterna-
plankton yet still control it within acceptable lev- tion of bottom–up production and exploitative,
els. The belief is that the persistence of high algal top–down responses is probably the norm in such
populations is the consequence of overstocking highly managed systems. That similar behaviour
with fish and the zooplankton is simply squeezed is not so apparent in less impacted systems,
out. The truth is that even the three-component which may support fish production but without
system is extremely difficult to balance. If a stable being subject to algal blooms of a consistency
phytoplankton is desired, it needs to be cropped approaching that of pea soup, requires us to look
as fast as it self-replicates. If it is able to dou- more closely at the trophic pathways in other
ble once per day, then the zooplankton needs to types of small lake. Stability of water quality, at
filter about half the water volume each day. If least at the time scales of season to season and,
the requisite biomass (roughly 0.5 mg C L−1 ) is in many instances, from year to year, is usually
associated with a strong presence of macrophytic In this way, different types of macrophyte
angiosperms and pteridophytes. tend to be associated with particular habitats. A
A number of families is represented in the full spectrum of life forms may be encountered in
aquatic macrophytic flora (for an up-to-date clear, oligotrophic lakes. Lobelia–Littorella stands
account, see Pokorný and Kvet, 2003). All show develop offshore, where, simultaneously, the sed-
secondary adaptations to one or other of the iments are sufficiently fine and the light pene-
habits that they have adopted. These broadly sep- tration is good. Deeper and further out, dense
arate among species that are either: rooted and stands of naiads and elodeids may develop, merg-
fully submerged; rooted with emergent foliage; ing finally with the isoetids, the deepest-growing
rooted with floating leaves; unrooted and float- macrophytes. Emergent flowering plants (reeds
ing. Besides supplying one of its key needs in and reed-like plants such as Phragmites, Typha,
abundance, water offers to submerged macro- Schoenoplectus and some associated herbs) may
phytes the benefit of Archimedean support and develop on more silted shores, while stands of
the equability of temperature fluctuation that fully submerged plants (such as Myriophyllum and
surpasses that experienced by their terrestrial rel- Najas and various Potamogeton spp.) and plants
atives. Aerating the submerged tissues (especially with floating leaves (nymphaeid water lilies) form
the roots) was a major constraint to returning in front of them in quieter bays. In more enriched
to water but its achievement was a major evo- lowland lakes, the reedswamp may be prolific
lutionary advance. The large interconnected air but the submerged plants are compressed into
passages and the aerenchymatous tissue that dis- narrow depth ranges. In small, shallow ponds, a
tinguish the structures of most aquatic species submerged vegetation of Myriophyllum or Cerato-
are illustrative of the remarkable power of adap- phyllum may carpet the entire bottom. In clear,
tation. Rooted macrophytes vie with microphytic calcareous shallows, the dominant plants may be
algae for light and nutrients but it is plain that macroalgae – the stoneworts, such as Chara and
they have inferior rates of growth and repli- Nitella.
cation to autotrophic microorganisms. However, In their respective localities, macrophytes
most macrophytes enjoy two crucial advantages may be regarded as ‘system engineers’ (Jones et
over microalgae: through their roots, they have al., 1994), fulfilling a keystone role to many of
access to nutrients in the sediments that are associated species (see Section 7.3.4). Dense plant
not normally available to algae; and they are stands suppress turbulence and create a calm,
able to develop tissues for the internal storage cryptic habitat to a trophic network of inverte-
of excess photosynthetic products and reserves brates. It is an environment that traps fine silt
of potentially limiting nutrients. Aquatic macro- and organic debris, whose deposition leads to
phytes are able to colonise standing water in all a truncation of the water depth and a succes-
climatic zones. Their distribution within lakes sion of swamp, marsh, and, finally, woodland
is restricted to shallow water, to depths depend- species, whose centripetal invasion potentially
ing mostly upon its clarity (the typical Secchi- obliterates open water (Tansley, 1939). The macro-
disk reading is often a good guide to the lim- phytes or, more particularly, the fungal fragmen-
its of colonisation by submerged plants; see Blin- tation and bacterial decomposition of seasonal
dow, 1992). In clear, deep lakes, there is usually dieback, provide some direct source of nourish-
a vertical zonation of the principal species and ment to the community of benthic invertebrates
life forms (Pokorný and Kvĕt, 2003). In the hor- but their primary role in the energy flow of the
izontal, there may be many shores from which lake lies in the provision of habitat to many other
they are excluded, for reasons to do with the producers. Surface-growing (epiphytic) algae and
suitability of the substratum and its exposure fungi (together, the Aufwuchs) constitute a source
to wind or wave action. Species also vary in the of food to snails, microcrustaceans and larval
degrees of acidity or alkalinity they will tolerate, ephemeropterans. These have a variety of poten-
and many are susceptible to changing levels of tial predators, including flatworms, hirudineans
productivity. and beetles. Accumulating organic debris and
detritus is exploited by a variety of malacostra- This explanation is now widely accepted. It has
cans and larval hemipterans and by dipterans, been supported by some direct measurements
oligochaetes and bivalves. There is another layer of the light attenuation by epiphytes on eel-
of predatory carnivores including larval odonates grass (Brush and Nixon, 2002). The critical photon
and coleopterans. Macrophyte beds play host to a flux transmitted to the photosynthetic apparatus
very complex web of trophic interactions. Ideally, to the host plant, before its own performance
their part in the economy of the lake increases is impaired, will vary with its own depth and
relatively and inversely to the size and depth the levels of harvestable light. However, Brush
of the lake. Quite plainly, in small or shallow and Nixon (2002) developed regressions predict-
lakes, macrophyte beds are the focus of the most ing epiphyte densities of 10 mg dry mass per
intense production and the location of most of cm2 reduced PAR transmission by 33–70%. This
its producers. is slightly less than some of the values sug-
gested in earlier studies. The nature and habit
Effects of eutrophication on macrophytes of the epiphytes also influences light transmissi-
In many instances of eutrophication of shal- bility: the biomass achieved by crustose growths
low lakes, the first symptom of change has of naviculoid diatoms is less than what is attain-
been the disruption and decline of macrophyte able by erect cells of Synedra-like diatoms or the
dominance. Charophyte communities are con- branching arbuscular formations of Gomphonema
sidered to be the most susceptible to replace- or many chlorophyte forms. Nevertheless, the
ment by more diverse assemblages of species expectation is that epiphyte densities of 20 mg
of richer lakes, before these too become elim- dry mass per cm2 would exceed their tolerance
inated (De Nie, 1987). The complete transition, by macrophytes. Actually, such densities become
to a turbid, phytoplankton-dominated system, is unstable and slough off periodically, although
regressive on grounds of loss of diversity and the host plant would, by then, be unlikely to
amenity, as well as a probable deterioration in derive much benefit. Depending on species com-
the yield of fish. The mechanism of such tran- position, the chlorophyll a complement of the
sitions was supposed to be light-mediated: as epiphytes might be 0.5–2.0% of the dry mass: the
nutrients in solution increased, so did the phyto- critical epiphyte density of 10–20 mg dry mass
plankton and, in consequence, light penetration per cm2 is equivalent to 0.05–0.4 mg chla cm−2 ,
decreased. Even the intermediate replacement or 0.5–4 g chla m−2 of macrophyte leaf area.
of low-growing charophytes by tall-stalked Pota- To compare this with the areal concentration
mogeton (Moss, 1983) fits this explanation. How- of phytoplankton, or even of epilithic films on
ever, it is only part of the story and it does the solid surface of rocks and stones, allowance
not explain why species with floating leaves and must be made for the leaf-area index (the area
the emergent reedswamp species should also fail of leaf surface per unit water area) (LAI). For a
to survive. Work by Sand-Jensen (1977) had sug- LAI of ∼5, potential densities of epiphytes (on
gested that the shading of epiphytes growing on this logic, up to 20 g chla m−2 ) would be far in
eelgrass (Zostera) significantly reduces light pen- excess of supportable phytoplankton or epiliths.
etration to the leaf blade. At about the same Turning the calculation around, the theoretical
time, Phillips et al. (1978) reported field observa- light-limited maximum active photoautotrophic
tions and some supporting tank experiments in biomass that can be supported (∼120 g C m−2 )
which the effect of epiphytic algal growth was to Fig. 3.8, Section 3.5.3) and the maximum phyto-
smother the photosynthetic surfaces of the host plankton crops that are observed (range 30–50 g
plant. Although a film of epiphytic algae is proba- C m−2 , say, 0.8–1 g chla m−2 ) would leave sub-
bly a normal occurrence on the surfaces of water merged macrophytes already light-deficient.
plants, Phillips et al. (1978) considered that the Of course, the epiphytic community of the
increasing availability of nutrients in the water Aufwuchs is not exclusively algal and its chloro-
column is as beneficial to epiphytes as to planktic phyll content is normally dynamic owing to
algae. grazers. Macrophytes may be able to suppress
epiphytic and some planktic algae through the This hysteretic behaviour of forward and
production of allelopathic substances (reviewed reverse switches has been successfully modelled
in Scheffer, 1998), although, realistically, the by Scheffer (1990; see also 1998), using graphical
principal influence over dominance is the compe- analysis. The model has stimulated further inves-
tition for space, light and nutrients. This struggle tigative studies of the factors that most influ-
is further influenced by the role of consumers – ence site-to-site differences in the trigger levels
especially planktic filter-feeders and Aufwuchs of phosphorus and the response of the whole
browsers. The complexity of these interactions system.
make it difficult to predict their many potential
outcomes. Nevertheless, some experiences of the Macrophytes and trophic relationships
change from macrophyte to microplankton dom- The influence of fish on the operation of the
inance of shallow lakes, and the reverse, focus on lower trophic levels is no less profound in the lit-
the critical switches. toral world of shallow lakes than it is in the
The eventual loss of submerged macrophytes pelagic zone (Gliwicz, 2003b). In the absence
as a consequence of progressive eutrophication of fish predation, filter-feeding crustacean zoo-
of shallow lakes has often been found to be plankton can become sufficiently numerous to
quite abrupt, presumably as a result of near- clear the water of algae. In macrophyte-rich
simultaneous light inadequacy across a relatively ponds, however, alternations with algal abun-
flat bottom. It is often the case that macro- dance are considerably damped, both in inten-
phytes are either extensive across the lake, or sity and through time. The basis of this relative
they are very nearly absent from a water col- stability seems to be that a low concentration of
umn that is turbid with phytoplankton. Thus, large-bodied, filter-feeding cladocerans, involving
these small, shallow lakes seem typically to exist genera such as Simocephalus, Sida, Diaphonosoma
in one of two alternative steady states – either and the large Daphnia magna, is able to keep the
they are macrophyte-dominated with clear water; water reasonably clear of phytoplankton. The sup-
or they are phytoplankton-dominated, with fre- ply of locally generated and/or trapped organic
quent high turbidity and a dearth of macro- debris, dislodged epiphytes and abundant detri-
phytes (Blindow et al., 1993; Scheffer et al., tus and bacteria from near the bottom provides
1993). an alternative and sufficient resource of filterable
From syntheses based on the eutrophication organic carbon to support the stock of cladocera
of a large number of shallow lakes in Europe, (Gulati et al., 1990).
it appears that the change from macrophyte The presence of fish does not automatically
to phytoplankton dominance occurs anywhere disrupt this stability. Cropping of zooplankton
within a wide range of aquatic total phospho- by mature fish and 0+ (i.e. fry under 1 year
rus concentrations (50–650 µg P L−1 ). These are in age) is not necessarily less intense in shal-
arguably capable of sustaining phytoplankton low lakes than it is in deeper lakes. Indeed,
and/or epiphytic algae of chlorophyll a at con- without the refuge provided by vertical migra-
centrations likely to deprive submerged macro- tion, zooplankton is potentially subject to even
phytes of adequate light. According to Jeppesen heavier predation. Several studies have revealed
et al. (1990), many upward transitions occur in a that the behaviour of free-swimming cladocer-
range 120–180 µg P L−1 , in which the phosphate ans alters under the threat of fish predation,
sequestering power of macrophytes is saturated exploiting the alternative refuge from ready vis-
(Søndergaard and Moss, 1998). However, it does ibility offered by macrophyte beds. Horizontal
not follow that a downward shift in phospho- migrations were first noted by Timms and Moss
rus availability will trigger a return to macro- (1984), the zooplankton moving into macrophytes
phyte dominance in the same range. Indeed, it by day and moving into more open water in the
seems that phosphorus concentrations need to hours of darkness. Moreover, subsequent stud-
be rather lower than 120 µg P L−1 for the switch ies have demonstrated that the variety of species
to work in the opposite direction (Moss, 1990). and the size classes of the cladocerans behaving
thus is directly related to the potential intensity duction (usually by large and persistent popu-
of planktivory (Lauridsen and Buenk, 1996; Lau- lations of Planktothrix agardhii), robust, micro-
ridsen et al., 1999). The experimental investiga- bially mediated recycling of autochthonous and
tion of the behaviour of Daphnia magna (Laurid- allochthonous organic matter, and a food web
sen and Lodge, 1996) showed a natural tendency comprising little but the larvae of Chironomus
to avoid the Myriophyllum plants was reversed in plumosus or C. anthracinus and benthic-foraging
the presence of sunfish (Lepomis cyanellus) and of carp, such as Cyprinus carpio. Most people find
chemicals in the water in which the fish had this unattractive and believe its ecosystem health
been recently present. In effect, the cladocerans to be poor but it is, nonetheless, a stable
benefit from the presence of macrophyte beds state and also one that is ‘depressingly sustain-
through reduced exposure to planktivory but able’ (Reynolds, 2000b). It is for these reasons
remain sufficiently active to regulate the concen- that it is also very difficult to manage sites
tration of phytoplankton and other forms of fine away from this kind of species structure with-
particulate organic carbon. These are powerful out applying some fairly drastic environmental
contributions to the short-term ecological stabil- engineering.
ity of macrophyte-dominated systems.
However, these are not continuous forces, nei- Biomanipulative management
ther are they insensitive to changing fish den- The prospects for preventing and reversing the
sities, species composition and feeding refuges, worst symptoms of eutrophication in small, shal-
nor are they independent of macrophyte den- low lakes probably depend upon first recognising
sity. High densities of facultative, opportunist the present state. Traditionally, the art of suc-
and obligate planktivores in open water (in small, cessful biomanipulation is built upon the con-
shallow lakes, this may refer mainly to young trol of phytoplankton abundance, which really
percids and cyprinids, as well, of course, as means protecting the zooplankton. There are
the 0+ recruits of most species) will eventually thus threshold levels of planktivory that should
overcome the capacity of cladocerans to clear not be exceeded. Even then, phosphorus availabil-
the water. Encouraging piscivorous predation on ity may still bias against macrophytes, whose col-
planktivores (for instance, by stocking, with pike, lapse will precipitate the loss of a viable agent in
Esox lucius) should assist in keeping their impact maintaining water quality.
on the zooplankon within its critical threshold Quite clearly, if habitat quality once slips to
intensity and, thus, in upholding and enhancing that level, then recovery demands that the tight
the stable, macrophyte-dominated state (Grimm linking among the minimal components of the
and Backx, 1990; Bean and Winfield, 1995). Other- whole system is broken. This may require drastic
wise, phytoplankton may be released from effec- reductions by netting of stocks of cyprinid fish
tive control and the balance in favour of macro- (roach, Rutilus; bleak, Alburnus; bream, Abramis;
phytes, supposing that they really do dominate carp, Cyprinus carpio), as well as small individu-
the material transfers within the existing sys- als of non-cyprinid species. It may also require
tem, may quickly be turned against them. More- the dredging out of the unoxidised sediments
over, with the foraging of the facultative plank- and/or routine flushing with less nutrient-rich
tivores focused on benthic invertebrates (see Sec- water. The assisted re-establishment of appropri-
tion 8.2.2), there is an increasing risk that the ate macrophytes and their protection from graz-
weakened macrophytes are uprooted and dis- ing birds (such as coot, Fulica atra) may prepare
lodged. Sediments are increasingly liable to tur- the way for restocking with non-cyprinid fish.
bation with further feedbacks towards macro- Such restorations are cumbersome and expen-
phyte exclusion by turbidity and loss of refuge sive to apply and the precedents have achieved
for zooplankton. The whole system can quickly only varying degrees of success. Needless to say,
experience a rapid ‘regime shift’ (Carpenter, instances of successful, self-sustaining biomanip-
2001), towards a simplified, low-diversity struc- ulative management have been attained before
ture based upon copious planktic primary pro- the undesirable phytoplankton-dominated steady
state has become too firmly established (Hos- ies, the concentrations of acidic gases and radi-
per and Meijer, 1993). Often, it is sufficient cals and the concentrations of neutralising bases
to intervene before the ‘forward switches’ have determine the hydrogen ion concentration in
operated. Over-management of macrophytes and lakes and seas. Lakes are, or become, acidic when
overstocking with fingerling fish, both perpetra- the input of hydrogen ions exceeds the amounts
tions of single-minded but misinformed anglers, of neutralising bases gleaned from the water-
have predictably deleterious impacts on zoo- shed through the weathering of rocks. Phyto-
plankton stocks and the objective of limpid water. plankton production in the lake also helps to
Once the stability of the macrophyte- reduce acidity through the uptake of nitrate, sul-
dominated state has been threatened, greater phate and carbon dioxide gas dissolved in the
control over planktivore stocks must be exer- water.
cised. Based on the investigations of Gliwicz The simple methodology for estimating the
and Preis (1977) on the direct impacts of plank- hydrogen-ion concentration (expressed as a nega-
tivory on zooplankton, it would seem unwise tive logarithm, the pH) allows us to record it with
to allow the fresh biomass to exceed 300 kg some diffidence. The effects of photosynthetic
ha−1 (30 g m−2 , about 6 g C m−2 ); to do so withdrawal of carbon dioxide and the buffer-
would impair the sustainability of secondary pro- ing provided by the dissociation of the weakly
duction yields (see Sections 6.4.2, 8.2.3). Stock- acidic bicarbonate ion have been discussed ear-
ing of benthic foragers (carp, bream and tench, lier (see, especially, Sections 3.4.2, 3.4.3). The car-
Tinca tinca) should be avoided. It may be nec- bon dioxide dissolved in precipitation (rainfall,
essary to exclude bottom-feeding ducks, coots snow, condensation) comes from the atmosphere.
and other rails. Other ‘reverse switches’, includ- The natural pH of rainwater may be between 5.0
ing the reduction of nutrient loading, should be and 5.8, depending upon contemporary temper-
applied if feasible. For fuller advice and guidance ature and pressure conditions and the extent of
on applying biomanipulative techniques, any of carbon dioxide saturation. The particular prob-
the excellent manuals now available should be lem of ‘acid rain’ begins with the anthropogenic
consulted (Hosper et al., 1992; Moss et al., 1996). enhancement of the products of oxidation –
oxides of carbon, sulphur and nitrogen – in the
8.3.7 Phytoplankton and acidification atmosphere. There, solution and reaction with
Acidic waters are not unnatural but anthro- liquid water results in the enhanced formation
pogenic acidification of rainfall and land of strong acids (H2 SO3 , H2 SO4 , HNO3 ). On contact
drainages has impacted upon aquatic ecosystems with the ground, acidic precipitation may be neu-
with effects quite as deleterious as those imposed tralised or rendered more or less alkaline by the
by nutrient enrichment. However, the behaviour bases (carbonates of Ca and Mg, K+ , Na+ and NH3 )
of phytoplankton is rather less central to the that it dissolves. Alternatively, rainwater pass-
perceived environmental damage through acid- ing across impervious catchments dominated by
ification than it is in the case of eutrophication. granitic, dioritic and other weathering-resistant
The restoration of acidified water bodies is not a rocks acquires little base or alkaline buffering
major concern of the present section; rather, the capacity. Accordingly, lakes and rivers draining
relationships between phytoplankton and envi- catchments in which the buffering capacity is
ronmental excesses in the hydrogen ion concen- generally poor, achieving alkalinities of ≤0.2 meq
tration are the main focus. L−1 , are particularly sensitive to ‘acid rain’.
Natural water supplies to lakes and seas At first, the symptoms of acidity generation
are not pure. Characteristically and distinctively, were recognised mainly in the immediate down-
they contain varying amounts of numerous wind localities of industrial and domestic fuel-
solutes, leached from the atmosphere and from burning: areas of northern England and western
the terrestrial catchments with which the water Germany were once very badly affected with dam-
has been in contact. In much the same way age to trees and buildings, loss of soil fertility
as nutrients are loaded on receiving water bod- and lowered pH of drainage water. Through much
of the industrial period, these effects were sec- aquatic ecosystems have come from base-poor,
ondary to (but they compounded) the more obvi- high-rainfall upland areas, where the inherent
ous consequences of the deposition of unburned neutralising capacity is least: northern Europe
carbon (soot). Analogous problems arose from (Scotland, Fennoscandia and, especially, the Tele-
ore-smelting in the area of Sudbury, Ontario. As mark area of Norway) and certain Shield areas of
much to deal with local air-pollution problems Canada.
as for any other reason, a progressive switch was The impacts of acidity on aquatic ecosystems
made to alternative, more completely burning invoke several mechanisms. Besides having to
fuels that are combusted at higher temperatures, cope with an excess of hydrogen ions, low pH
whilst waste gases were vented higher into the affects the chemistry of several elements of bio-
atmosphere. At the same time, the total con- logical importance. One obvious consequence of
sumption of carbon-based fuels fuel has greatly the direct sensitivity of calcium carbonate sol-
expanded, with the result that local problems ubility to pH affects the tolerances of shell-
of poor air quality and noxious deposition were building molluscs and carapace formation in the
replaced by a global one of acidified precipita- crustaceans (Økland and Økland, 1986). Also crit-
tion. It is not, however, a straightforward case ical is the solubility and aluminium ions which
of a universal and systematic lowering of the are extremely toxic to organisms not equipped to
pH of rain: depending upon the provenance of deal with them (Brown and Sadler, 1989; Herr-
a particular airstream and its recent history of mann et al., 1993). Metals, including iron and
rainfall elution, precipitation is much more sub- manganese, may be activated to harmful levels
ject to what are sometimes termed ‘acid events’ through the disruption of dissolved humic com-
(Brodin, 1995). Since the mid-1980s, some ame- plexes (DHM) (see Section 3.5.4).
lioration has come through better understand- Aquatic organisms vary in their sensitivity
ing and recent international concord has secured to low pH and elevated concentrations of alu-
cuts in the emissions of strong acids to the atmo- minium and manganese. Examples of species
sphere. The expedient of burning natural gas in (or races) of fish and invertebrate that are tol-
preference to oil or coal also generates less sul- erant of high acidity levels have been docu-
phate per unit energy yield. Nevertheless, there mented (by e.g. Almer et al., 1978). However,
can be few locations on the Earth where rain- those close to the extremes of physiological tol-
fall does not, at times, continue to deposit abnor- erance may suffer catastrophic mortalities as
mally enhanced loads of acid. a consequence of a relatively small downward
In the present context, the concern lies in the drift in pH during a single acid event. Among
precipitation after it has reached the ground. It the macroinvertebrates, only insects seem tol-
is early on in its percolation into a well-formed erant of high acidity (Henrikson and Oscar-
soil and its throughflow to the surface drainage son, 1981). Investigations of known acid-tolerant
that the chemical composition of the drainage to species of phytoplankton have revealed a bio-
rivers and lakes is mainly determined. It is quite chemical mechanism in Chlorella pyrenoidosa,
apparent that the natural acidity of rain is nor- Scenedesmus quadricauda (Chlorophyta) and Euglena
mally quickly neutralised by the bases present mutabilis (Euglenophyta) for regulating internal
and which, generally, are derived from the weath- pH against high external H+ concentrations (Lane
ering of parent bedrocks. However, where rocks and Burris, 1981). However, the ability of E. muta-
are hard or slopes are steep, soil cover is thin bilis to deal with mobilised aluminium species
and bases are deficient, the acidity is not over- (Nakatsu and Hutchinson, 1988) may be the
come and it registers in the lake water. Acidity decisive specialist defence against extreme acid
in drainage may actually be enhanced by cer- enviromnents.
tain types of vegetation exchanging hydrogen Some algae, including desmids (Cosmarium,
ions for valuable nutrients. Thus, as already indi- Cylindrocystis) and eustigmatophytes (Chlorobotrys),
cated, almost all the reported instances of pro- are characteristically associated with acidic pools
gressive acidification rising to levels damaging in wet Sphagnum bog. Among the freshwater
phytoplankton, tolerance of natural high acid- carbonate/bicarbonate reserve. Another interest-

ity is shared among a number of other species ing and controversial alternative approach that
(see, for instance, Swale, 1968; Nixdorf et al., has been devised recently involves the deliber-
2001), many of which do happen to be flag- ate addition of phosphorus to the water (Davison,
ellated chlorophytes or euglenophytes. Species 1987). It was successfully applied to secure the
of Chlamydomonas, Sphaerella, Euglena and Lep- restoration of a series of highly acidified former
ocinclis are frequently encountered in small sand workings in eastern England (Davison et al.,
acid lakes, one or other usually dominating 1989) and, later, to the restoration of an acidified
a low-diversity population. Cryptophytes (Pla- upland tarn in the English Lake District (Davison
gioselmis, Cyathomonas), chrysophytes (Ochromonas, et al., 1995). Davison’s (1987) original stoichiomet-
Chromulina) and dinoflagellates (Gymnodinium, ric calculations led him to propose that phospho-
Peridinium umbonatum) are also recorded, as are rus is 47 times more effective in generating base
certain non-motile chlorococcalean species of than the molecular equivalent of calcium carbon-
Chlorella and Ankistrodesmus. Some species of the ate. In extreme cases, lime may still be needed to
diatom genera Eunotia and Nitzschia are also bring the system to a pH level at which phos-
known to be acid tolerant but most planktic phorus addition will promote the development
species of Cyanobacteria do not grow at pH levels of a biotic community and the consumption of
much below 6.0 (H+ ≥ 1 µmol L−1 ). acidic anions. In each of the restorations thus
Swale (1968) recorded occasional large pop- attempted, a single or pulsed addition of phos-
ulations of Lagerheimia genevensis and a 14- phate was sufficient to maintain an alternative
month period of dominance by Botryococcus brau- planktic community, often with new species of
nii in Oak Mere, England (contemporaneous pH algae and crustaceans, submerged macrophytes
4.7–4.9). In the extremely acid lakes left by lig- and fish, for long periods of time measurable in
nite mining in Lusatia, Germany (pH 2.3–2.9), units of hydraulic residence time.
Lessmann et al. (2000) noted the presence of This may not be a universally acceptable
Scourfieldia and a Nannochloris, and the eugleniod approach to managing lakes, especially among
E. mutabilis. In the highly acidic caldera Lago managers who have craved the virtues of phos-
Caviahue, Patagonia, Argentina (pH 2.5, i.e. acid- phorus reduction as a route to high water qual-
ity ≥4 mmol H+ L−1 ), the phytoplankton was ity (Reynolds, 1992a). Their perplexity is easy
found by Pedrozo et al. (2001) to be dominated to understand; phosphorus addition could eas-
by a single green alga, Keratococcus raphidioides. ily contribute to a worse problem than the one
Indeed, this alga, a subdominant Chlamydomonas to be overcome. On the other hand, overzealous
and a single bdelloid rotifer comprised the full pursuit of phosphorus limitation strategies has
list of recorded planktic species. Species selec- sometimes reduced the ability of base-poor lakes
tion in the lake may also be influenced by the to resist the effects of strong-acid precipitation
low nitrogen content of the water and its rela- and the loss of their fish populations. Careful
tively high concentrations of the metals Fe, Cr, Ni analysis of the respective loads and fates of hydro-
and Zn. gen and orthophosphate ions could lead to more
The application of restorative treatments to imaginative strategies for management offer-
lakes that have become artificially acidified is ing longer-lasting benefits to biological water
generally considered if it benefits commercially quality.
important fish species or viable fisheries. The
usual method is to apply lime to the water 8.3.8 Anthropogenic effects on the sea
or to relevant parts of the catchment (Henrik- Nutrient enrichment of the oceans, at least as
son and Brodin, 1995; Hindar et al., 1998), Lakes a by-product of human activities, has not been
can be assisted to move back towards neutral- widely considered as an issue on the scale that
ity and to acquire a fully functional, healthy, it has become among inland waters. This does
pelagic ecosystem. Acidity is consumed and pro- not mean that anthropogenic eutrophication of
gressively more of the carbon dioxide enters the the sea is necessarily a lesser problem, nor that
we should not question a popular notion of rais- Away from the shelves and distant from ter-
ing the fertility of the ocean in order to enhance restrial influences, the nutrient content of the
its role as a sink for atmospheric carbon diox- upper oceans is plainly subject to depletion
ide. However, both are probably overshadowed through uptake by autotrophs but the variations
by problems arising from gross anthropogenic remain within levels representing approximate
distortion of the structure and geometry of the contemporary maxima (∼2 µM P, 20–40 µM N
marine food webs. Let us examine these issues in and ∼160 µM Si (see Sections 4.3.1, 4.4.1 and 4.7).
sequence. Actual concentrations in surface waters are gen-
erally rather lower than these levels, with, typi-
cally, the residual concentrations suggesting that
Eutrophication of the seas available nitrogen is more vulnerable to exhaus-
Given the vastness of the oceans, compared both tion by autotrophic growth than phosphorus.
to the tiny total area of inland standing waters However, as acknowledged and verified by exper-
and to the scale of human impacts on global imentation, the capacity-regulating element in
nutrient cycles, it is probably not at all surprising the upper ocean is iron, where maximum lev-
that oceanic systems remain resolutely ultraolig- els may be one or two orders of magnitude less
otrophic and support such low levels of biomass. concentrated than 10−3 µM (Martin et al., 1994;
On the other hand, the modest levels of biomass Section 4.5.2).
and production that they do support are not Thus, the open oceans remain steadfastly olig-
controlled primarily by the fluxes of nitrogen otrophic in potential and in their performance.
and phosphorus but, more probably, by the sup- On the principle that, at low levels of limiting
ply of bioavailable iron or of inorganic carbon nutrient availability, any increase in supply will
and, for long periods in most latitudes, by the raise the carrying capacity by a corresponding
hydrodynamics of vertical mixing and the depri- margin, then the pelagic biomass is most sen-
vation of light. It is in the relative shallows of sitive to a change in iron availability. Without
the continental shelves and at the coastal inter- this, indeed, enrichment with other nutrients
faces with the land masses, where the iron, car- will not stimulate classic eutrophication effects.
bon and light-dilution constraints are simulta- One corollary is that autotrophic production and
neously assuaged, that the enriching effects of the biomass-carrying capacity of the ocean (and
terrestrial sources of nitrogen and phosphorus thus its role as a carbon sink) can be stimulated
are already well known. For instance, in their by seeding the ocean with bioavailable iron (see
review of transport processes and fates of ter- pp. 428--31 below).
restrial phosphorus, Howarth et al. (1995) esti-
mated that current inputs to the sea have tre- Anthropogenic impacts upon marine
bled since the pre-agricultural period. There are communities and food webs
likely to have been significant increases in inputs At present, the greater anthropogenic impacts to
of inorganic nitrogen (chiefly as nitrate) during the health and functionality of marine ecosys-
the twentieth century but, as concentrations in tems come from a series of devastating assault
shelf waters and oceanic surface waters typically on the structure of marine ecosystems. These
continue to be depleted to <2 µmol N L−1 (<30 include destructive alterations to coastal habi-
µg N L−1 ), increasing biomass capacity remains tats, pollution and increasingly industrialised
in the control of nitrogen availability. Interest- methods of protein harvesting from the sea.
ingly, recent increases in planktic biomass in the Mostly, they are relatively local in concept and
Baltic Sea have been among the nitrogen-fixing in execution but, cumulatively, their impacts are
Cyanobacteria (notably Anabaena lemmermannii, alarming. In a thoughtful review, Jackson and
Aphanizomenon flos-aquae and Nodularia spumige- Sala (2001) summarised the effects of widespread
nea) in response to progressive increases in phos- coastal engineering, pollution, turbidity and fish-
phorus loads from adjacent land masses (Kuosa ing activities in damaging coral-reef habitats, the
et al., 1997). truncation of temperate kelp forests and their
replacement by ‘sea urchin barrens’ of crustose thus, the lack of an economic incentive to bet-
algae. In Europe, competing demands for the ter manage it in a way that does not lead to its
exploitation of the land surface in valley flood overexploitation. The third is that the industrial
plains are driving the winning of sands, grav- momentum is in advance of the ecological knowl-
els and other building aggregates into increas- edge to set sustainable levels of exploitation
ing intensities of offshore coastal dredging. It is on the underpinning relationships of biomass,
inevitable that the process removes entire ben- production and energy flow through pelagic
thic food webs and generates clouds of turbid- ecosystems.
ity in its wake. These may be short-lived impacts It is not even clear that, were such knowl-
while there are adjacent ‘islands’ whence biota edge available, informed guidance would be fol-
can re-establish but it is not clear how well cur- lowed. Long traditions in the study of growth
rent practices allow this. and recruitment in fish populations (Beverton
At another level, Jackson and Sala (2001) cat- and Holt, 1957; Ricker, 1958; Cushing, 1971, 1988:
alogued the diminution in large animals that Cushing and Horwood, 1977) and their match or
were once ‘keystone species’, linking structure otherwise to environmental conditions (Cushing,
to energy flow (see Section 8.2.4) and maintain- 1982, 1990, 1995) have not been able to avert
ing the pristine coastal ecosystems of which well-documented stock-recruitment collapses of
they were once part. Loss of these animals the herring (Clupea harengus) in the North Sea or
has contributed to significant habitat changes. of the northern cod in the north-west Atlantic,
Declines in grazing and foraging by manatees around the Labrador Grand Banks (see Cushing,
(Trichechus spp.) and dugongs (Dugong dugon) have 1996). The contested imposition of quotas now
allowed former sea-grass meadows to develop looks like actions that were ‘too little, too late’
into dense stands. The top layer of predators in as the symptoms of collapse show very early.
coral reefs (tiger sharks (Galeocerdo cuvievi), monk The ultimate truth of the maxim that ‘fisheries
seals (Monachus spp.)) has been reduced with cas- that are unlimited become unprofitable’ (Gra-
cading effects on coral consumers. Destruction of ham, 1943) is upheld.
cod (Gadus morhua) stocks and sea otters (Enhydra In recent decades, the issues have only intensi-
lutris) has released from control the sea urchins fied. Fishing fleets now scour continental shelves.
that now interfere with kelp regrowth. Trawl nets are built to be dragged along the sea
Some of these changes are the direct effects of floor where they sweep up all in their path for the
human overexploitation of the animals in ques- prize of a few more demersal fish. The imposition
tion, either for food or fur. Others are the indirect quotas for size and species catches can only be
consequences of the diminution of the exploita- controlled after sorting, when most of the ‘illegal
tion, either through the vacation of a foraging excess’ is returned to the water dead. Indeed the
niche to competitors, or to the relaxation of crop- catching methods are still so coarse that fishing
ping of lower trophic levels. It is perhaps 4000 can scarcely be directed to catch preferentially
years (4 ka) ago that Homo sapiens entered the the relatively plentiful fish. Conventional ecolog-
marine food web by catching predatory fish and ical theory suggests that, where interventions are
mammals in their natural environments (Cush- targetted successfully at a particular dominant
ing, 1996). However, it is mainly within the last species, then another with similar environmen-
century and, especially, the last 30 years that tal requirements and dietary preferences and the
industrial scales of sea fishing, backed by such next highest exergy potential (i.e. the next best
technological sophistications as sonar and satel- competitor) should be poised to assume that role
lite tracking that the hunter–gatherer approach (Wardle et al., 1999; Elliott et al., 2001b) (expla-
to foraging has descended to plunder and loot- nation in Section 7.3.4). This is roughly what
ing. The trouble is threefold. One is the economic happened in the North Sea: following the severe
drive for a return (through the sale of catches) reduction in herring, there was an ‘outburst’ of
on the investment in ships and crews. The sec- several gadoid species, including cod and had-
ond is the lack of ownership of the resource and, dock (Megalogrammus aeglifinus). The reasons are
complicated, however, and include the possibility rent trends in global warming has been referred
of reduced herring predation on cod larvae and to in Section 3.5.2. Burning coal, gas and oil
the greater availability of Calanus larvae to young is generating a flux of carbon dioxide from
cod, now relatively freed from consumption by the Earth to the atmosphere equivalent to 7.1
herring. One of the problems of the largely uns- (±0.5) Pg a−1 . Despite the Kyoto Accord, this
elective trawling of those same cod stocks is that is likely to go on increasing until demand
it is unlikely to leave many demersal contenders or depletion of the relatively readily accessi-
to fulfil its role. Thus, a further trophic linkage ble parts of the global reserve (estimated to be
is compromised. ∼5000 Pg) raise the cost sufficiently to make
Following similar logic, the industrial fish- it competitive to exploit other energy sources.
ing of species with other commercial value (for Apart from its contribution to acidity, carbon
instance, for the purpose of manufacturing fish- dioxide is one of the atmospheric gases (with
meal and the extraction of fish oils and pro- water vapour and ozone) that absorb much of
teins) may be depleting shelf waters of their diver- the long-wave terrestrial radiation back-reflected
sity and of the ability to support other mar- from the Earth (chiefly in the wavebands 2–8 and
itime species (birds, seals, whales). Of yet fur- >14 µm). Thus, its accumulation in the atmo-
ther concern should be the ‘creaming off’ of the sphere might enhance the so-called ‘greenhouse
higher levels of the oceanic food chain of the olig- effect’ and contribute to global warming. The
otrophic oceans; catches of such species as tuna extent of effects directly attributable to the accel-
(Thunnus), barracuda (Sphyraena), sea-bass (Morone) erated oxidation of organic carbon is still not
and others are commercially very attractive but clear. On the other hand, such a scale of inter-
probably unsustainable at present rates. vention into the natural planetary cycling of car-
The science and the economics of overex- bon is unlikely to avoid significant consequences
ploitation of fisheries are not the principal con- on average planetary temperatures, climatic pat-
cern of this book, though it does recognise that terns and sea levels.
urgent action is required if the excellent source Many insist that the changes have already
of healthy protein and oils is to remain available. commenced and there is a growing, general (but
It is clear that farming of commercially impor- not universal) political consensus that ‘some-
tant fish species (as opposed hunting and gather- thing must be done’. Simply cutting back on the
ing them) is no more sustainable while it requires consumption of fossil fuels would be an obvious
the continued fishing of other species to provide step but cheap and available energy is a drug,
feed. Nothing less than the total exclusion of all addiction to which is economically too painful
fishing fleets from large stock-recruiting areas is to abandon. There are problems of political equi-
going to assist the survival of viable and function- tability over the consequences of therapies and
ally intact food webs. The consequence of doing over the cost-sharing of the alternative energy
nothing is likely to be a fairly rapid slide towards sources.
a pelagic ecosystem that effectively comprises lit- Another idea, seriously advanced as a means
tle more than phytoplankton, mesozooplankton of stabilising atmospheric carbon dioxide lev-
and macroplankton. els, is that we could augment the flux of car-
bon dioxide from atmosphere to ocean. The sug-
gested mechanism is to raise the fertility of the
Anthropogenic countermeasures to the sea, so that more carbon might be fixed in pho-
atmospheric accumulation of greenhouse tosynthesis into more biomass, and with more
gases being consigned as export to the deep ocean. By
The irrefutable evidence for the increase in implication, the key players invoked are members
atmospheric carbon dioxide during the indus- of the marine phytoplankton, so the idea must
trial period (caused by the oxidation of fos- command the attention of readers of this book,
sil fuels and of humic matter in agricultur- requiring a careful consideration of its logic and
ally improved land) and its likely role in cur- the likely consequences of its implementation.
Starting with what we know, the global stores supposed to increase the capacity for its storage
and fluxes between atmosphere and ocean are in the sea?’
already impressive. Using the tabulated data of One part of the answer is another question –
Margalef (1997), the atmospheric store (presently ‘If so little does so much, then how much more
around about 650 Pg C) is currently increasing by could a little more do?’ Put another way, ‘Will
about 3.3 Pg a−1 . Thus, the net burden of anthro- a bigger primary producer biomass not increase
pogenic carbon emissions to the atmosphere is the flux of carbon dioxide to the ocean, per-
currently countered by an annual removal of car- haps to the point of balancing the anthropogenic
bon dioxide of some 3.8 (±0.5) Pg C a−1 . These excesses?’ How, indeed, can we be sure that
are not trivial amounts but, on the scale of the adding nutrients does not simply accelerate the
natural biogenic fluxes, they are comparable with cycle, with the wheel spinning faster rather than
the error terms of primary production (Table 3.3). increasing in mass? This must be a serious pos-
Terrestrial net primary production (56 Pg C a−1 ) sibility if we follow the deduction that the net
maintains a biomass (including wood) of about yield of exportable primary product across most
800 Pg C and a store of other organic necromass of the open ocean is tolerably close to the likely
of about twice that (1600 Pg C). These are not invasion rate of carbon dioxide across the sur-
static amounts but present rates of forest clear- face of the sea (Sections 3.4.1, 8.2.2). A rising
ance and land drainage probably impede any cur- flux of carbon dioxide, driven by an increasing
rent net increment to the storage capacity. By partial pressure in the atmosphere, might raise
deduction, much of the missing 3.8 Pg C a−1 is the sedimentary flux by the same few percent-
already dissolving in the sea, assisted by increas- age points but, mainly dependent upon microbial
ing partial pressure. In any case, the quantity processing, enhanced exports depend as much
of carbon dioxide in solution in the sea (some upon faster turnover through the web. Then, rais-
40 000 Pg) makes it by far the largest store of car- ing the fertility of the medium might lift the
bon in the biosphere. This does not mean that the supportable biomass at successive trophic levels,
additional input to the dissolved carbon dioxide including the larger consumer categories that
pool is necessarily insignificant. We should bear are likely to provide the desired increases in car-
in mind that the additional gaseous load is to bon exports. Moreover, the deep oceans are capa-
the immediate surface layer, where its involve- ble of storing a lot more carbon dioxide in solu-
ment in the carbon dioxide–bicarbonate system tion without the net ‘outgassing’ evident among
(see Section 3.4.1 and Fig. 3.17) and a net depres- small lakes (see Section 8.2.3). What is required
sive impact upon pH will make it increasingly is the minimisation of C cycling in the micro-
difficult for carbonate-deposting phytoplankters bial loop of the surface waters and the numerous
(such as coccolithophorids) and animals to form opportunities it offers for the venting of carbon
their calcareous shells. dioxide as a primary metabolite and, instead, to
The desired fate of the additional carbon diox- direct fixed carbon into exportable, sedimentary
ide flux is that it should be taken up by enhanced biomass
phytoplankton photosynthesis. Currently, net pri- This is easily stated but it is a formidable hur-
mary production in the sea (48 Pg C a−1 ), impres- dle to overcome. Let us recall the viscous envi-
sively generated by a producer mass equivalent to ronment of oceanic primary producers (Section
≤0.7 Pg C, supports an average oceanic biomass of 2.2.1), in which viscosity is a more significant
∼10 Pg C and which is, self-evidently from these force than gravity, and vertical transport is rel-
figures, turned over very rapidly. A large propor- evant primarily in the context of exposure to
tion of the organic carbon fixed in phytoplankton underwater irradiance (Section 3.3.3). The supply
photosynthesis is respired back to carbon dioxide of nutrients (some albeit very scarce) is medi-
without leaving the euphotic zone. Our first ques- ated primarily by molecular diffusion (Section
tion might be: ‘How is such a small part of the 4.2.1). Though species composition may vary, the
global biomass, itself already involved in 40% of structure of the food web and the relative pro-
the short-term movements of biospheric carbon, portionality of mass among its components (the
producers, heterotrophs and microphagotrophs) solution of the siliceous walls) before they have
are, within certain limits, highly conserved (a settled more than a few tens to hundreds of
point elegantly emphasised by Smetacek, 2002). metres through the ocean (Reynolds, 1986a). Of
The effect of pulsed enrichment by an other- course, the aggregation of dying cells into larger
wise limiting resource should benefit the growth flocs (up to and including ‘marine snow’ (All-
rates of all microbes alike. However, photoau- dredge and Silver, 1988; see also Section 2.5.4)
totrophs are uniquely able to invest resource may accelerate the rates of sinking and slow the
in new biomass whilst simultaneously cutting rates of decay (Smayda, 1970). In this way, the net
the overspill of carbon to the heterotrophs. export of diatom and other POC in the form of
At the same time, bacterial populations suffer ‘phytodetritus’ is enhanced (Legendre and Rivki,
the ongoing constraints of protist grazing and 2002a).
viral pathogens. Thus, for a variety of reasons, While oceanic production is concentrated in
the structure of the fertilised microbial web the picoplanktic prokaryotes and is intimately
is distorted in favour of pico- and nanoplank- coupled to the microbial web, it is easy to under-
tic producers and disproportionately against the stand why diatom-based export of unoxidised
smooth flow of carbon through to an exportable POC beyond the upper ocean is abnormal and
flux of mesoplanktic consumers or food to fish. event-led (Karl, 2002; Karl et al., 2002). In this con-
More and more sustainable phytoplankton is not text, the phytoplankton biomass of the coastal
at all an ideal carbon sink while it largely stim- and shelf waters, which is responsible for about
ulates only producer biomass. Above unspeci- a quarter of the net primary production of the
fied thresholds (though probably in the range sea (Table 3.3), is more continuously and more
0.01–0.1 mg C L−1 ), the concentration of phy- conspicuously populated by diatoms, dinoflagel-
toplankton carbon (together with other detrital lates and coccolithophorids. Globally, coastal and
sources of POC) may permit the short-circuiting shelf systems are already likely to be playing a
of the microbial loop and the imposition of direct significant part in the export of biogenic carbon
mesoplanktic herbivory on algae. Above 0.1 mg C out of the surface circulation (Bienfang, 1992).
L−1 , this is still more likely to be true, as the Moreover, parts of these same shelves are already
threshold concentrations of many types of plank- sites for the accumulation of inwashed terres-
tic filter-feeder are saturated (Section 6.4.2). trial detritus and POC, and where the concern
Thus, the critical cue (singularity: Legendre over nutrient limitation of carrying capacity has
and LeFevre, 1989; see also Legendre and Ras- been more to prevent its increase rather than to
soulzadegan, 1996) to enhanced carbon export encourage it.
from surface waters is to promote sustainable To recap, the uncertainties about the likely
pelagic structures producing either readily graze- effects of oceanic fertilisation on the atmospheric
able levels of nanoplankton (and the attendant carbon dioxide content are too many and too
flux of faecal pellets) or predominantly sedimen- great to advocate its deployment as a strategy.
tary microplankton. In the contemporary ocean, The impressive results of the IRONEX and SOIREE
large-celled diatoms are usually the main agents fertilisations of the iron-deficient Pacific (Martin
of the sedimentary export flux of organic car- et al., 1994; Bowie et al., 2001; see also Section
bon (Falkowski, 2002), especially at times of accel- 4.5.2) might offer a persuasive case for emula-
erated production, beyond the control of the tion on the basin scale. However, supplementing
microbial food web (Legendre and Rassoulzade- the iron supply in a low-latitude ocean would
gan, 1996). only raise the ceiling on producer mass to the
This behaviour is generally attributable to the scale of the next capacity limitation (nitrogen)
relatively high sinking rates of diatoms (Smayda, and not at all where iron is not already limit-
1970; see also Section 2.5). However, the fastest ing growth. Under these circumstances, the most
sinking rates of settling individual cells are prob- likely consequence of iron enrichment would
ably inadequate to prevent the death and decom- be, presumably, the promotion of nitrogen-fixing
position of the protoplast (and even the re- prokaryotes, with little benefit either to the
carbon-exporting elements of the pelagic food come. Even the extrapolations on which the cur-
web or to the growth of diatoms. According to rent fears about climate change are based may
Legendre and Rivkin (2002b), the frequency, quan- be quite wrong, with the response of global tem-
tity and composition of nutrient additions would peratures to the rise in greenhouse gases being
greatly bias the structure of the planktic com- either overestimated or frighteningly underesti-
munity. Frequent additions of nutrients offering mated in its effects on human civilisations. The
high ratios of Si to both N and to P are necessary latter might yet persuade us to adopt appropriate
to promote diatom dominance. For the produc- countermeasures. However, doing nothing but
tion of blooms to sink to depth ungrazed, any allow the planet to adjust its carbon fluxes and
matching of nutrient additions to grazer demand stores might be a prudent option, even though it
must be avoided. It is clear that there is no simple probably carries enormous social and economic
strategy for enriching tropical seas that does not discomforts to humankind.
risk chaotic or undesirable consequences. Out- The contemporary view of the role of the
side the tropics, the most likely capacity con- oceanic biota in the global carbon cycles has
straint upon phytoplankton is an insufficiency of been elegantly encapsulated in Falkowski’s (2002)
light, which no amount of fertilisation will over- overview. The diatoms are the major exporters
come. Shallow coastal waters offer the only real of organic carbon (with silica) to the sediments;
prospect for accepting more carbon into biomass the coccolithophorids are major exporters of
carbon and then transporting it to depth. It is calcite (calcium carbonate). This arrangement
not clear that such waters are strongly nutrient has persisted for over 100 Ma, really since tec-
deficient. However, the grave distortions to the tonic upheavals and continental drift effectively
food webs of the continental shelves, mentioned opened up new niches for exploitation (see also
in the previous section, make the impacts of fur- Section 1.3). Throughout that period, there has
ther fertilisation on the structure and function been a progressive reduction in the atmospheric
of their surviving components still more difficult carbon dioxide content (from about 700 to 280
to predict. p.p.m.). This may be due largely to the rate of
A better candidate for taking forward the subduction of carbonate-rich oceanic sediments
principle of enhanced carbon dioxide sequestra- exceeding the rate of volcanic and orogenic car-
tion by marine primary producers is surely fur- bon dioxide generation, throughout the Cenozoic
nished by kelps and other large seaweeds. These period. The planet has cooled sufficiently for the
at least have the merit of accumulating (rather ice caps to form, for the sea level to fall, exposing
than simply metabolising) the carbon they absorb areas of continental shelf, and for the planet to
and, moreover, in a form that is conveniently become drier.
harvestable, compostable and combustible as bio- Following the establishment of polar ice,
fuel. Seaweed growth is naturally even more con- the Milankovitch cycle of climatic oscillations,
fined to unpolluted shallow shelf waters than is caused by rhythmic orbital variations of the
phytoplankton. However, if the provision of suit- Earth around the Sun (with a frequency of 90–120
able artificial substrata could be devised, such as ka) and resultant variation in solar radiative
floating mats on which the seaweeds could estab- forcing, has been marked by significant glacia-
lish, then the potential of the marine system to tions. The last four, at least, saw the expan-
remove of more carbon dioxide from the atmo- sion of the Arctic ice cover over large parts of
sphere might begin to be achievable. the northern hemisphere. The interactive role
of the biotic oceanic carbon cycle with climatic
Phytoplankton and the future variations through the last 400–500 ka are now
The balance of wisdom should be against delib- well understood, through the combination of
erate manipulation of the marine phytoplank- radiocarbon dating and stable-isotope analysis of
ton to mop up anthropogenic carbon dioxide. biogenic deposits with atmospheric fingerprints
Present knowledge provides an inadequate basis of contemporaneous layers in ice cores. With
for reliable and precise anticipation of the out- the onset of a glacial period, terrestrial plant
production decreases and significant amounts overtake the scale of Cenozoic fluctuations and
of carbon are transferred from the land to the match the temperature rises of the late Permian,
sea. The process is reversed during interglacials. which brought life on earth close to extinction
Sea levels fall as ice volume increases, exposing (Section 1.3).
coastal lowlands to oxidation and erosion and the Neither prospect – being baked or suffering
release of nitrogen and other nutritive elements violent climate change, lowland inundation and
to marine ecosystems. Increased nutrient delivery massive economic disruption – is at all comfort-
to inshore waters stimulates production of large- ing to humankind. However, the prospects for the
celled phytoplankton and the export of more car- microbial engineers of atmospheric and oceanic
bon to depth. Drier climates also allow more composition, including the phytoplankton, are
wind-blown dust and the aeolian fluxes of min- mostly good. Their ability to go on regulating the
erals, including iron, to the sea are raised. These planet, while other species have come through
changes in the nutrient availability drive accel- and caused desolation, seems to be assured.
erated carbon dioxide exchanges, with increased
leakage back to the atmosphere. Falkowski (2002)
analogised these exchanges to the ocean ‘breath- 8.4 Summary
ing’ in and out on a 100-ka cycle, inhaling carbon
dioxide from terrestrial systems during glacial The chapter seeks to evaluate the importance
periods and exhaling it during the interglacials. of phytoplankton to pelagic function and to
Because these exchanges feed back positively on the biogeochemical role of pelagic domains to
the Milankovich-driven variations, it is reason- the behaviour of various categories of aquatic
able to attribute to biological processes a dom- ecosystem. It moves on to examine the responses
inant contribution in the contemporary carbon of pelagic systems to changes, especially those
cycle. wrought through deliberate and unthinking
Just as certainly, they will also impose a bal- anthropogenic activities.
anced distribution of the contemporary anthro- In the open water of the sea and of the largest
pogenic contribution to carbon dioxide ‘inhala- lakes (those >500 km2 in area), phytoplankton
tion’ by the sea. This being a warm (and warming) are the only photoautotrophic primary produc-
phase, the rate of inhalation may be less rapid ers, upon which everything else in the water is
than during a glacial phase. On the other hand, dependent for its nutrition (energy as fixed car-
the flooding of coastal plains rich in nutrients bon and the other elements of biomass). This
may accelerate carbon withdrawal into diatoms long-standing view is correct but the secondary
and other ‘large’ phytoplankters. Moreover, the processing of primary product depends less on
rate of exploitation of the finite and increas- the grazing of phytoplankton, as once supposed,
ingly inaccessible remnants of the fossil-fuel car- so much as on the microbial uptake of fixed car-
bon is, in any case, bound to diminish, so the bon released into the medium by the producers.
fluxes from atmosphere to ocean will, inevitably, Severe poverty of nutrient resources controls the
reduce. Within the context of the last 100 Ma, the biomass at all trophic levels but carbon contin-
present carbon crisis would seem relatively trivial ues to be traded through the food-web compo-
and unlikely to disturb the pattern of prokaryote- nents. The bacteria → nanoflagellate → ciliate
mediated carbon cycling in the open ocean and → copepod pathway delivers efficiently a yield,
eukaryote-mediated exports from shelf waters. perhaps nearly 10% of the original primary prod-
The pre-industrial quasi-steady state might be uct. A further 10–20%, up to 100 mg C m−2 a−1 ,
re-established within 100–200 years and, if not, may be exported by sedimentation. The balance
before the onset of the next glacial phase. of the carbon budget is simply recycled through
It is prudent, nevertheless, to follow my own the pelagic network, which thus constitutes a
stricture and recognise the weak base for ‘reli- fairly closed cycle. The export is argued to be sim-
able and precise anticipation’. Nobody can be ilar in magnitude to the net annual invasion of
sure that the rate of global warming will not atmospheric carbon. The system is a quasi-steady
SUMMARY 433
state, a low, stable, average biomass being run on ter feed) are geared to the availability of their
the available flux of carbon. The activity is analo- nutritional requirements. The corollary is that
gised to a small wheel spinning rather fast (20 to the relatively most abundant forms will be those
70 revolutions per year, according to the criteria most capable of fulfilling their requirements.
used). This is another way of saying that species compo-
In the pelagic zones of smaller lakes, coastal sition (at least at the level of functional types)
seas, shelf waters and upwelling areas of the depends, in part, on the resource base. For
ocean, primary production benefits from addi- example, greater nutrient availability raises the
tional supplies of inorganic carbon, including autotrophic supportive capacity, enabling pho-
that dissolved in inflowing streams and rivers toautotrophic phytoplankton to retain more of
and the metabolic gases released through the their own photosynthate in their own biomass,
oxidation of sedimentary organic carbon. The and with a lesser proportion garnered by het-
delivery to the water-body margins of terres- erotrophic plankton. Greater nutrient availability
trial particulate organic carbon can provide a makes it easier for larger species of algae to flour-
direct input to littoral and sub-littoral consump- ish in the pelagic. Mesozooplanktic feeding on
tion. The interaction with the adjacent terrestrial absolutely larger nano- and microplanktic algal
ecosystems contributes to enhanced fixation of fractions short-circuit the microbial web; filter-
inorganic carbon (net yields of primary produc- feeding zooplankton are capable of its destruc-
tion range are frequently in the range 200–800 g tion. The optimal pelagic pathway changes to
C m−2 ) as well as providing an additional income microalgae → cladoceran.
of organic carbon to limnetic heterotrophs. The How the pelagic community processes its
shallow littoral margins of lakes are important in resources and directs the flow of energy is shown
other ways, not least because of their ability to to be strongly conditioned by the activities of the
support macrophytic plant production. Stands of main species (that is, how the community func-
aquatic plants accumulate organic debris (some tions is a consequence of what is there). Exam-
of it by filtration from the circulation), offer shel- ples are presented that show that the princi-
ter and microhabitats to invertebrates. Linked to pal components of the phytoplankton and zoo-
the pelagic through the horizontal movements of plankton, as well as the foraging activities of
animals, especially fish, macrophyte stands influ- fish, can be matched to the carbon concentration
ence the material and energy flow to a consider- (subsidised or otherwise) and the energy trans-
able distance into the lake. The additional car- ferred among trophic levels. Critical boundaries
bon flux represents a supportive subsidy with of ∼0.01 mmol C L−1 (∼0.12 mg C L−1 ) distin-
respect to the truly pelagic system. The smaller guish classically eutrophic lakes from classically
is the water body then the relatively greater is oligotrophic lakes. A similar abundance of car-
the potential subsidy to the overall function of bon separates those lakes that support Cyanobac-
the lake. More carbon, proportionately and abso- teria from those in which chrysophytes are abun-
lutely, may reside in the biomass of aquatic pro- dant; those whose zooplankton is dominated
ducers, phagotrophic consumers and other het- by calanoids from those in which cladocerans
erotrophs of smaller lakes and ponds, provided are abundant; and those supporting coregonids
always that it is within the stoichiometric capac- from those in which benthivorous fish dominate.
ity of the limiting bioavailable nutrient(s). It is also suggested that, in small to medium-
Resource availability and processing con- sized lakes where the carbon balance exceeds this
straints determine the relationship between threshold, a large, possibly dominant, part of the
energy flow and the structure (sensu the func- authochthonous organic carbon is transmitted to
tional groupings of the main participants). The higher trophic levels by way of the sub-littoral
abilities of phytoplankton to gather resources sediments, benthic invertebrates and browsing
(whether through diffuse demand, high uptake benthivorous fish.
affinities or scavenging propensity) and of zoo- The chapter also considers the suscepti-
plankton to forage (to encounter, to hunt, to fil- bility and dimensions of changes in pelagic
communities consequential upon anthropogenic ing of the shorelines are unwelcome symptoms of
changes. Eutrophication of lakes through enhan- eutrophication. Phytoplankton abundance preju-
ced phosphorus loading is considered in some dicial to water quality, to treatment for potability
detail, developing the underlying processes link- and to the quality of fishing is likely to inspire
ing catchment sources (‘loads’) of phospho- and justify the costs of reversal of eutrophica-
rus to in-lake concentrations and the mean tion (‘oligotrophication’), and the restoration of
algal biomass (as chlorophyll a concentration) past quality or rehabilitation of the water body
in the lake, as represented in the well-known to an acceptable ecological state. A series of suc-
Vollenweider–OECD regression. cessful schemes in which phosphorus-load reduc-
The relationship remains a powerful descriptions have led to reduced plankton biomass and
tor of the biomass capacity of the deep, high- better perceived water quality is highlighted. The
latitude lakes that dominated the original biological response is usually dramatic but it
dataset, but it is not a general management is never manifest before the soluble, MRP frac-
model. Mean phytoplankton abundance in low- tions, readily available to support further phy-
latitude lakes, in shallow lakes, especially those toplankton growth, are effectively drawn down
experiencing fast rates of hydraulic exchange, to the limits of detection. Phosphorus-reduction
and those in which phosphorus is manifestly not schemes that, so far, have been unsuccessful in
the capacity-regulating factor, is not well simu- invoking the desired biomass response have not
lated by the regression. Special problems aris- achieved, or have progressed too slowly towards,
ing from the abundance of particular types of the satisfaction of this criterion. These poor
phytoplankton are considered here. Even though responses are observed frequently among shallow
their dominance may have only tenuous connec- lakes, where the persistent recycling of phospho-
tions with nutrient eutrophication, their abun- rus from already enriched sediments goes on sus-
dance is potentially greater where there is signifi- taining phytoplankton independently of external
cant enrichment. Conspicuous among the organ- loads.
isms generating nuisance on aesthetic grounds Methods for anticipating the sensitivity of
are the bloom-forming Cyanobacteria. These have individual sites to altered external loadings (up
the added hazard of often being highly toxic. as well as down) are available. If control over
They are difficult to eradicate from lakes but, the internal loads (or adequate contol over the
with an understanding of their preferences and external loads) is impractical, other rehabilitative
weaknesses, some kinds of lake and reservoir can approaches are available. Provided the water col-
be managed in ways that keep these algae in umn is deep enough to make a difference, arti-
check. Cyanobacterial blooms are not confined to ficial circulation of water bodies (chiefly reser-
inland waters, having become common in recent voirs) to break or to prevent thermal stratifi-
years in the Baltic Sea. Elsewhere, harmful algal cation reduces the light-carrying capacity and
blooms in the sea comprise ‘brown tides’ of Aureo- rates of growth of phytoplankton. Subject to the
coccus, ‘green tides’ of Phaeocystis and, especially, same provision, the overriding imposition of defi-
‘red tides’ of toxic dinoflagellates. In all events, cient insolation pushes species composition in
local abundances may depend upon concentra- favour of functional types with acknowledged
tion by water movements but the incidence of light-harvesting specialisms (diatoms, filamen-
such events is increasing and are considered tous green, yellow-green and blue-green algae,
indicative of nutrient enrichment of the shelf of trait-separated groups P, T and, especially, S).
waters from which they are best known. Several examples are presented where mixing
There are few instances where eutrophication of deep reservoirs assures reasonable supplies of
is considered beneficial; mostly it is abhorred on water for treatment and within predictable lim-
aesthetic grounds of appearance, the detriment its of phytoplankton abundance.
usually being directly attributable to increased There is a good prospect of restoring or reha-
biomass of phytoplankton and littoral epiliths. bilitating those ‘small’ lakes (i.e. up to 10 km2
Greening of the water, loss of clarity and sully- in area) and, especially, ‘shallow’ lakes (in which
A LAST WORD 435
>50% of the area is <5 m in depth) through <2.5. The problem of metal (Al, Zn, Cu) toxicity
the biomanipulation of the ecosystem compo- is generally critical.
nents. Investment of carrying capacity is moved Apart from some estuaries, coastal waters
from phytoplankton to macrophyte and/or fish and part-landlocked shelf areas, the open seas
biomass. The necessary changes in the consump- have suffered less from anthropogenic nutrient
tion pressures may be imposed from the top enrichment than from severe food-web distor-
down by adjusting the abundance of fish pop- tions as a consequence of exploitative industrial
ulation (lower the density of planktivores, raise fisheries. Indeed, deliberate fertilisation of the
the density of piscivores; both supposedly reduce oceans, with a view to increasing carbon fluxes
the controls on zooplankton which, obligingly, from atmosphere to sea water, has been seriously
graze down the phytoplankton). Managers still suggested as a counter to the net accumulation
find themselves having to repeat treatments to of greenhouse gases from the oxidation of fos-
enforce the imposed, non-steady state. Left to sil fuels. Such action is argued to be misguided.
themselves, water bodies will often gravitate to While the carbon storage capacity of the ocean
a system characterised by high POC/dissolved is not exhausted, the net carbon flux is possibly
nutrient concentrations → Planktothrix agardhii as fast as it can be. If fertilisation could secure
→ chironomids → cyprinids. The task is to the rapid deep transport of POC (for instance,
induce the system into a valid alternative, macro- in diatoms or faecal pellets) balances might be
phyte → benthos → fish, steady state, in which altered a little. Possible outcomes are consid-
better habitat and water quality are more nearly ered. The consequences for human societies are
self-maintaining. Even here, there are approxi- uncomfortable to contemplate.
mate critical boundaries (in terms of phosphorus
rather than carbon). These may not be easy to
pass in the downward direction, usually requir- 8.5 A last word
ing quite drastic interventions into the numbers
of planktivorous and benthivorous fish. Below On the time and space scales of the biogeochem-
the critical boundaries, however, handbook guid- istry of the Earth, the current ‘carbon crisis’ is
ance extols the importance of macrophyte domi- modest, and more serious shifts in the planetary
nance to attractive ecosystems. Put simply, macro- distribution of carbon have occurred in the Meso-
phyte beds support macroinvertebrate popula- zoic. Indeed, the groups of phytoplankton that
tions adequate to interest many species of adult evolved during that era – diatoms, dinoflagellates
fish, and a sufficient refuge for cladocerans to and coccolithophorids – may have done so in
keep the water reasonably clear of phytoplank- response to contemporaneous ocean–atmosphere
ton. The boundary (‘switch’) in the upward direc- interactions, have persisted for 100 million years
tion is mediated by the algal biomass spon- and may still be best equipped to operate through
sored by bioavailable phosphorus concentrations the worst anthropogenically induced changes
exceeding 120–180 µg P L−1 , whether it is present that can be anticipated.
as phytoplankton or epiphytic growth. Further It is perhaps humbling that precise extrapola-
stability may be brought by stocking with pisci- tions cannot be made. Through the past century
vores such as pike. and a half, there have been several reasons for
Brief consideration is directed towards the studying phytoplankton, not least for the beauty
selective responses of phytoplankton to anthro- (that first attracted Haeckel), the desire to cata-
pogenic acidification. The impacts on some logue and name many thousands of species and
waters affected by acid precipitation are cata- the challenge of sorting out their phylogenetic
logued. However, in this instance, algal behaviour affinities and evolutionary development. These
in highly acidic waters filling some Lusatian min- persist today. The physiology of cell growth and
ing hollows and a natural, sulphate-rich caldera replication and collective impacts in the global
in Argentina illustrate the remarkable tolerances carbon and oxygen cycles represent extremes of
of a few green-algal species down to pH values of a scale covering a dozen orders of magnitude and
nearly as many biological disciplines. The ecology somes are providing ever greater dimensions of
of populations and communities is relevant to precision to the knowledge base. Yet it is quite
many aspects of human existence, from the safety evident that, far from knowing all there is to
of drinking water to the sustainability of fish- know about phytoplankton, relevant and polit-
eries. The accumulated knowledge is both broad ically sensitive questions persist, such as pre-
and deep but it is far from complete. Common dicting and understanding the impacts of global
tenets and basic understanding of the pelagic warming on the fundamental life-supporting sys-
ecosystem have undergone comprehensive revi- tems of the planet. The answers elude us. Much
sion more than once, the most recent occasion needs to be done; the scientific study of phyto-
having been only in the last two decades. Com- plankton will continue. Tomorrow’s leaders need
pletely new organisms and new organisations are an appreciation of the breadth and limits of exist-
being found, even now, in the deep oceans and ing knowledge of our home planet. It is my fer-
in the vicinity of hydrothermal vents. New tools, vent hope that this book may contribute to the
from satellite-based remote sensing to the anal- base of reference and the stimulus for the future
ysis of gene sequences from individual chromo- research.
Glossary
Text boxes are used to explain the usage of certain terms specific to plankton (Box 1.1) and their ecology
(Box 7.3). The meanings of some other less familiar terms used in the book are noted below.
abyssal pertaining to the abyss, or the very deep parts chemotrophy the capacity of organisms to obtain
of the ocean energy through chemical oxidation and using
aeolian of the wind, referring to the transport and either inorganic compounds as electron donors
deposition of dust particles from the land to lakes (chemolithotrophs) or preformed organic
and oceans compounds (chemoorganotrophs)
amictic of lakes that are scarcely wind mixed or chromophore intracellular organelle containing
which are weakly and incompletely mixed for very photosynthetic pigments
long periods coccoliths flattened, often delicately fenestrated
anoxygenic photosynthesis carbon fixation that takes scales, impregnated with calcium carbonate
place in the absence of oxygen and during which coenobium (plural coenobia) a group of monospecific
no oxygen is produced; photosynthesis that does cells forming a single, often distinctive unit
not split water to generate oxygen compensation point the point in a water column at
atelomictic of water columns, usually in low latitudes, which the rate of photosynthesis of an alga just
in which the depth of wind mixing varies balances its respiration; or the notional point when
conspicuously in extent but at frequencies the daily depth-integrated photosynthesis
of days to weeks rather than at diel or annual compensates the daily depth integral of respiration
cycles dimictic of lakes and water bodies, generally at mid
autotrophy the capacity of organisms to grow and to high latitudes, that are fully mixed during two
reproduce independently of an external supply of separate periods of the year (typically spring and
organic carbon; the ability to generate organic autumn)
carbon by reduction of inorganic sources using emergy the total amount of energy required to
light or chemical energy produce the usable energy of a final product. In the
auxospore sexually reproduced propagule of diatoms ecological context, a grown elephant contains
that also enables organismic size to be recovered biomass equivalent to a potential yield of direct
after a period of asexual replication energy but a lot more energy has been lost in
bathypelagic of the deep open water of the ocean foraging for food, as well as in its production
benthic of the solid surfaces at the physical bottom Emergy represents the sum of these contributions
of aquatic habitats; this can be very shallow as well endosymbiosis a symbiosis between two organisms in
as far beneath the water surface which one lives entirely within the body of the
bioassay a technique for identifying the supportive other but to the mutual benefit of both partners
capacity of water and for identifying nutritive epilimnion the upper part of a seasonally stratified
components that are deficient therein; if the lake; the part between the water surface and the
addition of a particular nutrient raises the response first seasonal pycnocline
relative to the unmodified control then that eukaryote a cellular organism having a
nutrient is deemed to have been limiting the full membrane-bound nucleus within which the
response chromosomes are carried
biogenic of the ability of living materials to create or euphotic zone that (upper) part of the water column
generate a particular substance; of substances thus wherein there is sufficient light to support net
generated by living organisms photosynthetic gain (often approximated as
cadavers the corpses and remains of dead organisms requiring 1% of surface light intensity but, in
carboxylation the essential step in photosynthesis reality, variable according to species and light
involving the combination of carbon dioxide, water history)
and ribulose biphosphate and yielding the initial exergy the extent of high-quality, short-wave energy
fixation product that incorporates the high-energy that an individual, population or community can
phosphate bond invest in the synthesis of biomass (see also Box 7.3)
438 GLOSSARY
form resistance the resistance to movement through mixis wind- or convection-driven integration of the
water effected by shape distortion with respect to a water masses of a lake; complete mixing is called
sphere of similar volume and density holomictic; frequent complete mixing is described as
frustule the siliceeous case of a diatom, comprising polymictic
two similar valves mixolimnion the upper part of a meromictic lake; the
haptonema distinctive additional appendage, said to part between the water surface and the perennial
be characteristic of the Haptophyta, carried pycnocline, that is actually or potentially mixed
anteriorally, between the flagella during the year
heterocysts (or heterocytes) specialised cells of mixotrophy the capacity of an autotroph to
Cyanobacteria (Nostocales) dedicated to the fixation supplement its carbon and/or nutrient requirement
of nitrogen (see Fig. 1.4h) from externally produced organic carbon
heterotrophy the means of nutrition of organisms compounds
that grow and reproduce that is dependent upon neritic pertaining to the shallow, inshore regions of
externally produced organic carbon compounds as the sea
a source of carbon and energy osmotrophy the capacity of certain microorganisms to
hypolimnion the lower part of a seasonally stratified absorb selected dissolved organic compounds across
lake; the part between the seasonal pycnocline and the cell surface
the lake floor oxidation chemical reaction involving the removal of
isopycny the condition of one entity having the same electrons and/or hydrogen atoms with, usually, a
density as another or as the medium in which it is release of energy
suspended phagotrophy case of heterotrophy in which organic
karyokinesis the apportionment of the mitotically carbon is ingested in the form of another organism
reproduced genetic material of a cell; nuclear or part thereof, requiring digestion prior to
division assimilation
kataglacial of processes or deposits associated with photoautotrophy autotrophy using light energy to
the melting of glaciers generate organic carbon by reduction of inorganic
kinase an enzyme that mediates the transfer of sources
phosphate groups from (e.g.) ATP to a specific phycobilin accessory biliprotein pigment of
substrate or target molecule Cyanobacteria, also found in Rhodophyta, some
laminar flow ordered fluid motion, molecules cryptophytes and glaucophytes
moving in layers, one upon another prokaryote cellular organism lacking
lorica hard part of the surface of certain flagellates, membrane-bound nucleus and organelles
especially euglenophytes and chrysophytes; from phytoplanktont or phytoplankter terms applied to an
the Latin word for breastplate individual organism of the phytoplankton
meromixis condition of (usually) deep tropical lakes in pseudotissue tissue-like structure of a fungus but
which mixing energy is inadequate to overcome comprising hyphae in close mutual application
stratification and which, therefore, remain pycnocline a gradient (usually vertical) separating
stratified for many years on end water masses of different densities. The density
meroplankton organisms that are planktic for a short difference may be due to a difference in
part of their life history, the rest of which they pass temperature or solute (e.g. salt) content
in the benthos or in the periphyton reduction chemical reaction involving the addition of
metalimnion the intermediate part of the water electrons and/or hydrogen atoms with, usually, an
column, separating epilimnion and hypolimnion investment of energy; the opposite of
and characterised by density gradients and much oxidation
weaker vertical diffusivity than either of the layers redox potential the balance of electrochemical
it separates potential between oxidative and reducing reactants
monomictic of lakes and water bodies, generally at reductant chemical substance that will yield
low to mid latitudes that are fully mixed during electrons and hydrogen ions
only one part of the year (typically between solar constant the energy flux from the Sun to the
autumn and spring) Earth, estimated as that reaching a notional surface
monimolimnion the lower part of a meromictic lake, held perpendicular to the solar rays and at a point
between the perennial pycnocline and the lake floor above Earth’s atmosphere, before there is any
GLOSSARY 439
reflection, absorption or consumption. It is not tychoplankton organisms that are not planktic
constant but an average, about 1.36 kW m−2 but which may, fortuitously, be introduced
thermocline a gradient (usually vertical) separating to the plankton from adjacent habitats. The
water masses of different temperature distinction from meroplankton is narrow but
thylakoid membrane supporting the light-harvesting the criterion is whether the suspended
complexes and sites of phase is essential to the life cycle or purely
photosynthesis incidental
Units, symbols and abbreviations
UNITS
Dimension Unit Notation Equivalent

mass kilogram kg
gram g 10−3 kg
gram molecule mol
length metre m
area square metre m2
volume cubic metre m3
litre L 10−3 m3
density kilogram per cubic metre kg m−3
time second s
day d 86 400 s
year a
velocity metres per second m s−1
acceleration metres per second per m s−2
second
force newton N kg m s−2
pressure pascal Pa N m−2
= kg m−1 s−2
viscosity poise P N m−2 s−1
= kg m−1 s−1
work joule J Nm
= kg m2 s−2
power watt W J s−1
absolute temperature kelvin K
◦
customary temperture degree Celsius C K + 273
Arrhenius temperature reciprocal Kelvin A 1000/K
electrical potential volt V
molecular weight dalton Da
concentration mass per volume, e.g. kg m−3
M mol L−1
Multiple units Symbols
Many units are managed numerically by the use A reciprocal kelvins × 1000
of prefixes, as in Pg (for petagrams), µm (for Da molecular weight, especially of large
micrometres) or nM (for nanomols L−1 ). The SI molecules
notations are used throughout: P = 1015 times Eq equivalent mass of a reactant ion or
the basic unit; T = 1012 ; G = 109 ; M = 106 ; k(or substance
K) = 103 ; m = 10−3 ; µ = 10−6 ; n = 10−9; p = J joule, the basic unit for expressing work
10−12 ; a = 10−15 . K kelvin, the basic unit for expressing
absolute temperature
UNITS, SYMBOLS AND ABBREVIATIONS 441
L litre, a customary unit of volume; 1 m3 = D demand for a resource

1000 L Dx coefficient of horizontal diffusivity
M molar, being a concentration of 1 gram (see Eq. 2.37)
molecule per litre of solvent (herein, E rate of energy dissipation, in m2 s−3
always water) Es equitability or eveness of species
N newton, the basic unit for expressing representation in a multi-species
force community
N nitrogen Fa solute flux to the vicinity of a cell
P poise, the basic unit for expressing (see Eq. 4.4)
viscosity FC flux of CO2 across the water surface,
P phosphorus in mol m−2 s−1 (see Eq. 3.18)
[P]i phosphorus concentration of the ith Fi filtration rate (sensu volume processed
inflow stream to a receiving water per unit time) of an individual
Pa pascal, the basic unit for expressing filter-feeding zooplankter
pressure F0 emitted chlorophyll fluorescence,
V volt, the basic unit for expressing electric when all light-harvesting centres are
potential open, compared to the fluorescence
a year following a subsequent saturating
d day flash (Fm )
d (as a prefix) indicating a very small Fm emitted chlorophyll fluorescence,
increment, as in: when all light-harvesting centres
dN population rise are closed by a saturating flash of
dt time increment light
dSe extrabiotic dissipative flux of solar Fv variable fluorescence (= Fm − F0 ),
energy G material removed by consumers from
dSi biotic dissipative energy flux a given volume and in a given time
e the base of natural logarithms period
g gram GC the gas exchange coefficient, or linear
ln natural logarithm (i.e. power to base e) migration rate m s−1 (see Eq. 3.18)
log standard logarithm (i.e. power to base 10) H the depth of water in a flow or lake
m metre H Shannon diversity in bits
pCO2 the partial pressure exerted by I irradiance, or the flux of visible light
atmospheric gaseous carbon dioxide (or PAR), or photon flux density, in
s second mol per unit area per unit time
I 0 instantaneous irradiance flux
penetrating the surface of the sea or a
Variables lake, usually expressed in µmol
photon m−2 s−1 or integrated over the
A a defined area of the Earth’s surface day as mol photon m−2 d−1
Ax the cross-sectional area of a river Ic notation for the intercept of
channel (in m2 ) photosynthetic rates (P) against (I)
B biomass derived in Fig. 3.3
Bmax biomass-supportive capacity of the Ik irradiance intensity that just saturates
available resources or, effectively, the photosynthetic rate
maximum biomass that can be Im residual irradiance intensity at the
supported by the available resources base of the contemporary mixed layer
BQ energy of penetrative convection IP=R irradiance intensity that sustains just
Co solute concentration, as in Fick’s enough photosynthesis to meet
equation (3.19). respirational demand, and which
442 UNITS, SYMBOLS AND ABBREVIATIONS
defines the water-column N population of particles or organisms

compensation point or their biomass per unit area or
Iz irradiance intensity at a given depth volume (as specified)
in the water column, z N0 initial population, N (i.e. at t = 0) and
I530 irradiance intensity at the subscripted subject to dynamic change
wavelength Nt population remaining after a period
I∗ instantaneous integral of fluctuating of time, t, has elapsed
irradiance intensities applying in a Nt/m specifically to settling particles, to
mixed layer show an alternative outcome at t, had
I∗∗ integral of fluctuating irradiance the settlement been subject to m
intensities applying in a mixed layer complete and instantaneous mixings
over the day NP communal photosynthetic rate, being
Jb buoyancy force in the surface mixed the product of the chlorophyll-specific
layer, owing to the difference in its photosynthetic rate (P) and the
density with that of the water population, as chlorophyll, present (N)
immediately beneath P individual, usually biomass-specific,
Jk the kinetic energy available to photosynthetic rate, expressed by
overcome buoyancy of the surface mass of carbon fixed or oxygen
mixed layer generated per unit biomass carbon or
Ki Steady-state concentration of the ith chlorophyll per unit time
nutrient resource P primary production
Kr half-saturation constant of Pg gross primary production
resource-limited growth rate (i.e. the Pn net primary production
concentration of resource at which Pmax maximum observed or extrapolated
the rate of cell replication, r , is half photosynthetic rate
the maximum when growth rate PQ photosynthetic quotient, as mol O2
uptake is nutrient-saturated) evolved per mol CO2 assimilated
KU half-saturation constant of resource Pe Péclet number, being the ratio of the
uptake (i.e. the concentration of momentum of a moving particle to
resource at which the rate of diffusive transport in the medium
uptake, VU , is half the maximum Qs short-wave solar radiation
when uptake is nutrient-saturated QT net radiative heat flux reaching the
(VU max ) planetary surface (units, W m−2 or J
L Horizontal downwind distance over m−2 d−1 )
a lake or other defined water Q ∗T that fraction of Q T that penetrates
body beyond the top millimeter or so of
LD ‘light divisions’; Talling’s (1957c) the water column
integral of light intensity in a water Qz net radiative heat flux reaching a
column capable of supporting net given point in the water column
photosynthesis: LD = [ln (I 0 max / 0.5 Ik ) R rate of respiration (or biomass
/ ln 2] maintenance)
LDH Talling’s (1957c) ‘light division hours’, Ra rate of community respiration
expressing the daily integral contributed by photoautotrophs
irradiance received by a water body Ra20 steady Ra at 20 ◦ C
that is deemed to support Rh rate of respiration of heterotrophs
phtosynthesis (see Eq. 3.20) Re Reynolds number, being the ratio
L(P) aggregate areal rate of phosphorus between the driving forces of water
loading to a lake (in mg m−2 of lake flow and the viscous forces resisting
area per year) them
Rib bulk Richardson number, being the hw height of a theoretically static water
ratio between the driving forces of layer
motion in a water body and the i,j identifiers in a multiple component
buoyant resistance to entrainment of list of variables
deep waters k exponential coefficient of heat
S supply or concentration of a resource absorption (see Eq. 2.27)
Sh Sherwood number, being the ratio k exponent of invasiveness of dispersing
between the total flux of a nutrient organisms (see Eq. 7.3)
solute arriving at the surface of a cell ka area-specific light interception of an
in motion and the wholly diffusive alga
flux kn rate of population change in a
U wind velocity horizontal patch (see Eq. 2.37)
V volume of a lake or other defined la length dimension available for the
body of water dissipation of the energy, usually the
VU rate of uptake of a given nutrient depth of the flowing water layer
VU max nutrient-saturated uptake rate of a le length dimension of the largest
given nutrient turbulent eddies
W Wedderburn number, being the ratio lm smallest eddy size supported by the
between the Richardson number of a available mechanical energy before
structure and its aspect ratio (see it is overwhelmed by viscous
Eq. 2.34) forces
Wb dry mass of a zooplankter m number of mixing events
Wc dry mass of an algal cell (Section 2.6.2)
a factor of increased diameter, used in m maximum cell dimension
Eq. (2.17) m maximum size of particle that is
a area term used in Fick’s equation available to a given filter-feeder
(3.19) n number of moles of a solute that will
b number of beads in a chain as a diffuse across an area
variable in sinking rate (see Eq. 2.18) n(t) number of exploitable niches in the
bi biomass of the ith of s species present location at a time t
in a community or sample ñ number of occupied niches when
c speed of light t=∞
cd coefficient of frictional drag nsp number of species to be found in a
d diameter of a spherical cell defined area, A, as influenced by
dc diameter of a cylindrical cell arrivals and extinction rates
ds diameter of a sphere of equal volume p wetted perimeter of a vertical
to an irregularly shaped particle section through a stream
g gravitational acceleration (here taken channel
as a constant 9.8081 m s−2 ) q cell-specific content of a given
h Planck’s constant, having the value nutrient (or quota) (see Eq. 4.12)
6.63 × 10−34 J s q0 minimum cell quota of a given
hc height of a cylidrical cell nutrient, below which it is deemed to
hm height of the mixed layer, from its be no longer viable
base to the water surface qmax replete cell content of a given
hp height of the photic layer, being from nutrient
the depth of IP=R to the water surface qi the inflow rate, or inflow volume per
hs height of the water layer from the unit time (usually) to a lake
surface to the depth of Secchi-disk qs discharge, or outflow volume per unit
extinction (zs ) time (usually) from a lake
r exponent of the rate of cell tq theoretical hydraulic retention time

replication, or rate of cell recruitment (V/qs )
through growth tw measured hydraulic retention time
r max resource-saturated rate of cell (V/qs )
replication u vector or flow velocity in the
r 20 rate of cell replication at temperature horizontal (x) plane
20 ◦ C ū mean horizontal velocity of turbulent
r θ rate of cell replication at the field flow
temperature θ ± u velocity variations in turbulent flow
r θ ,I rate of cell replication at the field in relation to ū
temperature θ and the natural light us horizontal advective velocity around
conditions patch
r (Si) rate of cell replication of diatoms u∗ turbulent velocity, the friction
calculated from uptake of soluble velocity or the shear velocity, as
reactive silicon from solution derived in Section 2.3.2 and Eq. (2.4)
rc critical radius of a patch able to (u∗ )2 turbulent intensity, as derived in
maintain distinctive pelagic Section 2.3.2
populations (see Eq. 2.37) û arrival rate of colonist species at a
rp roughness of the stream bed given habitat
(sensu the height of projections) v volume of a phytoplankton cell or
(see Eq. 2.6) colony
rn exponent of the rate of increase of a w vector or flow velocity in the vertical
self-replicating population net of (z) plane
simultaneous rates of loss ± w velocity variations in turbulent flow
rL sum of the exponents detracting from in the vertical direction
a self-replicating population wc sinking rate of chain
rG exponent of the rate of loss from a ws sinking rate of algal cell or colony
population of cells to consumers wP rate of loss of phosphorus in the
rS exponent of the rate of loss from a outflow from a lake
population of cells to settlement x concentration of organisms in a
rW exponent of the rate of loss from a sample, used in the calculation of
population of cells to wash-out Lloyd’s crowding index (see Sections
s surface area of a cell or colony 2.7.1 and 2.7.2)
sb gradient of the stream bed (in m m−1 ) x̄ mean of concentration, x, in a series
s2 statistical variance of samples
t a period of time or a fixed point in x∗ Lloyd’s crowding index, based upon
time the variance in x among individual
t a shorter period of time within t or samples. The greater the variance, the
an intermediate fixed point greater is the difference among
tc limiting closure period of LHC individual samples. In Sections 2.7.1
intercepting light (in s photon−1 ) and 2.7.2, the index is used to gauge
te time to achieve total elimination (or a the effect of wind upon small-scale
defined proportion) of particles from patchiness
suspension z used to identify points in the vertical
tm probabilistic time of travel of an direction; a specific depth
entrained particle through a mixed z(I k ) depth beneath the water surface at
layer which Ik is located; depth below
tp photoperiod or time spent in which photosynthetic rate becomes
euphotic zone light-dependent
z(0.5I k ) depth beneath the water surface at γ temperature coefficient of thermal

which photosynthetic rate is 50% that expansion, given by − (1/ρ w) (dρ w /
at light saturation dθ) K−1
zeu the depth of the conventionally δ to indicate ‘increment in’
defined euphotic layer ε exponent of vertical light attenuation
zm the depth to which the water is in water (units, m−1 )
mixed; the base of the mixed layer εa exponent of that part of the vertical
zs the Secchi-disk depth, being the light attenuation due to the presence
distance beneath the surface at which of planktic algae
a Secchi disk becomes obscured from εp exponent of that part of the vertical
the observer above the water surface light attenuation due to the presence
(see Box 4.1) of tripton
day length from sunrise to sunset εw exponent of that part of the vertical
to indicate difference, as in ρw (of light attenuation due to absorption by
density between top and bottom the water
water in a stratified water body) εz exponent of vertical light attenuation
N integral of simultaneous at a given depth
concentration gradients (δ/δx, δ/δy, ε440 , ε 530 , exponent of attenuation at the
δ/δz) in the x, y and z planes (nabla etc. particular wavelengths
operator) (see Section 4.2.1) ε min coefficient of attenuation in
P mean availability of phosphorus in a least-absorbed waveband
lake taking account of loading and εav coefficient of attenuation averaged
hydraulic exchanges over a number of wavebands
to indicate ‘sum of’, as in Fi, the w
εav is a weighted average attenuation
cumulative filtration rate of coefficient average (see Section 3.3.3)
filter-feeders in a unit volume of ε the energy of a single photon; ε
medium, and tp varies with the wavelength
(= hp / hm) , the daily sum of (see Eq. 3.2)
photoperiods η absolute viscosity of a fluid,
NP area-integrated photosynthetic rate, in kg m−1 s−1
expressed as the sum of NP ν kinematic viscosity of a fluid,
NP daily area-integrated photosynthetic in m2 s−1
rate θ temperature, usually in ◦ C
NR area-integrated respiration rate, λ wavelength (e.g. of light, in nm)
expressed as the sum of NR ξ gas solubility coefficient (in mol m−3
NR daily area-integrated respiration rate atmosphere−1 )
quotient relating sinking behaviour to π the ratio of the circumference of a
turbulent diffusivity, used as an index circle to its diameter
of entrainment (explained in Section ρ density, usually in kg m−3
2.6.1) ρa density of air
α coefficient of slope of a regression of ρc density of a cell
y on x, used herein mainly to describe ρm density of mucilage
light dependence of photosynthesis ρw density of water, in kg m−3
and growth (α = P/I) σ specific heat of water, in J kg−1 K−1
αr growth efficiency on low light σs succession rate index
incomes, derived from the slope of σP phosphorus sedimentation rate
the regression of r on I σX cross-sectional area of a
β slope of maximum replication rate on light-harvesting centre
Arrhenius temperature scale τ force or stress, in N
ϕ quantum yield of photosynthesis, in HNF heterotrophic nanoflagellates

mol C (mol photon)−1 IDH intermediate disturbance hypothesis
ϕr coefficient of form resistance LAI leaf area index
(dimensionless) LHC light-harvesting complex
ω coefficient of selectivity of ingestible MLD maximum linear dimension
particles from the inhalant current of MPF maturation-promoting factor (see
a filter-feeder Section 5.2.1)
0
integral sign, as in ∞ : integrate MRP molybdate-reactive phosphorus
from zero to infinity, or to a large NADP nicotinamide adenine dinucleotide
number phosphate
NAO North Atlantic Oscillation
NPP net primary production
Abbreviations used in the text NTA nitrilotriacetic acid
OECD Organisation for Economic
A Photon acceptor for photochemical Co-Operation and Development
system I PAR photosynthetically active radiation
ANN artificial neural network PCR polymerase chain reaction
APAR absorbed photosynthetically active POC particulate organic carbon
radiation POM particulate organic matter
ATP adenosine triphosphate POOZ permanently open-ocean zone (of
BAP biologically available phosphorus high-latitude waters)
BOD biochemical oxygen demand PP particulate phosphorus
CCM carbon-concentration mechanism PQ plastoquinone
CRP cAMP receptor proteins PSI photochemical system I
CV coefficient of interannual variation PSII photochemical system II
DAPI 4,6-diamidino-2-phenylindole (see Q A , QB quinones in PSII
Section 5.5.1) RNA ribonucleic acid
DCM deep chlorophyll maximum RUBISCO ribulose 1,5-biphosphate carboxylase
DCMU 3-(3,4-dichlorophenyl)-1, 1-dimethyl RuBP ribulose 1, 5-biphospate
urea SEH size-efficiency hypothesis
DEU dissipative ecological unit SIZ seasonally ice-covered zone (of
DHM dissolved humic matter high-latitude seas)
DIC dissolved inorganic carbon SRP soluble reactive phosphorus
DIN dissolved inorganic nitrogen SRSi soluble reactive silicon
(combined; does not include STPP sodium tripolyphosphate
dissolved nitrogen gas) TFe total iron (sensu all chemical forms
DMS dimethyl sulphide present)
DMSP dimethyl-sulphonioproponiate TN total nitrogen (sensu all chemical
DNA deoxyribonucleic acid forms present)
DOC dissolved organic carbon TP total phosphorus (sensu all chemical
DOFe dissolved organic iron (complexed forms present)
with DOC) a.s.l above sea level
DON dissolved organic nitrogen cAMP cyclic adenosine monophosphate
EDTA ethylene diamine tetra-acetic acid chla chlorophyll a
FDC frequency of dividing cells [chla] concentration of chlorophyll a
GPP gross primary production [chla]max maximum supportable concentration
G3P glycerate 3-phosphate of chlorophyll a
G3AP glyceraldehyde 3-phosphate ppGpp guanosine 3 ,5 -bipyrophosphate
HAB harmful algal bloom tris trishydroxymethyl-aminomethane
References
Abeliovich, A. and Shilo, M. (1972). Photo-oxidative Alldredge, A. L. and Silver, M. W. (1988).

death in blue-green algae. Journal of Bacteriology, Characteristics, dynamics and significance of
111, 682–9. marine snow. Progress in Oceanography, 20, 41–82.
Achterberg, E. P., Berg, C. M. G. van den, Boussemart, Allen, E. D. and Spence, D. H. N. (1981). The
M. and Davison, W. (1997). Speciation and cycling differential ability of aquatic plants to utilise the
of trace metals in Esthwaite Water, a productive inorganic carbon supply in fresh waters. New
English lake with seasonal deep-water anoxia. Phytologist, 87, 269–83.
Geochimica et Cosmochimica Acta, 61, 5233–53. Allison, E. H., Irvine, K., Thompson, A. B. and
Ackleson, S. G. (2003). Light in shallow waters: a brief Ngatunga, B. P. (1996). Diets and food
research review. Limnology and Oceanography, 48, consumption rates of pelagic fish in Lake Malawi,
323–8. Africa. Freshwater Biology, 35, 489–515.
Adams, M. H. (1966). Bacteriophages, 3rd edn. New York: Almer, B., Dickson, W., Ekström, C. and Hörnström, E.
Interscience. (1978). Sulfur pollution and the aquatic
Agawin, N. S. R., Duarte, C. M. and Agustı́, S. (2000). ecosystem. In Sulfur in the Environment, ed.
Nutrient and temperature control of the O. Nriagu, pp. 271–86. New York: John Wiley.
contribution of picoplankton to phytoplankton Alvarez Cobelas, M., Velasco, J. L., Rubio, A. and Brook,
biomass and production. Limnology and A. J. (1988). Phased cell division in a field
Oceanography, 45, 591–600. population of Staurastrum longiradiatum
Ahlgren, G. (1977). Growth of Oscillatoria agardhii in (Conjugatophyceae: Desmidaceae). Archiv für
chemostat culture. I. Nitrogen and phosphorus Hydrobiologie, 112, 1–20.
requirements. Oikos, 29, 209–24. Amezaga, E. de, Goldman, C. R. and Stull, E. A. (1973).
(1978). Growth of Oscillatoria agardhii in chemostat Primary productivity and rate of change of
culture. II. Dependence of growth constants on biomass of various species of phytoplankyton in
temperature. Mitteillungen der internationalen Castle Lake, California. Verhandlungen der
Vereinigung für theoretische und angewandte internationalen Vereinigung für theoretische und
Limnologie, 21, 88–102. angewandte Limnologie, 18, 1768–75.
(1985). Growth of Oscillatoria agardhii in chemostat Anderies, J. M. and Beisner, B. E. (2000). Fluctuating
culture. III. Simultaneous limitation of nitrogen environments and phytoplankton community
and phosphorus. British Phycological Journal, 20, structure: a stochastic model. American Naturalist,
249–61. 155, 556–9.
Aladin, N. V., Plotnikov, I. S. and Filippov, A. A. (1993). Anderson, D. M., Cembella, A. D. and Hallegraeff,
Alteration of the Aral Sea ecosystem by human G. M. (1998). Physiological Ecology of Harmful Algal
impact. Hydrobiological Journal, 29(2), 22–31. Blooms. Berlin: Springer-Verlag.
(English translation of original paper in Anderson, L. W. J. and Sweeney, B. M. (1978). Role of
Gidrobiologeskii Zhurnal.) inorganic ions in controlling sedimentation rate
Alcaraz, M., Paffenhofer, G.-A. and Strickler, J. R. (1980) of a marine centric diatom, Ditylum brightwelli.
Catching the algae: a first account of visual Journal of Phycology, 14, 204–14.
obeservations on filter-feeding calanoids. In Antia, N. J., Harrison, P. J. and Oliveira, L. (1991). The
Evolution and Ecology of Zooplankton Communities, role of dissolved organic nitrogen in
ed. W. C. Kerfoot, pp. 241–8. Hanover, NH: phytoplankton nutrition, cell biology and
University of New England Press. ecology. Phycologia, 30, 1–89.
Algéus, S. (1950). Further studies on the utilisation of Armbrust, E. V., Bowen, J. D., Olsen, R. J. and
aspartic acid, succinamide and asparagine. Chisholm, S. W. (1989). Effect of light on the cell
Physiologia Plantarum, 3, 370–5. cycle of a marine Synechococcus strain. Applied and
Allanson, B. R. and Hart, R. C. (1979). Limnology of the Environmental Microbiology, 55, 425–32.
P. K. le Roux Dam. Reports of the Rhodes University Arrigo, K. R., Robinson, D. H., Worthern, D. L. et al.
Institute for Freshwater Studies, 11(7), 1–3. (1999). Phytoplankton community structure and
448 REFERENCES
the drawdown of nutrients and CO2 in the Lakes, ed. W. Stumm, pp. 189–205. New York: John
Southern Ocean. Science, 283, 365–8. Wiley.
Arruda, J. A., Marzolf, G. R. and Faulk, R. T. (1983). The Badger, M. R. and Gallacher, A. (1987). Adaptation of
role of suspended sediments in the nutrition of photosynthetic CO2 and HCO3 : accumulation by
zooplankton in turbid reservoirs. Ecology, 64, the cyanobacterium Synechococcus PCC6301 to
1225–35. growth at different inorganic carbon
Arvola, L. and Kankaala, P. (1989). Winter and spring concentrations. Australian Journal of Plant
variability in phyto- and bacterioplankton in Physiology, 14, 189–210.
lakes with different water quality. Aqua Fennica, Badger, M. R., Kaplan, A. and Berry, J. A. (1980).
19, 29–39. Internal inorganic carbon pool of Chlamydomonas
Assaf, G., Gerard, R. and Gordon, A. L. (1971). Some reinhardtii: evidence for a carbon-dioxide
mechanisms of oceanic mixing revealed in aerial concentrating mechanism. Plant Physiology, 66,
photographs. Journal of Geophysical Research, 76, 407–13.
6550–72. Badger, M. R., Andrews, T. J., Whitney, S. M. et al.
Ashton, P. (1985). Seasonality in southern-hemisphere (1998). The diversity and co-evolution of Rubisco,
phytoplankton assemblages. Hydrobiologia, 125, plastids, pyrenoids and chloroplast-based
179–90. CO2 -concentrating mechanisms in algae. Canadian
Atkin, D. J. (1949). Seasonal fluctuations in planktonic Journal of Botany, 76, 1052–71.
distribution in a tropical impounding reservoir. Bailey-Watts, A. E. (1978). A nine-year study of the
In Proceedings of the Conference on Biology and Civil phytoplankton of the eutrophic and
Engineering, 21–23 September 1948, pp. 235–8. non-stratifying Loch Leven (Kinross, Scotland).
London: Institution of Civil Engineers. Journal of Ecology, 66, 741–71.
Atkinson, K. M. (1980). Experiments in dispersal of Bailey-Watts, A. E. and Duncan, P. (1981). The
phytoplankton by ducks. British Phycological phytoplankton. In The Ecology of Scotland’s Largest
Journal, 15, 49–58. Lochs, ed. P. S. Maitland, pp. 91–118. The Hague:
Atlas, R. M. and Bartha, R. (1993). Microbial Ecology, 3rd W. Junk.
edn. Redwood City, CA: Benjamin-Cummings. Balch, W. M., Kilpatrick, K. A. and Trees, C. C. (1996).
Aubrey, D., Moncheva, S., Demirov, E., Diaconu, V. and The 1991 coccolithophore bloom in the central
Dimitrov, A. (1999). Environmental changes in North Atlantic. I. Optical properties and factors
the western Black Sea related to anthropogenic affecting distribution. Limnology and Oceanography,
and natural conditions. Journal of Marine Systems, 41, 1669–83.
7, 411–25. Bannister, T. T. and Weidemann, A. D. (1984). The
Aubriot, L., Wagner, F. and Falkner, G. (2000). The maximum quantum yield of phytoplankton
phosphate uptake behaviour of phytoplankton photosynthesis. Journal of Plankton Research, 6,
communities in eutrophic lakes reflect 275–94.
alterations in the phosphate supply. European Banse, K. (1976). Rates of growth, respiration and
Journal of Phycology, 38, 255–62. photosynthesis of unicellular algae as related to
Axler, R. P., Gersberg, R. M. and Goldman, C. R. (1980). cell size: a review. Journal of Phycology, 12, 135–40.
Stimulation of nitrate uptake and photosynthesis (1994). Grazing and zooplankton production as key
by molybdenum in Castle Lake, California. controls of phytoplankton production in the
Canadian Journal of Fisheries and Aquatic Sciences, 37, open ocean. Oceanography, 7, 13–19.
707–12. Barber, J. and Anderson, J. M. (2002). Introduction [to
Azam, F. and Chisholm, S. W. (1976). Silicic acid papers of a discussion meeting on photosystem
uptake and incorporation by natural marine II]. Philosophical Transactions of the Royal Society of
phytoplankton populations. Limnology and London B, 357, 1325–8.
Oceanography, 21, 427–35. Barber, J. and Nield, J. (2002). Organization of
Azam, F., Fenchel, T., Field, J. G. et al. (1983). The transmembrane helices in photosystem II:
ecological role of water-column microbes comparison of plants and cyanobacteria.
in the sea. Marine Ecology Progress Series, 10, Philosophical Transactions of the Royal Society of
257–63. London B, 357, 1329–35.
Baccini, P. (1985). Phosphate interactions at the Barber, R. T. and Hilting, A. K. (2002). History of the
sediment–water surface. In Chemical Processes in study of plankton productivity. In Phytoplankton
REFERENCES 449
Productivity, ed. P. J. leB. Williams, D. N. Thomas chlorophyll concentration. Limnology and

and C. S. Reynolds, pp. 16–43. Oxford: Blackwell Oceanography, 42, 1–20.
Science. Behrenfeld, M. J., Esaias, W. E. and Turpie, K. R. (2002).
Barbiero, R. P., James, W. F. and Barko, J. W. (1999). Assessment of primary production at the global
The effects of disturbance events on scale. In Phytoplankton Productivity, ed. P. J. leB.
phytoplankton community structure in a small Williams, D. N. Thomas and C. S. Reynolds,
temperate reservoir. Freshwater Biology, 42, 503–12. pp. 156–86. Oxford: Blackwell Science.
Barón, P. J. (2003). The paralarvae of two South Béjà, O., Suzuki, M. T., Heidelberg, J. F. et al (2002).
American sympatric squid: Loligo gahi and Loligo Unsuspected diversity among marine aerobic
sanpaulensis. Journal of Plankton Research, 25, anoxygenic phototrophs. Nature, 415,
1347–58. 630–3.
Barrett, P. R. F., Littlejohn, J. W. and Curnow, J. (1999). Belcher, J. H. (1968). Notes on the physiology of
Long-term algal control in a reservoir using Botryococcus braunii Kützing. Archiv für
barley straw. Hydrobiologia, 415, 309–13. Mikrobiologie, 61, 335–46.
Bartram, J. and Chorus, I. (1999). Toxic Cyanobacteria. Bellinger, E. (1974). A note on the use of algal sizes in
London: E. and F. Spon. estimates of population standing crop. British
Bautista, M. F. and Paerl, H. W. (1985). Diel N2 fixation Phycological Journal, 9, 157–61.
in an intertidal marine cyanobacterial mat Belyea, L. R. and Lancaster, J. (1999). Assembly rules
community. Marine Chemistry, 16, 369–77. within a contingent ecology. Oikos, 86,
Beadle, L. C. (1974). The Inland Waters of Tropical Africa: 402–16.
An Introduction to Tropical Limnology. London: Berger, C. (1984). Consistent blooming of Oscillatoria
Longman. agardhii Gom. in shallow hypertrophic lakes.
Bean, C. W. and Winfield, I. J. (1995). Habitat use and Verhandlungen der internationalen Vereinigung für
activity patterns of roach (Rutilus rutilus), rudd theoretische und angewandte Limnologie, 22,
(Scardinius erythrophthalmus (L.), perch (Perca 910–16.
fluviatilis) and pike (Esox lucius) in the laboratory: (1989). In situ primary production, biomass and
the role of predation threat and structural light regime in the Wolderwijd, the most stable
complexity. Ecology of Freshwater Fish, 4, 37–46. Oscillatoria agardhii lake in the Netherlands.
Beard, S. J., Handley, B. A., Hayes, P. K. and Walsby, Hydrobiologia, 185, 233–44.
A. E. (1999). The diversity of gas-vesicle genes in Bergh, O., Børsheim, K. Y., Bratbak, G. and Heldal, M.
Planktothrix rubescens from Lake Zürich. (1989). High abundances of viruses found in
Microbiology, 145, 2757–68. aquatic environments. Nature, 340, 467–8.
Beard, S. J., Davis, P. A., Iglesias-Rodriguez, D., Berman, T. and Bronk, D. A. (2003). Dissolved organic
Skulberg, O. M. and Walsby, A. E. (2000). Gas nitrogen: a dynamic participant in aquatic
vesicle genes in Planktothrix spp. from Nordic ecosystems. Aquatic Microbial Ecology, 31,
lakes: strains with weak gas vesicles possess a 279–305.
longer variant of gvpC. Microbiology, 146, 2009–18. Berman, T. and Shteinman, B. (1998). Phytoplankton
Beardall, J., Berman, T., Heraud, P. et al. (2001). A development and turbulent mixing in Lake
comparison of methods of detection of phosphate Kinneret (1992–1996). Journal of Plankton Research,
limitation in microalgae. Aquatic Sciences, 63, 20, 709–26.
107–21. Berman, T., Hadas, O. and Kaplan, B. (1977). Uptake
Beers, J. R., Reid, F. M. H. and Stewart, G. L. (1982). and respiration of organic compounds and
Seasonal abundance of the microplankton heterotrophic growth in Pediastrum duplex
population in the North Pacific cental gyre. (Meyen). Freshwater Biology, 7, 495–502.
Deep-Sea Research, 29, 227–45. Berman, T., Walline, P. D., Schneller, A., Rothenberg, J.
Beeton, A. M. and Hageman, J. (1992). Impact of and Townsend, D. W. (1985). Secchi-disk depth
Dreissena polymorpha on the zooplankton record: a claim for the eastern Mediterranean.
community of western Lake Erie. Verhandlungen Limnology and Oceanography, 30, 447–8.
der internationalen Vereinigung für theoretische und Berman, T., Yacobi, Y. Z. and Pollingher, U. (1992). Lake
angewandte Limnologie, 25, 2349. Kinneret phytoplankton: stability and variability
Behrenfeld, M. J. and Falkowski, P. G. (1997). during twenty years (1970–1989). Aquatic Sciences,
Photosynthetic rates derived from satellite-based 54, 104–27.
450 REFERENCES
Berner, R. A. (1980). Early Diagenesis: A Theoretical Bleiwas, A. H. and Stokes, P. M. (1985). Collection of
Approach. Princeton, NJ: Princeton University large and small food particles by Bosmina.
Press. Limnology and Oceanography, 30, 1090–2.
Bertilsson, S. and Tranvik, L. J. (1998). Photochemically Blindow, I. (1992). Decline of charophytes during
produced carboxylic acids xsas substrates for eutrophication: comparison with angiosperms.
freshwater bacterioplankton. Limnology and Freshwater Biology, 28, 9–14.
Oceanography, 43, 885–95. Blindow, I., Andersson, G., Hargeby, A. and Johansson,
(2000). Photochemical transformation of dissolved S. (1993). Long-term pattern of alternative stable
organic matter in lakes. Limnology and states in two shallow eutrophic lakes. Freshwater
Oceanography, 45, 753–62. Biology, 30, 159–67.
Besch, W. K., Ricard, M. and Cantin, R. (1972). Benthic Blomqvist, P., Pettersson, A. and Hyenstrand, P. (1994).
diatoms as indicators of mining pollution in the Ammonium nitrogen: a key regulatory factor
Northwest Miramichi River System, New causing dominance of non-nitrogen-fixing
Brunswick, Canada. Internationale Revue des Cyanobacteria in aquatic systems. Archiv für
gesamten Hydrobiologie, 57, 39–73. Hydrobiologie, 132, 141–64.
Beverton, R. J. H. and Holt, S. J. (1957). On the Booker, M. and Walsby, A. E. (1979). The relative form
dynamics of exploited fish populations. Fishery resistance of straight and helical blue-green algal
Investigations, 19, 1–532. filaments. British Phycological Journal, 14, 141–50.
Bhattacharya, D. and Medlin, L. (1998). Algal (1981). Bloom formation and stratification by a
phylogeny and the origin of land plants. Plant planktonic blue-green alga in an experimental
Physiology, 116, 9–15. water column. British Phycological Journal, 16,
Bidanda, B. P. and Cotner, J. B. (2002). Love handles in 411–21.
aquatic ecosystems: the role of dissolved organic Booker, M. J., Dinsdale, M. T. and Walsby, A. E. (1976).
carbon drawdown, resuspended sediments and A continuously monitored column for the study
terrigenous inputs in the carbon balance of Lake of stratification by planktonic organisms.
Michigan. Ecosystems, 5, 431–45. Limnology and Oceanography, 14, 915–19.
Biddanda, B. A., Ogdahl, M. and Cotner, J. B. (2001). Booth, K., Bellinger, E. D. and Sigee, D. (1987). Use of
Dominance of bacterial metabolism in X-ray microanalysis and atomic-absorption
oligotrophic relative to eutrophic waters. spectroscopy for monitoring metal levels in algal
Limnology and Oceanography, 46, 730–39. and bacterial plankton. British Phycological Journal,
Bienfang, P. K. (1992). The role of coastal high-latitude 22, 300–1.
ecosystems in global export production. In Bormans, M., Sherman, B. S. and Webster, I. T. (1999).
Primary Productivity and Biogeochemical Cycles in the Is buoyancy regulation in Cyanobacteria an
Sea, ed. P. G. Falkowski and A. Woodhead, adaptation to exploit separation of light and
pp. 285–97. New York: Plenum Press. nutrients? Marine and Freshwater Research, 50,
Biggs, B. J. F., Stevenson, R. J. and Lowe, R. L. (1998). A 897–906.
habitat matrix coeneptual model for stream Borradaile, L. A., Eastham, L. E. S., Potts, F. A. and
periphyton. Archiv für Hydrobiologie, 143, Saunders, J. T. (1961). The Invertebrata, 4th edn
21–56. (revised by G. A. Kerkut). Cambridge: Cambridge
Bilton, D. T., Freeland, J. R. and Okamura, B. (2001). University Press.
Dispersal in freshwater invertebrates. Annual Boström, B., Andersen, J. M., Fleischer, S. and Jansson,
Review of Ecology and Systematics, 32, 159–81. M. (1988). Exchange of phosphorus across the
Bindloss, M. E. (1974). Primary productivity of sediment–water interface. Hydrobiologia, 170,
phytoplankton in Loch Leven, Kinross. Proceedings 229–44.
of the Royal Society of Edinburgh B, 74, 157–81. Bottrell, H. H., Duncan, A., Gliwicz, Z. M. et al. (1976).
Bird, D. F. and Kalff, J. (1989). Phagotrophic A review of some problems in zooplankton
sustenance of a metalimnetic phytoplankton production studies. Norwegian Journal of Zoology,
peak. Limnology and Oceanography, 34, 155–62. 24, 419–56.
Björk, S. (1972). Swedish lake restoration program gets Boucher, P., Blinn, D. W. and Johnson, D. B. (1984).
results. Ambio, 1, 153–65. Phytoplankton ecology in an unusually stable
(1988). Redevelopment of lake ecosystems: a environment (Montezuma Wall, Arizona, USA).
case-study approach. Ambio, 17, 90–98. Hydrobiologia, 199, 149–60.
REFERENCES 451
Bowen, C. C. and Jensen, T. E. (1965). Blue-green algae: (2000). The daily integral of growth of Planktothrix
fine structure of the gas vacuoles. Science, 147, rubescens calculated from growth in culture and
1460–3. irradiance in Lake Zürich. New Phytologist, 146,
Bower, A. S., Le Cann, B., Rossby, T. et al. (2002). 301–16.
Directly measured mid-depth circulation in the Brodin, Y.-W. (1995). Acidification of lakes and water
north-eastern Atlantic Ocean. Nature, 419, courses in a global perspective. In Liming of
603–7. Acidified Surface Waters: A Swedish Synthesis, ed. L.
Bowie, A. R., Maldonado, M. T., Frew, R. D. et al. (2001). Henrikson and Y.-W. Brodin, pp. 45–62. Berlin:
The fate of added iron during a mesoscale Springer-Verlag.
fertilisation experiment in the Southern Ocean. Brook, A. J., Baker, A. L. and Klemer, A. R. (1971). The
Deep-Sea Reseach Part II, Topical Studies in use of turbidimetry in studies of the population
Oceanography, 48, 2703–43. dynamics of phytoplankton populations, with
Bowie, A. R., Achterberg, E. P., Sedwick, P. N., Ussher, special reference to Oscillatoria agardhii var.
S. and Worsfold, P. J. (2002). Real-time monitoring isothrix. Mitteilungen der internationalen Vereinigung
of picomolar concentrations of iron(II) in marine für theoretische und angewandte Limnologie, 19,
waters using automated flow injection- 244–52.
chemiluminescence instrumentation. Brooks, A. S., Warren, G. J., Boraas, M. E., Scale, D. B.
Environmental Science and Technology, 36, and Edington, D. N. (1984). Long-term
4600–7. phytoplankton shifts in Lake Michigan: cultural
Boyce, F. M. (1974). Some aspects of Great Lakes eutrophication or biotic shifts. Verhandlungen der
physics of importance to biological and chemical internationalen Vereinigung für theoretische und
processes. Journal of the Fisheries Research Board of angewandte Limnologie, 22, 452–9.
Canada, 31, 689–730. Brooks, J. L. and Dodson, S. I. (1965). Predation, body
Boyd, P. W. (2002). Environmental factors controlling size and composition of plankton. Science, 150,
phytoplankton processes in the Southern Ocean. 28–35.
Journal of Phycology, 38, 844–61. Brown, D. J. A. and Sadler, K. (1989). Fish survival in
Boyd, P. W., LaRoche, J., Gall, M., Frew, R. and McKay, acid waters. In Acid Toxicity and Aquatic Animals,
R. M. L. (1999). Role of iron, light and silicate in ed. R. Morris, E. W. Taylor, D. J. A. Brown and
controlling algal biomass in Subantarctic waters, J. A., Brown, pp. 31–44. Cambridge: Cambridge
south east of New Zealand. Journal of Geophysical University Press.
Research, 104, 13395–408. Brown, J. H. (1999). Macroecology: progress and
Braginskiy, L. P. and Sirenko, L. A. (2000). Vegetation prospect. Oikos, 87, 3–14.
cycle of Microcystis aeruginosa Kütz. emend. Elenk. Brown, J. H. and Maurer, B. A. (1989). Macroecology:
International Journal on Algae, 2(3), 78–91. (English the division of food and space among species on
translation of original publication in Russian, continents. Science, 243, 1145–50.
Algologiya, 7(2), 153–65.) Brownlee, E. F., Sellner, S. G. and Sellner, K. G. (2003).
Bratbak, G. and Heldal, M. (1995). Viruses – the new Effects of barley straw (Hordeum vulgare) on
players in the game: their ecological role and freshwater and brackish phytoplankton and
could they mediate genetic exchange by cyanobacteria. Journal of Applied Phycology, 15,
transduction. In Molecular Ecology of Aquatic 525–31.
Microbes, ed I. Joint, pp. 249–64. Berlin: Brunberg, A. K. and Blomqvist, P. (2003). Recruitment
Springer-Verlag. of Microcystis (Cyanophyceae) from lake sediments:
Braun-Blanquet, J. (1964). Pflanzensociologie. Vienna: the importance of littoral inocula. Journal of
Springer-Verlag. Phycology, 39, 58–63.
Braunwarth, C. and Sommer, U. (1985). Analyses of Bruning, K. (1991a). Infection of the diatom
the in-situ growth rates of Cryptophyceae by use Asterionella by a chytrid. I. Effects of light on
of the mitotic index. Limnology and Oceanography, reproduction and infectivty of the parasite.
30, 893–7. Journal of Plankton Research, 13, 103–17.
Bright, D. I. and Walsby, A. E. (1999). The relationship (1991b). Infection of the diatom Asterionella by a
between critical pressure and width of gas chytrid. II. Effects of light on survival and
vesicles in isolates of Planktothrix rubescens from epidemic development of the parasite. Journal of
Lake Zürich. Microbiology, 145, 2769–75. Plankton Research, 13, 119–29.
452 REFERENCES
(1991c). Effects of temperature and light on the quotients of marine algae. Marine Biology, 65,
population dynamics of the Asterionella– 215–19.
Rhizophydium association. Journal of Plankton Butcher, R. W. (1924). The plankton of the River
Research, 13, 707–19. Wharfe. The Naturalist, Hull, April–June 1924,
Brunner, P. H. and Lampert, C. (1997). The flow of 175–214.
nutrients in the Danube River Basin. EAWAG News, (1932). Studies in the ecology of rivers. II. The
43, 15–17. microflora of rivers with special reference to the
Brush, M. J. and Nixon, S. W. (2002). Direct algae on the river bed. Annals of Botany new series,
measurement of light attenuation by epiphytes 46, 813–61.
on eelgrass Zostera marina. Marine Ecology – Bykovskiy, V. I. (1978) The effect of current velocity on
Progress Series, 238, 73–9. the phytoplankton (a review). Hydrobiological
Brush, M. J., Brawley, J. W., Nixon, S. W. and Kremer, Journal, 14, 34–40.
J. N. (2002). Modelling phytoplankton production: Cabrera, S., Montecino, V., Vila, I. et al. (1977).
problems with the Eppley curve and an empirical Caracteristicas limnologicas del Embalse Rapel,
alternative. Marine Ecology – Progress Series, 238, Chile Central. Seminario sobre medio ambiente y
31–45. represas, 1, 40–61.
Bulgakov, N. G. and Levich, A. P. (1999). The nitrogen : Cáceres, O. and Reynolds, C. S. (1984). Some effects of
phosphorus ratio as a factor regulating artificially enhanced anoxia on the growth of
phytoplankton community structure. Archiv für Microcystis aeruginosa Kütz. emend. Elenkin, with
Hydrobiologie, 146, 3–22. special reference to the initiation of the annual
Burger-Wiersma, T., Stal, L. J. and Mur, L. R. (1989). growth cycle in lakes. Archiv für Hydrobiologie, 99,
Prochlorothrix hollandica gen. nov., sp. nov., a 379–97.
filamentous oxygenic photoautotrophic Callieri, C. and Stockner, J. G. (2002). Freshwater
prokaryote containing chlorophyll a and autotrophic picoplankton: a review. Journal of
chlorophyll b. International Journal of Systematic Limnology, 61, 1–14.
Bacteriology, 39, 250–7. Callieri, C., Karjalainen, S. M. and Passoni, S. (2002).
Burkhardt, S., Riebesell, U. and Zondervan, I. (1999). Grazing by ciliates and heterotrophic
Effects of growth rate, CO2 concentration and cell nanoflagellates on picocyanobacteria in Lago
size on the stable carbon isotope fractionation in Maggiore, Italy. Journal of Plankton Research, 24,
marine phytoplankton. Geochimica and 785–96.
Cosmochimica Acta, 63, 3729–41. Campbell, L., Nolla, H. A. and Vaulot, D. (1994). The
Burmaster, D. (1979). The continuous culture of importance of Prochlorococcus to community
phytoplankton: mathematical equivalence among structure in the central North Pacific Ocean.
three steady-state models. American Naturalist, 113, Limnology and Oceanography, 39, 954–61.
123–34. Canabeus, L. (1929). Über die Heterocysten und
Burns, C. W. (1968a). The relationship between body Gasvakuolen der Blaualgen. Pflanzenforschung, 13,
size of filter-feeding cladocera and the maximum 1–48.
size of particle ingested. Limnology and Canter, H. M. (1979). Fungal and protozoan parasites
Oceanography, 23, 675–8. and their importance in the ecology of
(1968b). Direct observations of mechanisms phytoplankton. Report of the Freshwater Biological
regulating feeding behaviour of Daphnia in Association, 47, 43–50.
lakewater. Internationale Revue des Gesamten Canter, H. M. and Jaworski, G. H. M. (1978). The
Hydrobiologie, 53, 83–100. isolation, maintenance and host range
(1969). Relation between filtering rate, temperature studies of a chytrid, Rhizophydium planktonicum
and body size in four species of Daphnia. Canter emend., parasitic on Asterionella
Limnology and Oceanography, 14, 693–700. formosa Hassall. Annals of Botany, 42, 967–
Burns, C. W. and Gilbert, J. J. (1993). Predation on 79.
ciliates by freshwater calanoid copepods: rates of (1979). The occurrence of a hypersensitive reaction
predation and relative vulnerabilities of prey. in the planktonic diatom Asterionella formosa
Freshwater Biology, 30, 377–93. Hassal parasitised by the chytrid Rhizophydium
Burris, J. (1981). Effects of oxygen and inorganic planctonicum Canter emend. in culture. New
carbon concentrations on the photosynthetic Phytologist, 82, 187–206.
REFERENCES 453
(1980). Some general observations on the zoospores diseases. Critical Reviews in Environmental Control,
of the chytrid Rhizophydium planctonicum Canter 15, 275–313.
emend. New Phytologist, 84, 515–31. Caron, D. A., Pick, F. R. and Lean, D. R. S. (1985).
(1981). The effect of light and darkness upon Chroococcoid Cyanobacteria in Lake Ontario:
infection of Asterionella formosa Hassal by the vertical and seasonal distributions during 1982.
chytrid, Rhizophydium planktonicum Canter emend. Journal of Phycology, 21, 171–5.
Annals of Botany, New Series, 47, 13–30. Carpenter, E. J. and Chang, J. (1988). Species-specific
(1982). Some observations on the alga Fragilaria phytoplankton growth rates via diel and
crotonensis Kitton and its parasitism by two DNA synthesis cycles. I. Concept of the
chytridaceous fungi. Annals of Botany, New Series, method. Marine Ecology – Progress Series, 43,
47, 13–30. 105–11.
(1986). A study on the chytrid Zygorhizidium Carpenter, E. J. and McCarthy, J. J. (1975). Nitrogen
planktonicum Canter, a parasite of the diatoms fixation and uptake of combined nitrogenous
Asterionella and Synedra. Nova Hedwigia, 43, 269–98. nutrients by Oscillatoria (Trichodesmium) thiebautii
Canter, H. M. and Lund, J. W. G. (1948). Studies on in the western Sargasso Sea. Limnology and
plankton parasites. I. Fluctuations in the Oceanography, 20, 389–401.
numbers of Asterionella formosa Hass. in relation Carpenter, E. J. and Price, C. C. (1976). Marine
to fungal epidemics. New Phytologist, 47, 238–61. Oscillatoria (Trichodesmium): explanation for aerobic
(1951). Studies on plankton parasites. III. Examples nitrogen fixation without heterocysts. Science, 191,
of the interaction between parsitism and other 1278–80.
factors determining the growth of diatoms. Carpenter, E. J. and Romans, K. (1991). Major role of
Annals of Botany, New Series, 15, 359–71. the Cyanobacterium, Trichodesmium, in nutrient
(1953). Studies on plankton parasites. II. The cycling in the North Atlantic Ocean. Science, 254,
parasitism of diatoms with special reference to 1356–8.
the lakes in the English Lake District. Transactions Carpenter, S. R. (2001). Alternate states of ecosystems:
of the British Mycological Society, 36, 13–37. evidence and some implications. In Ecology:
Canter-Lund, H. M. and Lund, J. W. G. (1995). Freshwater Achievement and Challenge, ed. M. C. Press, N. J.
Algae: Their Microscopic World Explored. Bristol: Huntly and S. Levin, pp. 357–83. Oxford:
Biopress. Blackwell Science.
Caperon, J. and Meyer, J. (1972). Nitrogen-limited Carpenter, S. R. and Kitchell, J. F. (1992). Trophic
growth of marine phytoplankton. II. Uptake cascade and biomanipulation: interface of
kinetics and their role in nutrient-limited growth research and management. A reply to the
of phytoplankton. Deep-Sea Research, 19, 619–32. comment by de Melo et al. Limnology and
Caraco, N. F. (1995). Influence of human poulations on Oceanography, 37, 208–13.
P transfers to aquatic systems: a regional-scale (1993). The Trophic Cascade in Lakes. Cambridge:
study using large rivers. In Phosphorus in the Global Cambridge Univerity Press.
Environment, ed. H. Tiessen, pp. 235–44. Carpenter, S. R., Kitchell, J. F. and Hodgson, J. R.
Chichester: John Wiley. (1985). Cascading trophic interactions and lake
Carling, P. A. (1992) In-stream hydraulics and productivity. BioScience, 35, 634–9.
sediment transport. In The Rivers Handbook, vol. 1, Carpenter, S. R., Ludwig, D. and Brock, W. A. (1999).
ed. P. Calow and G. E. Petts, pp. 101–25. Oxford: Management of eutrophication for lakes subject
Blackwell Scientific Publications. to potentially irrevresible change. Ecological
Carlsson, P. and Granéli, E. (1999). Effects of N : P : Si Applications, 9, 751–71.
ratios and zooplankton grazing on phytoplankton Carpenter, S. R., Cole, J. J., Hodgson, J. R. et al. (2001).
communities in the northern Adriatic Sea. II. Trophic cascades, nutrients and lake productivity:
Phytoplankton species composition. Aquatic whole-lake experiments. Ecological Monographs, 71,
Microbial Ecology, 18, 55–65. 163–86.
Carmichael, W. W. (1986). Algal toxins. In Advances in Carr, N. G. (1995). Microbial culturs and natural
Botanical Research, vol. 12, ed. J. Callow, poulations. In Molecular Ecology of Aquatic Microbes,
pp. 467–501. London: Academic Press. ed. I. Joint, pp. 391–402. Berlin: Springer-Verlag.
Carmichael, W. W., Jones, C. L. A., Mahoud, N. A. and Carrick, H. J. and Schelske, C. L. (1997). Have we
Thiess, W. C. (1985). Algal toxins and water-based overlooked the importance of small
454 REFERENCES
phytoplankton in productive waters? Limnology nutrient-rich areas of the open sea? Limnology and
and Oceanography, 42, 1613–21. Oceanography, 36, i.
Carrick, H. J. F., Aldridge, J. and Schelske, C. (1993). Chisholm, S. W., Olson, R. J., Zettler, E. R. et al. (1988).
Wind influences phytoplankton biomass and A novel free-living prochlorophyte abundant in
composition in a shallow, productive lake. the the oceanic euphotic zone. Nature, 334,
Limnology and Oceanography, 38, 1179–92. 340–3.
Cassie, R. M. (1959). Micro-distribution of plankton. Chisholm, S. W., Frankel, S. L., Goericke, R. et al.
New Zealand Journal of Science, 3, 26–50. (1992). Prochlorococcus marinus nov. gen. nov. sp.: an
Cavicchioli, R., Ostrowski, M., Fegatella, F., Goodchild, oxyphototrophic marine prokaryote containing
A. and Guixa-Boixereu, N. (2003). Life under divinyl chlorophyll a and b. Archives of
nutrient limitation in oligotrophic marine Microbiology, 157, 297–300.
environments: an eco/physiological perspective of Chorus, I. and Schlag, G. (1993). Importance of
Sphingopyxis alaskensis (formerly Sphingomonas intermediate disturbances for species
alskensis). Microbial Ecology, 45, 203–17. composition and diversity in two very different
Cembella, A. D., Antia, N. J. and Harrison, P. J. (1984). Berlin lakes. Hydrobiologia, 249, 67–92.
The utilization of inorganic and organic Christie, W. J. (1974). Changes in the fish species
phosphorus compounds as nutrients by composition of the Great Lakes. Journal of the
eukaryotic microalgae: a multidisciplinary Fisheries Research Board of Canada, 31, 827–54.
perspective. Part I. Critical Reviews in Microbiology, Christoffersen, K. (1996). Ecological implications of
10, 317–91. cyanobacterial toxins in aquatic food webs.
Chandler, D. C. (1937). Fate of typical lake plankton in Phycologia, 35 (Suppl. 6), 42–50.
streams. Ecological Monographs, 7, 445–79. Chu, S. P. (1942). The influence of the mineral
Chang, J. and Carpenter, E. J. (1990). Species-specific composition of the medium on the growth of
phytoplankton growth rates via diel and DNA planktonic algae. I. Methods and culture media.
synthesis cycles. IV. Evaluation of the magnitude Journal of Ecology, 30, 284–325.
of error with computer-simulated cell populations. (1943). The influence of the mineral composition of
Marine Ecology – Progress Series, 65, 293–304. the medium on the growth of planktonic algae.
(1994). Active growth of the oceanic dinoflagellate II. The influence of the concentrations of
Ceratium teres in the Caribbean and Sargasso Seas inorganic nitrogen and phosphorus. Journal of
estimated by cell cycle analysis. Journal of Ecology, 31, 109–48.
Phycology, 30, 375–81. Clasen, J. (1979). Das Ziel der Phosphoreliminierung
Chapleau, F., Johansen, P. H. and Williams, M. M. am Zulauf der Wahnbachtalsperre in Hinblick auf
(1988). The distinction between pattern and die Oligotrophierung diese Gewässers. Zeitschrift
process in evolutionary biology: the use and für Wasser- und Abwasser-Forschung, 12, 65–77.
abuse of the term ‘strategy’. Oikos, 53, 136–8. Clements, F. E. (1916). Plant Succession: An Analysis of the
Chapman, D. V., Dodge, J. D. and Heaney, S. I (1980). Development of Vegetation, Publications of the
Light- and electron-microscope observations on Carnegie Institute No. 242. Washington, DC:
cysts and cyst formation in Ceratium hirundinella. Carnegie Institute.
British Phycological Journal, 15, 193. Clesceri, N. L. and Lee, G. F. (1965). Hydrolysis of
Chapman, D. V., Livingstone, D. and Dodge, J. D. condensed polyphospates. I. Non-sterile
(1981). An electron-microscope study of the environment. Air and Water Pollution, 9, 723–42.
excystment and early development of the Cloern, J. E. (1977). Effects of light intensity and
dinoflagellate Ceratium hirundinella. British temperature on Cryptomonas ovata (Cryptophyceae)
Phycological Journal, 16, 183–94. and nutrient uptake. Journal of Phycology, 13,
Chernousova, V. M., Sirenko, L. A. and Arendarchuk, 389–95.
V. V. (1968). Localisation and physiological state of Cloern, J. E., Grenz, C. and Vidergar-Lucas, L. (1995).
species of blue-green algae during late-autumn An empirical model of the phytoplankton
and spring periods. In Tsvetenie Vody, ed. A. V. chlorophyll : carbon ratio: the conversion factor
Topachvskiy, L. P. Braginskiy, N. V. Kondra’eva, between productivity and growth rate. Limnology
L. A. Kulskiy and L. A. Sirenko, pp. 81–91. Kiev: and Oceanography, 40, 1313–21.
Naukova Dumka. (In Russian.) Cmiech, H. A., Reynolds, C. S. and Leedale, G. F. (1984).
Chisholm, S. W. and Morel, F. M. M. (1991). What Seasonal periodicity, heterocyst differentiation
controls phytoplankton production in and sporulation of planktonic Cyanophyceae in a
REFERENCES 455
shallow lake, with special reference to Anabaena quality of surface and underground waters.
solitaria. British Phycological Jourmal, 19, 245–57. Technical Bulletin, Ministry of Agriculture Fisheries and
Codd, G. A. (1995). Cyanobacterial toxins: occurrence, Food, 32, 5–57.
properties and biological significance. Water Cordoba-Molina, J. F., Hudgins, R. R. and Silverston,
Science and Technology, 32(4), 149–56. P. L. (1978). Settling in continuous sedimentation
Coesel, P. F. M. (1994). On the ecological significance tanks. Journal of the Environmental Division of the
of a mucilaginous envelope in planktic desmids. American Society of Civil Engineers, 104(EE6),
Algological Studies, 73, 65–74. 1263–75.
Cohen, J. E., Briand, F. and Newman, C. M. (1990). Corner, E. D. S. and Davies, A. G. (1971). Plankton as a
Community Food Webs: Data and Theory. New York: factor in the nitrogen and phosphorus cycles in
Springer-Verlag. the sea. Advances in Marine Biology, 9, 101–204.
Cole, J. J. (1982). Interactions between bacteria and Cottingham, K. L., Brown, B. L. and Lennon, J. T.
algae in aquatic ecosystems. Annual Reviews of (2001). Biodiversity may regulate the the temporal
Ecology and Systematics, 13, 291–314. variability of ecological systems. Ecology Letters, 4,
Cole, J. J. and Caraco, N. F. (1998). Atmospheric 72–85.
exchange of carbon dioxide in a low-wind Coulter, G. W. (1994). Speciation and fluctuating
oligotrophic lake. Limnology and Oceanography, 43, environments, with special reference to ancient
647–56. African lakes. Ergebnisse der Limnologie, 44,
Cole, J. J., Likens, G. E. and Strayer, D. L. (1982). 127–37.
Photosynthetically produced dissolved organic Coveney, M. F. and Wetzel, R. G. (1995). Biomass,
carbon: an important carbon source for production and specific growth rate of
planktonic bacteria. Limnology and Oceanography, bacterioplankton and coupling to phytoplankton
27, 1080–90. in an oligotrophic lake. Limnology and
Cole, J. J., Findlay, S. and Pace, M. L. (1988). Bacterial Oceanography, 40, 1187–1200.
production in fresh and saltwater ecosystems: a Crawford, D. W. and Lindholm, T. (1997). Some
cross-system overview. Marine Ecology – Progress observations on vertical distribution and
Series, 43, 1–10. migration of the phototrophic ciliate Mesodinium
Cole, J. J., Caraco, N. F., Strayer, D. L., Ochs, C. and rubrum (Myrionecta rubra) in a stratified brackish
Nolan, S. (1989). A detailed carbon budget as an inlet. Aquatic Microbial Ecology, 13, 267–74.
ecosystem-level calibration of bacterial respiration Crawford, R. M. and Schmid, A.-M. M. (1986).
in an oligotrophic lake during midsummer. Ultrastructure of silica deposition in diatoms. In
Limnology and Oceanography, 34, 286–96. Biomineralization in the Lower Plants and Animals, ed.
Cole, J. J., Caraco, N. F., Kling, G. W. and Kratz, T. W. B. S. C. Leadbeater and R. Ridings, pp. 291–314.
(1994). Carbon dioxide supersaturation in the Oxford: Oxford University Press.
surface waters of lakes. Science, 265, 1568–70. Croome, R. L. and Tyler, P. A. (1984). Microbial
Colebrook, J. M. (1960). Plankton and water microstratification and crepuscular
movements in Windermere. Journal of Animal photosynthesis in meromictic Tasmanian lakes.
Ecology, 29, 217–40. Verhandlungen der internationalen Vereinigung für
Colebrook, J. M., Glover, R. S. and Robinson, G. A. theoretische und angewandte Limnologie, 22, 1216–23.
(1961). Continuous plankton records: Crumpton, W. G. and Wetzel, R. G. (1982). Effects of
contribution towards a plankton atlas of the differential growth and mortality in the seasonal
noth-eastern Atlantic and the North Sea: general succession of phytoplankton populations of
introduction. Bulletins of Marine Ecology, 5, 67–80. Lawrence Lake, Michigan. Ecology, 63, 1729–39.
Coleman, A. (1980). Enhanced detection of bacteria in Crusius, J. and Wanninkhof, R. (2003). Gas transfer
natural environments by fluorochrome staining velocities measured at low wind speed over a
of DNA. Limnology and Oceanography, 25, 948–51. lake. Limnology and Oceanography, 48, 1010–17.
Connell, J. H. (1978). Diversity in tropical rain forests Cryer, M., Pierson, G. and Townsend, C. R. (1986).
and coral reefs. Science, 199, 1302–10. Reciprocal interactions between roach, Rutilus
Conway, K. and Trainor, F. R. (1972). Scenedesmus rutilus, and zooplankton in a small lake: prey
morphology and flotation. Journal of Phycology, 8, dynamics and fish growth and recruitment.
138–43. Limnology and Oceanography, 31, 1022–38.
Cooke, G. W. (1976). A review of the effects of Csanady, G. T. (1978). Water circulation and dispersal
agriculture on the chemical composition and mechanisms. In Lakes: Chemistry, Geology and
456 REFERENCES
Physics, ed. A. Lerman, pp. 21–64. New York: Darwin, C. (1859). On the Origin of Species by Means of
Springer-Verlag. Natural Selection. London: John Murray.
Cuhel, R. L. and Lean, D. R. S. (1987a). Protein Dau, H. (1994). Molecular mechanisms and
synthesis by lake plankton measured using in quantitative models of variable photosystem II
situ carbon dioxide and sulfate assimilation. fluorescence. Photochemistry and Photobiology, 60,
Canadian Journal of Fisheries and Aquatic Sciences, 44, 1–23.
2102–17. Dauta, A. (1982). Conditions de développement du
(1987b). Influence of light intensity, light quality phytoplancton: étude comparative du
temperture and daylength on uptake and comportement de huit espèces en culture. I.
assimilation of carbon dioxide and sulfate by lake Détermination des paramètres de croissance en
plankton. Canadian Journal of Fisheries and Aquatic fonction de la lumière et de la température.
Sciences, 44, 2118–32. Annales de Limnologie, 18, 217–62.
Cummins, K. W. and Wuychek, J. C. (1971). Caloric Davey, M. C. and Walsby, A. E. (1985). The form
equivalents for investigations in ecological resistance of sinking algal chains. British
energetics. Mitteilungen der internationalen Phycological Journal, 20, 243–8.
Vereinigung für theoretische und angewandte Davies, P. S. (1997). Fluxes of phosphorus in lakes
Limnologie, 18, 1–158. relevant to phytoplankton ecology. Ph.D. thesis,
Cushing, D. H. (1971). The dependence of recruitment University of Leeds.
on parent stock in different groups of fishes. Davis, P. A., Dent, M. M., Parker, J. M., Reynolds, C. S.
Journal du Conseil internationale pour l’Exploration de and Walsby, A. E. (2003). The annual cycle of
la Mer, 33, 340–62. growth rate and biomass change in Planktothrix
(1982). Climate and Fisheries. London: Academic Press. spp. in Blelham Tarn, English Lake District.
(1988). The study of stock and recruitment. In Fish Freshwater Biology, 48, 852–67.
Population Dynamics, ed. J. A. Gulland, pp. 105–28. Davis, R. B. (1974). Tubificids alter profiles of redox
Chichester: John Wiley. potential and pH in profundal lake sediment.
(1990). Plankton production and year-class strength Limnology and Oceanography, 19, 342–6.
in fish poulations: an update on the Davis, R. B., Thurlow, D. L. and Brewster, F. E. (1975).
match/mismatch hypothesis. Advances in Marine Effects of burrowing tubificid worms on the
Biology, 26 249–93. exchange of phosphorus between lake sediment
(1995). The long-term relationship between and overlying water. Verhandlungen der
zooplankton and fish. ICES Journal of Marine internationalen Vereinigung für theoretische und
Science, 50, 611–26. angewandte Limnologie, 19, 382–94.
(1996). Towards a Science of Recruitment of Fish Davison, W. (1987). Internal element cycles affecting
Populations. Oldendorf: ECI. the long-term alkalinity status of lakes:
Cushing, D. E. and Horwood, J.W (1977). Development implications for lake restoration. Schweizerische
of a model of stock and recruitment. In Fisheries Zeitschrift für Hydrologie, 49, 186–201.
Mathematics, ed. J. H. Steele, pp. 21–35. London: Davison, W., Reynolds, C. S., Tipping, E. and Needham,
Academic Press. R. F. (1989). Reclamation of acid waters using
Daft, M. J., McCord, S. B. and Stewart, W. D. P. (1975). sewage sludge. Environmental Pollution, 57,
Ecological studies on algal-lysing bacteria in fresh 251–74.
waters. Freshwater Biology, 5, 577–96. Davison, W., George, D. G. and Edwards, N. J. A. (1995).
Dahl, E., Lindahl, O., Paasche, E. and Throndsen, J. Controlled reversal of lake acidification by
(1989). The Chrysochromulina polylepis bloom in treatment with phosphate fertiliser. Nature, 377,
Scandinavian waters during spring 1988. In Novel 504–7.
Phytoplankton Blooms: Causes and Impacts of Deacon, G. E. R. (1982). Physical and biological
Recurrent Brown Tides and Other Unusual Blooms, ed. zonation in the Southern Ocean. Deep-Sea Research,
E. M. Cosper, M. Bricelj and E. J. Carpenter, 29, 1–16.
pp. 383–405. Berlin: Springer-Verlag. Dejoux, C. and Iltis, A. (1992). Lake Titicaca. Dordrecht:
Dale, B. (2001). The sedimentary record of Kluwer.
dinoflagellate cysts: looking back into the future Dekker, A. C., Malthus, T. J. and Hoogenboom, H. J.
of phytoplankton blooms. Scientia Marina, 65 (1995). The remote sensing of inland water
(Suppl. 2), 257–72. quality. In Advances in Environmental Remote
REFERENCES 457
Sensing, ed. F. M. Danson and S. E. Plummer, Dillon, P. J. and Rigler, F. H. (1974). The
pp. 123–42, New York: John Wiley. phosphorus–chlorophyll relationship in lakes.
Delwiche, C. (2000). Tracing the thread of plastid Limnology and Oceanography, 19, 767–73.
diversity through the tapestry of life. American Dinsdale, M. T and Walsby, A. E. (1972). The
Naturalist, 154, 164–77. interrelations of cell turgor pressure, gas
De Melo, R., France, R. and McQueen, D. J. (1992). vacuolation and buoyancy in a blue-green alga.
Biomanipulation: hit or myth? Limnology and Journal of Experimental Botany, 23, 561–70.
Oceanography, 37, 192–207. DiTullio, G. R., Geesey, M. E., Jones, D. R. et al. (2003).
De Mott, W. R. (1982). Feeding selectivities and Phytoplankton assemblage structure and primary
relative ingestion rates of Daphnia and Bosmina. productivity along 170◦ W in the South Pacific
Limnology and Oceanography, 27, 518–27. Ocean. Marine Ecology – Progress Series, 255, 55–80.
(1986). The role of taste in food selection by Dobbins, W. E. (1944). Effect of turbulence on
freshwater zooplankton. Oecologia, 69, 334–40. sedimentation. Transactions of the American Society
De Mott, W. R. and Kerfoot, W. C. (1982). Competition of Civil Engineers, 109, 629–56.
among cladocerans: nature of the interaction Dodson, S. I., Arnott, S. E. and Cottingham, K. L.
between Bosmina and Daphnia. Ecology, 63, (2000). The relationship in lake communities
1949–66. between primary productivity and species
De Nie, H. W. (1987). The Decrease in Aquatic Vegetation richness. Ecology, 81, 2662–79.
in Europe and its Consequences for Fish Populations, Dokulil, M. T. and Teubner, K. (2000). Cyanobacterial
EIFAC/CECPI Occasional Papers No. 19. Rome: dominance in lakes. Hydrobiologia, 438, 1–12.
Food and Agriculture Organisation. (2003). Steady-state phytoplankton assemblages
Demmig-Adams, B. and Adams, W. W. (1992). during thermal stratification in deep alpine
Photoprotection and other responses of plants to lakes: do they occur? Hydrobiologia, 502, 65–72.
high light stress. Annual Reviews in Plant Physiology Donato-Rondon, J. C. (2001). Fitoplancton de los lagos
and Plant Molecular Biology, 43, 599–626. andinos del norte de Sudamérica, Colección Jorge
Denman, K. and Gargett, A. E. (1983). Time and space Alvares Lleras No. 19. Bogotá: Academia
scales of vertical mixing and advection of Colombiana de Ciencias Exactas, Fisicas y
phytoplankton in the upper ocean. Limnology and Naturales.
Oceanography, 28, 801–15. Donk, E. van and Voogd, H. (1998). Control of Volvox
Denman, K. L. and Platt, T. (1975). Coherences in the blooms by Hertwigia, a rotifer. Verhandlungen der
horizontal distributions of phytoplankton and internationalen Vereinigung für theoretische und
temperature in the upper ocean. Mémoirs de la angewandte Limnologie, 26, 1781–84.
Société royale des Sciences, Liège, 6e Série, 7, 19–30. Donner, J. (1966). Rotifers. (Translated by H. G. S.
Descy, J.-P., Everbecq, E., Gosselain, V., Viroux, L. and Wright.) London: Frederick Warne.
Smitz, J. S. (2003). Modelling the impact of Doolittle, W. F., Boucher, Y., Nesbo, C. L. et al. (2003).
benthic filter-feeders on the composition and How big is the iceberg of which organellar genes
biomass of river plankton. Freshwater Biology, 48, in nuclear genomes are but the tip? Philosophical
404–17. Transactions of the Royal Society of London B, 358,
Detmer, A. E. and Bathmann, U. V. (1997). Distribution 39–57.
patterns of autotropic pico- and nano-plankton Dortch, Q., Roberts, T. L., Clayton, J. R. and Ahmed,
and their relative contribution to algal biomass S. I. (1983). RNA/DNA ratios and DNA
during spring in the Atlantic sector of the concentrations as indicators of growth rate and
Southern Ocean. Deep-Sea Research Part II Topical biomass in planktonic marine organisms. Marine
Studies in Oceanography, 44, 299–325. Ecology – Progress Series, 13, 61–71.
Diaz, M. M. and Pedrozo, F. L. (1996). Nutrient Doty, M. S. (ed.) (1961). Proceedings of the Conference on
limitation in Andean–Patagonian lakes at Primary Productivity Measurement, Marine and
latitude 40–41◦ S. Archiv für Hydrobiologie, 138, Freshwater. Washington, DC: US Atomic Energy
123–43. Commission.
Diaz, M., Pedrozo, F. L. and Baccala, N. (2000). Summer Douglas, A. E. and Raven, J. A. (2003). Genomes at the
classification of southern-hemispere tropical interface between bacteria and organelles.
lakes (Patagonia, Argentina). Lakes and Reservoirs, Philosophical Transactions of the Royal Society of
5, 213–29. London B, 358, 5–17.
458 REFERENCES
Douglas, G. B., Coad, D. N. and Adeney, J. A. (1998). (1999). The species richness of reservoir plankton
Reducing phosphorus in aquatic systems using and the effect of reservoirs on plankton dispersal
modified clays. Water, 25(3), 42–3. (with particular emphasis on rotifers and
Downing, J. A., Watson, S. B. and McCauley, E. (2001). cladocerans). In Theoretical Reservoir Ecology and its
Predicting cyanobacteria dominance in lakes. Applications, ed. J. G. Tundisi and M. Strakraba,
Canadian Journal of Fisheries and Aquatic Sciences, 58, pp. 477–91. São Carlos: Institute of Ecology.
1905–8. Dumont, H. and Segers, H. (1996). Estimating
Dring, M. J. and Jewson, D. H. (1979). What does lacustrine zooplankton species richness and
14
C-uptake by phytoplankton really measure? A complementarity. Hydrobiologia, 341, 125–32.
fresh approach using a theoretical method. British Dunst, R. C., Born, S. M., Uttormark, P. D. et al. (1974).
Phycological Journal, 14, 122–3. Survey of Lake Rehabilitation Techniques and
(1982). What does 14 C-uptake by phytoplankton Experiences, Technical Bulletin No. 75. Madison,
really measure? A theoretical modelling WI: Wisconsin Department of Natural
approach. Proceedings of the Royal Society of London Resources.
B, 214, 351–68. Duthie, H. C. (1965). Some observations on the ecology
Droop, M. R. (1973). Some thoughts on nutrient of desmids. Journal of Ecology, 53, 695–703.
limitation in algae. Journal of Phycology, 9, 264–72. Duthie, H. C. and Hart, C. J. (1987). The phytoplankton
(1974). The nutrient status of algae cells in of subarctic Canadian great lakes. Ergebnisse der
continuous culture. Journal of the Marine Biological Limnologie, 25, 1–9.
Association of the United Kingdom, 54, 825–55. Duysens, L. N. M. (1956). The flattening of the
Drum, R. W. and Pankratz, H. S. (1964). Post-mitotic absorption spectrum of suspensions as compared
fine structure of Gomphonema parvulum. Journal of to that of solutions. Biochimica et Biophysica Acta,
Ultrastructural Research, 10, 217–23. 19, 1–12.
Dryden, R. C. and Wright, S. J. L. (1987). Predation of Eberley, W. R. (1959). The metalimnetic oxygen
cyanobacteria by protozoa. Canadian Journal of maximum in Myers lake. Investigations of Indiana
Microbiology, 33, 471–82. Lakes and Streams, 5, 1–46.
Ducklow, H. W. (2000). Bacterial production and (1964). Primary production in McLish Lake
biomass in the Ocean. In Microbial Ecology of the (northern Indiana), an extreme plus-heterograde
Oceans, ed. D. L. Kirchman, pp. 47–84. New York: lake. Verhandlungen der internationalen Vereinigung
Wiley-Liss. für theoretische und angewandte Limnologie, 15,
Ducklow, H. W., Kirchman, D. L. and Anderson, T. R. 394–401.
(2002). The magnitude of spring bacterial Eddy, S. (1931). The plankton of the Sangamon River
production in the North Atlantic Ocean. in the summer of 1929. Bulletin of the Natural
Limnology and Oceanography, 47, 1684–93. History Survey of Illinois, 19, 469–86.
Dufrêne, M. and Legendre, P. (1997). Species Edler, L. (1979). Phytoplankton succesion in the Baltic
assemblages and indicator species: the need for a Sea. Acta Botanica Fennica, 110, 75–8.
flexible asymmetric approach. Ecological Edmondson, W. T. (1970). Phosphorus, nitrogen and
Monographs, 67, 345–66. algae in Lake Washington after diversion of
Dugdale, R. C. (1967). Nutrient limitation in the sea: sewage. Science, 169, 690–1.
dynamics, identification and significance. (1972). The present condition of Lake Washington.
Limnology and Oceanography, 12, 685–95. Verhandlungen der internationalen Vereinigung für
Dugdale, R. C., Dugdale, V. A., Neess, J. C. and theoretische und angewandte Limnologie, 19, 606–15.
Goering, J. J. (1959). Nitrogen fixation in lakes. Edmondson, W. T. and Lehman, J. T. (1981). The effect
Science, 130, 859–60. of changes in the nutrient income on the
Dumont, H. (1972). The biological cycle of condition of lake Washington. Limnology and
molybdenum in relation to primary production. Oceanography, 26, 1–29.
Verhandlungen der internationalen Vereinigung für Edmondson, W. T. and Litt, A. H. (1982). Daphnia in
theoretische und angewandte Limnologie, 18, 84–92. Lake Washington. Limnology and Oceanography, 27,
(1992). The regulation of plant and animal species 272–93.
and communities in African shallow lakes and Edvardsen, B. and Paasche, E. (1998). Bloom dynamics
wetlands. Revue d’Hydrobiologie Tropicale, 25, 303 and physiology of Prymnesium and
346. Chrysochromulina. In Physiological Ecology of Harmful
REFERENCES 459
Algal Blooms, ed. D. M. Anderson, A. D. Cembella incompletely mixed eutrophic water column
and G. M. Hallegraeff, pp. 193–208. Berlin: using the PROTECH model. Freshwater Biology, 47,
Springer-Verlag. 433–40.
Ehrlich, P. R. and Ehrlich, A. H. (1981). Extinction: The Elliott, J. M. (1975a). The growth rate of brown trout
Causes and Consequences of the Disappearance of (Salmo trutta L.) fed on maximum rations. Journal
Species. New York: Random House. of Animal Ecology, 44, 805–21.
Einsele, W. (1936). Über die Beziehungen den (1975b). The growth rate of brown trout (Salmo
Eisenkrieslaufs zum Phosphatkrieslauf im trutta L.) fed on reduced rations. Journal of Animal
eutrophen See. Archiv für Hydrobiologie, 33, Ecology, 44, 823–42.
361–87. Elliott, J. M. and Hurley, M. A. (1998). A new functional
Einsele, W. and Grim, J. (1938). Über den model for estimating the maximum amount of
Kieselsäurgehalt planktischer Diatomeen und invertebrate food consumed per day by brown
dessen Bedeutung für einiger Fragen ihrer trout, Salmo trutta. Freshwater Biology, 39, 339–49.
Ökologie. Zeitschrift für Botanik, 32, (1999). A new energetics model for brown trout,
545–90. Salmo trutta. Freshwater Biology, 42, 235–46.
Eker, E., Georgieva, L., Senichkina, L. and Kideys, A. E. (2000a). Daily energy intake and growth of
(1999). Phytoplankton distribution in the western piscivorous brown trout, Salmo trutta. Freshwater
and eastern Black Sea in spring and autumn Biology, 44, 237–45.
1995. ICES Journal of Marine Science, 56 (Suppl.), (2000b). Optimum energy intake and gross
15–22. efficiency of energy conversion for brown trout,
Eker-Develi, E. and Kideys, A. E. (2003). Distribution of Salmo trutta, feeding on invertebrates or fish.
phytoplankton in the southern Black Sea in Freshwater Biology, 44, 605–15.
summer 1996, spring and autumn 1998. Journal of Eloranta, P. (1993). Diversity and succession of the
Marine Systems, 39, 203–11. phytoplankton in a small lake over a two-year
Elliott, J. A., Reynolds, C. S., Irish, A. E. and Tett, P. period. Hydrobiologia, 249, 25–32.
(1999a). Exploring the potential of the PROTECH Elser, J. J. (1999). The pathway to noxious
model to investigate phytoplankton community cyanobacteria blooms in lakes: the food web as
theory. Hydrobiologia, 414, 37–43. the final turn. Freshwater Biology, 42, 537–43.
Elliott, J. A., Irish, A. E., Reynolds, C. S. and Tett, P. Elser, J. J., Sterner, R. W., Gorokhova, E. et al. (2000).
(1999b). Sensitivity analysis of PROTECH, a new Biological stoichiometry from genes to
approach in phytoplankton modelling. ecosystems. Ecology Letters, 3, 540–50.
Hydrobiologia, 414, 45–51. Elser, J. J., Kyle, M., Makino, W., Yosshida, T. and
(2000a). Modelling freshwater phytoplankton Urabe, J., (2003). Ecological stoichiometry in the
communities: an exercise in validation. Ecological microbial food web: a test of the light: nutrient
Modelling, 128, 19–26. hypothesis. Aquatic Microbial Ecology, 31, 49–65.
Elliott, J. A., Reynolds, C. S. and Irish, A. E. (2000b). Elton, C. S. (1927). Animal Ecology. London: Macmillan.
The diversity and succession of phytoplankton (1958). The Ecology of Invasions by Animals and Plants.
communities in disturbance-free environments, London: Methuen.
using the model PROTECH. Archiv für Elzenger, J. T., Prins, H. B. A. and Stefels, J. (2000). The
Hydrobiologie, 149, 241–58. role of extracellular carbonic anhydrase activity
Elliott, J. A., Irish, A. E. and Reynolds, C. S. (2001a). in inorganic utilization of Phaeocystis globosa
The effects of vertical mixing on a phytoplankton (Prymnesiophyceae): a comparison with other
community: a modelling approach to the marine algae using an isotopic disequilibrium
intermediate disturbance hypothesis. Freshwater technique. Limnology and Oceanography, 45,
Biology, 46, 1291–7. 372–80.
Elliott, J. A., Reynolds, C. S. and Irish, A. E. (2001b). An Emsley, J. (1980). The phosphorus cycle. In The
investigation of dominance in phytoplankton Handbook of Environmental Chemistry, vol. 1A, ed. O.
using the PROTECH model. Freshwater Biology, 46, Hutzinger, pp. 147–67. Berlin: Springer-Verlag.
99–108. Eppley, R. W. (1970). Relationships of phytoplankton
Elliott, J. A., Irish, A. E. and Reynolds, C. S. (2002). species distribution to the depth distribution of
Predicting the spatial dominance of nitrate. Bulletin of the Scripps Institute of
phytoplankton in a light-limited and Oceanography, 17, 43–9.
460 REFERENCES
(1972). Temperature and phytoplankton growth rate Falkowski, P. G. and Raven, J. A. (1997). Aquatic
in the sea. Fisheries Bulletin, 70, 1063–85. Photosynthesis. Oxford: Blackwell Science.
Eppley, R. W., Holmes, R. W. and Strickland, J. D. H. Falkowski, P. G., Barber, R. T. and Smetacek, V. (1998).
(1967). Sinking rates of marine phytoplankton Biogeochemical controls and feedbacks on ocean
measured with a fluorometer. Journal of primary production. Science, 281, 200–6.
Experimental Marine Biology and Ecology, 1, 191–208. Faller, A. J. (1971). Oceanic turbulence and the
Eppley, R. W., Coatsworth, J. L. and Solorzano, L. Langmuir circulation. Annual Review of Ecology and
(1969a). Studies of nitrate reductase in marine Systematics, 2, 201–34.
phytoplankton. Limnology and Oceanography, 14, Fallon, R. D. and Brock, T. D. (1981). Overwintering of
194–205. Microcystis in Lake Mendota. Freshwater Biology, 11,
Eppley, R. W., Rogers, J. N. and McCarthy, J. N. (1969b). 217–26.
Half-saturation constants for uptake of nitrate Fay, P., Stewart, W. D. P., Walsby, A. E. and Fogg, G. E.
and ammonium by marine phytoplankton. (1968). Is the heterocyst the site of nitrogen
Limnology and Oceanography, 14, 912–20. fixation in blue-green algae? Nature, 220,
Evans, G. T. and Taylor, F. J. R. (1980). Phytoplankton 810–12.
accumulation in Langmuir cells. Limnology and Fenchel, T. and Finlay, B. J. (1994). Ecology and Evolution
Oceanography, 25, 840–5. in Anoxic Worlds. Oxford: Oxford University Press.
Evans, M. S. and Jude, D. J. (1986). Recent shifts in Ferguson, A. J. D. and Harper, D. M. (1982). Rutland
Daphnia community structure in south-eastern Water phytoplankton: the development of an
Lake Michigan: a comparison of the inshore and asset or a nuisance? Hydrobiologia, 88, 117–33.
offshore regions. Limnology and Oceanography, 31, Ferguson, A. J. D., Thompson, J. M. and Reynolds, C. S.
56–67. (1982). Structure and dynamics of zooplankton
Everall, N. C. and Lees, D. R. (1996). The use of barley maintained in closed systems, with special
straw to control general and blue-green algal reference to the food supply. Journal of Plankton
growth in a Derbyshire reservoir. Water Research, Research, 4, 523–43.
30, 269–76. Fernando, C. H. (1980). The species and size
Fabbro, L. D. and Duivenvorden, L. J. (2000). A two-part composition of tropical freshwater zooplankton
model linking multi-dimensional environmental with special reference to the oriental region
gradients and seasonal succession of (South East Asia). Internationale Revue des gesamten
phytoplankton assemblages. Hydrobiologia, 438, Hydrobiologie, 76, 149–67.
13–24. Field, C. B., Behrenfeld, M. J., Randerson, J. T. and
Fahnenstiel, G. L., Sicko-Goad, L., Scavia, D. and Falkowski, P. G. (1998). Primary production of the
Stoermer, E. F. (1986). Importance of picoplankton biosphere: integrating terrestrial and oceanic
in Lake Superior. Canadian Journal of Fisheries and components. Science, 281, 237–40.
Aquatic Sciences, 51, 2769–83. Findenegg, I. (1943). Untersuchungen über die
Falconer, R. A., George, D. G. and Hall, P. (1991). Ökologie und die Produktionsverhältnisse des
Three-dimensional numerical modelling of Planktons im Kärnter Seengebeite. Internationale
wind-driven circulation in a shallow Revue des gesamten Hydrobiologie, 43, 388–429.
homogeneous lake. Journal of Hydrology, 124, (1947). Über die Lichtanspruche planktischer
59–79. Süsswasseralgen. Sitzungsberichte der Akademie der
Falkner, G., Falkner, R. and Schwab, A. (1989). Wissenschaften in Wien, 155, 159–71.
Bioenergetic state characterization of transient Finenko, Z. Z., Piontkovski, S. A., Williams, R. and
state phosphate uptake by the cyanobacterium Mishonov, A. V. (2003). Variability of
Anacystis nidulans. Archives of Microbiology, 152, phytoplankton and mesozooplankton biomass in
353–61. the subtropical and tropical Atlantic Ocean.
Falkowski, P. G. (1997). Evolution of the nitrogen cycle Marine Ecology – Progress Series, 250, 125–44.
and its influence on the biological sequestration Finkel, Z. V. (2001). Light absorption and size scaling
of CO2 in the ocean. Nature, 387, 272–5. of light-limited metabolism in marine diatoms.
(2002). On the evolution of the carbon cycle. In Limnology and Oceanography, 46, 86–94.
Phytoplankton Productivity, ed. P. J. leB. Williams, Finlay, B. J. (2001). Protozoa. In Encyclopedia of
D. N. Thomas and C. S. Reynolds, pp. 318–49. Biodiversiy, vol. 4, ed. S. R. Levin, pp. 569–99. San
Oxford: Blackwell Science. Diego, CA: Academic Press.
REFERENCES 461
(2002). Global dispersal of free-living microbial stoichiometry. Scientia Marina, 65 (Suppl. 2),
eukaryote species. Science, 296, 1061–3. 153–69.
Finlay, B. J. and Clarke, K. J. (1999). Ubiquitous Frankignoulle, M. (1988). Field measurements of
dispersal of microbial species. Nature, 400, air–sea CO2 exchange. Limnology and Oceanography,
828. 33, 313–22.
Finlay, B. J., Esteban, G. F. and Fenchel, T. (1998). Freeland, J. R., Rimmer, V. K. and Okamura, B. (2001).
Protozoan diversity: converging estimates of the Genetic changes within freshwater bryozoan
global number of free-living ciliate species. Protist, populations suggest temporal gene flow from
149, 29–37. statoblast banks. Limnology and Oceanography, 46,
Fish, G. R. (1952). Hydrology and algology. In Annual 1121–9.
Report of the East African Fisheries Research Freeman, C., Evans, C. D. and Monteith, D. T. (2001).
Organisation (1951), pp. 6–11. Export of organic carbon from peat soils. Nature,
Fleming, R. H. (1940). Composition of plankton and 412, 785.
units for reporting populations and production. Freire-Nordi, C. S., Vieira, A. A. H. and Nascimento,
In Proceedings of the 6th Pacific Science Congress, vol. O. R. (1998). Selective permeability of the
3, pp. 535–40. extracellular envelope of the microalga
Flöder, S. (1999). The influence of mixing frequency Spondylosium panduriforme (Chlorophyceae) as
and depth on phytoplankton mobility. Inter- revealed by electron paramagnetic resonance.
national Review of Hydrobiology, 84, 119–28. Journal of Phycology, 34, 631–7.
Flöder, S., Urabe, J. and Kawabata, Z. (2002). The Frenette, J.-J., Vincent, W. F., Legendre, L. et al. (1996).
influence of fluctuating light intensities on Biological responses to typhoon-induced mixing
species composition and diversity of natural in two morphologically distinct basins of Lake
phytoplankton communities. Oecologia, 133, Biwa. Japanese Journal of Limnology, 57, 501–10.
395–401. Friedman, M. M. (1980). Comparative morphology and
Fogg, G. E. (1971). Extracellular products of algae in functional significance of copepod receptors and
freshwater. Ergebnisse der Limnologie, 5, 1–25. oral structures. In Evolution and Ecology of
(1975). Algal Cultures and Phytoplankton Ecology, 2nd Zooplankton Communities, ed. W. C. Kerfoot,
edn. London: University of London Press. pp. 185–97. Hanover, NH: University of New
(1977). Excretion of organic matter by England Press.
phytoplankton. Limnology and Oceanography, 22, Frempong, E. (1984). A seasonal sequence of diel
576–7. distribution patterns for the planktonic
Fogg, G. E. and Thake, B. (1987). Algal Cultures and dinoflagellate Ceratium hirundinella in a eutropic
Phytoplankton Ecology, 3rd edn. London: University lake. Freshwater Biology, 14, 301–21.
of London Press. Fromme, P., Kern, J., Loll, B. et al. (2002). Functional
Forsberg, B. (1985). The fate of planktonic primary implications on the mechanism of the function
production. Limnology and Oceanography, 30, of photosystem II including water oxidation
807–19. based on the structure of photosystem II.
Fox, J. F. and Connell, J. H. (1979). Intermediate Philosophical Transactions of the Royal Society of
disturbance hypothesis. Science, 204, 1344–5. London B, 357, 1337–45.
Foy, R. H. and Gibson, C. E. (1982). Photosynthetic Fryer, G. (1987). Morphology and the classification of
characteristics of planktonic blue-green algae: the so-called Cladocera. Hydrobiologia, 145, 19–28.
changes in photosynthetic capacity and Fuhrman, J. A., Griffith, J. F. and Schwalbach, M. S.
pigmentation of Oscillatoria redekei Van Goor (2002). Prokaryotic and viral diversity patterns in
under high and low light intensities. British marine plankton. Ecological Research, 17,
Phycological Journal, 17, 183–93. 183–94.
Foy, R. H., Gibson, C. E. and Smith, R. V. (1976). The Gaarder, T. and Gran, H. H. (1927). Investigations of
influence of daylength, light intensity and the production of plankton in the Oslo Fjord.
temperature on the growth rates of planktonic Rapports et Procès-Verbaux des Réunions, Conseil
blue-green algae. British Phycological Journal, 11, international pour l’exploration de la Mer, 42, 1–48.
151–63. Gaedke, U. (1998). Functional and taxonomical
Fraga, F. (2001). Phytoplanktonic biomass synthesis: properties of the phytoplankton community of
application to deviations from Redfield large and deep Lake Constance: interannual
462 REFERENCES
variability and response to re-oligotrophication Garcı́a de Emiliani, M. O. (1993). Seasonal succession

(1979–93). Ergebnisse der Limnologie, 53, 119–41. of phytoplankton in a lake of the Paraná flood
Gaedke, U. and Sommer, U. (1986). The influence of plain, Argentina. Hydrobiologia, 264, 101–14.
the frequency of periodic disturbances on the Garcia-Pichel, F. and Castenholz, R. W. (1991).
maintenance of phytoplankton diversity. Characterisation and biological implications of
Oecologia, 71, 98–102. scytonemin, a cyanobacterial sheath pigment.
Gaedke, U. and Straile, D. (1994a). Seasonal changes in Journal of Phycology, 27, 395–409.
the quantitative importance of protozoans in a Gasol, J. M. and del Giorgio, P. A. (2000). Using flow
large lake: an ecosystem approach using cytometry for counting ntural planktonic
mass-balanced carbon-flow diagrams. Marine bacteria and understanding the structure of
Microbial Food Webs, 8, 163–88. planktonic bacterial communities. Scientia Marina,
(1994b). Seasonal changes of trophic transfer 64, 197–224.
efficiencies in a plankton food web derived from Gasol, J. M. and Pedrós-Alió, C. (1991). On the origin of
biomass size distributions and network analysis. deep chlorophyll maxima: the case of Lake Cisó.
Ecological Modelling, 75/76, 435–45. Verhandlungen der internationalen Vereinigung für
Gálvez, J. A., Niell, F. X. and Lucena, J. (1988). theoretische und angewandte Limnologie, 24, 1024–28.
Description and mechanism of formation of a Gasol, J. M., Guerrero, R. and Pedrós-Alió, C. (1992).
deep chlorophyll maximum due to Ceratium Spatial and temporal dynamics of a metalimnetic
hirundinella (O. F. Müller) Bergh. Archiv für Cryptomonas peak. Journal of Plankton Research, 14,
Hydrobiologie, 112, 143–56. 1565–79.
Ganf, G. G. (1974a). Diurnal mixing and the vertical Gattuso, J.-P., Peduzzi, S., Pizay, M.-D. and Tonolla, M.
distribution of phytoplankton in a shallow (2002). Changes in freshwater bacterial
equatorial lake (Lake George, Uganda). Journal of community composition during measurements of
Ecology, 62, 611–29. microbial and community respiration. Journal of
(1974b). Incident solar radiation and underwater Plankton Research, 24, 1197–206.
light penetration as factors controlling the Geider, R. J. and La Roche, J. (2000). Redfield revisited:
chlorophyll a content of a shallow equatorial lake variability of C : N : P in marine microalgae and
(Lake George, Uganda). Journal of Ecology, 62, its biochemical basis. European Journal of Phycology,
593–609. 37, 1–17.
(1975). Photosynthetic production and Geider, R. J. and MacIntyre, H. L. (2002). Physiology
irradiance–photosynthesis relationships of the and biochemistry of photosynthesis and algal
phytoplankton from a shallow equatorial lake (L. carbon acquisition. In Phytoplankton Productivity,
George, Uganda). Oecologia, 18, 165–83. ed. P. J. leB. Williams, D. N. Thomas and C. S.
(1980). Factors Controlling the Growth of Phytoplankton Reynolds, pp. 44–77. Oxford: Blackwell Science.
in Mount Bold Reservoir, South Australia, Technical Geider, R. J. and Osborne, B. A. (1992). Algal
Paper No. 48. Canberra: Australian Water Photosynthesis: The Measurement of Algal Gas
Research Council. Exchange. New York: Chapman and Hall.
Ganf, G. G. and Oliver, R. L. (1982) Vertical separation Geider, R. J., Delucia, E. H., Falkowski, P. G. et al.
of light and available nutrients as a factor (2001). Primary productivity of planet Earth:
causing replacement of green algae by blue-green biological determinants and physical constraints
algae in the plankton of a stratified lake. Journal in terrestrial and aquatic habitats. Global Change
of Ecology, 70, 829–44. Biology, 7, 849–82.
Ganf, G. G. and Shiel, R. J. (1985). Feeding behaviour Geller, W. and Müller, H. (1981). The filtration
and limb morphology of the cladocerans with apparatus of Cladocera: filter mesh sizes and
small intersetular distances. Australian Journal of their implications on food selectivity. Oecologia,
Marine and Freshwater Research, 36, 69–86. 49, 316–21.
Ganf, G. G., Heaney, S. I. and Corry, J. (1991). Light Gentry, D. R., Hernandez, V. J., Nguyen, L. H., Jensen,
absorption and pigment content in natural D. B. and Cashel, M. (1993). Synthesis of stationary
populations and cultures of a non-gas-vacuolate phase sigma factor sS is positively regulated by
cyanobacterium, Oscillatoria bourrelleyi ppGpp. Journal of Bacteriology, 175, 5710–13.
(= Tychonema bourrellyi) in Windermere. Journal of George, D. G. (1981). Zooplankton patchiness. Report of
Plankton Research, 13, 1101–21. the Freshwater Biological Association, 49, 32–44.
REFERENCES 463
(2002). Regional scale influences on the long-term (1975). Effect of zooplankton grazing on
dynamics of plankton. In Phytoplankton photosynthetic activity and composition of the
Productivity, ed. P. J. leB. Williams, D. N. Thomas phytoplankton. Verhandlungen der internationalen
and C. S. Reynolds, pp. 265–90. Oxford: Blackwell Vereinigung für theoretische und angewandte
Science. Limnologie, 19, 1490–97.
George, D. G. and Edwards, R. W. (1973). Daphnia (1977). Food size selection and seasonal succession
distributions within Langmuir circulations. of filter-feeding zooplankton in an eutrophic lake.
Limnology and Oceanography, 18, 798–800. Ekologia Polska, Seria A, 25, 179–225.
(1976). The effect of wind on the distribution of (1980). Filtering rates, food size selection and
chlorophyll a and crustacean zooplankton in a feeding rates in cladocerans: another aspect of
shallow eutrophic reservoir. Journal of Applied interspecific competition in filter-feeding
Ecology, 13, 667–90. zooplankton. In Evolution and Ecology of Zooplankton
George, D. G. and Harris, G. P. (1985). The effect of Communities, ed. W. C. Kerfoot, pp. 282–91.
climate on long-term changes in the crustacean Hanover, NH: University of New England Press.
zooplankton biomass of Lake Windermere, UK. (1990). Food thresholds and body size in
Nature, 316, 536–9. cladocerans. Nature, 343, 638–40.
George, D. G. and Heaney, S. I. (1978). Factors (2003a). Zooplankton. In The Lakes Handbook, vol. 1,
influencing the spatial distribution of ed. P. O’Sullivan and C. S. Reynolds, pp. 461–516.
phytoplankton in a small, productive lake. Journal Oxford: Blackwell Science.
of Ecology, 66, 133–55. (2003b). Between Hazards of Starvation and Risk of
George, D. G. and Taylor, A. H. (1995). UK lake Predation: The Ecology of Offshore Animals.
plankton and the Guf Stream. Nature, 378, 139. Oldendorf: ECI.
Gerloff, G. C., Fitzgerald, G. P. and Skoog, F. (1952). Gliwicz, Z. M. and Hillbricht-Ilkowska, A. (1975).
The mineral nutrition of Microcystis aeruginosa. Ecosystem of Mikoajskie Lake: elimination of
American Journal of Botany, 39, 26–32. phytoplankton biomass and is subsequent fate in
Gervais, F., Padisák, J. and Koschel, R. (1997). Do light the lake through the year. Polskie Archiwum
quality and low nutrient concentration favour Hydrobiologii, 22, 39–52.
picocyanobacteria below the thermocline of the Gliwicz, Z. M. and Preis, A. (1977). Can planktivorous
oligotrophic Lake Stechlin? Journal of Plankton fish keep in check planktonic crustacean
Research, 19, 771–81 populations? A test of the size–efficiency
Gibson, C. E. (1971). Nutrient limitation. Journal of the hypothesis. Ekologia Polska, Seria A, 25, 567–91.
Water Pollution Control Federation, 43, Gliwicz, Z. M. and Siedlar, E. (1980). Food-size
2436–40. limitation and algae interfering with food
Gibson, C. E., Wood, R. B., Dickson, E. L. and Jewson, collection in Daphnia. Archiv für Hydrobiologie, 88,
D. M. (1971). The succession of phytoplankton in 155–77.
L. Neagh, 1968–1971. Mitteilungen der Glover, H. E. (1977). Effects of iron deficiency on
internationalen Vereinigung für theoretische und Isochrysis galbana (Chrysophyceae) and
angewandte Limnologie, 19, 146–60. Phaeodactylum cornutum (Bacillariophyceae). Journal
Gismervik, I. (1997). Stoichiometry of some marine of Phycology, 13, 208–12.
planktonic cladocerans. Journal of Plankton Rearch, Goericke, R. (2002). Bacteriochlorophyll a in the
19, 279–85. ocean: is anoxygenic bacterial photosynthesis
Gjessing, E. T. (1970). Ultrafiltration of aquatic humus. important? Limnology and Oceanography, 47, 290–5.
Environmental Science and Technology, 4, 437–8. Goldman, C. R. (1960). Molybdenum as a factor
Gleason, H. (1917). The structure and development of limiting primary productivity in Castle Lake,
the plant association. Bulletin of the Torrey Botanical California. Science, 132, 1016–17.
Club, 44, 463–81. (1964). Primary productivity and micro-nutrient
(1927). Further views on the succession concept. limiting factors in some North American and
Ecology, 8, 299–326. New Zealand lakes. Verhandlungen der
Gleick, J. (1988). Chaos: Making a New Science. London: internationalen Vereinigung für theoretische und
Heinemann. angewandte Limnologie, 15, 365–74.
Gliwicz, Z. M. (1969). The food sources of lake Goldman, C. R. and Jassby, A. D. (2001). Primary
zooplankton. Ekologia Polska, Seria B, 15, 205–23. productivity, phytoplankton and nutrient status
464 REFERENCES
in Lake Baikal. In The Great Lakes of the World Gosselain, V. and Descy, J.-P. (2000). Phytoplankton of
(GLOW): Food Web, Health and Integrity, ed. M. the River Meuse (1994–1996): form and size
Munawar and R. E. Hecky, pp. 111–25. Leiden: structure analysis. Verhandlungen der
Backhuys. internationalen Vereinigung für theoretische und
Goldman, J. C. (1977). Steady-state growth of angewandte Limnologie, 27, 1031.
phytoplankton in continuous culture: comparison Gosselain, V., Hamilton, P. B. and Descy, J.-P. (2000).
of internal and external nutrient concentrations. Estimating phytoplankton carbon from
Journal of Phycology, 13, 251–8. microscopic counts: an application for riverine
(1980). Physiological processes, nutrient availability systems. Hydrobiologia, 438, 75–90.
and the concept of relative growth rate in marine Goussias, C., Boussac, A. and Rutherford, A. W. (2002).
phytoplankton ecology. In Primary Productivity in Photosystem II and photosynthetic oxidation of
the Sea, ed. P. G. Falkowski, pp. 179–94. New York: water: an overview. Philosophical Transactions of the
Plenum Press. Royal Society of London B, 357, 1369–81.
Goldman, J. C. and Dennett, M. R. (2000). Growth of Grace, J. B. (1991). A clarification of the debate
marine bacteria in batch and continuous culture between Grime and Tilman. Functional Ecology, 5,
under carbon and nitrogen limitation. Limnology 583–7.
and Oceanography, 45, 789–800. Graham, M. (1943). The Fish Gate. London: Faber.
Goldman, J. C., McCarthy, J. J. and Peavey, D. G. (1979). Gran, H. H. (1912). Pelagic plant life. In The Depths of
Growth rate influence on the chemical the Ocean, ed. J. Murray and J. Hjort, pp. 307–87.
composition of phytoplankton in oceanic waters. London: Macmillan.
Nature, 279, 210–15. (1931). On the conditions for the production of
Golterman, H. L., Bakels, C. C. and Jakobs-Möglin, J. plankton in the sea. Rapports du Conseil
(1969). Availability of mud phosphates for the internationale pour l’exploration de la Mer, 75,
growth of algae. Verhandlungen der internationalen 37–46.
Vereinigung für theoretische und angewandte Gray, C. B., Neilsen, M., Johannsson, O. et al. (1994).
Limnologie, 17, 467–79. Great lakes and the St. Lawrence lowland lakes.
González, N., Anadón, R. and Viesca, L. (2003). Carbon In The Book of Canadian Lakes, ed. R. J. Allan, M.
flux through the microbial community in a Dickman, C. B. Gray and V. Cromie, pp. 14–98.
temperate sea during summer: role of bacterial Burlington: Canadian Association on Water
metabolism. Aquatic Microbial Ecology, 33, 117–26. Quality.
Gorham, E. (1958). Observations on the formation and Greenwood, P. H. (1963). A History of Fishes. London:
breakdown of the oxidised microzone at the mud Ernest Benn.
surface in lakes. Limnology and Oceanography, 3, Grey, J., Jones, R. I. and Sleep, D. (2001). Seasonal
291–8. changes in the importance of the source of
Gorham, E. and Boyce, F. M. (1989). Influence of lake organic matter to the diet of zooplankton in Loch
surface area and depth upon thermal Ness, as indicated by stable isotope analysis.
stratification and the depth of the summer Limnology and Oceanography, 46, 505–13.
thermocline. Journal of Great Lakes Research, 15, Griffiths, B. M. (1939). Early references to water
233–45. blooms in British lakes. Proceedings of the Linnean
Gorham, E., Lund, J. W. G., Sanger, J. E. and Dean, Society of London, 151, 12–19.
W. E. (1974). Some relationships between algal Grime, J. P. (1973). Competitive exclusion in
standing crop, water chemistry and sediment herbaceous vegetation. Nature, 242, 344–7.
chemistry in the English lakes. Limnology and (1977). Evidence for the existence of three primary
Oceanography, 19, 601–17. strategies in plants and its relevance to ecological
Gorham, P. R., MacLachlan, J., Hammer, U. T. and Kim, and evolutionary theory. American Naturalist, 111,
W. W. (1964). Isolation and culture of toxic strains 1169–94.
of Anabaena flos-aquae (Lyngb.) Bréb. Verhandlungen (1979). Plant Strategies and Vegetation Processes.
der internationalen Vereinigung für theoretische und Chichester: John Wiley.
angewandte Limnologie, 15, 796–804. (2001). Plant Strategies, Vegetation Processes and
Gorlenko, V. M. and Kuznetsov, S. I. (1972). Über die Ecosystem Properties. Chichester: John Wiley.
photosynthesirenden Bakterien der Grimm, M. P. and Backx, J. J. G. M. (1990). The
Kononjer-Sees. Archiv für Hydrobiologie, 70, 1–13. restoration of shallow lakes and the role of
REFERENCES 465
northern pike, aquatic vegetation and nutrient Haberyan, K. A. and Hecky, R. E. (1987). The late
concentration. Hydrobiologia, 200/201, 557–66. Pleistocene and Holocene stratigraphy and
Gross, F. and Zeuthen, E. (1948). The buoyancy of palaeolimnology of Lakes Kivu and Tanganyika.
planktonic diatoms. a problem of cell physiology. Palaeo, 61, 169–97.
Proceedings of the Royal Society of London B, 135, Häder, D.-P. (1986). Effect of solar and artificial
382–9. UV-radiation on motility and phototaxis in the
Grossman, A. R., Schaefer, M. R., Chiang, G. C. and flagellate Euglena gracilis. Photochemistry and
Collier, J. L. (1993). The phycobilisome, a Photobiology, 44, 651–6.
light-harvesting complex responsive to Haeckel, E. (1904). Kunstformen der Natur. Leipzig:
environmental conditions. Microbiological Reviews, Bibliographisches Institut.
57, 725–49. Haffner, G. D., Harris, G. P. and Jarai, M. K. (1980).
Grover, J. P. and Chrzanowski, T. H. (2000). Seasonal Physical variability and phytoplankton
patterns of substrate utilisation by communities. III. Vertical structure in
bacterioplankton: case studies in four temperate phytoplankton populations. Archiv für
lakes of different latitudes. Aquatic Microbial Hydrobiologie, 89, 363–81.
Ecology, 23, 41–54. Halbach, U. and Halbach-Keup, G. (1974). Quantitaive
Grünberg, H. (1968). Über das Zeta-Potential zweier Beziehungen zwischen Phytoplankton und der
synchron kultiviert Chlorella-Stämme. Archiv für Populationsdynamik des Rotators Brachionus
Hydrobiologie (Suppl.), 33, 331–62. calyciflorus Pallas. Befunde aus
Guerrero, R. and Mas-Castellà, J. (1995). The problem Laboratoriumsexperimenten und Freiland
of excess and/or limitation of the habitat untersuchungen. Archiv für Hydrobiologie, 73,
conditions: do natural assemblages exist? In 273–309.
Molecular Ecology of Aquatic Microbes ed. I. Joint, Halldal, P. (1953). Phytoplankton investigations from
pp. 191–203. Berlin: Springer-Verlag weather ship M in the Norwegian Sea (1948–49).
Guikema, J. A. and Sherman, L. A. (1984). Influence of Hvalrdets Skrifter, 38, 1–91.
iron deprivation on the membrane composition Hambright, K. D., Gophen, M. and Serruya, S. (1994).
of Anacystis nidulans. Plant Physiology, 74, 90–5. Influence of long-term climatic changes on the
Guillen, O., Rojas de Mendiola, B. and Izaguirre de stratification of a subtropical, warm monomictic
Rondan, R. (1971) Primary productivity and lake. Limnology and Oceanography, 39, 1233–42.
phytoplankton. In Fertility of the Sea, vol. 1, ed. Hammer, U. T., Walker, K. F. and Williams, W. D.
J. D. Costlow, pp. 187–96. New York: Gordon and (1973). Derivation of daily phytoplankton
Breach. production estimates from short-term
Guinasso, N. L. and Schink, D. R. (1975). Quantitative experiments in some shallow eutrophic
estimates of biological mixing rates in abyssal Australian saline lakes. Australian Journal of Marine
sediments. Journal of Geophysical Research, 80, and Freshwater Research, 24, 259–66.
3022–43. Haney, J. F. (1973). An in situ examination of the
Gulati, R. D., Lammens, E. H. R. R., Meijer, M.-L. and grazing activities of natural zooplankton
Donk, E. van (1990). Biomanipulation: Tool for Water communities. Archiv für Hydrobiologie, 72, 88–132.
Management. Dordrecht: Kluwer. Hanson, M. J. and Stefan, H. G. (1984). Side effects of
Gurung, T. B. and Urabe, J. (1999). Temporal and 58 years of copper sulphate treatment of the
vertical differences in factors limiting growth Fairmont Lakes, Minnesota. Water Resources
rate of heterotrophic bacteria in Lake Biwa. Bulletin, 20, 889–900.
Microbial Ecology, 38, 136–45. Happey-Wood, C. M. (1988). Ecology of freshwater
Gurung, T. B., Urabe, J. and Nakanishi, M. (1999). planktonic green algae. In Growth and Reproductive
Regulation of the relationship between Strategies of Freshwater Phytoplankton, ed. C. D.
phytoplankton Scenedesmus acutus and Sandgren, pp. 175–226. Cambridge: Cambridge
heterotrophic bacteria by the balance of light University Press.
and nutrients. Aquatic Microbial Ecology, 17, Hardin, G. (1960). The competitive exclusion principle.
27–35. Science, 131, 1292–7.
Gusev, M. V. (1962). Effect of dissolved oxygen on the Harding, W. R. (1997). Phytoplankton primary
development of blue green algae. Doklady production in a shallow, well-mixed, hypertrophic
Akademia Nauk SSSR, 147, 947–50. (In Russian.) South African Lake. Hydrobiologia, 344, 87–102.
466 REFERENCES
Hardy, A. C. (1939). Ecological investigations with the Haworth, E. Y. (1976). The changes in composition of
Continuous Plankton Recorder: objects, plans and the diatom assemblages found in the surface
methods. Hull Bulletin of Marine Ecology, 1, 1–57. sediments of Blelham Tarn in the English Lake
(1956). The Open Sea: The World of Plankton. London: District during 1973. Annals of Botany, 40,
Collins. 1195–205.
(1964). The Open Sea, Part 1, The World of Plankton. (1985). The highly nervous system of the English
revd edn. London: Collins. Lakes: aquatic sensitivity to external changes, as
Harris, G. P. (1978). Photosynthesis, productivity and demonstrated by diatoms. Report of the Freshwater
growth. Ergebnisse der Limnologie, 10, 1–163. Biological Association, 53, 60–79.
(1986). Phytoplankton Ecology: Structure, Function and Hayes, P. K. and Walsby, A. E. (1986). The inverse
Fluctuation. London: Chapman and Hall. correlation between width and strength of gas
Harris, G. P. and Lott, J. N. A. (1973). Observations of vesicles in Cyanobacteria. British Phycological
Langmuir circulations in Lake Ontario. Limnology Journal, 21, 191–7.
and Oceanography, 18, 584–9. Haygarth, P. M. (1997). Agriculture as a source of
Harris, G. P. and Piccinin, B. B. (1977). Photosynthesis phosphorus transfer to water: sources and
by natural phytoplankton populations. Archiv für pathways. Scope Newsletter, 21, 1–16.
Hydrobiologie, 80, 405–57. Hayward, T. L., Venrick, E. L. and McGowan, J. A. (1983).
Harris, G. P., Heaney, S. I. and Talling, J. F. (1979). Environmental heterogeneity and plankton
Physiological and environmental constraints in community structure in the central North Pacific.
the ecology of the planktonic dinoflagellate Journal of Marine Research, 41, 711–29.
Ceratium hirundinella. Freshwater Biology, 9, Healey, F. P. (1973). Characteristics of phosphorus
413–28. deficiency in Anabaena. Journal of Phycology, 9,
Hart, R. C. (1996). Naupliar and copepodite growth 383–94.
and survival of two freshwater calanoids at Heaney, S. I. (1976). Temporal and spatial distribution
various food levels: demographic contrasts, of the dinoflagellate Ceratium hirundinella O. F.
similarities and food needs. Limnology and Müller within a small productive lake. Freshwater
Oceanography, 41, 648–58. Biology, 6, 531–42.
Hart, T. J. (1934). On the phytoplankton of the Heaney, S. I. and Furnass, T. I. (1980). Laboratory
Southwest Atlantic and the Bellingshausen Sea, models of diel vertical migration in the
1929–1931. Discovery Reports, 8, 1–268. dinoflagellate, Ceratium hirundinella. Freshwater
Hartmann, H. J. and Kunkel, D. D. (1991). Mechanisms Biology, 10, 163–70.
of food selection in Daphnia. Hydrobiologia, 225, Heaney, S. I. and Talling, J. F. (1980a). Dynamic aspects
129–54. of dinoflagellate distribution patterns in a small,
Hartmann, H. J., Taleb, H., Aleya, L. and Lair, N. (1993). productive lake. Journal of Ecology, 68, 75–94.
Predation on ciliates by the suspension-feeding (1980b). Ceratium hirundinella: ecology of a complex,
calanoid copepod Acanthodiaptomus denticornis. mobile and successful plant. Report of the
Canadian Journal of Fisheries and Aquatic Sciences, 50, Freshwater Biological Association, 48, 27–40.
1382–93. Heaney, S. I., Chapman, D. V. and Morison, H. R.
Harvey J. G. (1976). Atmosphere and Ocean: Our Fluid (1981). The importance of the cyst stage in the
Environments, London: Artemis Press. seasonal growth of the dinoflagellate, Ceratium
Hastwell, G. T. and Huston, M. A. (2001). On hirundinella, in a small productive lake. British
disturbance and diversity: a reply to Mackey and Phycological Journal, 16, 136.
Currie. Oikos, 92, 367–71. Heaney, S. I., Smyly, W. J. P. and Talling, J. F. (1986).
Havens, K. E., Aumen, N. G., James, R. T. and Smith, Interactions of physical, chemical and biological
V. H. (1996). Rapid ecological changes in a large processes in depth and time within a productive
subtropical lake undergoing cultural English lake during summer stratification.
eutrophication. Ambio, 25, 150–5. Internationale Revue des gesamten Hydrobiologie, 71,
Hawlay, G. R. W. and Whitton, B. A. (1991). Seasonal 441–94.
changes in chlorophyll-containing picoplankton Heaney, S. I., Lund, J. W. G., Canter, H. M. and Gray, K.
populations of ten lakes in northern England. (1988). Population dynamics of Ceratium spp. in
Internationale Revue des gesamten Hydrobiologie, 76, three English lakes, 1945–85. Hydrobiologia, 161,
545–54. 133–48.
REFERENCES 467
Heaney, S. I., Canter, H. M. and Lund, J. W. G. (1990). (1990). Distribution of the world’s largest lakes. In
The ecological significance of grazing planktonic Large Lakes: Ecological Structure and Function, ed.
populations of cyanobacteria by the ciliate M. M. Tilzer and C. Serruya, pp. 3–38. New York:
Nassula. New Phytologist, 114, 247–63. Springer-Verlag.
Heaney, S. I., Parker, J. E., Butterwick, C. and Clarke, Herrmann, J., Degerman, E., Gerhardt, A. et al. (1993).
K. J. (1996). Interannual variability of algal Acid-stress effects on stream biology. Ambio, 22,
populations and their influence on lake 298–307.
metabolism. Freshwater Biology, 35, 561–77. Hessen, D. O. and Lyche, A. (1991). Inter- and
Hecky, R. E. (1993). The eutropication of Lake Victoria. intra-specific variations in zooplankton elemental
Verhandlungen der internationalen Vereinigung für composition. Archiv für Hydrobologie, 121,
theoretische und angewandte Limnologie, 25, 39–48. 343–53.
Hecky, R. E. and Fee, E. J. (1981) Primary production Hessen, D. O. and van Donk, E. (1993). Morphological
and rates of algal growth in Lake Tanganyika. changes in Scenedesmus induced by substances
Limnology and Oceanography, 26, 532–46. released from Daphnia. Archiv für Hydrobiologie,
Hecky, R. E. and Kling, H. J. (1981). The phytoplankton 127, 129–40.
and protozooplankton of the euphotic zone of Hildrew, A. G. and Townsend, C. R. (1987).
Lake Tanganyika: species composition, biomass, Organisation in freshwater benthic communities.
chlorophyll content and spatio-temporal In Organization of Communities, Past and Present, ed.
distribution. Limnology and Oceanography, 26, J. H. R. Gee and P. S. Giller, pp. 347–71. Oxford:
548–64. Blackwell Scientific Publications.
Hecky, R. E., Spigel, R. H. and Coulter, G. (1991). The Hill, R. and Bendall, F. (1960). Function of two
nutrient regime. In Lake Tanganyika and its Life, ed. cytochrome components in chloroplasts: a
G. Coulter, pp. 76–89. Oxford: Oxford University working hypothesis. Nature, 186, 136–7.
Press. Hillbricht-Ilkowska, A. and Weglenska, T. (1978).
Hegewald, E. (1972). Untersuchungen zum Experimentally increased fish stock in the
Zetapotential von Planktonalgen. Archiv für pond-type Lake Warniak. VII. Numbers, biomass
Hydrobiologie (Suppl.), 42, 14–90. and production of zooplankton. Ekologia Polska,
Heinbokel, J. F. (1986). Occurrence of Richelia Seria A, 21, 533–51.
intracellularis (Cyanophyta) within the diatoms Hillbricht-Ilkowska, A., Spodniewska, I. and Wegleska,
Hemiaulus haukii and H. membranaceus of Hawaii. T. (1979). Changes in the phytoplankton–
Journal of Phycology, 22, 399–403. zooplankton relationship connected with the
Heller, M. (1977). The phased division of the eutrophication of lakes. Symposia Biologica
freshwater dinoflagellate Ceratium hirundinella and Hungariae, 19, 59–75.
its use as a method for assessing growth in a Hilton, J. (1985). A conceptual framework for
natural population. Freshwater Biology, 7, 527–33. predicting the occurrence of sediment focussing
Henrikson, L. and Brodin, Y.-W. (1995). Liming of and sediment redistribution in small lakes.
surface waters in Sweden: a synthesis. In Liming of Limnology and Oceanography, 30, 1131–43.
Acidified Surface Waters: A Swedish Synthesis, ed. L. Hilton, J., Rigg, E. and Jaworski, G. H. M. (1989). Algal
Henrikson and Y.-W. Brodin, pp. 1–44. Berlin: identification using in vivo fluorescence spectra.
Springer-Verlag. Journal of Plankton Research, 11, 65–74.
Henrikson, L. and Oscarson, H. G. (1981). Corixids Hindar, A., Henrikson, L., Sandøy, S. and Romundstad,
(Hemiptera – Heteroptera): the new top predators A. J. (1998). Critical load concept to set
of acidified lakes. Verhandlungen der internationalen restoration goals for liming acidified Norwegian
Vereinigung für theoretische und angewandte waters. Restoration Ecology, 6(4), 1–13.
Limnologie, 21, 1616–20. Hino, K., Tundisi, J. G. and Reynolds, C. S. (1986).
Hensen, V. (1887). Über die Bestimmung des Planktons Vertical distribution of phytoplankton in a
oder des im Meere treibenden materiels an stratified lake (Lago Dom Helvécio, south eastern
Pflanzen und Tieren. Bericht des deutschen Brasil), with special reference to the metalimnion.
wissenschaftlichen Kommission für Meerenforschung, 5, Japanese Journal of Limnology, 47, 239–46.
1–107. Hino, S., Mikami, H., Arisue, J. et al. (1998).
Herdendorf, C. E. (1982). Large lakes of the world. Limnological characteristics and vertical
Journal of Great Lakes Research, 8, 379–412. distribution of phytoplankton in oligotrophic
468 REFERENCES
Lake Akan-Panke. Japanese Journal of Limnology, 59, Horne, A. J. and Commins, M. L. (1987). Macronutrient
263–79. controls on nitrogen fixation in planktonic
Hobson, L. A., Morris, W. J. and Pirquet, K. T. (1976). cyanobacterial populations. New Zealand
Theoretical and experimental analysis of the 14 C Journal of Marine and Freshwater Research, 21,
technique and its use in studies of primary 413–23.
production. Journal of the Fisheries Research Board of Hosper, S. H. (1984). Restoration of Lake Veluwe, The
Canada, 33, 1715–21. Netherlands, by reduction of phosphorus loading
Hoge, F. and Swift, R. (1983). Airborne dual laser and flushing. Water Science and Technology, 17,
excitation and mapping of phytoplankton 757–68.
photopigments in a Gulf-Stream warm core ring. Hosper, S. H. and Meijer, M.-L. (1993). Biomanipulation:
Applied Optics, 22, 2272–81. will it work in your lake? A simple test for the
Holland, R. (1993). Changes in planktonic diatoms and assessment of chances for clear water following
water transparency in Hatchery Bay, Bass Island drastic fish-stock reduction in shallow, eutrophic
area, western Lake Erie, since the establishment lakes. Ecological Engineering, 2, 63–72.
of the zebra mussel. Journal of Great Lakes Research, Hosper, S. H., Meijer, M.-L. and Walker, P. A. (1992).
16, 121–9. Handleidung actief biologisch Beheer: Beoordeeling van
Holligan, P. M. and Harbour, D. S. (1977). The vertical de mogelijkhedenvan Visstandbeheer bij het Herstel van
distribution and succession of phytoplankton in meren en plassen. Lelystad: Rijksinstituut voor
the Western English Channel in 1975 and 1976. integraal Zoetwaterbeheer en
Journal of the Marine Biological Association of the Afwalwaterdehanling – Organasitie ter
United Kingdom, 57, 1075–93. Verbetering van de Binnenvisserij.
Holm, N. P. and Armstrong, D. E. (1981). Role of Howard, A. and Pelc, S. R. (1951). Nuclear
nutrient limitation and competition in incorporation of 32 P as demonstrated by auto-
controlling the populations of Asterionella formosa radiographs. Experimental Cell Research, 2,
and Microcystis aeruginosa in semicontinuous 178–87.
culture. Limnology and Oceanography, 26, 622–34. Howarth, R. W. and Michaels, A. F. (2000). The
Holmes, R. W. (1956). The annual cycle of measurement of primary production in aquatic
phytoplankton in the Labrador Sea, 1950–51. ecosystems. In Methods in Ecosystem Science, ed.
Bulletin of the Bingham Oceanographic College, 16, O. E. Sala, R. B. Jackson, H. A. Mooney and R. W.
1–74. Howarth, pp. 72–85. New York: Springer-Verlag.
Holmgren, S. (1968). Fytoplankton och primarproduktion i Howarth, R. W., Jensen, H. J., Marino, R. and Postma,
Nedre Laksjon i Abisko National Park under aren H. (1995). Transport to and processing of P in
1962–1965. Uppsala: Limnologiska Institutionene. near-shore and oceanic waters. In Phosphorus in the
(1983). Phytoplankton Biomass and Algal Composition in Global Environment, ed. H. Tiessen, pp. 323–45.
Natural, Fertilised and Polluted Subarctic Lakes, Acta Chichester: John Wiley.
Universitatis Upsaliensis No. 674. Uppsala: Hrbáček, J., Dvořáková, M., Kořı́nek, V. and
University of Uppsala. Procházková, L. (1961). Demonstration of the
Holzmann, R. (1993). Seasonal fluctuations in the effect of fish stock on the species composition of
diversity and compositional stability of the zooplankton and the intensity of metabolism
phytoplankton communities in small lakes in of the whole plankton association. Verhandlungen
upper Bavaria. Hydrobiologia, 249, 101–9. der internationalen Vereinigung für theoretische und
Hoogenhout, H. and Amesz, J. (1965). Growth rates of angewandte Limnologie, 14, 192–5.
photosynthetic microorganisms in laboratory Huber, G. and Nipkow, F. (1922). Experimentelle
cultures. Archiv für Mikrobiologie, 50, 10–25. Untersuchungen über die Entwicklung von
Hopson, A. J. (1982). Lake Turkana: A Report of the Ceratium hirundinella. Zeitschrift für Botanik, 14,
Findings of the Lake Turkana Project, 1972–1975. 337–71.
London: Government of Kenya and Ministry of Hudson, J., Schindler, D. W. and Taylor, W. (2000).
Overseas Development. Phosphate concentrations in lakes. Nature, 406,
Horne, A. J. (1979). Nitrogen fixation in Clear Lake, 54–6.
California. IV. Diel studies of Aphanizomenon and Hughes, J. C. and Lund, J. W. G. (1962). The rate of
Anabaena blooms. Limnology and Oceanography, 24, growth of Asterionella in relation to its ecology.
329–41. Archiv für Mikrobiologie, 42, 117–29.
REFERENCES 469
Huisman, J. and Sommeijer, B. (2002). Maximal Huszar, V. L. M. and Caraco, N. (1998). The
sustainable sinking velocity of phytoplankton. relationship between phytoplankton composition
Marine Ecology – Progress Series, 244, 39–48. and physical-chemical variables: a comparison of
Huisman, J. and Weissing, F. J. (1994). Light-limited taxonomic and morphological–functional
growth and competition for light in well-mixed approaches in six temperate lakes. Freshwater
aquatic environments: an elementary model. Biology, 40, 1–18.
Ecology, 75, 507–20. Huszar, V., Kruk, C. and Caraco, N. (2003). Steady-state
(1995). Competition for nutrients and light in a assemblages of phytoplankton in four temperate
mixed water column: a theoretical analysis. lakes (NE USA). Hydrobiologia, 502, 97–109.
American Naturalist, 146, 536–64. Hutchinson, G. E. (1941). Limnological studies in
(2001). Fundamental unpredictability in multi- Connecticut. IV. Mechanisms of intermediary
species competition. American Naturalist, 157, metabolism in stratified lakes. Ecological
488–94. Monographs, 11, 21–60.
Huisman, J., van Oostveen, P. and Weissing, F. J. (1999). (1959). Homage to Santa Rosalia: or, why are there
Critical depth and critical turbulence: two so many kinds of animals? American Naturalist, 93,
different mechanisms for the development of 145–59.
phytoplankton blooms. Limnology and (1961). The paradox of the plankton. American
Oceanography, 44, 1781–7. Naturalist, 95, 137–47.
Huisman, J., Johanson, A. M., Folmer, E. O. and (1967). A Treatise on Limnology, vol. 2, Introduction to
Weissing, P. J. (2001). Towards a solution of the Lake Biology and the Limnoplankton. New York: John
plankton paradox: the importance of physiology Wiley.
and life history. Ecology Letters, 4, 408–11. Ibelings, B. W. and Maberly, S. C. (1998). Photo-
Huisman, J., Arrayas, M., Ebert, U. and Sommeijer, B. inhibition and the availability of inorganic
(2002). How do sinking phytoplankton species carbon restrict photosynthesis by surface blooms
manage to persist? American Naturalist, 159, of Cyanobacteria. Limnology and Oceanography, 43,
245–54. 408–19.
Hulburt, E. M., Ryther, J. H. and Guillard, R. R. L. Ibelings, B. W., Kroon, B. M. A. and Mur, L. (1994).
(1960). The phytoplankton of the Sargasso Sea off Acclimation of photosystem II in a cyano-
Bermuda. Journal du Conseil permanent international bacterium and a eukaryotic green alga to high
de l’exploration de la Mer, 25, 115–27. and fluctuating photosynthetic photon flux
Humphries, S. E. and Imberger, J. (1982). The Influence densities, simulating light regimes induced by
of the Internal Structure and Dynamics of Burrinjuck mixing in lakes. New Phytologist, 128, 407–24.
Reservoir on Phytoplankton Blooms. Nedlands: Centre Ichimura, S., Nagasawa, S. and Tanaka, T. (1968). On
for Water Research. the oxygen and chlorophyll maxima found in the
Huntley, M. E. and Lopez, M. D. G. (1992). metalimnion of a mesotrophic lake. Botanical
Temperature-dependent production of marine Magazine, Tokyo, 81, 1–10.
copepods: a global synthesis. American Naturalist, Ihlenfeldt, M. J. A. and Gibson, J. (1975). Phosphate
140, 201–42. utilisation and alkaline phosphatase activity in
Huntsman, S. A. and Sunda, W. G. (1980). The role of Anacystis nidulans (Synechococcus). Archives of
trace metals in regulating phytoplankton growth. Microbiology, 102, 23–8.
In The Physiological Ecology of Phytoplankton, ed. I. Imberger, J. (1985). Thermal characteristics of
Morris, pp. 285–328. Oxford: Blackwell Scientific standing waters: an illustration of dynamic
Publications. processes. Hydrobiologia, 125, 7–29.
Hurrell, J. W. (1995). Decadal trends in the North (ed.) (1998). Physical Processes in Lakes and Oceans,
Atlantic Oscillation: regional temperature and Coastal and Estuarine Studies No. 54.
precipitation. Science, 269, 676–9. Washington, DC: American Geophysical Union.
Huston, M. (1979). A general theory of species Imberger, J. and Hamblin, P. F. (1982). Dynamics of
diversity. American Naturalist, 113, 81–101. lakes, reservoirs and cooling ponds. Annual Review
(1999). Local processes and regional patterns: of Fluid Mechanics, 14, 153–87.
appropriate scales for understanding variation in Imberger, J. and Ivey, G. N. (1991). On the nature of
the diversity of plants and animals. Oikos, 86, turbulence in a stratified fluid. II. Application to
393–401. lakes. Journal of Physical Oceanography, 21, 659–80.
470 REFERENCES
Imberger, J. and Spigel, R. H. (1987). Circulation and a zooplankton community to experimental

mixing in Lake Rotongaio and Lake Okaro under manipulations of planktivory. Ecology, 80,
conditions of light to moderate winds: 1405–21.
preliminary results. New Zealand Journal of Marine Ives, A. R., Klug, J. L. and Gross, K. (2000). Stability and
and Freshwater Research, 21, 515–19. species richness in complex communities. Ecology
Imboden, D. M. and Wüest, A. (1995). Mixing Letters, 3, 399–411.
mechanisms in lakes. In Lakes: Chemistry, Geology Ives, K. J. (1956). Electrokinetic phenomena of
and Physics, 2nd edn, ed. A. Lerman, D. M. planktonic algae. Proceedings of the Society for Water
Imboden and J. R. Gat, pp. 83–138. New York: Treatment and Examination, 5, 41–58.
Springer-Verlag. Izaguirre, I., Mataloni, G., Allende, L. and Vinocur, A.
Irish, A. E. (1980). A modified 1-m Friedinger sampler: (2001). Summer fluctuations of microbial comm-
a description and some selected results. unities in a eutrophic lake – Cierva Point,
Freshwater Biology, 10, 135–9. Antarctica. Journal of Plankton Research, 23,
Irish, A. E. and Clarke, R. T. (1984). Sampling designs 1095–1109.
for the estimation of phytoplankton abundance Jackson, J. B. C. and Sala, E. (2001). Unnatural oceans.
in limnetic environments. British Phycological Scientia Marina, 65 (Suppl. 2), 273–81.
Journal, 19, 57–66. Jacobsen, B.A. and Simonsen, P. (1993). Disturbance
Irvine, K., Patterson, G., Allison, E. H., Thompson, A. B. events affecting phytoplankton biomass, com-
and Menz, A. (2001). The pelagic system of Lake position and species diversity in a shallow,
Malawi, Africa: trophic structure and current eutrophic, temperate lake. Hydrobiologia, 249,
threats. In The Great Lakes of the World (GLOW): Food 9–14.
Web, Health and Integrity, ed. M. Munawar and Jähnichen, S., Petzoldt, T. and Benndorf, J. (2001).
R. E. Hecky, pp. 3–30. Leiden: Backhuys. Evidence of control of microcystin dynamics in
Irwin, W. H., Symons, J. M. and Roebuck, G. G. (1969). Bautzen Reservoir (Germany) by cyanobacterial
Water quality in impoundments and population growth rates and dissolved inorganic
modifications from destratification. In Water carbon. Archiv für Hydrobiologie, 150, 177–96.
Quality Behaviour in Reservoirs, ed. J. M. Symons, James, W. F., Taylor, W. D. and Barko, J. W. (1992).
pp. 363–88. Cincinnati, OH: US Department of Production and vertical migration of Ceratium
Health, Education and Welfare. hirundinella in relation to phosphorus availability
Ishikawa, K., Kumagai, M., Vincent, W. F., Tsujimura, in Eau Galle Reservoir, Wisconsin. Canadian
S. and Nakahara, H. (2002). Transport and Journal of Fisheries and Aquatic Sciences, 49, 694–700.
accumulation of bloom-forming Cyanobacteria in Jannasch, H. W. and Mottl, M. J. (1985). Geomicro-
a large, mid-latitude lake: the gyre–Microcystis biology of deep-sea hydrothermal vents. Science,
hypothesis. Limnology, 3, 87–96. 229, 717–25.
Ishikawa, K., Tsujimura, S., Kumagai, M. and Janzen, D. H. (1968). Host plants as islands in
Nakahara, H. (2003). Surface water recruitment of evolutionary and contemporary time. American
bloom-forming Cyanobacteria from deep water Naturalist, 102, 592–5.
sediments of Lake Biwa, Japan. In Sediment Quality Jassby, A. D. and Goldman, C. R. (1974a). Loss rates
Assessment and Management: Insight and Progress, ed. from a phytoplankton community. Limnology and
M. Munawar, pp. 287–305. Burlington: Aquatic Oceanography, 19, 618–27.
Ecosystem Health and Management Society. (1974b). A quantitative measure of succession rate
Istvánovics, V., Pettersen, K., Rodrigo, M. A., Pierson, D. and its application to the phytoplankton of lakes.
and Padisák, J. (1993). Gloeotrichia echinulata, a American Naturalist, 108, 688–93.
colonial Cyanobacterium with unique Jassby, A. D. and Platt, T. (1976). Mathematical
phosphorus uptake and life strategy. Journal of formulation of the relationship between
Plankton Research, 15, 531–52. photosynthesis and light for phytoplankton.
Ives, A. R. and Hughes, J. B. (2002). General Limnology and Oceanography, 21, 540–7.
relationships between species diversity and Javorničky, P. (1958). Revise nekterých metod pro
stability in competitive systems. American zjištováni kvantity fytoplanktonu. Sbornk vysoké
Naturalist, 159, 388–95. Skoly chemicko-technologické v Praze, 2, 283–367.
Ives, A. R. Carpenter, S. R. and Dennis, B. (1999). Jaworski, G. H. M., Talling, J. F. and Heaney, S. I.
Community interaction webs and the response of (1981). The influence of carbon dioxide depletion
REFERENCES 471
on growth and sinking rate of two planktonic sensing. Journal of Experimental Marine Biology and
diatoms in culture. British Phycological Journal, 16, Ecology, 250, 233–55.
395–410. Jónasson, P. M. (1978). Zoobenthos of lakes.
Jaworski, G. H. M., Wiseman, S. W. and Reynolds, C. S. Verhandlungen der internationalen Vereinigung für
(1988). Variability in the sinking rate of the theoretische und angewandte Limnologie, 20, 13–37.
freshwater diatom, Asterionella formosa: the (2003). Hypolimnetic eutrophication of the
influence of colony morphology. British N-limited dimictic lake Esrom. Archiv für
Phycological Journal, 23, 167–76. Hydrobiologie Suppl., 139, 449–512.
Jaworski, G. H. M., Talling, J. F. and Heaney, S. I. Jónasson, P. M. and Kristiansen, J. (1967). Primary and
(2003). Potassium dependence and phytoplankton secondary production in Lake Esrom: growth of
ecology: an experimental study. Freshwater Biology, Chironomus anthracinus in relation to seasonal
48, 833–40. cycles of phytoplankton and dissolved oxygen.
Jeppesen, E., Jensen, J. P., Kristensen, P. et al. (1990). Internationale Revue des gesamten Hydrobiologie, 52,
Fish manipulation as a lake restoration tool in 163–217.
shallow, eutrophic, temperate lakes. II. Threshold Jones, C. G., Lawton, J. H. and Shachak, M. (1994).
levels, long-term stability and conclusions. Organisms as ecosystem engineers. Oikos, 69,
Hydrobiologia, 200/201, 219–28. 373–86.
Jeppesen, E., Søndergaard, M., Jensen, J. P. et al. (1998). Jones, R. I. (1978). Adaptation to fluctuating irradiance
Cascading trophic interactions from fish to by natural phytoplankton communities. Limnology
bacteria and nutrients after reduced sewage and Oceanography, 23, 920–6.
loading: an 18-year study of a shallow Jones, R. I., Grey, J., Quarmby, C. and Sleep, D. (2001).
hypertrophic lake. Ecosystems, 1, 250–97. Sources and fluxes of inorganic carbon in a deep,
Jewell, W. J. and McCarty, P. L. (1971). Aerobic oligotrophic lake (Loch Ness, Scotland). Global
decomposition of algae. Environmental Science and Biogeochemical Cycles, 15, 863–70.
Technology, 5, 1023–31. Jordan, P., Fromme, P., Witt, H. T., Klukas, O., Saenger,
Jewson, D. H. (1976). The interaction of components W. and Krauss, N. (2001). Three-dimensional
controlling net phytoplankton photosynthesis in structure of cyanobacterial photosystem 1 at
a well-mixed lake (Lough Neagh, Northern 2.5 Å resolution. Nature, 411, 909–17.
Ireland). Freshwater Biology, 6, 551–76. Jørgensen, S.-E. (1995). State-of-the-art management
Jewson, D. H. and Wood, R. B. (1975). Some effects on models for lakes and reservoirs. Lakes and
integral photosynthesis of artifiial circulation of Reservoirs, Research and Management, 1, 79–87.
phytoplankton through light gradients. Verhand- (1997). Integration of Ecosystem Theory: A Pattern, 2nd
lungen der internationalen Vereinigung für theoretische edn. Dordrecht: Kluwer.
und angewandte Limnologie, 19, 1037–44. (1999). State-of-the-art of ecological modelling
Jiménez Montealegre, R., Verreth, J., Steenbergen, K., with emphasis on development of structural
Moed, J. and Machiels, M. (1995). A dynamic dynamic models. Ecological Modelling, 120,
simulation model for the blooming of Oscillatoria 75–96.
agardhii in a monomictic lake. Ecological Modelling, (2002). Explanation of ecological rules and
78, 17–24. observation by application of ecosystem theory
John, D. M., Whitton, B. A. and Brook, A. J. (2002). The and ecological models. Ecological Modelling, 158,
Freshwater Algal Flora of the British Isles. Cambridge: 241–8.
Cambridge University Press. Jørgensen, S.-E., Nielsen, S. N. and Mejer, H. F. (1995).
Johnson, L. (1975). Physical and chemical Emergy, environ, exergy and ecological
characteristics of Great Bear Lake. Journal of the modelling. Ecological Modelling, 77, 99–109.
Fisheries Research Board of Canada, 32, 1971–87. Joseph, J. and Sendner, H. (1958). Über die horizontale
(1994). Great Bear. In The Book of Canadian Lakes, ed. Diffusion im Meere. Deutsches hydrographische
R. J. Allan, M. Dickman, C. B. Gray and V. Cromie, Zeitschrift, 11, 51–77.
pp. 549–59. Burlington: Canadian Association on Juday, C. (1934). The depth distribution of some
Water Quality. aquatic plants. Ecology, 15, 325–35.
Joint, I. R. and Groom, S. B. (2000). Estimation of Juhász-Nagy, P. (1992). Scaling problems almost
phytoplankton production from space: current everywhere: an introduction. Abstracta Botanica,
status and future potential of satellite remote 16, 1–5.
472 REFERENCES
(1993). Notes on compositional biodiversity. Karp-Boss, L., Boss, E. and Jumars, P. A. (1996).
Hydrobiologia, 249, 173–82. Nutrient fluxes to planktonic osmotrophs in the
Jürgens, K., Arndt, H. and Rothaupt, K. O. (1994). presence of fluid motion. Oceanography and Marine
Zooplankton-mediated changes of microbial Biology, 34, 71–107.
food-web structure. Microbial Ecology, 27, Kasprzak, P., Lathrop, R. C. and Carpenter, S. R. (1999).
27–42. Influence of different-sized Daphnia species on
Kadiri, M. O. and Reynolds, C. S. (1993). Long-term chlorophyl concentration and summer
monitoring of the conditions of lakes: the phytoplankton community structure in eutropic
example of the English Lake District. Archiv für Wisconsin lakes. Journal of Plankton Research, 21,
Hydrobiologie, 129, 157–78. 2164–74.
Kahn, N. and Swift, E. (1978). Positive buoyancy Kauffman, Z. S. (1990). The Ecosystem of Onega Lake and
through ionic control in the non-motile marine the Trends of its Changes. Leningrad: Nauka. (In
dinoflagellate Pyrocystis noctiluca Murray ex Russian.)
Schuett. Limnology and Oceanography, 23, 649–58. Keddy, P. A. (1992). Assembly and response rules: two
Kajak, Z., Hillbricht-Ilkowska, A. and Piczyska, E. goals for predictive community ecology. Journal of
(1972). The production processes in several Polish Vegetation Science, 3, 157–64.
lakes. In Productivity Problems of Freshwaters (2001). Competition, 2nd edn. Dordrecht: Kluwer.
(Proceedings of the IBP-UNESCO Symposium, Kazimierz Kemp, P. F., Lee, S. and LaRoche, J. (1993). Estimating
Dolny, Poland, May 6–12, 1970), ed. Z. Kajak and A. the growth rate of slowly growing marine
Hillbricht-Ilkowska, pp. 129–147. Warsaw: Polstie bacteria from RNA content. Applied and Environ-
Wydawnictwo Naukowe. mental Microbiology, 59, 2594–601.
Källqvist, T. and Berge, D. (1990). Biological availability Kerby, N. W., Rowell, P. and Stewart, W. D. P. (1987).
of phosphorus in agricultural runoff compared to Cyanobacterial ammonium transport,
other sources. Verhandlungen der internationale ammonium assimilation and nitrogenase
Vereinigung für theoretische und angewandte regulation. New Zealand Journal of Marine and
Limnologie, 24, 214–17. Freshwater Research, 21, 447–55.
Kamp-Nielsen, L. (1975). A kinetic approach to the Kerkhof, L. and Ward, B. B. (1993). Comparison of
aerobic sediment–water exchange in Lake Esrom. nucleic-acid hybridization and fluorometry for
Ecological Modelling, 1, 153–60. measurement of the relationship between
Kamykowski, D., Reed, R. E. and Kirkpatrick, G. I. RNA/DNA ratio and growth rate in a marine
(1992). Comparison of sinking velocity, swimming bacterium. Applied and Environmental Microbiology,
velocity, rotation and path characteristics among 59, 1303–9.
six marine dinoflagellate species. Marine Biology, Ketchum, B. H. and Redfield, A. C. (1949). Some
113, 319–28. physical and chemical characteristics of algal
Kappers, F. I. (1984). On the Population Dynamics of the growth in mass cultures. Journal of Cellular and
Cyanobacterium Microcystis aeruginosa. Comparative Physiology, 13, 373–81.
Amsterdam: University of Amsterdam. Kibby, H. V. (1971). Energetics and population
Karl, D. M. (1999). A sea of change: biogeochemical dynamics of Diaptomus gracilis. Ecological
variability in the North Pacific sub-tropical gyre. Monographs, 41, 311–27.
Ecosystems, 2, 181–214. Kiene, R. P., Linn, L. J. and Bruton, J. A. (2000). New
(2002). Nutrient dynamics in the deep blue sea. and important roles for DMSP in marine
Trends in Microbiology, 10, 410–18. microbial communities. Journal of Sea Research, 43,
Karl, D. M., Wirsen, C. O. and Jannasch, H. W. (1980). 209–24.
Deep-sea primary production at the Galapagos Kilham, P. and Kilham, S. S. (1980). The evolutionary
hydrothermal vents. Science, 207, 1345–7. ecology of phytoplankton. In The Physiological
Karl, D. M., Bidigare, R. R. and Letelier, R. M. (2002). Ecology of Phytoplanktons, ed. I. Morris,
Sustained and aperiodic variability in organic pp. 571–97. Oxford: Blackwell Scientific
matter production and phototrophic microbial Publications.
community structure in the North Pacific Kilham, P. and Kilham, S. S. (1990). Endless summer:
Subtropical Gyre. In Phytoplankton Productivity, ed. internal loading processes dominate nutrient
P. J. leB. Williams, D. N. Thomas and C. S. cycles in tropical lakes. Freshwater Biology, 23,
Reynolds, pp. 222–64. Oxford: Blackwell Science. 379–89.
REFERENCES 473
Kierstead, H. and Slobodkin, L. B. (1953). The size of Klemer, A. R., Feuillard, J. and Feuillard, M. (1982).
water masses containing plankton blooms. Journal Cyanobacterial blooms: carbon and nitrogen have
of Marine Science, 12, 141–7. opposite effects on the buoyancy of Oscillatoria.
Kirchman, D. L. (1990). Limitation of bacterial growth Science, 215, 1629–31.
by dissolved organic matter in the subarctic Klemer, A. R., Pierson D. C. and Whiteside, M. C.
Pacific. Marine Ecology – Progress Series, 62, 47–54. (1985). Blue-green algal (cyanobacterial) nutrition,
Kirk, J. T. O. (1975a). A theoretical analysis of the buoyancy and bloom formation. Verhandlungen der
contribution of algal cells to the attenuation of internationale Vereinigung für theoretische und
light within natural waters. I. General treatment angewandte Limnologie, 22, 2791–8.
of suspensions of pigmented cells. New Phytologist, Kling, H. (1975). Phytoplankton Successions and Species
75, 11–20. Distributions in Prairie Ponds of the Erickson-
(1975b). A theoretical analysis of the contribution Elphinstone District, South-West Manitoba. Technical
of algal cells to the attenuation of light within Report No. 512. Ottawa: Fisheries and Marine
natural waters. II. Spherical cells. New Phytologist, Sciences of Canada.
75, 21–36. Kling, H. J., Mugidde, R. and Hecky, R. E. (2001).
(1976). A theoretical analysis of the contribution of Recent changes in the phytoplankton community
algal cells to the attenuation of light within of Lake Victoria in response to eutrophication. In
natural waters. III. Cylindrical and sphaeroidal The Great Lakes of the World (GLOW): Food Web,
cells. New Phytologist, 77, 341–58. Health and Integrity, ed. M. Munawar and R. E.
(1994). Light and Photosynthesis in Aquatic Ecosystems, Hecky, pp. 47–65. Leiden: Backhuys.
2nd edn. Cambridge: Cambridge University Press. Knapp, C. W., deNoyelles, F., Graham, D. W. and
(2003). The vertical attenuation of irradiance as a Bergin, S. (2003). Physical and chemical
function of the optical properties of the water. conditions surrounding the diurnal vertical
Limnology and Oceanography, 48, 9–17. migration of Cryptomonas spp. (Cryptophyceae) in
Kirke, B. K. (2000). Circulation, destratification, mixing a seasonally stratified reservoir (USA). Journal of
and aeration: why and how? Water, 27(4), 24–30. Phycology, 39, 855–61.
Kitchell, J. F., Koonce, J. F., O’Neill, R. V. et al. (1974). Knoechel, R. and Kalff, J. (1975). Algal sedimentation:
Model of fish biomass dynamics. Transactions of the the cause of a diatom–bluegreen succession.
American Fisheries Society, 103, 786–98. Verhandlungen der internationalen Vereinigung für
Kitchell, J. F., Stewart, D. J. and Weininger, D. (1977). theoretische und angewandte Limnologie, 19, 745–54.
Applications of a bioenergetic model to yellow (1978). An in situ study of the productivity and
perch (Perca flavescens) and walleye (Stizostedion population dynamics of five freshwater diatom
vitreum vitreum). Jounal of the Fisheries Research Board species. Limnology and Oceanography, 23, 195–218.
of Canada, 34, 1922–35. Knox, R. S. (1977). Photosynthetic efficiency and
Kjelleberg, S., Albertson, N., Flärdh, K. et al. (1993). exciton trapping. In Primary Production Processes of
How do non-differentiated bacteria adapt to Photosynthesis, ed. J. Barber, pp. 55–97. Amsterdam:
starvation? Antonie van Leeuwenhoek, 63, 333–41. Elsevier.
Klaveness, D. (1988). Ecology of the Cryptomonadida: a Knudsen, B. M. (1953). The diatom genus Tabellaria. II.
first review. In Growth and Reproductive Strategies of Taxonomy and morphology of the plankton
Freshwater Phytoplankton, ed. C. D. Sandgren, varieties. Annals of Botany, 17, 131–5.
pp. 105–33. Cambridge: Cambridge University Kofoid, C. A. (1903). The phytoplankton of the lower
Press. Illinois River and its basin. I. Quantitative
Klebahn, H. (1895). Gasvakuolen, ein Bestandtiel der investigations and general results. Bulletin of the
Zellender Wasserblütebildenden Illinois State Laboratory for Natural History, 6,
Phycochronaceen. Flora, Jena, 80, 241–82. 95–269.
Klemer, A. R. (1976). The vertical distribution of Kolber, Z. and Falkowski, P. G. (1993). Use of active
Oscillatoria agardhii var. isothrix. Archiv für fluorescence to estimate phytoplankton
Hydrobiologie, 78, 343–62. photosynthesis in situ. Limnology and
(1978). Nitrogen limitation of growth and Oceanography, 38, 1646–65.
gas-vacuolation in Oscillatoria rubescens. Kolber, Z. S., Barber, R. T., Coale, H. et al. (1994). Iron
Verhandlungen der internationale Vereinigung für limitation of phytoplankton photosynthesis in
theoretische und angewandte Limnologie, 20, 2293–7. the equatorial Pacific Ocean. Nature, 371, 145–9.
474 REFERENCES
Kolber, Z. S., Plumley, F. G., Lang, A. S. et al. (2001). Kozhov, M. (1963). Lake Baikal and its Life. The Hague:
Contribution of photoheterotrophic bacteria to W. Junk.
the carbon cycle in the ocean. Science, 292, Kozhova, O. M. (1987). Phytoplankton of Lake Baikal:
2492–4. structural and functional characteristics.
Kolmogorov, A. N. (1941). The local structure of Ergebnisse der Limnologie, 25, 19–37.
turbulence in incompressible viscous fluid for Kozhova, O. M. and Izmest’eva, L. R. (1998). Lake Baikal:
very high Reynolds numbers. Doklady Akademii Evolution and Biodiversity. Leiden: Backhuys.
nauk SSSR, 30, 301. (English summary of paper in Kratz, W. A. and Myers, J. (1955). Nutrition and growth
Russian.) of several blue-green algae. American Journal of
Kolpakova, A., Meshcheryakova, A. I. and Votintsev, Botany, 42, 282–7.
K. K. (1988). Characteristics of seasonal dynamics Krienitz, L. (1990). Coccale Grünalgen der mittleren
of chlorophyll-a, primary production and Elbe. Limnologica, 21, 165–231.
biogenic elements in Lake Baikal. Hydrobiological Kristiansen, J. (1996). Dispersal of freshwater algae: a
Journal, 24, 1–6. (English translation of original review. Hydrobiologia, 336, 151–7.
paper in Gidrobiologeskii Zhurnal.) Krivtsov, V., Tien, C. and Sigee, D (1999). X-ray
Komar, D. (1980). Settling velocities of circular microanalytical study of the protozoan Ceratium
cylinders at low Reynolds numbers. Journal of hirundinella from Rostherne Mere (Cheshire, UK):
Geology, 88, 327–36. dynamics of intracellular elemental
Komárek, J. (1996). Towards a combined approach to concentrations, correlations and implications for
the taxonomy and species limitation of overall ecosystem functioning. Netherlands Journal
picoplanktic cyanoprokaryotes. Algological Studies, of Zoology, 49, 263–74.
83, 377–401. Kromkamp, J. and Mur, L. (1984). Buoyant density
Komárková, J. and Šimek, K. (2003). Unicellular and changes in the cyanobacterium Microcystis
colonial formations of picoplanktonic aeruginosa due to changes in the cellular
cyanobacteria under variable environmental carbohydrate content. FEMS Microbial Letters, 25,
conditions and predation pressure. Algological 105–9.
Studies, 109, 327–40. Kruk, C., Mazzeo, N., Lacerot, G. and Reynolds, C. S.
Kondoh, M. (2001). Unifying the relationships of (2002). Classification schemes for phytoplankton:
species richness to productivity and disturbance. a local validation of a functional approach to the
Proceedings of the Royal Society of London B, 268, analysis of species temporal replacement. Journal
269–71. of Plankton Research, 24, 1191–216.
Konopka, A. E. and Brock, T. D. (1978). Effect of Kuentzel, L. E. (1969). Bacteria, carbon dioxide and
temperature on blue-green algae (Cyanobacteria) algal blooms. Journal of the Water Pollution Control
in Lake Mendota. Applied and Environmental Federation, 41, 1737–47.
Microbiology, 36, 572–6. Kühlbrandt, W. (2001). Chlorophylls galore. Nature,
Konopka, A., Brock, T. D. and Walsby, A. E. (1978). 411, 896–8.
Buoyancy regulation by Aphanizomenon in Lake Kühlbrandt, W. and Wang, D. N. (1991). Three-
Mendota. Archiv für Hydrobiologie, 83, 524–37. dimensional structure of plant light-harvesting
Kořínek, V., Fott, J., Fuksa, J., Lellák, J. and Prazáková, complex determined by electron crystallography.
M. (1987). Carp ponds of central Europe. In Nature, 351, 130–1.
Ecosystems of the World, vol. 29, Managed Aquatic Kühlbrandt, W., Wang, D. N. and Fujiyoshi, Y. (1994).
Ecosystems, ed. R. G. Michael, pp. 29–62. Atomic model of plant light-harvesting complex
Amsterdam: Elsevier. by electron crystallography. Nature, 367, 614–21.
Korneva, L. G. (2001). Ecological aspects of mass Kunkel, W. B. (1948). Magnitude and character of
development of Gonyostomum semen (Ehr.) Dies. errors in Stokes’ Law estimates of particle radius.
(Raphidophyta). International Journal of Algae, 3, Journal of Applied Physics, 19, 1066–8.
40–54. (Originally published, in Russian, in Kuosa, H., Autio, R., Kuuppo, P., Setälä, O. and
Algologiya, 10, 265–77.) Tanskanen, S. (1997). Nitrogen, silicon and
Koroleva, N. N. (1993). Phytoplankton in the northern zooplankton controlling the Baltic spring bloom:
part of the Aral Sea in in September, 1991. In an experimental study. Estuarine, Coast and Shelf
Ekologichekiy krizis na Aral’skom more, ed. O. A. Science, 45, 813–21.
Skarlato and N. V. Aladin, pp. 52–6. St Petersburg: Kusch, J. (1995). Adaptations to inducible defense in
Zoological Institute, Russian Academy of Sciences. Euplotes daidaleos (Ciliophora) to predation risks
REFERENCES 475
by various predators. Microbial Ecology, 30, internationale Vereinigung für theoretische und
79–88. angewandte Limnologie, 20, 969–74.
Kustka, A., Sañudo-Wilhelmy, S., Carpenter, E. J., (1987). Feeding and nutrition in Daphnia. Memorie
Capone, D. G. and Raven, J. A. (2003). A revised dell’Istituto italiano di Idrobiologia, 45, 143–92.
estimate of the iron-use efficincy of nitrogen Lampert, W. and Schober, U. (1980). The importance of
fixation, with special reference to the marine threshold food concentrations. In Evolution and
Cyanobacterium Trichodesmium spp. (Cyanophyta). Ecology of Zooplankton Communities, ed. W. C.
Journal of Phycology, 39, 12–25. Kerfoot, pp. 264–7. Hanover, NH: University of
Kutsch, W. L., Steinborn, W., Herbst, M. et al. (2001). New England Press.
Environmental indication: a field test of an Lampert, W. and Sommer, U. (1993). Limnoökologie.
ecosystem approach to quantify self-organisation. Stuttgart: Georg Thieme Verlag.
Ecosystems, 4, 49–66. Lampert, W., Fleckner, W., Rai, H. and Taylor, B. E.
Kuuppo, P., Samuelsson, K., Lignell, R. et al. (2003). Fate (1986). Phytoplankton control by grazing
of increased production in late-summer plankton zooplankton. Limnology and Oceanography, 31,
communities due to nurient enrichment of the 478–90.
Baltic proper. Aquatic Microbial Ecology, 32, 47–60. Lampert, W., Rothaupt, K. O. and von Elert, E. (1994).
Kyewalyanga, M., Platt, T. and Sathyendranath, S. Chemical induction of colony formation in a
(1992). Ocean primary production calculated by green alga (Scenedesmus acutus) by grazers
spectral and broad-band models. Marine Ecology – (Daphnia). Limnology and Oceanography, 39,
Progress Series, 85, 171–85. 1543–50.
Lack, T. J. and Lund, J. W. G. (1974). Observations and Lane, A. E. and Burris, J. E. (1981). Effects of
experiments on the phytoplankton of Blelham environmental pH on the internal pH of Chlorella
Tarn, English Lake District. I. The experimental pyrenoidosa, Scenedesmus quadricauda and Euglena
tubes. Freshwater Biology, 4, 399–415. mutabilis. Plant Physiology, 68, 439–42.
Lacroix, G., Lescher-Moutoué, F. and Bertolo, A. (1999). Lange, W. (1976). Speculations on a possible essential
Biomass and production of plankton in shallow function of the gelatinous sheath of blue-green
lakes and deep lakes: are there general patterns? algae. Canadian Journal of Microbiology, 22, 1181–5.
Annales de Limnologie, 35, 111–22. Langenberg, V. T., Mwape, L. M., Tshibangu, K. et al.
Lamouroux, N., Olivier, J.-M., Persat, H. et al. (1999). (2002). Comparison of the thermal stratification,
Predicting community characteristics from light attenuation and chlorophyll-a dynamics
habitat conditions: fluvial fish and hydraulics. between the ends of Lake Tanganyika. Aquatic
Freshwater Biology, 42, 275–99. Ecosystem Health and Management, 5, 255–65.
Lampert, W. (1977a). Studies on the carbon balance of Langmuir, I. (1938). Surface motion of water induced
Daphnia pulex De Geer as related to by wind. Science, 87, 119–23.
environmental conditions. II. The dependence of Larkum, A. W. D. and Barrett, J. (1983). Light-
carbon assimilation on animal size, temperature, harvesting processes in algae. In Advances in
food concentration and diet species. Archiv für Botanical Research, vol. 10, ed. H. W. Woolhouse,
Hydrobiologie (Suppl.), 48, 310–35. pp. 1–219. London: Academic Press.
(1977b). Studies on the carbon balance of Daphnia Larsson, U. and Hagström, A. (1979). Phytoplankton
pulex De Geer as related to environmental exudate release as an energy source for the
conditions. III. Production and production growth of pelagic bacteria. Marine Biology, 52,
efficiency. Archiv für Hydrobiologie (Suppl.), 48, 199–206.
336–60. Lauridsen, T. L. and Buenk, I. (1996). Diel changes in
(1977c). Studies on the carbon balance of Daphnia the horizontal distribution of zooplankton in the
pulex De Geer as related to environmental littoral zone of two eutrophic lakes. Archiv für
conditions. IV. Determination of the threshold Hydrobilogie, 137, 161–76.
concentration as a factor controlling the Lauridsen, T. L. and Lodge, D. M. (1996). Avoidance by
abundance of zooplankton species. Archiv für Daphnia magna of fish and macrophytes: chemical
Hydrobiologie (Suppl.), 48, 361–8. cues and predator-mediated use of macrophyte
(1978). Climatic conditions and planktonic beds. Limnology and Oceanography, 41, 794–8.
interactions as factors controlling the regular Lauridsen, T. L., Jeppesen, E. and Mitchell, S. F. (1999).
succession of spring algal bloom and extremely Diel variation in horizontal distribution of
clear water in Lake Constance. Verhandlungen der Daphnia and Ceriodaphnia in oligotrophic and
476 REFERENCES
mesotrophic lakes with contrasting fish densities. Legendre, L. and Rassoulzadegan, F. (1996). Food-web
Hydrobiologia, 409, 241–50. mediated export of biogenic carbon in oceans:
Laurion, I., Lami, A. and Sommaruga, R. (2002). hydrodynamic control. Marine Ecology – Progress
Distribution of mycosporine-like amino-acids and Series, 145, 179–93.
photoprotective carotenoids among freshwater Legendre, L. and Rivkin, R. B. (2002a). Fluxes of carbon
phytoplankton assemblages. Aquatic Microbial in the upper ocean: regulation by food-web
Ecology, 26, 283–94. control nodes. Marine Ecology – Progress Series, 242,
Laws, E. A. (1991). Photosynthetic quotients, new 95–109.
production and net communiy production in the (2002b). Pelagic food webs: responses to
open ocean. Deep-Sea Research, 38, 143–67. environmental processes and effects on the
(2003). Partitioning of microbial biomass in pelagic environment. Ecological Research, 17, 143–9.
aquatic communities: maximum resiliency as a Lehman, J. T. (1975). Reconstructing the rate of of
food-web organising construct. Aquatic Microbial accumulation of lake sediment: the effect of
Ecology, 32, 1–10. sediment focusing. Quaternary Research, 5,
Lawton, J. H. (1997). The role of species in ecosystems: 541–50.
aspects of ecological complexity and biological (1976). Ecological and nutritional studies on
diversity. In Biodiversity: An Ecological Perspective, ed. Dinobryon Ehrenb.: seasonal periodicity and the
T. Abe, S. R. Levin and M. Higashi, pp. 215–28. phosphate toxicity problem. Limnology and
New York: Springer-Verlag. Oceanography, 21, 646–58.
(2000). Community Ecology in a Changing World. Lehman, J. T. and Cáceres, C. E. (1993). Food-web
Oldendorf: ECI. responses to species invasion by a predatory
Lawton, J. H and Brown, V. K. (1993). Redundancy in invertebrate: Bythotrephes in Lake Michigan.
ecosystems. In Biodiversity and Ecosystem Function, Limnology and Oceanography, 38, 879–91.
ed. E.-D. Schulze and H. A. Mooney, pp. 255–70. Lehman, J. T., Botkin, D. B. and Likens, G. E. (1975).
Berlin: Springer-Verlag. The assumptions and rationales of computer
Lawton, J. H. and Schröder, D. (1977). Effects of plant model of phytoplankton population dynamics.
type, size of geographical range and taxonomic Limnology and Oceanography, 20, 343–64.
isolation on number of insect species associated Leibovich, S. (1983). The form and dynamics of
with British plants. Nature, 265, 137–40. Langmuir circulations. Annual Review of Fluid
Lazier, J. R. N. and Mann, K. H. (1989). Turbulence and Mechanics, 15, 391–427.
the diffusive layers around organisms. Deep-Sea Leitão, M., Morata, S. M., Rodriguez, S. and Vergon,
Research, 36, 1721–33. J. P. (2003). The effect of perturbations on
Le Cren, E. D. (1987). Perch (Perca fluviatilis) and pike phytoplankton assemblages in a deep reservoir
(Esox lucius) in Windermere from 1940 to 1985: (Vouglans, France). Hydrobiologia, 502, 73–83.
studies in poulation dynamics. Canadian Journal of Lessmann D., Fyson, A. and Nixdorf, B. (2000).
Fisheries and Aquatic Sciences, 44 (Suppl. 2), 216–28. Phytoplankton of the extremely acidic mining
Leadbeater, B. S. C. (1990). Ultrastructure and assembly lakes of Lusatia (Germany) with pH ≤3.
of the scale case in Synura (Synurophyceae Hydrobiologia 433, 123–8.
Andersen). British Phycological Journal, 25, 117–32. Letcher, B. H., Rice, J. A., Crowder, L. B. and Rose, K. A.
Lee, S. and Fuhrman, J. (1987). Relationships between (1996). Variability in survival of larval fish:
biovolume and biomass of naturally derived disentangling components with a generalised
marine bacterioplankton. Applied Environmental individual-based model. Canadian Journal of
Microbiology, 53, 1298–303. Fisheries and Aquatic Sciences, 53, 787–801.
Legendre, L. and LeFevre, J. (1989). Hydrodynamical Letelier, R. M., Bidigare, R. R., Hebel, D. V. et al. (1993).
singularities as controls of recycled versus export Temporal variability of phytoplankton
production in oceans. In Productivity of the Ocean, community structure based on pigment analysis.
Present and Past, ed. W. H. Berger, V. Smetacek and Limnology and Oceanography, 38, 1420–37.
G. Wefer, pp. 49–63. New York: Wiley Interscience. Lévêque, C., Carmouze, J. P., Dejoux, C. et al. (1972).
Legendre, L. and Michaud, J. (1999). Chlorophyll a to Recherches sur les biomasses et la productivité
estimate the particulate organic carbon available du Lac Tchad. In Productivity Problems of
as food to large zooplankton in the euphotic zone Freshwaters, ed. Z. Kajak and A.
of oceans. Journal of Plankton Research, 21, 2067–83. Hillbricht-Ilkowska, pp. 165–81. Warsaw:
REFERENCES 477
Levich, A. P. (1996). The role of nitrogen : phosphorus Philosophical Transactions of the Royal Society of
ratio in selecting for dominance of London B, 304, 519–28.
phytoplankton by Cyanobacteria or green algae Li, W. K. W. (2002). Macroecological patterns of
and its application to reservoir management. phytoplankton in the northwestern North
Journal of Aquatic Ecosystem Health, 5, 55–61. Atlantic Ocean. Nature, 419, 154–7.
Levich, V. G. (1962). Physicochemical Hydrodynamics. Li, W. K. W. and Platt, T. (1982). Distribution of carbon
Englewood Cliffs, NJ: Prentice Hall. among photosynthetic end products in
Levin, S. A. (2000). Multiple scales and the phytoplankton of the eastern Canadian Arctic.
maintenance of biodiversity. Ecosystems, 3, Journal of Phycology, 58, 135–50.
498–506. Liebig, J. von (1840). Organic Chemistry and its
Levins, R. (1966). The strategy of model building in Application to Agriculture and Physiology. London:
population ecology. American Scientist, 54, 421–31. Taylor and Walton.
Lewin, J. C. (1962). Silicification. In Physiology and Liere, L. van (1979). On Oscillatoria agardhii Gomont:
Biochemistry of the Algae, ed. R. A. Lewin, Experimetal Ecology and Physiology of a Nuisance
pp. 445–55. New York: Academic Press. Bloom-Forming Cyanobacterium. Zeist: De Niuwe
Lewin, J. C., Lewin, R. A. and Philpott, D. E. (1958). Schouw.
Observations on Phaeodactylum tricornutum. Journal Lijklema, L. (1977). The role of iron in the exchange of
of General Microbiology, 18, 418–26. phosphate between water and sediments. In
Lewin, R. A. (1981). The Prochlorophytes. In The Interactions between Sediments and Freshwater, ed.
Prokaryotes: A Handbook on Habitats, Isolation and H. L. Golterman, pp. 313–17. The Hague:
Identification of Bacteria, ed. M. P. Starr, H. Stolp, W. Junk.
H. G. Trüper, A. Ballows and H. G. Schlegel, Likens, G. E. and Davis, M. B. (1975). Postglacial history
pp. 257–66. Berlin: Springer-Verlag. of Mirror Lake and its watershed in New
Lewin, R. A. (2002). Prochlorophyta: a matter of class Hampshire, USA: an initial report. Verhandlungen
distinction. Photosynthesis Research, 73, 59–61. der internationale Vereinigung für theoretische und
Lewis, D. M., Elliott, J. A., Lambert, M. J. and Reynolds, angewandte Limnologie, 19, 982–93.
C. S. (2002). The simulation of an Australian Likhoshway, Ye. V., Kuzmina, A. Ye., Potyemkina, T. G.,
reservoir using a phytoplankton community Potyemkin, V. L. and Shimarev, M. N (1996). The
model, PROTECH. Ecological Modelling, 150, 107–16. distribution of diatoms near a thermal bar in
Lewis, W. M. (1976). Surface/volume ratio: implications Lake Baikal. Great Lakes Research, 22, 5–14.
for phytoplankton morphology. Science, 192, Lin, C. K. (1972). Phytoplankton succession in a
885–7. eutrophic lake with special reference to
(1978). Dynamics and succession of the blue-green algal blooms. Hydrobiologia, 39,
phytoplankton in a tropical lake: Lake Lanao, 321–34.
Philippines. Journal of Ecology, 66, 849–80. Lindley, J. A., Robins, B. B. and Williams, R. (1999). Dry
(1983). A revised classification of lakes based on weight, carbon and nitrogen content of some
mixing. Canadian Journal of Fisheries and Aquatic euphausiids from the north Atlantic Ocean and
Sciences, 40, 1779–87. the Celtic Sea. Journal of Plankton Research, 21,
(1986). Phytoplankton succession in Lake Valencia, 2053–66.
Venezuela. Hydrobiologia, 138, 9–204. Lindström, E. S. (2000). Bacterioplankton community
Lewitus, A. J. and Kana, T. M. (1994). Responses of composition in five lakes differing in trophic
estuarine phytoplankton to exogenous glucose: status and humic content. Microbial Ecology, 40,
stimulation versus inhibition of photosynthesis 104–13.
and respiration. Limnology and Oceanography, 39, Link, J. S. (2002). What does ecosystem-based fisheries
182–9. management mean? Fisheries, 27(4), 18–21.
Li, A. S., Stoecker, D. K. and Coats, D. W. (2000). Litchman, E. (2000). Growth rates of phytoplankton
Mixotrophy in Gyrodinium galatheanum under fluctuating light. Freshwater Biology, 44,
(Dinophyceae): grazing responses to light 223–35.
intensity and inorganic nutrients. Journal of (2003). Competition and coexistence of
Phycology, 36, 33–45. phytoplankton under fluctuating light:
Li, C.-W. and Volcani, B. E. (1984). Aspects of experiments with two Cyanobacteria. Aquatic
silicification in wall morphogenesis of diatoms. Microbial Ecology, 31, 241–8.
478 REFERENCES
Liti, D., Källqvist, T. and Lien, L. (1991). Limnological Lucas, W. J. and Berry, J. A. (1985). Inorganic Carbon
aspects of Lake Turkana, Kenya. Verhandlungen der Uptake by Aquatic Photosynthetic Organisms.
internationale Vereinigung für theoretische und Rockville, MD: American Society of Plant
angewandte Limnologie, 24, 1108–11. Physiologists.
Livingstone, D. and Cambray, R. S. (1978). Luftig, R. and Haselkorn, R. (1967). Morphology of a
Confirmation of 137 Cs dated by algal stratigraphy. virus of blue-green algae and properties of its
Nature, 276, 259–60. deoxyribonucleic acid. Journal of Virology, 1, 344–61.
Livingstone, D. and Jaworski, G. H. M. (1980). The Lund, J. W. G. (1949). Studies on Asterionella. I. The
viability of akinetes of blue-green algae recovered origin and nature of the cells producing seasonal
from the sediments of Rostherne Mere. British maxima. Journal of Ecology, 37, 389–419.
Phycological Journal, 15, 357–64. (1950). Studies on Asterionella formosa Hass. II.
Lloyd, M. (1967). ‘Mean crowding’. Journal of Animal Nutrient depletion and the spring maximum.
Ecology, 36, 1–30. Journal of Ecology, 38, 1–35.
Loehle, C. (1988). Problems with the triangular model (1954). The seasonal cycle of the plankton diatom
for representing plant strategies. Ecology, 69, 284–6. Melosira italica subsp. subarctica O. Mull. Journal of
Löffler, H. (1988). Natural hazards and health risks Ecology, 42, 151–79.
from lakes. Water Resources Development, 4, (1957). Chemical analysis in ecology illustrated from
276–83. Lake District tarns and lakes. II. Algal differences.
Long, S. P., Humphries, S. and Falkowski, P. G. (1994). Proceedings of the Linnean Society of London, 167,
Photoinhibition of photosynthesis in nature. 165–71.
Annual Reviews in Plant Physiology and Plant (1959). Buoyancy in relation to the ecology of the
Molecular Biology, 45, 633–62. freshwater phytoplankton. British Phycological
Longhurst, A., Sathyendranath, S., Platt, T. and Bulletin, 1(7), 1–17.
Caverhill, C. (1995). An estimate of global primary (1961). The algae of the Malham Tarn District. Field
production in the ocean from satellite radiometer Studies, 1(3), 85–115.
data. Journal of Plankton Research, 17, 1245–71. (1965). The ecology of the freshwater phyto-
Loreau, M., Naeem, S., Inchausti, P. et al. (2001). plankton. Biological Reviews, 40, 231–93.
Biodiversity and ecosystem functioning: current (1966). The importance of turbulence in the
knowledge and future challenges. Science, 294, periodicity of certain freshwater species of the
804–8. genus Melosira. Botanicheskii Zhurnal SSSR, 51,
Lorenzen, C. J. (1966). A method for the continuous 176–87. (In Russian.)
measurement for in-vivo chlorophyll (1970). Primary production. Water Treatment and
concentration. Deep-Sea Research, 13, 223–7. Examination, 19, 332–58.
Lorenzen, M. W. (1974). Predicting the effects of (1971). An artificial alteration of the seasonal cycle
nutrient diversion on lake recovery. In Modelling of the plankton diatom Melosira italica subsp.
the Eutrophication Process, ed. E. J. Middlebrooks, subarctica in an English Lake. Journal of Ecology, 59,
D. H. Falkenborg and T. E. Maloney, pp. 205–21. 421–33.
Ann Arbor, MI: Ann Arbor Science. (1972). Eutrophication. Proceedings of the Royal Society
Lovelock, J. E. (1979). Gaia: A New Look at Life on Earth. of London B, 180, 371–82.
Oxford: Oxford University Press. (1975). The use of large experimental tubes in lakes.
Lowe-McConnell, R. (1996). Fish communities in In The Effects of Storage on Water Quality, ed. R. E.
African great lakes. Environmental Biology of Fishes, Youngman, pp. 291–311. Medmenham: Water
45, 219–35. Research Centre.
(2003). Recent research in the African great lakes: (1978). Changes in the phytoplankton of an English
fisheries, biodiversity and cichlid evolution. lake, 1945–1977. Hydrobiological Journal, 14, 6–21.
Freshwater Forum, 20, 1–64. Lund, J. W. G. and Reynolds, C. S. (1982). The
Lucas, L. V., Koseff, J. R., Monismith, S. G., Cloern, J. E. development and operation of large limnetic
and Thompson, J. K. (1999). Processes governing encosures and their contribution to
phytoplankton blooms in estuaries II. The role of phytoplankton ecology. In Progress in Phycological
horizontal transport. Marine Ecology – Progress Research, vol. 1, ed. F. E. Round and D. J. Chapman,
Series, 187, 17–30. pp. 1–65. Amsterdam: Elsevier.
REFERENCES 479
Lund, J. W. G., Kipling, C. and Le Cren, E. D. (1958). demethylsulphoniopropionate in the northeast

The inverted microscope method of estimating Atlantic during the summer coccolithophore
algal numbers and the statistical basis of bloom. Deep-Sea Research, 40, 1487–508.
estimation by counting. Hydrobiologia, 11, 143–70. Malinsky-Rushansky, N., Berman, T. and Dubinsky, Z.
Lund, J. W. G., Mackereth, F. J. H. and Mortimer, C. H. (1995). Seasonal dynamics of picophytoplankton
(1963). Changes in depth and time of certain in Lake Kinneret, Israel, Freshwater Biology, 34,
chemical and physical conditions and of the 241–51.
standing crop of Asterionella formosa Hass. in the Maloney, T. E., Miller, W. R. and Blind, N. L. (1973). Use
north basin of Windermere in 1947. Philosophical of algal bioassays in studying eutrophication
Transactions of the Royal Society of London B, 246, problems. In Advances in Water Pollution Research,
255–90. ed. S. H. Jenkins, pp. 205–15. Oxford: Pergamon
Lund, J. W. G., Jaworski, G. H. M. and Butterwick, C. Press.
(1975). Algal bioassay of water from Blelham Mann, K. H. and Lazier, J. R. N. (1991). Dynamics of
Tarn, English Lake District, and the growth of Marine Ecosystems. Oxford: Blackwell Scientific
planktonic diatoms. Archiv für Hydrobiologie Publications.
(Suppl.), 49, 49–69. Mann, N. H. (1995). How do cells express limitation at
Lyck, S. and Christoffersen, K. (2003). Microcystin the molecular level? In Molecular Ecology of Aquatic
quota, cell division and microcystin net Microbes, ed I. Joint, pp. 171–90. Berlin:
production of precultured Microcystis aeruginosa Springer-Verlag.
CYA 228 (Chrococcales, Cyanophyceae) under field Maranger, R. and Bird, D. F. (1995). Viral abundance in
conditions. Phycologia, 42, 667–74. aquatic systems: a comparison between marine
Lynch, M., Wieder, L. and Lampert, W. (1986). and freshwaters. Marine Ecology – Progress Series,
Measurement of the carbon balance in Daphnia. 121, 217–26.
Limnology and Oceanography, 31, 17–33. Margalef, R. (1957). Nuevos aspectos del problema de
Maberly, S. C. (1996). Diel, episodic and seasonal la suspensión en los organismos planctónicos.
changes in pH and concentrations of inorganic Investigación pesquera, 7, 105–16.
carbon in a productive lake. Freshwater Biology, 35, (1958). Temporal succession and spatial
579–98. heterogeneity in phytoplankton. In Perspectives in
Maberly, S. C., Hurley, M. A., Butterwick, C. et al. Marine Biology, ed. A. A. Buzatti-Traverso,
(1994). The rise and fall of Asterionella formosa in pp. 323–49. Berkeley, CA: University of California
the South Basin of Windermere: analysis of a Press.
45-year series of data. Freshwater Biology, 31, 19–34. (1963). Succession in marine populations. Advancing
MacArthur, R. H. and Wilson, J. O. (1967). The Theory of Frontiers in Plant Science, 2, 137–88.
Island Biogeography. Princeton, NJ: Princeton (1964). Correspondence between the classic types of
University Press. lake and the structural and dynamic properties
Mackereth, F. J. H. (1953). Phosphorus utilisation by of their populations. Verhandlungen der
Asterionella formosa Hass. Journal of Experimental internationale Vereinigung für theoretische und
Botany, 4, 296–313. angewandte Limnologie, 15, 169–75.
Mackey, R. L. and Currie, D. J. (2000). A re-examination (1967). Some concepts relative to the organisation
of the expected effects of disturbance on of plankton. Annual Review of Oceanography and
diversity. Oikos, 88, 483–93. Marine Biology, 5, 257–89.
Maeda, H. (1993). A new perspective for studies of (1978). Life-forms of phytoplankton as survival
plankton community: ecology and taxonomy of alternatives in an unstable environment.
picophytoplankton and flagellates. Japanese Journal Oceanologia Acta, 1, 493–509.
of Limnology, 54, 155–9. (1997). Our Biosphere. Oldendorf: ECI.
Maguire, B. (1963). The passive dispersal of small Margalef, R., Estrada, M. and Blasco, D. (1979).
aquatic organisms and their colonisation of Functional morphology of organisms involved in
isolated bodies of water. Ecological Monographs, 33, red tides, as adapted to decaying turbulence. In
161–85. Toxic Dinoflagellate Blooms, ed. D. L. Taylor and
Malin, G., Turner, S., Liss, P., Holligan, P. and Harbour, H. H. Seliger, pp. 89–94. Amsterdam:
D. (1993). Dimethylsulfide and Elsevier-North Holland.
480 REFERENCES
Margulis, L. (1970). Origin of Eukaryotic Cells. New McCarthy, J. J., Taylor, W. R. and Taft, J. L. (1975). The
Haven, CT: Yale University Press. dynamics of nitrogen and phosphorus cycling in
(1981). Symbiosis in Cell Evolution. San Francisco, CA: the open waters of the Chesapeake Bay. In Marine
W. H. Freeman. Chemistry in the Coastal Environment, ed. T. M.
Marra, J. (1978). Phytoplankton photosynthetic Church, pp. 664–81. Washington, DC: American
response to vertical movement in a mixed layer. Chemical Society.
Marine Biology, 46, 203–8. McCauley, E., Nisbet, R. M., Murdoch, W. W., de Roos,
Marra, J. (2002). Approaches to the measurement of A. M. and Gurney, W. S. C. (1999). Large-amplitude
plankton production. In Phytoplankton Productivity, cycles of Daphnia and its algal prey in enriched
ed. P. J. leB. Williams, D. N. Thomas and C. S. environments. Nature, 402, 653–6.
Reynolds, pp. 78–108. Oxford: Blackwell Science. McDermott, G. A., Prince, S. M., Freer, A. A. et al.
Marra, J., Haas, L. W. and Heinemann, K. R. (1988). (1995). Crystal structure of an integral membrane
Time course of C assimilation and microbial food light-harvesting complex from phoyosynthetic
webs. Journal of Experimental Marine Biology and bacteria. Nature, 374, 517–21.
Ecology, 115, 263–80. McGowan, J. A. and Walker, P. W. (1985). Dominance
Marsden, M. W. (1989). Lake restoration by reducing and diversity maintenance in an oceanic
external phosphorus loading: the influence of ecosystem. Ecological Monographs, 55, 103–18.
phosphorus release. Freshwater Biology, 21, 139–62. Mackenthun, K. M., Keup, L. E. and Stewart, R. K.
Marti, D. E. and Imboden, D. M. (1986). Thermische (1968). Nutrients and algae in Lake Sebasticook,
Energieflüsse an der Wasseroberfläche: beispiel Maine. Journal of the Water Pollution Control
Sempachersee. Schweizerische Zeitschrift für Federation, 40, R72–81.
Hydrologie, 48, 196–229. McMahon, J. W. and Rigler, F. H. (1963). Mechanisms
Martin, J. H. (1992). Iron as a limiting factor in regulating the feeding rate of Daphnia magna
oceanic productivity. In Primary Productivity and Strauss. Canadian Journal of Zoology, 41, 321–32.
Biogeochemical Cycles in the Sea, ed. P. G. Falkowski McNown, J. S. and Malaika, J. (1950). Effect of particle
and A. Woodhead, pp. 137–55. New York: Plenum shape on settling velocity at low Reynolds
Press. numbers. Transactions of the American Geophysical
Martin, J. H., Coale, K. H., Johnson, K. S. et al. (1994). Union, 31, 74–82.
Testing the iron hypothesis in ecosystems of the Meffert, M.-E. (1971). Cultivation and growth of two
equatorial Pacific Ocean. Nature, 371, 123–43. planktonic Oscillatoria species. Mitteilungen der
Massana, R. and Jürgens, K. (2003). Composition and internationalen Vereinigung für theoretische und
population dynamics of planktonic bacteria and angewandte Limnologie, 19, 189–205.
bacterivorous flagellates in seawater chemostat Megard, R. O. (1972). Phytoplankton photosynthesis
cultures. Aquatic Microbial Ecology, 32, 11–22. and phosphorus in lake Minnetonka, Minnesota.
May, R. M. (1973). Stability and Complexity in Model Limnology and Oceanography, 17, 68–87.
Ecosystems. Princeton, NJ: Princeton University Mejer, H. and Jørgensen, S.-E. (1979). Exergy and
Press. ecological buffer capacity. In State-of-the-Art
(1977). Thresholds and breakpoints in ecosystems Ecological Modelling, vol. 7, ed. S.-E. Jørgensen,
with a multiplicity of stable states. Nature, 269, pp. 829–46. Copenhagen: International Society for
471–7. Ecological Modelling.
May, V. (1972). Blue-Green Algal Blooms at Braidwood, New Meybeck. M. (1995). Global distribution of lakes. In
South Wales (Australia). Sydney: New South Wales Physics and Chemistry of Lakes, 2nd edn, ed. A.
Department of Agriculture. Lerman, D. M. Imboden and J. R. Gat, pp. 1–35.
McAlice, B. J. (1970). Observations on the the Berlin: Springer-Verlag.
small-scale distribution of estuarine Mills, C. A. and Hurley, M. A. (1990). Long-term studies
phytoplankton. Marine Biology, 7, 100–11. on the Windermere populations of perch, Perca
McCarthy, J. J. (1972). The uptake of urea by natural fluviatilis, pike, Esox lucius, and arctic charr,
poulations of marine phytoplankton. Limnology Salvelinus alpinus. Freshwater Biology, 23, 119–36.
and Oceanography, 17, 738–48. Mills, E. L. and Forney, J. L. (1983). Impact on Daphnia
(1980). Nitrogen. In The Physiological Ecology of pulex of predation by young perch in Oneida
Phytoplankton, ed. I. Morris, pp. 191–233, Oxford: Lake, New York. Transactions of the American
Blackwell Scientific Publications. Fisheries Society, 112, 154–61.
REFERENCES 481
(1987). Trophic dynamics and development of medium and application to specific absorption of
freshwater pelagic food webs. In Complex phytoplankton. Deep-Sea Research, 28A, 1375–93.
Interactions in Lake Communities, ed. S. R. Morel, A. and Smith, R. C. (1974). Relation between
Carpenter, pp. 11–30. New York: Springer–Verlag. total quanta and total energy for aquatic
Mischke, U. and Nixdorf, B. (2003). Equilibrium-phase photosynthesis. Limnology and Oceanography, 19,
conditions in shallow German lakes: how 591–600.
Cyanoprokaryota species establish a steady-state Moroney, J. V. (2001). Carbon concentrating
phase in late summer. Hydrobiologia, 502, mechanisms in aquatic photosynthetic
123–32. organisms: a report on CCM 2001. Journal of
Mitchison, G. J. and Wilcox, M. (1972). A rule Phycology, 37, 928–31.
governing cell division in Anabaena. Nature, 239, Moroney, J. V. and Chen, Z. Y. (1998). The role of the
110–11. chloroplast in inorganic carbon uptake by
Moed, J. R. (1973). Effect of combined action of light eukaryotic algae. Canadian Journal of Botany, 76,
and silicon depletion on Asterionella formosa Hass. 1025–34.
Verhandlungen der internationale Vereinigung für Morse, G. K., Lester, J. N. and Perry, R. (1993). The
theoretische und angewandte Limnologie, 18, 1367–74. Economic and Environmental Impact of Phosphorus
Molen, D. T. van der and Boers, P. C. M. (1994). Removal from Waste Water in the European
Influence of internal loading on phosphorus Community. London: Selper Publications.
concentration in shallow lakes before and after Mortimer, C. H. (1941). The exchange of dissolved
reduction of the external loading. Hydrobiologia, substances between mud and water in lakes. I.
275/276, 279–389. Journal of Ecology, 29, 280–329.
Moll, R. A. and Rohlf, F. J. (1981). Analysis of temporal (1942). The exchange of dissolved substances
and spatial variability in a Long Island saltmarsh. between mud and water in lakes. II. Journal of
Journal of Experimental Marine Biology and Ecology, Ecology, 30, 147–201.
51, 133–44. (1974). Lake hydrodynamics. Mitteilungen der
Moll, R. A. and Stoermer, E. F. (1982). A hypothesis internationale Vereinigung für theoretische und
relating trophic status and subsurface angewandte Limnologie, 29, 124–97.
chlorophyll maxima in lakes. Archiv für Moss, B. (1967). Vertical heterogeneity in the water
Hydrobiologie, 94, 425–40. column of Abbot’s Pond. II. The influence of
Montagnes, D. J. S., Berges, J. A., Harrison, P. J. and physical and chemical conditions on the spatial
Taylor, F. J. R. (1994). Estimating carbon, nitrogen, and temporal distribution on the phytoplankton
protein and chlorophyll a from volume in marine and of a community of epipelic algae. Journal of
phytoplankton. Limnology and Oceanography, 39, Ecology, 57, 397–414.
1044–60. (1972). The influence of environmental factors on
Moore, J. W. (1980). Seasonal distribution of the distribution of freshwater algae: an
phytoplankton in Yellowknife Bay, Great Slave experimental study. I. Introduction and the
Lake. Internationale Revue des gesamten influence of calcium concentration. Journal of
Hydrobiologie, 65, 283–93. Ecology, 60, 917–32.
Moore, L. R., Rocap, G. and Chisholm, S. (1998). (1973a). The influence of environmental factors on
Physiology and molecular phylogeny of coexisting the distribution of freshwater algae: an
Prochlorococcus ecotypes. Nature, 393, 464–7. experimental study. II. The role of pH and the
Morabito, G., Ruggiu, D. and Panzani, P. (2002). Recent carbon dioxide–bicarbonate system. Journal of
dynamics (1995–1999) of the phytoplankton Ecology, 61, 157–77.
assemblages in Lago Maggiore as a basic tool for (1973b). The influence of environmental factors on
defining association patterns in the Italian deep the distribution of freshwater algae: an
lakes. Journal of Limnology, 61, 129–45. experimental study. III. Effects of temperature,
Morel, A. (1991). Light and marine photosynthesis: a vitamin requirements and inorganic nitrogen
spectral model with geochemical and compounds on growth. Journal of Ecology, 61,
climatological implications. Progress in 179–92.
Oceanography, 26, 263–306. (1973c). The influence of environmental factors on
Morel, A. and Bricaud, A. (1981). Theoretical results the distribution of freshwater algae: an
concerning light absorption in a discrete experimental study. IV. Growth of test species in
482 REFERENCES
natural lake waters and conclusions. Journal of für theoretische und angewandte Limnologie, 21,
Ecology, 61, 193–211. 1695–716.
(1983). The Norfolk Broadland: experiments in the (1986). The seasonality of phytoplankton in the
restoration of a complex wetland. Biological North American Great Lakes: a comparative
Reviews, 58, 521–61. synthesis. Hydrobiologia, 138, 85–115.
(1990). Engineering and biological approaches to (1996). Phytoplankton Dynamics in the North American
the restoration from eutropication of shallow Great Lakes, vol. 1, Lakes Ontario, Erie and St. Clair.
lakes in which aquatic plant communities are Amsterdam: SPB Science Publishers.
important components. Hydrobiologia, 200/201, (2000). Phytoplankton Dynamics in the North American
367–77. Great Lakes, vol. 2, The Upper lakes, Huron, Superior
(1992). The scope of biomanipulation for improving and Michigan. Leiden: Backhuys.
water quality. In Eutrophication: Research and (2001). An overview of the changing flora and fauna
Application to Water Supply, ed. D. W. Sutcliffe and of the North American Great Lakes. I.
J. G. Jones, pp. 73–81. Ambleside: Freshwater Phytoplankton and microbial food web. In The
Biological Association. Great Lakes of the World (GLOW): Food Web, Health
Moss, B. and Balls, H. (1989). Phytoplankton and Integrity, ed. M. Munawar and R. E. Hecky,
distribution in a temperate floodplain lake and pp. 219–75. Leiden: Backhuys.
river system. II. Seasonal changes in the Munk, W. H. and Riley, J. A. (1952). Absorption of
phytoplankton communities and their control by nutrients by aquatic plants. Journal of Marine
hydrology and nutrient availability. Journal of Research, 11, 215–40.
Plankton Research, 11, 839–67. Murphy, J. and Riley, J. P. (1962). A modified
Moss, B., Madgwick, J. and Phillips, G. L. (1996). A single-solution method for the determination of
Guide to the Restoration of Nutrient-Enriched Shallow phosphate in natural waters. Analytica Chimica
Lakes. Norwich: Environment Agency and Broads Acta, 27, 1–36.
Authority. Murphy, T. P., Lean, D. R. S. and Nalewajko, C. (1976).
Moustaka-Gouni, M. (1993). Phytoplankton succession Blue-green algae: their excretion of iron-selective
and diversity in a warm, monomictic, relatively chelators enables them to dominate other algae.
shallow lake: Lake Volvi, Macedonia, Greece. Science, 192, 900–2.
Hydrobiologia, 249, 33–42. Murray, A. and Hunt, T. (1993). The Cell Cycle: An
Mugidde, R. (1993). The increase in phytoplankton Introduction. New York: W. H. Freeman.
primary productivity and biomass in Lake Murray, A. W. and Kirschner, M. W. (1991). What
Victoria (Uganda). Verhandlungen der internationale controls the cell cycle? Scientific American, 264,
Vereinigung für theoretische und angewandte 34–41.
Limnologie, 25, 846–9. Murrel, M. C. (2003). Bacterioplankton dynamics in a
Mullin, M. M., Sloane, P. R. and Eppley, R. W. (1966). subtropical estuary: evidence for substrate
Relationships between carbon content, cell limitation. Aquatic Microbial Ecology, 32,
volume and area in phytoplankton. Limnology and 239–50.
Oceanography, 11, 307–11. Nadin-Hurley, C. M. and Duncan, A. (1976). A
Munawar, M. and Fahnenstiel, G. L. (1982). The comparison of daphnid gut particles with the
Abundance and Significance of Ultraplankton and sestonic particles in two Thames Valley reservoirs
Microalgae at an Offshore Station in Central Lake throughout 1970 and 1971. Freshwater Biology, 6,
Superior, Canadian Technical Report No. 1153. 109–23.
Ottawa: Fisheries and Marine Sciences of Canada. Nakamura, M. and Chang, W. Y. B. (2001). Lake Biwa:
Munawar, M. and Munawar, I. F. (1975). Some largest lake in Japan. In The Great Lakes of the
observations on the growth of diatoms in Lake World (GLOW): Food Web, Health and Integrity, ed. M.
Ontario, with emphasis on Melosira binderana Munawar and R. E. Hecky, pp. 151–82. Leiden:
Kütz. during thermal bar conditions. Archiv für Backhuys.
Hydrobiologie, 75, 490–9. Nakanishi, M. (1984). Phytoplankton. In Lake Biwa, ed.
(1981). A general comparison of the taxonomic S. Horie, pp. 154–74. Dordrecht: W. Junk.
composition and size analyses of the Nakano, S.-I., Ishi, N., Mange, P. M. and Kawabata, Z.
phytoplankton of the North American Great (1998). Trophic roles of heterotrophic
Lakes. Verhandlungen der internationale Vereinigung nanoflagellates and ciliates among planktonic
REFERENCES 483
organisms in a hypereutrophic pond. Aquatic the English Lake District: irradiance-dependent

Micobial Ecology, 16, 153–61. phytoplankton dynamics during the spring
Nakano, S.-I., Mitamura, O., Sugiyama, M. et al. (2003). diatom maximum. Limnology and Oceanography, 36,
Vertical planktonic structure in the central basin 751–60.
of Lake Baikal in Summer, 1999, with special Neale, P. J., Heaney, S. I. and Jaworski, G. H. M. (1991b).
reference to the microbial food web. Limnology, 4, Responses to high irradiance contribute to the
155–60. decline of the spring diatom maximum. Limnology
Nakatsu, C. and Hutchinson, T. (1988). Extreme metal and Oceanography, 36, 761–8.
and acid tolerance of Euglena mutabilis and an Newman, J. R. and Barrett, P. R. F. (1993). Control of
associated yeast from Smoking Hills, Northwest Microcystis aeruginosa by decomposing barley straw.
Territories, and their apparent mutualism. Journal of Aquatic Plant Management, 31, 203–6.
Microbial Ecology, 16, 213–31. Nielsen, S. L. and Sand-Jensen, K. (1990). Allometric
Nalepa, T. F. and Fahnenstiel, G. L. (1995). Dreissena scaling of maximal photosynthetic growth rate to
polymorpha in Saginaw Bay, Lake Huron surface-to-volume ratio. Limnology and
ecosystem: overview and perspective. Journal of Oceanography, 35, 177–81.
Great Lakes Research, 21, 411–16. Nielsen, S. N. (1992). Application of Maximum Energy in
Nalewajko, C. (1966). Dry weight, ash and volume data Structural Dynamic Models. Copenhagen:
for some freshwater planktonic algae. Journal of Miljøministeriet.
the Fisheries Research Board of Canada, 23, 1285–8. Nilssen, J.-P. (1984). Tropical lakes – functional ecology
Nalewajko, C. and Lean, D. R. S. (1978). Phosphorus and future development: the need for a
kinetics–algal growth relationships in batch process-oriented approach. Hydrobiologia, 113,
cultures. Mitteilungen der internationale Vereinigung 231–42.
für theoretische und angewandte Limnologie, 21, Nimer, N. A., Iglesias-Rodriguez, M. D. and Merrett,
184–92. M. J. (1997). Bicarbonate utilisation by marine
Naselli-Flores, L. and Barone, R. (2003). Steady-state phytoplankton species. Journal of Phycology, 33,
assemblages in a Mediterranean hypertrophic 625–31.
reservoir: the role of Microcystis ecomorphological Nixdorf, B., Fyson, A. and Krumbeck, H. (2001).
variability in maintaining an apparent Review: plant life in extremely acidic waters.
equilibrium. Hydrobiologia, 502, 133–43. Environmental and Experimental Botany, 46,
Naselli-Flores, L., Padisák, J., Dokulil, M. T. and 203–11.
Chorus, I. (2003). Equilibrium/steady-state concept Nixon, S. W. (1988). Physical energy inputs and
in phytoplankton ecology. Hydrobiologia, 502, comparative ecology of lake and marine
395–403. ecosystems. Limnology and Oceanography, 33,
Nasev, D., Nasev, S. and Guiard, V. (1978). Statistiche 1005–25.
Auswertung von Planktonuntersuchungen. II. Noy-Meir, I. (1975). Stability of grazing systems:
Räumliche Verteilung des Planktons. application of predator–prey graphs. Journal of
Konfidenzintervalle für die Individuendichte Ecology, 63, 459–81.
(Ind./l) des Phytoplanktons. Wissenschaftliche Nyholm, N. (1977). Kinetics of phosphorus-limited
Zeitschrift der Wilhem-Pieck-Universitat, 27, 357–61. growth. Biotechnology and Bioengineering, 19,
Naumann, E. (1919). Några synpunkter angende 467–72.
limnoplanktons ökologgi med särskild hänsyn Obernosterer, I., Reitner, B. and Herndl, G. J. (1999).
till fytoplankton. Svensk botanisk Tidskrift, 13, Contrasting effects of solar radiation on dissolved
129–63. organic matter and its bioavailability to marine
Nauwerck, A. (1963). Die Beziehungen zwischen bacterioplankton. Limnology and Oceanography, 44,
zooplankton und phytoplankton in See Erken. 1645–54.
Symbolae botanicae Upsaliensis, 17(5), 1–163. Obernosterer, I., Kawasaki, N. and Benner, R. (2003).
Neale, P. J. (1987). Algal photoinhibition and P-limitation of respiration in the Sargasso Sea
photosynthesis in the aquatic environment. In and uncoupling of bacteria from P regeneration
Photoinhibition, ed. D. Kyle, C. J. Arntzen and B. in size-fractionation experiments. Aquatic
Osmond, pp. 39–65. Amsterdam: Elsevier. Microbial Ecology, 32, 229–37.
Neale, P. J., Talling, J. F., Heaney, S. I., Lund, J. W. G. O’Brien, K. R., Ivey, G. N., Hamilton, D. P., Waite, A. M.
and Reynolds, C. S. (1991a). Long time series from and Visser, P. M. (2003). Simple mixing criteria for
484 REFERENCES
the growth of negatively buoyant phytoplankton. Plankton Communities, NATO Conference Series, IV,
Limnology and Oceanography, 48, 1326–37. No. 3, ed. J. Steele, pp. 21–42. New York: Plenum
Odum, E. P. (1969). The strategy of ecosystem Press.
development. Science, 164, 262–70. (1980). Diffusion and Ecological Problems. New York:
Odum, H. T. (1986). Energy in ecosystems. In Ecosystem Springer-Verlag.
Theory and Application, ed. N. Polunin, pp. 337–69. Oliver, R. L. (1994). Floating and sinking in
New York: John Wiley. gas-vacuolate Cyanobacteria. Journal of Phycology,
(1988). Self-organization, transformity and 30, 161–73.
information. Science, 242, 1132–8. Oliver, R. L. and Whittington, J. (1998). Using
(2002). Explanations of ecological relationships with measurements of variable chlorophyll-a
energy systems concepts. Ecological Modelling, 158, fluorescence to investigate the influence of water
201–11. movement on the photochemistry of
Oglesby, R. T. and Schaffner, W. R. (1975). The phytoplankton. In Physical Processes in Lakes and
response of lakes to phosphorus. In Nitrogen and Oceans, Coastal and Estuarine Studies No. 54, ed.
Phosphorus: Food Production, Waste and the J. Imberger, pp. 517–34. Washington, DC:
Environment, ed. K. S. Porter, pp. 23–57. Ann Arbor, American Geophysical Union.
MI: Ann Arbor Science. Oliver, R. L., Kinnear, A. J. and Ganf, G. G. (1981).
Ogutu-Ohwayu, R. (1990). The decline of native fishes Measurements of cell density of three freshwater
in Lake Victoria and Kyoga (East Africa) and the phytoplankters by density gradient
impact of introduced species, especially the Nile centrifugation. Limnology and Oceanography, 26,
perch, Lates niloticus, and the Nile tilapia, 285–94.
Oreochromis niloticus. Environmental Biology of Fishes, Oliver, R. L., Thomas, R. N., Reynolds, C. S. and Walsby,
27, 81–96. A. E. (1985). The sedimentation of buoyant
O’Gorman, R., Bergtedt, R. A. and Eckert, T. H. (1987). Microcystis colonies caused by precipitation with
Prey fish dynamics and salmonine predator an iron-containing colloid. Proceedings of the Royal
growth in Lake Ontario, 1978–84. Canadian Journal Society of London B, 223, 511–28.
of Fisheries and Aquatic Sciences, 44 (Suppl. 2), Olrik, K. (1981). Succession of phytoplankton in
390–403. response to environmental factors in Lake Arresø,
Ohmori, M. and Hattori, A. (1974). Effect of ammonia North Zealand, Denmark. Schweizerische Zeitschrift
on nitrogen fixation by the blue-green alga für Hydrologie, 43, 6–15.
Anabaena cylindrica. Plant and Cell Physiology, 15, (1994). Phytoplankton Ecology. Copenhagen:
131–42. Miljøministeriet.
Ohnstad, F. R. and Jones, J. G. (1982). The Jenkin Omata, T., Takahashi, Y., Yamaguchi, O. and
Surface-Mud Sampler, Occasional Publication No. Nishimura, T. (2002). Structure, function and
15. Ambleside: Freshwater Biological regulation of the cyanobacterial high-affinity
Association. bicarbonate transporter, BCT-1. Functional Plant
Ojala, A. and Jones, R. I. (1993). Spring development Biology, 29, 151–9.
and mitotic division pattern of a Cryptomonas sp. Organisation for Economic Co-operation and
in an acidified lake. European Journal of Phycology, Development (1982). Eutrophication of Waters:
28, 17–24. Monitoring, Assessment and Control. Paris: OECD.
Okada, M. and Aiba, S. (1983). Simulation of water Orr, P. T. and Jones, G. J. (1998). Relationship between
bloom in a eutrophic lake. III. Water Research, 17, microcystin production and cell division rates in
883–93. nitrogen-limited Microcystis aeruginosa cultures.
Okino, T. (1973). Studies on the blooming of Microcystis Limnology and Oceanography, 43, 1604–14.
aeruginosa. Japanese Journal of Botany, 20, 381–402. Osmond, C. B. (1981). Photorespiration and
Økland, J. and Økland, K. A. (1986). The effects of acid photosynthesis: some implications for the
deposition on benthic animals in lakes and energetics of photosynthesis. Biochimica et
streams. Experimentia, 42, 471–86. Biophysica Acta, 639, 77–98.
Okubo, A. (1971). Oceanic diffusion diagrams. Deep-Sea O’Sullivan, P. (2003). Palaeolimnology. In The Lakes
Research, 18, 789–802. Handbook, vol. 1, ed. P. E. O’Sullivan and C. S.
(1978). Horizontal dispersion and critical scales of Reynolds, pp. 609–66. Oxford: Blackwell
phytoplankton patches. In Spatial Pattern in Science.
REFERENCES 485
Owens, O. van H. and Esaias, W. (1976). Physiological and the relation of form resistance to
responses of phytoplankton to major morphological diversity of plankton: an
environmental factors. Annual Review of Plant experimental study. Hydrobiologia, 500, 243–57.
Physiology, 27, 461–83. Padisák, J., Scheffler, W., Sı́pos, C. et al. (2003b). Spatial
Paasche, E. (1964). A tracer study of the inorganic and temporal pattern of development and decline
carbon uptake during coccolith formation and of the spring diatom populations in Lake Stechlin
photosynthesis in the coccolithophorid Coccolithus in 1999. Advances in Limnology, 58, 135–55.
huxleyi. Physiologia Plantarum, Suppl. III, 1–82. Padisák, J., Borics, J., Fehér, G. et al. (2003c). Dominant
Paasche, E. (1973a). Silicon and the ecology of marine species, functional assemblages and frequency of
plankton diatoms. I. Thalassiosira pseudonana equilibrium phases in late summer
growth in a chemostat with silicate as limiting phytoplankton assemblages in Hungarian small,
nutrient. Marine Biology, 19, 117–26. shallow lakes. Hydrobiologia, 502, 169–76.
(1973b). Silicon and the ecology of marine plankton Paerl, H. W. (1988). Growth and reproductive strategies
diatoms. I. Silicate uptake kinetics in five diatom of freshwater blue-green algae (Cyanobacteria). In
species. Marine Biology, 19, 262–9. Growth and Reproductive Strategies of Freshwater
(1980). Silicon. In The Physiological Ecology of Phytoplankton, ed. C. D. Sandgren, pp. 261–315.
Phytoplankton, ed. I. Morris, pp. 259–84. Oxford: Cambridge: Cambridge University Press.
Blackwell Scientific Publications. Paerl, H. W., Tucker, J. and Bland, P. T. (1983).
Padan, E. and Shilo, M. (1973). Cyanophages: viruses Carotenoid enhancement and its role in
attacking blue-green algae. Bacteriological Reviews, maintaining blue-green algal (Microcystis
37, 343–70. aeruginosa) surface blooms. Limnology and
Padisák, J. (1992). Seasonal succession of Oceanography, 28, 847–57.
phytoplankton in a large shallow lake (Balaton, Pahl-Wostl, C. (1995). The Dynamic Nature of Ecosystems.
Hungary): a dynamic approach to ecological Chichester: John Wiley.
memory, its possible role and mechanisms. (2003). Self-regulation of limnetic ecosystems. In The
Journal of Ecology, 80, 217–30. Lakes Handbook, vol. 1, ed. P. O’Sullivan and C. S.
(1993). The influence of different disturbance Reynolds, 583–608. Oxford: Blackwell Science.
frequencies on the species richness, diversity and Pahl-Wostl, C. and Imboden, D. M. (1990). DYPHORA: a
equitability of phytoplankton of shallow lakes. dynamic model of the rate of photosynthesis of
Hydrobiologia, 249, 135–56. alga. Journal of Plankton Research, 12, 1207–21.
(1997). Cylindrospermopsis raciborskii (Wolszynska) Paine, R. T. (1966). Food-web complexity and species
Seenaayya et Subba Raju, an expanding, highly diversity. American Naturalist, 100, 65–75.
adaptive Cyanobacterium: worldwide distribution (1980). Food webs: linkage, interaction, strength and
and a review of its ecology. Archiv für Hydrobiologie community infrastructure. Journal of Animal
(Suppl.), 107, 563–93. Ecology, 49, 667–85.
(2003). Phytoplankton. In The Lakes Handbook, vol. 1, Paine, R. T., Tegner, M. J. and Johnson, E. A. (1998).
ed. P. E. O’Sullivan and C. S. Reynolds, Compound perturbations yield ecological
pp. 251–308. Oxford: Blackwell Science. surprises. Ecosystems, 1, 535–45.
Padisák, J. and Reynolds, C. S. (1998). Selection of Park, Y.-S., Céréghino, R., Compin, A. and Lek, S.
photoplankton associations in Lake Balaton, (2003). Applications of artificial neural networks
Hungary, in response to eutrophication and for patterning and predicting aquatic insect
restoration measures, with special reference to species richness in rinning waters. Ecological
the cyanoprokaryotes. Hydrobiologia, 384, 41–53. Modelling, 160, 265–80.
(2003). Shallow lakes: the absolute, the relative, the Parsons, T. R. and Takahashi, M. (1973). Environmental
functional and the pragmatic. Hydrobiologia, control of phytoplankton cell size. Limnology and
506/509, 1–11. Oceanography, 18, 511–15.
Padisák, J., Krienitz, L., Scheffler, W. et al. (1998). Pasciak, W. J. and Gavis, J. (1974). Transport limitation
Phytoplankton succession in the oligotrophic of nutrient uptake in phytoplankton. Limnology
Lake Stechlin (Germany) in 1994 and 1995. and Oceanography, 19, 881–9.
Hydrobiologia, 370, 179–97. (1975). Transport-limited nutrient-uptake rates in
Padisák, J., Soróczki-Pintér, É. and Rezner, Z. (2003a). Ditylum brightwelli. Limnology and Oceanography, 20,
Sinking properties of some phytoplankton shapes 604–17.
486 REFERENCES
Passow, U., Engel, A. and Ploug, H. (2003). The role of Petr, T. (1977). Bioturbation and exchange of
aggregation for the dissolution of diatom chemicals in the mud–water interface. In
frustules. FEMS Microbiology Ecology, 46, 247–55. Interactions between Sediments and Freshwater, ed.
Pavoni, M. (1963). Die Bedeutung des Nannoplanktons H. L. Golterman, pp. 216–26. The Hague: W. Junk.
im Vergleich zum Netzplankton. Schweizerische Petrova, N. A. (1986). Seasonality of Melosira plankton
Zeitschrift für Hydrologie, 25, 219–341. in great northern lakes. Hydrobiologia, 138, 65–73.
Pearsall, W. H. (1921). The development of vegetation Pfiester, L. A. and Anderson, D. M. (1987).
in the English lakes, considered in relation to the Dinoflagellate reproduction. In The Biology of
general evolution of glacial lakes and rock basins. Dinoflagellates, ed. F. J. R. Taylor, pp. 611–44.
Proceedings of the Royal Society of London B, 92, Oxford: Blackwell Scientific Publications.
259–84. Phillips, G. L., Eminson, D. F. and Moss, B. (1978). A
(1922). A suggestion as to factors influencing the mechanism to account for macrophyte decline in
distribution of free-floating vegetation. Journal of progressively eutrophicated freshwaters. Aquatic
Ecology, 9, 241–53. Botany, 4, 103–26.
(1924). Phytoplankton and environment in the Pick, F. R. (2000). Predicting the abundance and
English Lake District. Revue d’Algologie, 50, 1–5. production of photosynthetic picoplankton in
(1932). Phytoplankton in the English Lakes. II. The temperate lakes. Verhandlungen der internationale
composition of the phytoplankton in relation to Vereinigung für theoretische und angewandte
dissolved substances. Journal of Ecology, 20, Limnologie, 27, 1884–9.
241–58. Pick, F. R., Nalewajko, C. and Lean, D. R. S. (1984). The
Pedrós-Alió, C., Gasol, J. M. and Guerrero, R. (1987). On origin of a metalimnetic chrysophyte peak.
the ecology of a Cryptomonas phaseolus population Limnology and Oceanography, 29, 125–34.
forming a melimnetic bloom in Lake Cisó, Spain: Pickett, S. T. A., Kolasa, I., Armesto, I. I. and Collins,
annual distribution and loss factors. Limnology S. L. (1989). The ecological concept of disturbance
and Oceanography, 32, 285–98. and is expression at various hierarchical levels.
Pedrozo, F., Kelly, L., Diaz, M. et al. (2001). First results Oikos, 54, 129–36.
on the water chemistry, algae and trophic status Pickett-Heaps, J. D., Schmid, A.-M. M. and Edgar, L. E.
of an Andean acidic lake system of volcanic (1990). The cell biology of diatom valve
origin in Patagonia (Lake Caviahue). Hydrobiologia, formation. In Progress in Phycological Research, vol.
452, 129–37. 7, ed. F. E. Round and D. J. Chapman, pp. 1–168.
Pennington, W. (1943). Lake sediments: the bottom Bristol: Biopress.
deposits of the North Basin of Windermere. New Pielou, E. C. (1974). Population and Community Ecology.
Phytologist, 42, 1–27. New York: Gordon and Breach.
(1947). Studies of the post-glacial history of the (1975). Ecological Diversity. New York: John Wiley.
British vegetation. VII. Lake sediments: pollen Pinevich, A. B., Velichko, N. B. and Bazanova, A. V.
diagrams from the bottom deposits of the North (2000). Prochlorophytes twenty years on. Russian
Basin of Windermere. Philosophical Transactions of Journal of Plant Physiology, 47, 639–43.
the Royal Society of London B, 233, 137–75. Pirie, N. W. (1969). Food Resources, Conventional and
Pennington, W. (1969). The History of the British Novel. Harmondsworth: Penguin Books.
Vegetation. London: English Universities Press. Plisnier, P.-D. and Coenens, E. J. (2001). Pulsed and
Perry, J. J. (2002). Diversity of microbial heterotrophic dampened annual limnological fluctuations in
metabolism. In Biodiversity of Microbial Life, ed. J. T. Lake Tanganyika. In The Great Lakes of the World
Stanley and A.-L. Reysenbach, pp. 155–80. New (GLOW): Food Web, Health and Integrity, ed. M.
York: Wiley-Liss. Munawar and R. E. Hecky, pp. 83–96. Leiden:
Peters, R. H. (1987). Metabolism in Daphnia. Memorie Backhuys.
dell’Istituto italiano di Idrobiologia, 45, 193– Platt, T. and Denman, K. (1980). Patchiness in
243. phytoplankton distribution. In The Physiological
Petersen, R. (1975). The paradox of the plankton: an Ecology of Phytoplankton, ed. I. Morris, pp. 413–31.
equilibrium hypothesis. American Naturalist, 109, Oxford: Blackwell Scientific Publications.
35–49. Platt, T. and Sathyendranath, S. (1988). Oceanic
Peterson, B. J. (1978). Radiocarbon uptake: its relation primary production: estimation by remote
to net particulate carbon production. Limnology sensing at local and regional scales. Science, 241,
and Oceanography, 23, 179–84. 1613–20.
REFERENCES 487
(1999). Spatial structure of pelagic ecosystem Porter, K. G., Sherr, E. B., Sherr, B. F., Pace, M. and
processes in the global ocean. Ecosystems, 2, Sanders, R. W. (1985). Protozoa in planktonic
384–94. food webs. Journal of Protozoology, 32, 409–
Pokorný, J. and Květ, J. (2003). Aquatic plants and lake 15.
ecosystems. In The Lakes Handbook, vol. 1, ed. P. E. Post, A. F., Wit, R. de and Mur, L. R. (1985).
O’Sullivan and C. S. Reynolds, pp. 309–40. Oxford: Interactions between temperature and light
Blackwell Science. intensity on growth and photosynthesis of the
Polishchuk, L. V. (1999). Contribution analysis of cyanobacterium, Oscillatoria agarrdhii. Journal of
disturbance-caused changes in phytoplankton Plankton Research, 7, 487–95.
diversity. Ecology, 80, 721–5. Post, D. M. (2002a). Using stable isotopes to estimate
Pollingher, U. (1988). Freshwater armoured trophic position: models, methods and
dinoflagellates: growth, reproductive strategies assumptions. Ecology, 83, 703–18.
and population dynamics. In The Growth and (2002b). The long and short of food chains. Trends in
Reproductive Strategies of Freshwater Phytoplankton, Ecology and Evolution, 17, 269–77.
ed. C. D. Sandgren, pp. 134–74. Cambridge: Post, D. M., Conners, M. E. and Goldberg, D. S. (2000a).
Cambridge University Press. Prey preference by a top predator and the
(1990). Effects of latitude on phytoplankton stability of linked food chains. Ecology, 81,
composition and abundance in large lakes. In 8–14.
Large Lakes: Ecological Structure and Function, ed. Post, D. M., Pace, M. L. and Hairston, N. G. (2000b).
M. M. Tilzer and C. Serruya, pp. 368–402. New Ecosystem size determines food-chain length in
York: Springer-Verlag. lakes. Nature, 405, 1047–9.
Pollingher, U. and Berman, T. (1977). Quantitative and Post, J. R. (1990). Metabolic allometry of larval and
qualitative changes in the phytoplankton in Lake juvenile yellow perch (Perca flavescens): in situ
Kinneret, Israel, 1972–1985. Oikos, 29, 418–28. estimates and bioenergetic models. Canadian
Pollingher, U. and Serruya, C. (1976). Phased division Journal of Fisheries and Aquatic Sciences, 47, 554–60.
in Peridinium cinctum f. westii (Dinophyceae) and Postgate, J. R. (1987). Nitrogen Fixation, 2nd edn.
development of the Lake Kinneret (Israel) bloom. London: Edward Arnold.
Journal of Phycology, 12, 163–70. Pourriot, R. (1977). Food and feeding habits of
Pomeroy, L. (1974). The ocean’s food web: a changing Rotifera. Ergebnisse der Limnologie, 8, 243–60.
paradigm. BioScience, 24, 499–504. Prairie, Y. T., Duarte, C. M. and Kalff, J. (1989).
Poole, H. H. and Atkins, W. R. G. (1929). Photoelectric Unifying nutrient : chlorophyll relationships in
measurements of submarine illumination lakes. Canadian Journal of Fisheries and Aquatic
throughout the year. Journal of the Marine Biological Sciences, 46, 1176–82.
Association of the United Kingdom, 16, 297–324. Preisendorfer, R. W. (1976). Hydrological Optics.
Popovskaya, G. I. (2001). Ecological monitoring of Washington, DC: US Department of Commerce.
phytoplankton in Lake Baikal. Aquatic Ecosystem Preston, T., Stewart, W. D. P. and Reynolds, C. S. (1980).
Health and Management, 3, 215–25. Bloom-forming Cyanobacterium Microcystis
Porter, J. and Jost, M. (1973). Light-shielding by gas aeruginosa overwinters on sediment surface.
vacuoles in Microcystis aeruginosa. Journal of General Nature, 288, 365–7.
Microbiology, 75, xxii. Pretorius, G. A., Krüger, G. H. J. and Eloff, J. N. (1977).
(1976). Physiological effects of the presence and The release of nanocytes durin the growth of
absence of gas vacuoles in Microcystis aeruginosa Microcystis. Journal of the Limnological Society of
Kuetz. emend. Elenkin. Archives of Microbiology, Southern Africa, 3, 17–20.
110, 225–31. Price, G. D., Maeda, S., Omata, T. and Badger, M. R.
Porter, K. G. (1976). Enhancement of algal growth and (2002). Modes of active inorganic carbon uptake
productivity by grazing zooplankton. Science, 192, in the cyanobacterium Synechococcus sp. PCC7942.
1332–4. Functional Plant Biology, 29, 131–49.
Porter, K. G. and Feig, Y. S. (1980). The use of DAPI for Prigogine, I. (1955). Thermodynamics of Irreversible
identifying and counting aquatic microflora. Processes. New York: Interscience.
Limnology and Oceanography, 25, 943–8. Pringsheim, E. G. (1946). Pure Cultures of Algae.
Porter, K. G., Pace, M. L. and Battey, J. F. (1979). Ciliate Cambridge: Cambridge University Press.
protozoans as links in freshwater planktonic food Provasoli, L. and Carlucci, A. F. (1974). Vitamins and
shains. Nature, 277, 563–5. growth regulators. In Algal Physiology and
488 REFERENCES
Biochemistry, ed. W. D. P. Stewart, pp. 741–87. Raven, J. A. (1982). The energetics of freshwater algae:
Oxford: Blackwell Scientific Publications. energy requirements for biosynthesis and volume
Provasoli, L., McLaughlin, J. J. A. and Droop, M. R. regulation. New Phytologist, 92, 1–20.
(1957). The development of artificial media for (1983). The transport and function of silicon in
marine algae. Archiv für Mikrobiologie, 25, plants. Biological Reviews, 58, 179–207.
392–428. (1984). A cost–benefit analysis of photon absorption
Prowse, G. A. and Talling, J. F. (1958). The seasonal by photosynthetic cells. New Phytologist, 98,
growth and succession of plankton algae in the 593–625.
White Nile. Limnology and Oceanography, 3, 222–38. (1991). Implications of inorganic carbon utilisation:
Pugh, G. J. F. (1980). Strategies in fungal ecology. ecology, evolution and geochemistry. Canadian
Transactions of the British mycological Society, 75, Journal of Botany, 69, 908–24.
1–14. (1997). Inorganic carbon acquisition by marine
Purcell, E. M. (1977). Life at low Reynolds number. autotrophs. Advances in Botanical Research, 27,
American Journal of Physics, 45, 3–11. 85–209.
Pushkar’, V. A. (1975). Diurnal distribution of (1998). Small is beautiful: the picophytoplankton.
phytoplankton in nursery ponds. Hydrobiological Functional Ecology, 12, 503–13.
Journal, 11(5), 18–22. (English Translation of Raven, J. A. and Richardson, K. (1984). Dinophyte
original paper in Gidrobiologeskii Zhurnal.) flagella: a cost–benefit analysis. New Phytologist,
Qiu, B. and Gao, K. (2002). Effects of CO2 enrichment 98, 259–76.
on the bloom-forming cyanobacterium Microcystis Raven, J. A., Kübler, J. E. and Beardall, J. (2000). Put
aeruginosa (Cyanophyceae): physiological out the light, then put out the light. Journal of the
responses and relationships with the availability Marine Biological Association of the United Kingdom,
of dissolved inorganic carbon. Journal of Phycology, 80, 1–25.
38, 721–9. Ravera, O. and Vollenweider, R. A. (1968). Oscillatoria
Ramamurthy, V. D. (1970). Antibacterial activity of the rubescens D.C. as an indicator of pollution.
marine blue-green alga Trichodesmium erythraeum Schweizerische Zeitschrift für Hydrologie, 30, 374–80.
in the gastro-intestinal contents of the sea gull, Rawson, D. S. (1956). Algal indictors of trophic lake
Larus brunicephalus. Marine Biology, 6, 74–6. types. Limnology and Oceanography, 1, 18–25.
Ramenskii, L. G. (1938). Introduction to the Geobotanical Ray, J. M., Bhaya, D., Block, M. A. and Grossman, A. R.
Study of Complex Vegetations. Moscow: Selkozgiz. (1991). Isolation, transcription and inactivation of
Ramlal, P. S., Kling, G. W., Ndawula, L. M., Hecky, R. E. the gene for an atypical alkaline phosphatase of
and Kling, H. J. (2001). Diurnal fluctuations in Synechococcus strain PPC 7942. Journal of
PCO2 , DIC, oxygen and nutrients at inshore sites Bacteriology, 173, 4297–309.
in Lake Victoria, Uganda. In The Great Lakes of the Raymond, P. A. and Cole, J. J. (2003). Increase in the
World (GLOW): Food Web, Health and Integrity, ed. M. export of alkalinity from North America’s largest
Munawar and R. E. Hecky, pp. 67–82. Leiden: river. Science, 301, 88–91.
Backhuys. Raymont, J. E. G. (1980). Plankton Productivity in the
Ramsbottom, A. E. (1976). Depth charts of the Cumbrian Oceans, vol. 1, Phytoplankton, 2nd edn. Oxford:
Lakes. Scientific Publication No. 33. Ambleside: Pergamon Press.
Freshwater Biological Association. Raymont, J. E. G. (1983). Plankton productivity in the
Rao, N. N. and Torriani, A. (1990). Molecular aspects of oceans, vol. 2, Zooplankton, 2nd edn. Oxford:
phosphate transport in Escherichia coli. Molecular Pergamon Press.
Microbiology, 4, 1083–90. Recknagel, F. (1997). ANNA: artificial neural network
Raspopov, I. M. (1985). Higher Aquatic Vegetation of Large model for predicting species abundance and
Lakes in the North-Wetern Area of the USSR. succession of blue-green algae. Hydrobiologia, 349,
Leningrad: Nauka. (In Russian.) 47–57.
Rast, W. and Lee, G. F. (1978). Summary Analysis of the Recknagel, F., French, M., Harkonenen, P. and
North American (US Portion) OECD Eutrophication Yabunaka, K.-L. (1997). Artificial neural network
Project: Nutrient Loading Lake Response Relationships approach for modelling and prediction of algal
and Trophic State Indices, Report No. blooms. Ecological Modelling, 96, 11–28.
EPA-600/3-78-008. Corvallis, OR: US Environment Redfield, A. C. (1934). On the proportions of organic
Protection Agency. derivatives in sea water and their relation to the
REFERENCES 489
composition of plankton. In The James Johnstone D. FitzSimons, pp. 201–28. Amsterdam: Excerpta
Memorial Volume, ed. University of Liverpool, Medica.
pp. 176–92. Liverpool: Liverpool University Press. (1978d). The Plankton of the North-West Midland
(1958). The biological control of chemical factors in Meres, Occasional Paper No. 2. Shrewsbury: Cardoc
the environment. American Scientist, 46, 205–21. and Severn Valley Field Club.
Reguera, B., Bravo, I. and Fraga, S. (1995). Autecology (1979a). Seston sedimentation: experiments with
and some life-history stages of Dinophysis acuta Lycopodium spores in a closed system. Freshwater
Ehrenberg. Journal of Plankton Research, 17, Biology, 9, 55–76.
999–1015. (1979b). The limnology of the eutrophic meres of
Reguera, B., Blanco, J., Fernández, M. L. and Wyatt, T. the Shropshire–Cheshire Plain. Field Studies, 5,
(1998). Harmful Algae. Vigo: Xunta de Galici and 93–173.
Intergovernmental Oceanographic Commission of (1980a). Phytoplankton assemblages and their
UNESCO. periodicity in stratifying lake systems. Holarctic
Reynolds, C. S. (1971). The ecology of planktonic Ecology, 3, 141–59.
blue-green algae in the North Shropshire meres. (1980b). Cattle deaths and blue-green algae. Journal
Field Studies, 3, 409–32. of the Institution of Water Engineers and Scientists, 34,
(1972). Growth, gas vacuolation and buoyancy in a 74–6.
natural population of a planktonic blue-green (1983a). A physiological interpretation of the
alga. Freshwater Biology, 2, 87–106. dynamic responses of populations of a planktonic
(1973a). The seasonal periodicity of planktonic diatom to physical variability of the
diatoms in a shallow, eutrophic lake. Freshwater environment. New Phytologist, 95, 41–53.
Biology, 3, 89–110. (1983b). Growth-rate responses of Volvox aureus
(1973b). Growth and buoyancy of Microcystis Ehrenb. (Chlorophyta, Volvocales) to variability in
aeruginosa Kütz. emend. Elenkin in a shallow the physical environment. British Phycological
eutrophic lake. Proceedings of the Royal Society of Journal, 18, 433–42.
London B, 184, 29–50. (1984a). The Ecology of Freshwater Phytoplankton.
(1973c). The phytoplankton of Crose Mere, Cambridge: Cambridge University Press.
Shropshire. British Phycological Journal, 8, (1984b). Phytoplankton periodicity: the interaction
153–62. of form, function and environmental variability.
(1975). Interrelations of photosynthetic behaviour Freshwater Biology, 14, 111–42.
and buoyancy regulation in a natural population (1984c). Artificial induction of surface blooms of
of a blue-green alga. Freshwater Biology, 5, 323–38. Cyanobacteria. Verhandlungen der internationale
(1976a). Succession and vertical distribution on Vereinigung für theoretische und angewandte
phytoplankton in response to thermal Limnologie, 22, 638–43.
stratification in a lowland lake, with special (1986a). Diatoms and the geochemical cycling of
reference to nutrient availability. Journal of silicon. In Biomineralization in the Lower Plants and
Ecology, 64, 529–51. Animals, ed. B. S. C. Leadbeater and R. Ridings,
(1976b). Sinking movements of phytoplankton pp. 269–89. Oxford: Oxford University Press.
indicated by a simple trapping method. II. (1986b). Experimental manipulations of the
Vertical activity ranges in a stratified lake. British phytoplankton periodicity in large limnetic
Phycological Journal, 11, 293–303. enclosures in Blelham Tarn, English Lake District.
(1978a). Notes on the phytoplankton periodicity of Hydrobiologia, 138, 43–64.
Rostherne Mere, Cheshire, 1967–1977. British (1987a) Cyanobacterial water blooms. In Advances in
Phycological Journal, 13, 329–35. Botanical Research, vol. 13, ed. J. Callow,
(1978b). Stratification in natural populations of pp. 67–143. London: Academic Press.
bloom-forming blue-green algae. Verhandlungen der (1987b). Community organization in the freshwater
internationale Vereinigung für theoretische und plankton. In Organization of Communities, Past and
angewandte Limnologie, 20, 2285–92. Present, ed. J. H. R. Gee and P. S. Giller, pp. 297–325.
(1978c). Phosphorus and the eutrophication of Oxford: Blackwell Scientific Publications.
lakes: a personal view. In Phosphorus in the (1987c). The response of phytoplankton
Environment: Its Chemistry and Biochemistry, CIBA communities to changing lake environments.
Foundation Symposium No. 57, ed. R. Porter and Schweizerische Zeitschrift für Hydrologie, 49, 220–36.
490 REFERENCES
(1988a). Functional morphology and the adaptive Ecology: Scale, Pattern and Process, ed. P. S. Giller,
strategies of freshwater phytoplankton. In Growth A. G. Hildrew and D. G. Raffaelli, pp. 141–87.
and Reproductive Strategies of Freshwater Oxford: Blackwell Scientific Publications.
Phytoplankton, ed. C. D. Sandgren, pp. 338–433. (1994b). The long, the short and the stalled: on the
Cambridge: Cambridge University Press. attributes of phytoplankton selected by physical
(1988b). Potamoplankton: paradigms, paradoxes, mixing in lakes and rivers. Hydrobiologia, 289,
prognoses. In Algae and the Aquatic Environment, ed. 9–21.
F. E. Round, pp. 285–311. Bristol: Biopress. (1994c). The ecological basis for the successful
(1988c). The concept of ecological succession biomanipulation of aquatic communities. Archiv
applied to seasonal periodicity of freshwater für Hydrobiologie, 130, 1–33.
phytoplankton. Verhandlungen der internationale (1995a). Succssional change in the planktonic
Vereinigung für theoretische und angewandte vegetation: species, structures, scales. In Molecular
Limnologie, 23, 683–91. Ecology of Aquatic Microbes, ed. I. Joint, pp. 115–32.
(1989a). Relationships among the biological Berlin: Springer-Verlag.
properties, distribution and regulation of (1995b). River plankton: the paradigm regained. In
production by planktonic Cyanobacteria. Toxicity The Ecological Basis of River Management, ed. D.
Assessment, 4, 229–55. Harper and A. J. D. Ferguson, pp. 161–74.
(1989b). Physical determinants of phytoplankton Chichester: John Wiley.
succession. In Plankton Ecology: Succession in (1996a). Plant life of the pelagic. Verhandlungen der
Plankton Communities, ed. U. Sommer, pp. 9–56. internationale Vereinigung für theoretische und
Madison, WI: Brock-Springer. angewandte Limnologie, 26, 97–113.
(1990). Temporal scales of variability in pelagic (1996b). Phosphorus recycling in lakes: evidence
environments and the responses of from large enclosures for the importance of
phytoplankton. Freshwater Biology, 23, 25–53. shallow sediments. Freshwater Biology, 35, 623–45.
(1991). Lake communities: an approach to their (1997a). Vegetation Processes in the Pelagic: A Model for
management for conservation. In The Scientific Ecosystem Theory. Oldendorf: ECI.
Management of Temperate Communities for (1997b). Successional development, energetics and
Conservation, ed. I. F. Spellerberg, F. B. Goldsmith diversity in planktonic communities. In
and M. G. Morris, pp. 199–225. Oxford: Blackwell Biodiversity: An Ecological Perspective, ed. T. Abe, S. R.
Scientific Publications. Levin and M. Higashi, pp. 167–202. New York:
(1992a). Eutrophication and the management of Springer.
planktonic algae: what Vollenweider couldn’t tell (1998a). What factors influence the species
us. In Eutrophication: Research and Application to composition of phytoplankton in lakes of
Water Supply, ed. D. W. Sutcliffe and J. G. Jones, different trophic status? Hydrobiologia, 369/370,
pp. 4–29. Ambleside: Freshwater Biological 11–26.
Association. (1998b). Linkages between atmospheric weather and
(1992b). Algae. In The Rivers Handbook, vol. 1, ed. P. the dynamics of limnetic phytoplankton. In
Calow and G. E. Petts, pp. 195–215. Oxford: Management of Lakes and Reservoirs during Global
Blackwell Scientific Publications. Climate Change, ed. D. G. George, J. G. Jones, P.
(1992c). Dynamics, selection and composition of Punčochář, C. S. Renolds and D. W. Sutcliffe,
phytoplankton in relation to vertical structure in pp. 15–38. Dordrecht: Kluwer.
lakes. Ergebnisse der Limnologie, 35, 13–31. (1998c). The control and management of
(1993a). Swings and roundabouts: engineering the Cyanobacterial blooms. Special Publications of the
environment of algal growth. In Urban Waterside Australian Society of Limnology, 12, 6–22.
Regeneration, Problems and Prospects, ed. K. H. (1999a). With or against the grain: responses of
White, E. G. Bellinger, A. J. Saul, M. Symes phytoplankton to pelagic variability. In Aquatic
and K. Hendry, pp. 330–49. Chichester: Ellis Life-Cycle Strategies, ed. M. Whitfield, J. Matthews
Horwood. and C. Reynolds, pp. 15–43. Plymouth: Marine
(1993b). Scales of disturbance and their role in Biological Association.
plankton ecology. Hydrobiologia, 249, 157–71. (1999b). Non-determinism to probability, or N : P in
(1994a). The role of fluid motion in the dynamics of the community ecology of phytoplankton. Archiv
phytoplankton in lakes and rivers. In Aquatic für Hydrobiologie, 146, 23–35.
REFERENCES 491
(2000a). Phytoplankton designer or how to predict Reynolds, C. S. and Elliott, J. A. (2002). Phytoplankton
compositional responses to trophic state change. diversity: discontinuous assembly responses to
Hydrobiologia, 424, 123–32. environmental forcing. Verhandlungen der
(2000b). Defining sustainability in aquatic systems: internationale Vereinigung für theoretische und
a thermodynamic approach. Verhandlungen der angewandte Limnologie, 28, 336–44.
internationale Vereinigung für theoretische und Reynolds, C. S. and Glaister, M. R. (1993). Spatial and
angewandte Limnologie, 27, 107–17. temporal changes in phytoplankton abundance in
(2001a). Emergence in pelagic communities. Scientia the upper and middle reaches of the River
Marina, 65(Suppl. 2), 5–30. Severn. Archiv für Hydrobiologie (Suppl.), 101, 1–22.
(2001b). Plankton, status and role. In Encyclopedia of Reynolds, C. S. and Irish, A. E. (1997). Modelling
Biodiversiy, vol. 4, ed. S. R. Levin, pp. 569–99. San phytoplankton dynamics in lakes and reservoirs:
Diego, CA: Academic Press. the problem of in-situ growth rates. Hydrobiologia
(2002a). On the interannual variability in 397, 5–17.
phytoplankton production in freshwaters. In (2000). The Phytoplankton of Windermere (English Lake
Phytoplankton Productivity, ed. P. J. leB. Williams, District). Ambleside: Freshwater Biological
D. N. Thomas and C. S. Reynolds, pp. 187–221. Association.
Oxford: Blackwell Science. Reynolds, C. S. and Jaworski, G. H. M. (1978).
(2002b). Ecological pattern and ecosystem theory. Enumeration of natural Microcystis populations.
Ecological Modelling, 158, 181–200. British Phycological Journal, 13, 269–77.
(2002c). Resilience in aquatic ecosystems: hysteresis, Reynolds, C. S. and Lund, J. W. G. (1988). The
homeostasis and health. Aquatic Ecosystem Health phytoplankton of an enriched, soft-water lake
and Management, 5, 3–17. subject to intermittent hydraulic flushing
(2003a). Pelagic community assembly and the (Grasmere, English Lake District). Freshwater
habitat template. Bocconea, 16, 323–39. Biology, 19, 379–404.
(2003b). Planktic community assembly in flowing Reynolds, C. S. and Maberly, S. C. (2002). A simple
water and the ecosystem health of rivers. method for approximating the supportive
Ecological Modelling, 160, 191–203. capacities and metabolic constraints in lakes and
(2003c). The development of perceptions of aquatic reservoirs. Freshwater Biology, 47, 1183–88.
eutrophication and its control. Ecohydrology and Reynolds, C. S. and Petersen, A. C. (2000). The
Hydrobiology, 3, 149–63. distribution of planktonic Cyanobacteria in Irish
Reynolds, C. S. and Allen, S. E. (1968). Changes in the lakes in relation to their trophic states.
phytoplankton of Oak Mere following the Hydrobiologia, 424, 91–9.
introduction of base-rich water. British Phycological Reynolds, C. S. and Reynolds, J. B. (1985). The atypical
Bulletin, 3, 451–62. seasonality of phytoplankton in Crose Mere, 1972:
Reynolds, C. S. and Bellinger, E. D. (1992). Patterns an independent test of the hypothesis that
abundance and dominance of of the variability in the physical environment regulates
phytoplankton of Rostherne Mere, England: community dynamics and structure. British
evidence from an 18-year data set. Aquatic Sciences, Phycological Journal, 20, 227–42.
54, 10–36. Reynolds, C. S. and Rodgers, M. W. (1983). Cell- and
Reynolds, C. S. and Butterwick, C. (1979). Algal colony-division in Eudorina (Chlorophyta:
bioassay of unfertilised and artficially fertilised Volvocales) and some ecological implications.
lake water maintained in Lund Tubes. Archiv für British Phycological Journal, 18, 111–19.
Hydrobiologie (Suppl), 56, 166–83. Reynolds, C. S. and Rogers, D. A. (1976). Seasonal
Reynolds, C. S. and Davies, P. S. (2001). Sources and variations in the vertical distribution and
bioavailability of phosphorus fractions in buoyancy of Microcystis aeruginosa Kütz. emend.
freshwaters: a British perspective. Biological Elenkin in Rostherne Mere, England.
Reviews of the Cambridge Philosophical Society, 76, Hydrobiologia, 48, 17–23.
27–64. Reynolds, C. S. and Walsby, A. E. (1975). Water blooms.
Reynolds, C. S. and Descy, J.-P. (1996). The production, Biological Reviews of the Cambridge Philosophical
biomass and structure of phytoplankton in Society, 50, 437–81.
large rivers. Archiv für Hydrobiologie (Suppl.), 113, Reynolds, C. S. and West, J. R. (1988). Stratification in
161–87. the Severn Estuary: physical aspects and
492 REFERENCES
biological consequences, unpublished report to Reynolds, C. S., Padisák, J. and Kóbor, I. (1993a). A
the Severn Tidal Power Group. Ambleside: localized bloom of Dinobryon sociale in Lake
Freshwater Biological Association. Balaton: some implications for the perception of
Reynolds, C. S. and Wiseman, S. W. (1982). Sinking patchiness and the maintenance of species
losses of phytoplankton in closed limnetic richness. Abstracta Botanica, 17, 251–60.
systems. Journal of Plankton Research, 4, 489–522. Reynolds, C. S., Padisák, J. and Sommer, U. (1993b).
Reynolds, C. S., Jaworski, G. H. M., Cmiech, H. A. and Intermediate disturbance in the ecology of
Leedale, G. F. (1981). On the annual cycle of the phytoplankton and the maintenance of species
blue-green alga Microcystis aeruginosa Kütz. emend. diversity: a synthesis. Hydrobiologia, 249, 183–8.
Elenkin. Philosophical Transactions of the Royal Reynolds, C. S., Desortová, B. and Rosendorf, P. (1998).
Society of London B, 293, 419–77. Modelling the responses of lakes to day-to-day
Reynolds, C. S., Thompson, J. M., Ferguson, A. J. D. and changes in weather. In Management of Lakes and
Wiseman, S. W. (1982a). Loss processes in the Reservoirs during Global Climate Change, ed. D. G.
population dynamics of phytoplankton George, J. G. Jones, P. Punčochář, C. S. Reynolds
maintained in closed systems. Journal of Plankton and D. W. Sutcliffe, pp. 289–95. Dordrecht: Kluwer.
Research, 4, 561–600. Reynolds, C. S., Reynolds, S. N., Munawar, I. F. and
Reynolds, C. S., Morison, H. R. and Butterwick, C. Munawar, M. (2000a). The regulation of
(1982b). The sedimentary flux of phytoplankton phytoplankton population dynamics in the
in the south basin of Windermere. Limnology and world’s great lakes. Aquatic Ecosystem Health and
Oceanography, 27, 1162–75. Management, 3, 1–21.
Reynolds, C. S., Tundisi, J. G. and Hino, K. (1983a). Reynolds, C. S., Dokulil, M. and Padisák, J. (2000b).
Observations on a limnetic Lyngbya population in Understanding the assembly of phytoplankton in
a stably stratified tropical lake (Lagoa Carioca, relation to the trophic spectrum: where are we
eastern Brasil). Archiv für Hydrobiologie, 97, 7–17. now? Hydrobiologia, 424, 147–52.
Reynolds, C. S., Wiseman, S. W., Godfrey, B. M. and Reynolds, C. S., Irish, A. E. and Elliott, J. A. (2001). The
Butterwick, C. (1983b). Some effects of artificial ecological basis for simulating phytoplankton
mixing on the dynamics of phytoplankton responses to environmental change (PROTECH).
populations in large limnetic enclosures. Journal Ecological Modelling, 140, 271–91.
of Plankton Research, 5, 203–34. Reynolds, C. S., Huszar, V. L., Kruk, C., Naselli-Flores, L.
Reynolds, C. S., Wiseman, S. W. and Clarke, M. J. O. and Melo, S. (2002). Towards a functional
(1984). Growth- and loss-rate responses to classification of the freshwater phytoplankton.
intermittent artificial mixing and their potential Journal of Plankton Research, 24, 417–28.
application to the control of planktonic algal Rhee, G.-Y. (1973). A continuous culture study of
biomass. Journal of Applied Ecology, 21, 11–39. phosphate uptake, growth rate and
Reynolds, C. S., Harris, G. P. and Gouldney, D. N. polyphosphate in Scenedesmus sp. Journal of
(1985). Comparison of carbon-specific growth rates Phycology, 9, 495–506.
and rates of cellular increase in large limnetic (1978). Effects of N : P atomic ratios and nitrate
enclosures. Journal of Plankton Research, 7, 791–820. limitation on algal growth, cell composition and
Reynolds, C. S., Montecino, V., Graf, M.-E. and Cabrera, nitrate uptake. Limnology and Oceanography, 23,
S. (1986). Short-term dynamics of a Melosira 10–25.
population in the plankton of an impoundment Rhee, G.-Y. and Gotham, I. J. (1980). Optimum N : P
in central Chile. Journal of Plankton Research, 8, ratios and co-existence of planktonic algae.
715–40. Journal of Phycology, 16, 486–9.
Reynolds, C. S., Oliver, R. L. and Walsby, A. E. (1987). Richardson, T. L., Gibson, C. E. and Heaney, S. I. (2000).
Cyanobacterial dominance: the role of buoyancy Temperature, growth and seasonal succession of
regulation in dynamic lake environments. New phytoplankton in Lake Baikal, Siberia. Freshwater
Zealand Journal of Marine and Freshwater Research, Biology, 44, 431–40.
21, 379–99. Richerson, P. J., Neale, P. J., Wurtsbaugh, W., Alfaro
Reynolds, C. S., Carling, P. A. and Beven, K. J. (1991). Tapia, R. and Vincent, W. F. (1986). Patterns of
Flow in river channels: new insights into temporal variation in Lake Titicaca. I.
hydraulic retention. Archiv für Hydrobiologie, 121, Background, physical and chemical processes.
171–9. Hydrobiologia, 138, 205–20.
REFERENCES 493
Richey, J. E., Melack, J. M. Aufdemkampe, A. K., P. J. leB. Williams, D. N. Thomas and C. S.

Ballester, V. M. and Hess, L. L. (2002). Outgassing Reynolds, pp. 291–317. Oxford: Blackwell Science.
from Amazonian rivers and wetlands as a large Rippka, R., Neilson, A., Kunisawa, R. and Cohen-Bazire,
tropical source of atmospheric CO2 . Nature, 416, C. (1971). Nitrogen fixation by unicellular
617–19. blue-green algae. Archiv für Mikrobiologie, 76,
Richman, S., Bohon, S. A. and Robbins, S. E. (1980). 341–48.
Grazing interactions among freshwater calanoid Rippka, R., Deruelles, J., Waterbury, J. B., Herdman, M.
copepods. In Evolution and Ecology of Zooplankton and Stanier, R. Y. (1979). Generic assignments,
Communities, ed. W. C. Kerfoot, pp. 219–33. strain histories and properties of pure cultures of
Hanover, NH: University of New England press. Cyanobacteria. Journal of General Microbiology, 111,
Ricker, W. E. (1958). Stock and recruitment. Journal of 1–61.
the Fisheries Research Board of Canada, 1, Robarts, R. D. and Howard-Williams, C. (1989). Diel
559–623. changes in fluorescence capacity, photosynthesis
Riddolls, A. (1985). Aspects of nitrogen in Lough and macromolecular synthesis by Anabaena in
Neagh. II. Competition between Aphanizomenon response to natural variations in solar
flos-aquae, Oscillatoria redekei and Oscillatoria irradiance. Ergebnisse der Limnologie, 32, 35–
agardhii. Freshwater Biology, 15, 299–306. 48.
Ridley, J. E. (1970). The biology and management of Robarts, R. D. and Zohary, T. (1987). Temperature
eutrophic reservoirs. Water Treatment and effects on photosynthetic capacity, respiration
Examination, 19, 374–99. and growth rates of bloom-forming
Riebesell, U. and Wolf-Gladrow, D. A. (2002). Supply Cyanobacteria. New Zealand Journal of Marine and
and uptake of inorganic nutrients. In Freshwater Research, 21, 391–9.
Phytoplankton Productivity, ed. P. J. leB. Williams, Robinson, G. A. (1961). Continuous plankton records:
D. N. Thomas and C. S. Reynolds, pp. 109–140. contribution towards a plankton atlas of the
Oxford: Blackwell Science. north-eastern Atlantic and the North Sea. I.
Riebesell, U., Wolf-Gladrow, D. A. and Smetacek, V. Phytoplankton. Bulletins of Marine Ecology, 5,
(1993). Carbon dioxide limitation of marine 81–9.
phytoplankton growth rates. Nature, 361, 249–51. (1965). Continuous plankton records: contribution
Riemann, B. and Christoffersen, K. (1993). Microbial towards a plankton atlas of the north-eastern
trophodynamics in temperate lakes. Marine Atlantic and the North Sea. IX. Seasonal cycles of
Microbial Food Webs, 7, 69–100. phytoplankton. Bulletins of Marine Ecology, 6,
Riemann, B., Havskum, H., Thingstad, F. and Bernard, 104–22.
C. (1995). The role of mixotrophy in pelagic Rodhe, W. (1948). Environmental requirements of
environments. In Molecular Ecology of Aquatic freshwater plankton algae: experimental studies
Microbes, ed. I. Joint, pp. 87–105. Berlin: in the ecology of phytoplankton. Symbolae
Springer-Verlag. Botanicae Upsaliensis, 10, 5–149.
Riemann, L. and Winding, A. (2001). Community Rodhe, W., Vollenweider, R. A. and Nauwerck, A.
dynamics of free-living and particle-associated (1958). The primary production and standing crop
bacterial assemblages during a freshwater of phytoplankton. In Perspectives in Marine Biology,
phytoplankton bloom. Microbial Ecology, 42, ed. A. A. Buzzati-Traverso, pp. 299–322. Berkeley,
274–85. CA: University of California Press.
Riley, G. A. (1944). Carbon metabolism and Roelofs, T. D. and Oglesby, R. T. (1970). Ecological
photosynthetic efficiency. American Scientist, 32, observations on the planktonic cyanophyte
132–4. Gloeotrichia echinulata. Limnology and Oceanography,
(1957). Phytoplankton of the north central Sargasso 15, 244–9.
Sea. Limnology and Oceanography, 2, 252–70. Rojo, C. and Alvares Cobelas, M. (2000). A plea for
Rinehart, K. L., Namikoshi, M. and Choi, B. (1994). more ecology in phytoplankton ecology.
Structure and biosynthesis of toxins from Hydrobiologia, 424, 141–6.
blue-green algae (Cyanobacteria). Journal of Applied Rojo, C., Ortega-Mayagoitia, E. and Alvares Cobelas, M.
Phycology, 6, 159–76. (2000). Lack of pattern among phytoplankton
Ripl, W. and Wolter, K.-D. (2003). Ecosystem function assemblages or what does the exception to the
and degradation. In Phytoplankton Productivity, ed. rule mean? Hydrobiologia, 424, 133–40.
494 REFERENCES
Romanovsky, Yu. I. (1985). Food limitation and Ryther, J. H. (1956). Photosynthesis in the ocean as a
life-history strategies in cladoceran crustaceans. function of light intensity. Limnology and
Ergebnisse der Limnologie, 21, 363–72. Oceanography, 1, 61–70.
Romero, O. E., Lange, C. B. and Wefer, G. (2002). Safferman, R. S. and Morris, M. E. (1963). Algal virus:
Interannual variability (1988–1991) of siliceous isolation. Science, 140, 679–80.
phytoplankton fluxes off northwest Africa. Journal Sakamoto, M. (1966). Primary production by
of Plankton Research, 24, 1035–46. phytoplankton community in some Japanese
Romme, W. H., Everham, E. H., Frelich, L. E., Moritz, lakes and its dependence upon lake depth. Archiv
M. A. and Sparks, R. E. (1998). Are large, für Hydrobiologie, 62, 1–28.
infrequent disturbances quantitatively different Salmaso, N. (2000). Factors affecting the seasonality
from small, frequent disturbances? Ecosystems, 1, and distribution of Cyanobacteria and
524–34. chlorophytes: a case study from the large lakes
Rossolimo, L. L. (1957). Temparatuniy rezhim Ozero south of the Alps, with special reference to Lake
Baykal. Trudy baykalskoi limnologeskoi Stantsii, 16, Garda. Hydrobiologia, 438, 43–63.
1–551. (2003). Life strategies, dominance patterns and
Rother, J. A. and Fay, P. (1977). Sporulation and the mechanisms promoting species coexistence in
development of planktonic blue-green algae in phytoplankton communities along complex
three Salopian meres. Proceedings of the Royal environmental gradients. Hydrobiologia, 502, 13–36.
Society of London B, 196, 317–32. Salmaso, N., Morabito, G., Mosello, R. et al. (2003). A
(1979). Blue-green algal growth and sporulation in synoptic study of phytoplankton in the deep
response to simulated surface-bloom conditions. lakes south of the Alps (Lakes Garda, Iseo, Como,
British Phycological Journal, 14, 59–68. Lugano and Maggiore). Journal of Limnology, 62,
Rothschild, B. J. and Osborn, T. R. (1988). Small-scale 207–27.
turbulence and plankton contact rates. Journal of Salomonsen, J. (1992). Examination of properties of
Plankton Research, 10, 465–74. exergy, power and ascendency along a
Round, F. E. (1965). The Biology of the Algae. London: eutrophication gradient. Ecological Modelling, 62,
Edward Arnold. 171–81.
Round, F. E., Crawford, R. M. and Mann, D. G. (1990). Samuelsson, K., Berglund, J., Haecky, P. and
The Diatoms: Biology and Morphology of the Genera. Andersson, A. (2002). Structural changes in an
Cambridge: Cambridge University Press. aquatic microbial food web caused by inorganic
Rüdiger, W. (1994). Phycobiliproteins and phycobilins. nutrient addition. Aquatic Microbial Ecology, 29,
In Progress in Phycological Research, vol. 10, ed. F. E. 29–38.
Round and D. J. Chapman, pp. 97–135. Bristol: Sand-Jensen, K. (1977). Effect of epiphytes on eelgrass
Biopress. photosynthesis. Aquatic Botany, 3, 55–63.
Rueter, J. G. and Petersen, R. R. (1987). Micronutrient Sanders, R. W., Leeper, D. A., King C. H. and Porter,
effects on cyanobacterial growth and physiology. K. G. (1994). Grazing by rotifers and crustacean
New Zealand Journal of Marine and Freshwater zooplankton on nanoplanktonic protists.
Research, 21, 435–45. Hydrobiologia, 288, 167–81.
Ruttner, F. (1938). Limnologische Studien an einigen Sandgren, C. D. (1988a). General introduction. In
Seen der Östalpen. Archiv für Hydrobiologie, 32, Growth and Reproductive Strategies of Freshwater
167–319. Phytoplankton, ed. C. D. Sandgren, pp. 1–8.
(1953). Fundamentals of Limnology. (Translated by Cambridge: Cambridge University Press.
D. G. Frey and F. E. D. Fry.) Toronto: University of (1988b). The ecology of chrysophyte flagellates: their
Toronto Press. growth and perennation as freshwater
Ruyter van Steveninck, E. D. de, Admiraal, W., phytoplankton. In Growth and Reproductive
Breebart, L., Tubbing, G. M. J. and Zanten, B. van Strategies of Freshwater Phytoplankton, ed. C. D.
(1992). Plankton in the River Rhine: structural Sandgren, pp. 9–104. Cambridge: Cambridge
and functional changes observe during University Press.
downstream transport. Journal of Plankton Research, Sargent, J. R. (1976). The structure, metabolism and
14, 1351–68. function of lipids in marine organisms. In
Ryding, S.-O. and Rast, W. (1989). The Control of Biochemical and Biophysical Perspectives in Marine
Eutrophication in Lakes and Reservoirs. Paris: Biology, ed. D. C. Malins and J. R. Sargent,
UNESCO. pp. 150–212. London: Academic Press.
REFERENCES 495
Sarmiento, J. L. and Wofsy, S. C. (1999). A US Global of the dissolved inorganic carbon system.
Carbon Cycle Plan. Washington, DC: US Global Phycologia, 37, 467–77.
Change Research Program. Scanlan, D. J. and Wilson, W. H. (1999). Application of
Sarno, D., Zingone, A., Saggiomo, V. and Carrada, G. C. molecular techniques to addressing the role of P
(1993). Phytoplankton biomass and species as a key effector in marine ecosystems.
composition in a Mediterranean coastal lagoon. Hydrobiologia, 401, 149–75.
Hydrobiologia, 271, 27–40. Scanlan, D. J., Mann, N. H. and Carr, N. G. (1993). The
Sarvala, J., Tarvainen, M., Salonen, K. and Mölsä, H. response of the picoplanktonic marine
(2002). Pelagic food web as the basis of fisheries cyanobacterium Synechococcus species WH7803 to
in Lake Tanganyika: a bioenergetic modeling phosphate starvation involves a protein
analysis. Aquatic Ecosystem Health and Management, homologous to the periplasmic phosphate-
5, 283–92. binding protein of Escherichia coli. Molecular
Sarvala, J., Rask, M. and Karjalainen, J. (2003). Fish Microbiology, 10, 181–91.
community ecology. In The Lakes Handbook, vol. 1, Scavia, D. and Fahnenstiel, G. L. (1988). From
ed. P. E.O ’Sullivan and C. S. Reynolds, pp. 538–82. picoplankton to fish: complex interactions in the
Oxford: Blackwell Science. Great Lakes. In Complex Interactions in Lake
Sas, H. (1989). Lake Restoration by Reduction of Nutrient Communities, ed. S. R. Carpenter, pp. 85–97. New
Loading: Expectations, Experiences, Extrapolations. St York: Springer-Verlag.
Augustin: Academia Verlag Richarz. Scheffer, M. (1989). Alternative stable states in
Satake, K. and Saijo, Y. (1974) Carbon content and eutrophic shallow freshwater systems: a minimal
metabolic activity of microorganisms in some model. Hydrobiological Bulletin, 23, 73–83.
acid lakes in Japan. Limnology and Oceanography, (1990). Multiplicity of stable states in freshwater
19, 331–8. system. Hydrobiologia, 200/201, 475–86.
Satapoomin, S. (1999). Carbon content of some (1998). Ecology of Shallow Lakes. London: Chapman
common tropical Andaman Sea copepods. Journal and Hall.
of Plankton Research, 21, 2117–23. Scheffer, M., Hosper, S. H., Meijer, M.-L., Moss, B. and
Saubert, S. (1957). Amino-acid utilisation by Nitzschia Jeppesen, E. (1993). Alternative equilibria in
thermalis and Scenedesmus bijugatus. South African shallow lakes. Trends in Ecology and Evolution, 8,
Journal of Science, 53, 335–9. 275–9.
Saunders, P. A., Porter, K. G. and Taylor, B. E. (1999). Scheffer, M., Rinaldi, S., Gragnani, A., Mur, L. R. and
Population dynamics of Daphnia spp. and van Nes, E. H. (1997). On the dominance of
implications for trophic interactions in a small, filamentous Cyanobacteria in shallow, turbid
monomictic lake. Journal of Plankton Research, 21, lakes. Ecology, 78, 272–82.
1823–45. Scheffer, M., Rinaldi, S. and Kuznetsov, Yu. A. (2000).
Savvaitova, K. and Petr, T. (1992). Lake Issyk-kul, Effects of fish on plankton dynamics: a
Kirghizia. International Journal of Salt Lake Research, theoretical analysis. Canadian Journal of Fisheries
1, 21–46. and Aquatic Sciences 57, 1208–19.
Sawyer, C. N. (1966). Basic concepts of eutrophication. Schindler, D. W. (1971). Carbon, nitrogen and the
Journal of the Water Control Federation, 38, 737–44. eutrophication of freshwater lakes. Journal of
Saxby, K. J. (1990). The physiological ecology of Phycology, 7, 321–9.
freshwater chrysophytes with special reference to (1977). Evolution of phosphorus limitation in lakes.
Synura petersenii. Ph.D. thesis, University of Science, 196, 260–2.
Birmingham. Schindler, D. W. and Holmgren, S. K. (1971). Primary
Saxby-Rouen, K. J., Leadbeater, B. S. C. and Reynolds, production and phytoplankton in the
C. S. (1996). Ecophysiological studies on Synura experimental lakes area, north-westerrn Ontario,
petersenii (Synurophyceae). Beiheft, Nova Hedwigia, and other low-carbonate waters and a liquid
114, 111–23. scintillation method for determining 14 C activity
(1997). The growth response of Synura petersenii in photosynthesis. Journal of the Fisheries Research
(Synurophyceae) to photon-flux density, Board of Canada, 28, 189–201.
temperature and pH. Phycologia, 36, Schindler, D. E., Carpenter, S. R., Cottingham, K. I. et
233–43. al. (1996). Food web structure and littoral zone
(1998). The relationship between the growth of coupling to trophic cascades. In Food Webs:
Synura petersenii (Synurophyceae) and components Integration of Patterns and Dynamics, ed. G. A. Polis
496 REFERENCES
and K. O. Winemiller, pp. 96–105. New York: Shimarev, M. N., Granin, N. G. and Zhdanov, A. A.
Chapman and Hall. (1993). Deep ventilation of Lake Baikal water
Scott, J. T., Myer, G. E., Stewart, R. and Walther, E. G. during spring thermal bars. Limnology and
(1969). On the mechanism of Langmuir Oceanography, 38, 1068–72.
circulations and their role in epilimnion mixing. Sicko-Goad, L. M., Schelske, C. L. and Stoermer, E. F.
Limnology and Oceanography, 14, 493–503. (1984). Estimation of intracellular carbon and
Sell, A. and Overbeck, J. (1992). Exudates: silica content of diatoms from natural
phytoplankto-bacterioplankton interactions in assemblages, using morphometric techniques.
Plusssee. Journal of Plankton Research, 14, 1199–215. Limnology and Oceanography, 29, 1170–78.
Serruya, C. (1978). Lake Kinneret, Monographie Sieburth, J.McN., Smetacek, V. and Lenz, J. (1978).
Biologiae Series No. 32. The Hague: W. Junk. Pelagic ecosystem structure: heterotrophic
Serruya, C. and Pollingher, U. (1983). Lakes of the Warm compartments of the plankton and their
Belt. Cambridge: Cambridge University Press. relationship to plankton size fractions. Limnology
Shaboonin, G. D. (1982). Long-term multi-year and Oceanography, 23, 1256–63.
characteristics of the regime of the temperature Siever, R. (1962). Silica solubility, 0–200 ◦ C, and the
in Issyk-kul. Trudy Akademia Nauk Kirghiziaskaya diagenesis of siliceous sediments. Journal of
SSR, 3, 39–46. Geology, 70, 127.
Shannon, C. E. (1948). A mathematical theory of Sigg, L. and Xue, H. B. (1994). Metal speciation:
communication. Bell Systems Technical Journal, 27, concepts, analysis and effects. In Chemistry of
623–56. Aquatic Systems: Local and Global Perspectives, ed. G.
Shannon, C. E. and Weaver, W. (1949). The Mathematical Bidoglio and W. Stumm, pp. 153–81. Dordrecht:
Theory of Communication. Urbana, IL: University of Kluwer.
Illinois Press. Šimek, K., Kojecká, P., Nedoma, J. et al. (1999). Shifts in
Shapiro, J. (1973). Blue-green algae: Why they become bacterial community composition associated with
dominant. Science, 179, 382–4. different microzooplankton size fractions in a
(1990). Current beliefs regarding dominance by eutrophic reservoir. Limnology and Oceanography,
blue-greens: the case for the importance of 44, 1634–44.
CO2 . Verhandlungen der internationalen Vereinigung Simmons, J. (1998). Algal control and destratification
für theoretische und angewandte Limnologie, 24, at Hanningfield Reservoir. Water Science Technology,
38–54. 37/2, 309–16.
Shapiro, J. and Wright, D. J. (1984). Lake restoration Simo, R. (2001). Production of atmospheric sulphur by
by biomanipulation: Round Lake, Minnesota, oceanic plankton: biogeochemical, ecological and
the first two years. Freshwater Biology, 14, evolutionary links. Trends in Ecology and Evolution,
371–83. 16, 287–94.
Shapiro, J., Lamarra, V. and Lynch, M. (1975). Simon, M. I. (1995). Signal transduction in
Biomanipulation: an ecosystem approach to lake microorganisms. In Molecular Ecology of Aquatic
restoration. In Proceedings of a Symposium on Water Microbes, ed. I. Joint, pp. 205–15. Berlin:
Quality Management through Biological Control, ed. Springer-Verlag.
P. L. Brezonik and J. L. Fox, pp. 85–89. Gainesville, Simpson, F. B. and Neilands, J. B. (1976). Sidero-
FL: University of Florida Press. chromes in Cyanophyceae: isolation and
Sharp, J. H. (1977). Excretion of organic matter by characterisation of schizokinen from Anabaena sp.
marine phytoplankton: do healthy cells do it? Journal of Phycology, 12, 44–8.
Limnology and Oceanography, 22, 381–99. Simpson, T. L. and Volcani, B. E. (1981). Introduction.
Sherr, E. B. and Sherr, B. F. (1988). Role of microbesin In Silicon and Siliceous Structures in Biological
pelagic food webs: a revised concept. Limnology Systems, ed. T. L. Simpson and B. E. Volcani,
and Oceanography, 33, 1225–7. pp. 3–12. New York: Springer-Verlag.
Shiba, T., Simidu, U. and Taga, N. (1979). Distribution Sinker, C. A. (1962). The North Shropsire meres and
of aerobic bacteria which contain mosses: a background for ecologists. Field Studies,
bacterichlorophyll a. Applied and Environmental 1(4), 101–38.
Microbiology, 38, 43–8. Sirenko, L. A. (1972). Physiological Basis of Multiplication
Shilo, M. (1970). Lysis of blue-green algae by of Blue-Green Algae in Reservoirs. Kiev: Naukova
Myxobacter. Journal of Bacteriology, 104, 453–61. Dumka. (In Russian.)
REFERENCES 497
Sirenko, L. A., Stetsenko, N. M., Arendarchuk, V. V. and recent models to harmful dinoflagellate blooms.
Kuz’menko, M. I. (1968). Role of oxygen conditions Journal of Plankton Research, 23, 447–62.
in the vital activity of certain blue-green algae. Smetacek, V. (2002). The ocean’s veil. Nature, 419, 565.
Microbiology, 37, 199–202. (English translation of Smetacek, V. and Passow, U. (1990). Spring-bloom
original paper in Microbiologiya.) initiation and Sverdrup’s critical-depth model.
Sirenko, L. A., Chernousova, V. M., Arendarchuk, V. V. Limnology and Oceanography, 35, 228–34.
and Kozitskaya, V. N. (1969). Factors of mass Smetacek, V., de Baar, H. J., Bathmann, U. V., Lochte,
development of blue-green algae. Hydrobiological K. and Rutgers van der Loeff, M. M. (1997).
Journal, 5(3), 1–8. (English translation of original Ecology and biochemistry of the Antarctic
paper in Gidrobiologeskii Zhurnal.) Circumpolar Current during austral spring: a
Skellam, J. G. (1951). Random dispersal in theoretical summary of Southern Ocean JGOFS cruise ANT
populations. Biometrika, 78, 196–218. X/6 of R. V. Polarstern. Deep-Sea Research Part II,
Skulberg, O. (1964). Algal problems related to Topical Studies in Oceanography, 44, 1–22.
eutrophication of Eurpoean water supplies and a Smetacek, V., Montresa, M. and Verity, P. (2002).
bioassay method to assess fertilsing influences of Marine productivity: footprints of the past and
pollution in inland waters. In Algae and Man, ed. steps into the future. In Phytoplankton Productivity,
D. F. Jackson, pp. 269–99. New York: Plenum Press. ed. P. J. leB. Williams, D. N. Thomas and C. S.
Smayda, T. J. (1970). The suspension and sinking of Reynolds, pp. 350–69. Oxford: Blackwell Science.
phytoplankton in the sea. Annual Review of Smith, E. L. (1936). Photosynthesis in relation to light
Oceanography and Marine Biology, 8, 353–414. and carbon dioxide. Proceedings of the National
(1973). The growth of Skeletonema costatum during a Academy of Sciences of the USA, 22, 504–11.
winter–spring bloom in Narragansett Bay, Rhode Smith, I. R. (1975). Turbulence in Lakes and Rivers,
Island. Norwegian Journal of Botany, 20, 219–47. Scientific Publication No. 29. Ambleside:
(1974). Some experiments on the sinking Freshwater Biological Association.
characteristics of two freshwater diatoms. (1982). A simple theory of algal deposition.
Limnology and Oceanography, 19, 628–35. Freshwater Biology, 12, 445–9.
(1980). Phytoplankton species succession. In The Smith, V. H. (1983). Low nitrogen to phosphorus ratios
Physiological Ecology of Phytoplankton, ed. I. Morris, favor dominance by blue-green algae in lake
pp. 493–570. Oxford: Blackwell Scientific phytoplankton. Science, 221, 669–71.
Publications. Smith, V. H. and Bennett, S. J. (1999). Nitrogen :
(2000). Ecological features of harmful algal blooms phosphorus supply ratios and phytoplankton
in the sea. Limnology and Oceamography, 42, community structure in lakes. Archiv für
1137–53. Hydrobiologie, 146, 37–53.
(2002). Turbulence, watermass stratification and Smol, J. P. (1992). Palaeolimnolgy: an important tool
harmful algal blooms: an alternative view and for effective ecosystem management. Journal of
frontal zones as ‘pelagic seed banks’. Harmful Aquatic Ecosystem Health, 1, 49–58.
Algae, 1, 95–112. Smol, J. P., Birks, H. J. B. and Last, W. M. (2001).
Smayda, T. J. and Boleyn, B. J. (1965). Experimental Tracking Environmental Change Using Lake Sediments,
observations on the flotation of marine diatoms. vol. 3, Terrestrial, Algal and Siliceous Indicators.
I. Thalassiosira cf. nana, Thalassiosira rotula and Dordrecht: Kluwer.
Nitzschia seriata. Limnology and Oceanography, 10, Smyly, W. J. P. (1976). Some effects of enclosure on the
499–509. zooplankton in a small lake. Freshwater Biology, 6,
(1966a) Experimental observations on the flotation 241–51.
of marine diatoms. II. Skeltonema costatum and Sommer, U. (1981). The role of r- and K-selection in
Rhizoslenia setigera. Limnology and Oceanography, 11, the succession of phytoplankton in Lake
18–34. Constance. Acta Oecologia, 2, 327–42.
(1966b). Experimental observations on the flotation (1984). The paradox of the plankton: fluctuations of
of marine diatoms. III. Bacteriastrum hyalinum and phosphorus availability maintain the diversity of
Chatoceros lauderi. Limnology and Oceanography, 11, phytoplankton in flow-through cultures. Limnology
35–43. and Oceanography, 29, 633–6.
Smayda, T. J. and Reynolds, C. S. (2001). Community (1986). The periodicity of phytoplankton in Lake
assembly in marine phytoplankton: application of Constance (Bodensee) in comparison to other
498 REFERENCES
deep lakes of central Europe. Hydrobiologia, 138, Søndergaard, M. and Moss, B. (1998). Impact of
1–7. submerged macrophytes on phytoplankton in
(1988a). Growth and survival strategies of shallow lake. In The Structuring Role of Submerged
planktonic diatoms. In Growth and Reproductive Macrophytes in Lakes, ed. E. Jeppesen, M.
Strategies of Freshwater Phytoplankton, ed. C. D. Søndergaard and K. Christoffersen, pp. 115–32.
Sandgren, pp. 227–60. Cambridge: Cambridge New York: Springer-Verlag.
University Press. Søndergaard, M., Kristensen, P. and Jeppesen, E.
(1988b). Some size relationships of phytoflagellate (1993). Eight years of internal phosphorus
motility. Hydrobiologia, 161, 125–31. loading and changes in the sediment phosphorus
(1988c). Does nutrient competition among profile of Lake Søbygaard. Hydrobiologia, 253,
phytoplankton occur in situ? Verhandlungen der 345–56.
internationalen Vereinigung für theoretische und Søndergaard, M., Borch, N. H. and Riemann, B. (2000).
angewandte Limnologie, 23, 707–12. Dynamics of biodegradable DOC produced by
(1989). The role of competition for resources in freshwater plankton communities. Aquatic
phytoplankton succession. In Plankton Ecology: Microbial Ecology, 23, 73–83.
Succession in Plankton Communities, ed. U. Sommer, Sorokin, Yu. I. (1999). Aquatic Microbial Ecology: A
pp. 57–106. Madison, WI: Brock-Springer. Textbook for Students in Environmental Sciences.
(1993). Disturbance–diversity relationships in two Leiden: Backhuys.
lakes of similar nutrient chemistry but contrasted (2002). Dynamics of inorganic phosphorus in
disturbance regimes. Hydrobiologia, 249, 59–65. pelagic communities of Sea of Okhotsk. Journal of
(1994). Planktologie. Berlin: Springer-Verlag. Plankton Research, 24, 1253–63.
(1995). An experimental test of the intermediate Sorokin, Yu. I. and Sorokin, P. Yu. (2002).
disturbance hypothesis using cultures of marine Microplankton and primary production in the
phytoplankton. Limnology and Oceanography, 40, Sea of Okhotsk in summer 1994. Journal of
1271–7. Plankton Research, 24, 453–70.
(1996). Plankton ecology: the past two decades of Soto, D., Campos, H., Steffen, W., Parra, O. and Zuñiga,
progress. Naturwissenschaften, 83, 293–301. L. (1994). The Torres del Paine lake district
Sommer, U. and Gliwicz, Z. M. (1986). Long-range (Chilean Patagonia): a case of potentially
vertical migration of Volvox in tropical lake, N-limited lakes and ponds. Archiv für Hydrobiologie
Cabora Bassa (Mozambique). Limnology and (Suppl.) 99, 181–97.
Oceanography, 31, 650–3. Sournia, A. (1978). Phytoplankton Manual. Paris:
Sommer, U. and Stibor, H. (2002). Copepoda– UNESCO.
Cladocera–Tunicata: the role of the three major Sournia, A., Chrétiennot-Dinet, M.-J. and Ricard, M.
mesozooplankton groups in pelagic food webs. (1991). Marine plankton: how many species in the
Ecological Research, 17, 161–74. world oceans? Journal of Plankton Research, 13,
Sommer, U., Gliwicz, Z. M., Lampert, W. and Duncan, 1093–9.
A. (1986). The PEG-model of seasonal succession Sousa, W. P. (1984). The role of disturbance in natural
of planktonic events in lakes. Archiv für communities. Annual Reviews of Ecology and
Hydrobiologie, 106, 433–71. Systematics, 15, 353–91.
Sommer, U., Padisák, J., Reynolds, C. S. and Southwood, T. R. E. (1977). Habitat, the templet for
Juhász-Nagy, P. (1993). Hutchinson’s heritage: the ecological strategies. Journal of Animal Ecology, 46,
diversity–disturbance relationship in 337–65.
phytoplankton. Hydrobiologia, 249, 1–7. (1996). Natural communities: structure and
Sommer, U., Sommer, F., Santer, B. et al. (2001). dynamics. Philosophical Transactions of the Royal
Complementary impacts of copepods and Society of London B, 351, 1113–29.
cladocerans on phytoplankton. Ecology Letters, 4, Sparrow, F. K. (1960). Aquatic Phycomycetes, 2nd edn.
545–50. Ann Arbor, MI: University of Michigan Press.
Søndergaard, M. (2002). A biography of Einar Steeman Spencer, C. N. and King, D. L. (1985). Role of light,
Nielsen: the man and his science. In Phytoplankton carbon dioxide and nitrogen in the regulation of
Productivity, ed. P. J. leB. Williams, D. N. Thomas buoyancy, growth and bloom formation of
and C. S. Reynolds, pp. 1–15. Oxford: Blackwell Anabaena flos-aquae. Journal of Plankton Research, 11,
Science. 283–96.
REFERENCES 499
Spencer, M. and Warren, P. H. (1996). The effects of Steele, J. H. (1976). Patchiness. In The Ecology of the Seas,
habitat size and productivity on food web ed. D. H. Cushing and J. J. Walsh, pp. 98–115.
structure in small aquatic microcosms. Oikos, 75, Philadelphia, PA: W. B. Saunders.
419–30. Steemann Nielsen, E. (1952). The use of radioactive
Spencer, R. (1955). A marine bacteriophage. Nature, carbon (C14 ) for measuring organic production in
175, 690. the sea. Journal du Conseil pour l’exploration de la
Spigel, R. H. and Imberger, J. (1987). Mixing processes Mer, 18(2), 117–40.
relevant to phytoplankton dynamics in lakes. New (1955). The interaction of photosynthesis and
Zealand Journal of Marine and Freshwater Research, respiration and its importance for the
21, 367–77. determination of 14-C discrimination in
Spijkerman, E. and Coesel, P. F. M. (1996a). Phosphorus photosynthesis. Physiologia Plantarum, 8, 945–53.
uptake and growth kinetics of two planktonic Steemann Nielsen, E. and Jensen, Å.E. (1957). Primary
desmid species. European Journal of Phycology, 31, oceanic production: the autotrophic production
53–60. of organic matter in the oceans. Galathea Reports,
(1996b). Competition for phosphorus among 1, 49–136.
planktonic desmid species in continuous-flow Stein, J. R. (1973) Handbook of Phycological Methods:
culture. Journal of Phycology, 32, 939–48. Cultural Methods and Growth Measurements.
Spiller, S. C., Castelfranco, A. M. and Castelfranco, Cambridge: Cambridge University Press.
P. A. (1982). Effects of iron and oxygen on Stephenson, G. L., Hamilton, P., Kaushik, N. K.,
chlorophyll biosynthesis. I. In vivo observations on Robinson, J. B. and Solomon, K. R. (1984). Spatial
iron- and oxygen-deficient plants. Plant Physiology, distribution of phytoplankton in enclosures of
69, 107–11. three sizes. Canadian Journal of Fisheries and Aquatic
Stadlemann, P., Moore, J. E. and Pickett, E. (1974). Sciences, 41, 1048–54.
Primary production in relation to temperature (1989). The role of grazers in phytoplankton
structure, biomass concentration and light succession. In. Plankton Ecology, ed. U. Sommer,
conditions at an inshore and offshore station in pp. 107–70. Madison, WI: Brock-Springer.
Lake Ontario. Journal of the Fisheries Research Board (1993). Daphnia growth on varying quality of
of Canada, 31, 1215–32. Scenedesmus: mineral limitation of zooplankton.
Starkwether, P. L., Gilbert, J. J. and Frost, T. M. (1979). Ecology, 74, 2351–60.
Bacteria feeding by Brachionus calyciflorus: Sterner, R. W. and Elser, J. J. (2002). Ecological
clearance and ingestion rates, behaviour and Stoichiometry: The Biology of Elements from Molecules
population dynamics. Oecologia, 44, 26–30. to the Biosphere. Princeton, NJ: Princeton
Steeg, P. F. T., Hanson, P. J. and Paerl, H. W. (1986). University Press.
Growth-limiting quantities and accumulation of Sterner, R. W., Elser, J. J., Fee, E. J., Guildford, S. J. and
molybdenum in Anabaena oscillaroides Chrzanowski, T. H. (1997). The light : nutrient
(Cyanobacteria). Hydrobiologia, 140, 143–7. ratio in lakes: the balance of energy and
Steel, J. A. (1972). The application of fundamental materials affects ecosystem structure and process.
limnological research in water supply and American Naturalist, 150, 663–84.
management. Symposia of the Zoological Society of Stewart, K. M. (1976). Oxygen deficits, clarity and
London, 29, 41–67. eutrophication in some Madison lakes.
(1973). Reservoir algal productivity. In The Use of Internationale Revue des gesamten Hydrobiologie, 61,
Mathematical Models in Water Pollution Control, ed. 563–79.
A. James, pp. 107–35. Newcastle upon Tyne: Stewart, W. D. P. and Alexander, G. (1971). Phosphorus
University of Newcastle upon Tyne. availability and nitrogenase activity in aquatic
(1975). The management of Thames Valley blue-green algae. Freshwater Biology, 1, 389–404.
Reservoirs. In The Effects of Storage on Water Quality, Stewart, W. D. P. and Lex, M. (1970). Nitrogenase
ed. R. E. Youngman, pp. 371–419. Medmenham: activity in the blue-green alga Plectonema
Water Research Centre. boryanum Strain 594. Archiv für Mikrobiologie, 73,
Steel, J. A. and Duncan, A. (1999). Modelling the 250–60.
ecological aspects of bankside reservoirs and Stewart, W. D. P., Fitzgerald, G. P. and Burris, R. H.
implications for management. Hydrobiologia, (1967). In-situ studies on N2 fixation using the
395/396, 133–47. acetylene-reduction technique. Proceedings of the
500 REFERENCES
National Academy of Sciences of the USA, 58, non-volatile dissolved organic carbon in seawater
2071–8. by direct injection of liquid sample. Marine
Stibor, H. (1992). Predator-induced life-history shifts in Chemistry, 24, 105–31.
a freshwater cladoceran. Oecologia, 92, 162–5. Sullivan, C. W. and Volcani, B. E. (1981). Silicon in the
Stock, J. B., Ninfa, A. J. and Stock, A. J. (1989). Protein cellular metabolism of diatoms. In Silicon and
phosphorylation and regulation of adaptive Siliceous Structures in Biological Systems, ed. T. L.
responses in bacteria. Microbiological Reviews, 53, Simpson and B. E. Volcani, pp. 15–42. New York:
450–90. Springer-Verlag.
Stocker, R. and Imberger, J. (2003). Horizontal Sültemeyer, D. (1998). Carbonic anhydrase in
transport and dispersion in the surface layer of a eukaryotic algae: characterisation, regulation and
medium-sized lake. Limnology and Oceanography, possible function during photosynthesis.
48, 971–82. Canadian Journal of Botany, 76, 962–72.
Stockner, J. G. and Antia, N. J. (1986). Algal Sültemeyer, D. F., Fock, H. P. and Canvin, D. T. (1991).
picoplankton from marine and freshwater Active uptake of inorganic carbon by
ecosystems: a multidisciplinary perspective. Chlamydomonas reinhardtii: evidence of
Canadian Journal of Fisheries and Aquatic Sciences, 43, simultaneous transport of HCO− 3 and CO2 .
2472–503. Canadian Journal of Botany, 69, 995–1002.
Stoermer, E. F. (1968). Near-shore phytoplankton Sutcliffe, D. W., Carrick, T. R., Heron, J. et al. (1982).
populations in the Grand Haven, Michigan, Long-term and seasonal changes in the chemical
vicinity during thermal bar conditions. In composition of precipitation and surface waters
Proceedings of the 11th Conference on Great Lakes of lakes and tarns in the English Lake District.
Research, pp. 137–50. Ann Arbor, MI: International Freshwater Biology, 12, 451–506.
Association for Great Lakes Research. Sverdrup, H. U. (1953). On conditions of vernal
Stokes, G. G. (1851). On the effect of internal friction blooming of phytoplankton. Journal du Conseil
of fluids on the motion of pendulums. Transactions international pour l’exploration de la Mer, 18,
of the Cambridge Philosophical Society, 9 (II), 8–14. 287–95.
Stoyneva, M. P. (1994). Shallows of the lower Danube Sverdrup, H. U., Johnson, M. W. and Fleming, R. H.
as additional sources of potamoplankton. (1942). The Oceans: Their Physics, Chemistry and
Hydrobiologia, 289, 97–108. General Biology. New York: Prentice Hall.
Straile, D. (2000). Meteorological forcing of plankton Swale, E. M. F. (1963). Notes on Stephanodiscus hantzschii
dynamics in a large and deep continental lake. Grun. in culture. Archiv für Mikrobiologie, 45,
Oecologia, 122, 44–50. 210–16
Straškraba, M. T., Jørgensen, S.-E. and Patten, B. C. Swale, E. M. F. (1968). The phytoplankton of Oak Mere,
(1999). Ecosystems emerging II. Dissipation. Cheshire, 1963–1966. British Phycological Bulletin, 3,
Ecological Modelling, 117, 3–39. 441–9.
Strathmann, R. R. (1967). Estimating the organic Swift, D. G. (1980). Vitamins and phytoplankton
carbon content of phytoplankton from cell growth. In The Physiological Ecology of Phytoplankton,
volume or plasma volume. Limnology and ed. I. Morris, pp. 329–68. Oxford: Blackwell
Oceanography, 12, 411–18. Scentific Publications.
Strickler, J. R. (1977). Observations on swimming Szeligiwicz, W. (1998). Phytoplankton blooms
performances of planktonic copepods. Limnology predictions: a new turn for Sverdrup’s critical
and Oceanography, 22, 165–70. depth concept. Polskie Archiwum Hydrobiologii, 45,
Strickler, J. R. and Twombley, S. (1975). Reynolds’ 501–11.
numbers, diapause and predatory copepods. Takamura, N., Otsuki, A., Aizaki, M. and Nojiri, Y.
Verhandlungen der internationalen Vereinigung für (1992). Phytoplankton species shift accompanied
theoretische und angewandte Limnologie, 19, 2943–50. by transition from nitrogen dependence to
Stumm, W. and Morgan, J. P. (1981). Aquatic Chemistry, phosphorus dependence of primary production
2nd edn. New York: John Wiley. in Lake Kasumigaura, Japan. Archiv für
(1996). Aquatic Chemistry, 3rd edn. New York: John Hydrobiologie, 124, 129–48.
Wiley. Talling, J. F. (1951). The element of chance in pond
Sugimura, Y. and Suzuki, Y. (1988). A populations. The Naturalist, Hull,
high-temperature catalytic-oxidation method of October–December 1951, 157–70.
REFERENCES 501
(1957a). Diurnal changes of stratification and ‘Schwingineer’ structure. Journal of Ecology, 61,
photosynthesis in some tropical African waters. 537–67
Proceedings of the Royal Society of London B, 147, Tamiya, H., Iwamura, T., Shibata, K., Hase, E. and
57–83. Nihei, T. (1953). Correlation between
(1957b) Photosynthetic characteristics of some photosynthesis and light-independent
freshwater diatoms in relation to underwater metabolism in the growth of Chlorella. Biochimica
radiation. New Phytologist, 56, 29–50. Biophysica Acta, 12, 23–40.
(1957c). The phytoplankton population as a Tandeau de Marsac, N. (1977). Occurrence and nature
compound photosynthetic system. New Phytologist, of chromatic adaptation in Cyanobacteria. Journal
56, 133–49. of Bacteriology, 130, 82–91.
(1960). Self-shading effects in natural populations of Tang, E. P. Y. and Peters, R. H. (1995). The allometry of
a planktonic diatom. Wetter und Leben, 12, algal respiration. Journal of Plankton Research, 17,
235–42. 303–15.
(1965). The photosynthetic activity of phytoplankton Tansley, A. G. (1939). The British Isles and Their Vegetation.
in East African lakes. Internationale Revue des Cambridge: Cambridge University Press.
gesamten Hydrobiologie, 50, 1–32. Temporetti, P. F. and Pedrozo, F. L. (2000). Phosphorus
(1966). Photosynthetic behaviour in stratified and release rates from freshwater sediments affected
unstratified lake populations of a planktonic by fish farming. Aquaculture Research, 31, 447–55.
diatom. Journal of Ecology, 54, 99–127. Tennekes, H. and Lumley, J. L. (1972). A First Course in
(1971). The underwater light climate as a Turbulence. Cambridge, MA: Massachussetts
controlling factor in the production ecology of Institute of Technology Press.
freshwater phytoplankton. Mitteilungen der Tett, P. and Barton, E. D. (1995). Why are there about
internationalen Vereinigung für theoretische und 5000 species of phytoplankton in the sea? Journal
angewandte Limnologie, 19, 214–43. of Plankton Research, 17, 1693–704.
(1973). The application of some electro-chemical Tett, P., Heaney, S. I. and Droop, M. R. (1985). The
methods to the measurement of photosynthesis Redfield ratio and phytoplankon growth rate.
and respiration in freshwaters. Freshwater Biology, Journal of the Marine Biological Association, 65,
3, 335–62. 487–504.
(1976). The depletion of carbon dioxide from lake Thellier, M. (1970). An electrokinetic interpretation of
water by phytoplankton. Journal of Ecology, 64, the functioning of biological systems and its
79–121. application to the study of mineral salt
(1984). Past and contemporary trends and attitudes absorption. Annals of Botany, 34, 983–1009.
in work on primary productivity. Journal of Therriault, J.-C. and Platt, T. (1981). Environmental
Plankton Research, 6, 203–17. control of phytoplankton patchiness. Canadian
(1986). The seasonality of phytoplankton in African Journal of Fisheries and Aquatic Sciences, 38, 638–41.
lakes. Hydrobiologia, 138, 139–60. Thienemann, A. (1918). Untersuchungen über die
(1987). The phytoplankton of Lake Victoria (East Beziehungen zwischen dem Sauerstoffgehalt des
Africa). Ergebnisse der Limnologie, 25, 229–56. Wassers und der Zusammensetzung der Fauna in
Talling, J. F. and Heaney, S. I. (1988). Long-term norddeutschen Seen. Archiv für Hydrobiologie, 12,
changes in some English (Cumbrian) lakes 1–65.
subjected to increased nutrient inputs. In Algae Thingstad, F., Heldal, M., Bratbak, G. and Dundas, I.
and their Aquatic Environments, ed. F. E. Round, (1993). Are viruses important partners in pelagic
pp. 1–29. Bristol: Biopress. food webs? Trends in Ecology and Evolution, 8,
Talling, J. F. and Talling, I. B. (1965). The chemical 209–13.
composition of African lake waters. Internationale Thomas, E. A. (1949). Sprungschichtneigung in
Revue des gesamten Hydrobiologie, 50, 421–63. Zürichsee durch Strum. Schweizerische Zeitschrift für
Talling, J. F., Wood, R. B., Prosser, M. V. and Baxter, Hydrobiolgie, 11, 527–45.
R. M. (1973). The upper limit of photosynthetic (1950). Auffällige biologische Folgen von
productivity by phytoplankton: evidence from Sprungschichtneigung in Zürichsee. Schweizerische
Ethiopian soda lakes. Freshwater Biology, 3, 53–76. Zeitschrift für Hydrobiolgie, 12, 1–24.
Tallis, J. H. (1973). The terrestrialization of lake basins Thomas, J. D. (1997). The role of dissolved organic
in North Cheshire, with special reference to a matter, particularly free amino acids and humic
502 REFERENCES
substances, in freshwater ecosystems. Freshwater Titman, D. and Kilham, P. (1976). Sinking in

Biology, 38, 1–36. freshwater phytoplankton: some ecological
Thomas, R. H. and Walsby, A. E. (1985). Buoyancy implications of cell nutrient status and physical
regulation in a strain of Microcystis. Journal of mixing processes. Limnology and Oceanography, 21,
General Microbiology, 131, 799–809. 409–17.
Thomasson, K. (1963). Araucanian Lakes. Acta Tokeshi, M. (1997). Species co-existence and
Phytogeographica Suecica, 47, 1–139. abundance: patterns and processes. In Biodiversity:
Thompson, J. M., Ferguson, A. J. D and Reynolds, C. S. An Ecological Perspective, ed. T. Abe, S. R. Levin and
(1982). Natural filtration rates of zooplankton in a M. Higashi, pp. 35–55. New York: Springer-
closed system the derivation of a community Verlag.
grazing index. Journal of Plankton Research, 4, Tolland, H. G. (1977). Destratification/Aeration in
545–60. Reservoirs A Literature Review of the Techniques Used
Thorp, J. H. and Casper, A. F. (2002). Potential effects for Water Quality Management, Technical Report No.
on zooplankton from species shifts in 50. Medmenham: Water Research Centre.
planktivorous mussels: a field experiment in the Toms, I. P. (1987). Developments in London’s water
St Lawrence River. Freshwater Biology, 47, 107–19. supply system. Ergebnisse der Limnologie, 28,
Thouvenot, A., Richardot, M., Debroas, D. and Dévaux, 149–67.
J. (2003). Regulation of the abundance and Torella, F. and Morita, R. Y. (1979). Evidence for a high
growth of ciliates in a newly flooded reservoir. incidence of bacteriophage particles in the waters
Archiv für Hydrobiologie, 157, 185–93. of Yaquina Bay, Oregon: ecological and taxonomic
Tillmann, U. (2003). Kill and eat your predator: a implications. Applied and Environmental
winning strategy of the planktonic flagellate Microbiology, 37, 774–8.
Prymnesium parvum. Aquatic Microbial Ecology, 32, Tortell, P. D. (2000). Evolutionary and ecological
73–84. perspectives in carbon acquisition in
Tilman, D. (1977). Resource competition between phytoplankton. Limnology and Oceanography, 45,
planktonic algae: an experimental and 744–50.
theoretical approach. Ecology, 58, 338–48. Tow, M. (1979). The occurrence of Microcystis aeruginosa
(1987). On the meaning of competition and in the bottom sediments of a shallow eutrophic
community structure. Functional Ecology, 1, pan. Journal of the Limnological Society of Southern
304–15. Africa, 5, 9–10.
(1988). Plant Strategies and the Dynamics and Structure Townsend, C. R., Winfield, I. J., Peirson, G. and Cryer,
of Plant Communities. Princeton, NJ: Princeton M. (1986). The response of young roach Rutilus
University Press. rutilus to seasonal changes in the abundance of
Tilman, D. and Kilham, S. S. (1976). Phosphate and microcrustacean prey: a field determination of
silicate growth and uptake kinetics of the switching. Oikos, 46, 372–8.
diatoms Asterionella formosa and Cyclotella Tranvik, L. J. (1998). Degradation of dissolved organic
meneghiniana in batch and semi-continuous matter in humic waters by bacteria. In Aquatic
culture. Journal of Phycology, 12, 375–83. Humic Substances, ed. D. Hessen and L. J. Tranvik,
Tilman, D., Kilham, S. S. and Kilham, P. (1982). pp. 259–83. Berlin: Springer-Verlag.
Phytoplankton community ecology: the role of Tranvik, L. J. and Bertilsson, S. (2001). Contrasting
limiting nutrients. Annual Reviews of Ecology and effects of solar UV radiation on dissolved organic
Systematics, 13, 349–72. sources for bacterial growth. Ecology Letters, 4,
Tilzer, M. M. (1984). Estimation of phytoplankton loss 458–63.
rates from daily photosynthetic rates and Trevisan, R. (1978). Noto sull’uso dei volumi algali per
observed biomass changes in Lake Constance. la stima della biomassa. Rivista di Idrobiologia, 17,
Journal of Plankton Research, 6, 309–24. 345–58.
Timms, R. M. and Moss, B. (1984). Prevention of Trimbee, A. M. and Harris, G. P. (1983). Use of
growth of potentially dense phytoplankton time-series analysis to demonstrate advection
populations by zooplankton grazing in the rates of different variables in a small lake. Journal
presence of zooplanktivorous fish in a shallow of Plankton Research, 5, 819–33.
wetland ecosystem. Limnology and Oceanography, Tsujimura, S. (2003). Application of the frequency of
29, 472–86. dividing cells technique to estimate the in-situ
REFERENCES 503
growth rate of Microcystis (Cyanobacteria). Vadeboncoeur, Y., Vander Zanden, M. J. and Lodge,
Freshwater Biology, 48, 2009–24. D. M. (2002). Putting the lake back together:
Turner, M. G., Baker, W. L., Peterson, C. J. and Peet, reintegrating benthic pathways into lake food
R. K. (1998). Factors influencing succession: web models. BioScience, 52, 44–54.
lessons from large, infrequent natural Vadstein, O., Olson, H., Reinertsen, H. and Jensen, A.
disturbances. Ecosystems, 1, 511–23. (1993) The role of planktonic bacteria in lakes:
Turpin, D. H. (1988). Physiological mechanisms in sink and link. Limnology and Oceanography, 38,
phytoplankton resource competition. In Growth 539–44.
and Reproductive Strategies of Freshwater Vander Zanden, M.J, Shuter, B. J., Lesster, N. and
Phytoplankton, ed. C. D. Sandgren, pp. 316–68. Rasmussen, J. B. (1999). Patterns of food chain
Cambridge: Cambridge University Press. length in lakes: a stable isotope study. American
Tüxen, R. (1955). Das Systeme der nordwestdeutschen Naturalist, 154, 406–16.
Pflanzengesellschaft. Mitteilungen der Vanderploeg, H. A. and Paffenhöfer, H. A. (1985).
Florist-soziologische Arbeitsgemeinschaft, 5, 1–119. Modes of algal capture by the freshwater copepod
Tyler, J. E. and Smith, R. E. (1970). Measurements of Diaptomus sicilis and their relation to food-size
Spectral Irradiance Underwater. New York: Gordon selection. Limnology and Oceanography, 30, 871–85.
and Breach. Vanni, M. J., Luecke, C., Kitchell, J. F. and Magnusson,
Tyler, P. A. (1996). Endemism in freshwater algae, with J. (1990). Effects of planktivorous fish mass
special reference to the Australian region. mortality on the plankton community of Lake
Hydrobiologia, 336, 127–35. Mendota, Wisconsin: implications for
Uhlmann, D. (1971). Influence of dilution, sinking and biomanipulation. Hydrobiologia, 200/201, 329–36.
grazing on phytoplankton populations of Vaulot, D. (1995). The cell cycle of phytoplankton:
hyperfertilised ponds and micro-ecosystems. coupling cell growth to population growth. In The
Mitteilungen der internationalen Vereinigung für Molecular Ecology of Aquatic Microbes, ed. I. Joint,
theoretische und angewandte Limnologie, 19, 100–124. pp. 303–32. Berlin: Springer-Verlag.
Ulanowicz, R. E. (1986). Growth and Development: Vega-Palas, M. A., Flores, E. and Herrero, A. (1992).
Ecosystems Phenomenology. New York: Springer- NtcA, a global nitrogen regulator from the
Verlag. cyanobacterium Synechococcus that belongs to the
Upstill-Goddard, R. C., Watson, A. J., Liss, P. S. and CRP family of bacterial regulators. Molecular
Liddicoat, M. I. (1990). Gas transfer velocities in Microbiology, 6, 1853–9.
lakes measured with SF6 . Tellus, 42B, 364–77. Velikova, V., Moncheva, S. and Petrova, D. (1999).
Urbach, E., Scanlan, D. J., Distel, D. L., Waterbury, J. B. Phytoplankton dynamics and red tides
and Chisholm, S. W. (1998). Rapid diversification (1987–1997) in the Bulgarian Black Sea.
of marine picophytoplankton with dissimilar Oceanology, 37(3), 376–84. (English translation of
light-harvesting structures inferred from original article in Okeanologiya.)
sequences of Prochlorococcus and Synechococcus Venrick, E. L. (1990). Phytoplankton in an oligotrophic
(Cyanobacteria). Journal of Molecular Evolution, 46, ocean: species structure and interannual
188–201. variability. Ecology, 71, 1547–63.
Utermöhl, H. (1931). Neue Wege in der quantitativen (1995). Scales of variability in a stable environment:
Erfassung des Planktons. Verhandlungen der phytoplankton in the central North Pacific. In
internationalen Vereinigung für theoretische und Ecological Time Series, ed. T. M. Powell and J. H.
angewandte Limnologie, 5, 567–96. Steele, pp. 150–80. New York: Chapman and Hall.
Utkilen, H.-C., Oliver, R. L. and Walsby, A. E. (1985a). Ventelä, A.-M., Wiacowski, K., Moilanen, M. et al.
Buoyancy regulation in a red Oscillatoria unable to (2002). The effect of small zooplankton on the
collapse gas vesicles by turgor pressure. Archiv für microbial loop and edible algae during a
Hydrobiologie, 102, 319–29. cyanobacterial bloom. Freshwater Biology, 47,
Utkilen, H.-C., Skulberg, P. M. and Walsby, A. E. 1807–19.
(1985b). Buoyancy regulation and chromatic Verity, P. G., Robertson, C. Y., Tronzo, C. R. et al. (1992).
adaptation in planktonic Oscillatoria species: Relationships between cell volume and carbon
alternative strategies for optimising light and nitrogen of marine photosynthetic
absorption in stratified lakes. Archiv für nanoplankton. Limnology and Oceanography, 37,
Hydrobiologie, 104, 407–17. 1434–46.
504 REFERENCES
Vicente, E. and Miracle, M. R. (1988). Physicochemical to recurrence patterns and claimed relationship
and microbial stratification in a meromictic to sun activity cycles. Science of the Total
karstic lake in Spain. Verhandlungen der Environment, 165, 213–24.
internationalen Vereinigung für theoretische und Votintsev, K. K. (1992). On the characteristics of the
angewandte Limnologie, 23, 522–9. self-purification potential of Lake Baikal. Sibirkskiy
Vincent, W. F., Neale, P. J. and Richerson, P. J. (1984). biologicheskiy Zhurnal, 4, 36–40.
Photoinhibition: algal responses to bright light Vrind-de Jong, E. W. de, Emburg, P. R. van and Vrind,
during diel stratification and mixing in a J. P. M. de (1994). Mechanisms of calcification:
tropical alpine lake. Journal of Phycology, 20, Emiliana huxleyi as a model system. In The
201–11. Haptophyte Algae, ed. J. C. Green and B. S. C.
Viner, A. and Kemp, L. (1983). The effect of vertical Leadbeater, pp. 149–166. Oxford: Oxford
mixing on the phytoplankton of Lake Rotongaio University Press.
(July 1979 – January 1981). New Zealand Journal of Walker, B. H. (1992). Biodiversity and ecological
Marine and Freshwater Science, 17, 402–22. redundancy. Biological Conservation, 6, 18–23.
Volcani, B. E. (1981). Cell wall formation in diatoms: Walker, D. (1992). Excited leaves. New Phytologist, 121,
morphogenesis and biochemistry. In Silicon and 325–45.
Siliceous Structures in Biological Systems, ed. T. L. Wall, D. and Dale, B. (1968). Modern dinoflagellate
Simpson and B. E. Volcani, pp. 157–200. New York: cysts and evolution of the Peridiniales.
Springer-Verlag. Micropalaeontology, 14, 265–304.
Vollenweider, R. A. (1965). Calculation models of Wallis, S. G., Young, P. C. and Beven, K. (1989).
photosynthesis-depth curves and some Experimental investigation of an aggregated dead
implications regarding day rate estimates in zone model for longitudinal transport in stream
primary production. Memorie dell’Istituto italiano di channels. Proceedings of the Institute of Civil
Idrobiologia, 18 (Suppl.), 425–57. Engineers, 87, 1–22.
(1968). Scientific Fundamentals of the Eutrophication of Walsby, A. E. (1971). The pressure relationships of gas
Lakes and Flowing Waters, with Particular Reference to vacuoles. Proceedings of the Royal Society of London B,
Nitrogen and Phosphorus as Factors in Eutrophication, 178, 301–26.
Technical Report DAS/CSI/68.27. Paris: (1978). The properties and buoyancy-providing role
Organisation for Economic Co-Operation and of gas vauoles in Trichodesmium Ehrenberg. British
Dvelopment. Phycological Journal, 13, 103–16.
(1974). A Manual on Methods for Measuring Primary (1994). Gas vesicles. Microbiological Reviews, 58,
Production in Aquatic Environments. Oxford: 94–144.
Blackwell Scientific Publications. (2001). Determining the photosynthetic productivity
(1975). Input–output models with special reference of a stratified phytoplankton population. Aquatic
to the phosphorus-loading concept in limnology. Sciences, 63, 18–43.
Schweizerische Zeitschrift für Hydrologie, 37, 53–84. Walsby, A. E. and Bleything, A. (1988). The dimensions
(1976). Advances in defining critical load levels for of cyanobacterial gas vesicles in relation to their
phosphorus in lake eutrophication. Memorie efficiency in providing buoyancy. Journal of General
dell’Istituto italiano di Idrobiologia, 33, 53–83. Microbiology, 135, 2635–45.
Vollenweider, R. A. and Kerekes, J. (1980). The loading Walsby, A. E. and Klemer, A. R. (1974). The role of gas
concept as basis for controlling eutrophication vacuoles in the microstratification of a
philosophy and preliminary results of the OECD population of Oscillatoria agardhii var. isothrix in
programme on eutrophication. Progress in Water Deming Lake, Minnesota. Archiv für Hydrobiologie,
Technology, 12(2), 5–38. 74, 375–92.
Vollenweider, R. A., Munawar, M. and Stadelman, P. Walsby, A. E. and Reynolds, C. S. (1980). Sinking and
(1974). A comparative review of phytoplankton floating. In The Physiological Ecology of
and primary production in the Laurentian Great Phytoplankton, ed. I. Morris, pp. 371–412. Oxford:
Lakes. Journal of the Fisheries Research Board of Blackwell Scientific Publications.
Canada, 31, 739–62. Walsby, A. E. and Xypolyta, A. (1977). The form
Vollenweider, R. A., Montanari, G. and Rinaldi, A. resistance of chitan fibres attached to the cells of
(1995). Statistical inferences about the mucilage Thalassiosira fluviatilis Hustedt. British Phycological
events in the Adriatic Sea, with special reference Journal, 12, 215–23.
REFERENCES 505
Walsby, A. E., Utkilen, H. C. and Johnsen, I. J. (1983). Werner, D. (1977). Silicate metabolism. In The Biology of
Buoyancy changes in red-coloured Oscillatoria Diatoms, ed. D. Werner, pp. 110–49. Oxford:
agardhii in Lake Gjersjoen, Norway. Archiv für Blackwell Scientific Publications.
Hydrobiologie, 97, 18–38. Wershaw, R. L. (2000). The study of humic substances:
Walsby, A. E., Kinsman, R., Ibelings, B. W. and in search of a paradigm. In Humic Substances:
Reynolds, C. S. (1991). Highly buoyant colonies of Versatile Components of Plants, Soil and Water, ed.
the cyanobacterium Anabaena lemmermannii form E. A. Ghabbour and G. Davies, pp. 1–7. Cambridge:
persistent water blooms. Archiv für Hydrobiologie, Royal Society of Chemistry.
121, 261–80. Wesenberg-Lund, C. (1904). Studier over de dansk
Wardle, D. A., Bonner, K. I., Barker, G. M. et al. (1999). sersplankton, vol. 1, Tekst. Copenhagen:
Plant removals in perennial grassland: vegetation Glydendanske Boghandel.
dynamics, decomposers, soil biodiversity and West, R. G. (1977). Pleistocene Geology and Biology, 2nd
ecosystem properties. Ecological Monographs, 69, edn. London: Longman.
535–68. West, W. and West, G. S. (1909). The phytoplankton of
Wasmund, N. (1994). Phytoplankton periodicity in a the English Lake District. The Naturalist (in six
eutrophic coastal water of the Baltic Sea. parts), 626, 1115–22; 627, 134–41; 628, 186–93;
Internationale Revue des gesamten Hydrobiologie, 79, 629, 260–7; 631, 287–92; 632, 323–31.
259–85. Wetzel, R. G. (1995). Death, detritus and energy flow
Watson, A. J., Upstill-Goddard, R. C. and Liss, P. S. in aquatic ecosystems. Freshwater Biology, 35, 83–9.
(1991). Air–sea exchange in rough and stormy Whipple, G. C. (1899). The Microscopy of Drinking Water.
seas measured by a dual-tracer technique. Nature, New York: John Wiley.
349, 145–7. Whitton, B. A. (1975) Algae. In River Ecology, ed. B. A.
Watson, S. W. and Kalff, J. (1981). Relationships Whitton, pp. 81–105. Oxford: Blackwell Scientific
between nanoplankton and lake trophic status. Publications.
Canadian Journal of Fisheries and Aquatic Sciences, 38, Whitton, B. A. and Peat, A. (1969). On Oscillatoria redekei
960–7. Van Goor. Archiv für Mikrobiologie, 68, 362–76.
Weiher, E. and Keddy, P. A. (1995). Assembly rules, Wiackowski, K., Brett, M. T. and Goldman, C. R. (1994).
null models and trait dispersion: new questions Differential effects of zooplankton species on
from old patterns. Oikos, 74, 159–64. ciliate community structure. Limnology and
Weiher, E., Clarke, D. P. and Keddy, P. A. (1998). Oceanography, 39, 486–92.
Community assembly rules, morphological Wildman, R., Loescher, J. H. and Winger, C. L. (1975).
dispersion and the co-existence of plant species. Development and germination of akinetes of
Oikos, 81, 309–22. Aphanizomenon flos-aquae. Journal of Phycology, 11,
Weisse, T. (1994). Structure of microbial food webs in 96–104.
relation to the trophic state of lakes and fish Wilkinson, D. M. (1999). The disturbing history of
grazing pressure: a key role of Cyanobacteria. In intermediate disturbance. Oikos, 84, 145–7.
Ecology and Human Impact on Lakes and Reservoirs in Willén, E. (1976). A simplified method of
Minas Gerais with Special Reference to Future phytoplankton counting. British Phycological
Development and Management Strategies, ed. R. M. Journal, 11, 265–78.
Pinto-Coelho, A. Giani and E. von Sperling, Williams, D. B. (1975). On the concentrations of
pp. 55–70. Belo Horizonte: Sociedade Editoria e characterised dissolved organic compounds in
Grafı́ca de Acão Communitaria. seawater. In The Micropalaeontology of Oceans, ed.
(2003). Pelagic microbes: protozoa and the B. M. Funnell and W. R. Riedel, pp. 91–5.
microbial food web. In The Lakes Handbook, vol.1, Cambridge: Cambridge University Press.
ed. P. E. O’Sullivan and C. S. Reynolds, pp. 417–60. Williams, P. J. leB. (1970). Heterotrophic utilisation of
Oxford: Blackwell Science. dissolved organic compounds in the sea. I. Size
Weisse, T. and MacIsaac, E. (2000). Significance and distribution of population and relationship
fate of bacterial production in oligotrophic lakes between respiration and incorporation of growth
in British Columbia. Canadian Journal of Fisheries substances. Journal of the Marine Biological
and Aquatic Science, 57, 96–105. Association of the United Kingdom, 50, 859–70.
Welch, E. B. (1952) Limnology, 2nd edn. New York: (1981). Incorporation of microheterotrophic
McGraw-Hill. processes into the classical paradigm of the
506 REFERENCES
planktonic food web. Kieler meereforschungen Wollast, R. (1974). The silica problem. In The Sea, vol. 5,
Sonderheft, 1, 1–28. ed. E. D. Goldberg, pp. 359–92. New York: John
(1995). Evidence for the seasonal accumulation of Wiley.
carbon-rich dissolved organic material: its scale Woods, J. D. and Onken, R. (1983). Diurnal variation
in comparison with changes in particulate and primary production in the ocean:
material and the consequential effect on net preliminary results of a Lagrangian ensemble
C : N assimilation ratios. Marine Chemistry, 51, model. Journal of Plankton Research, 4, 735–56.
17–29. Wüest, A. and Lorke, A. (2003). Small-scale
Williams, P. J. leB. and Lefevre, D. (1996). Algal 14 C hydrodynamics in lakes. Annual Reviews in Fluid
uptake and total carbon metabolism. I. Models to Mechanics, 35, 273–412.
account for the physiological processes of Yates, M. G. (1977). Physiological aspects of nitrogen
respiration and recycling. Journal of Plankton fixation. In Recent Developments in Nitrogen Fixation,
Research, 18, 1941–59. ed. W. Newton, J. R. Postgate and C. Rodriguez-
Wilsey, B. J. and Potvin, C. (2000). Biodiversity and Barrueco, pp. 219–70. London: Academic Press.
ecosystem functioning: importance of species Yoshida, T., Gurung, T. B., Kagami, M. and Urabe, J.
eveness in an old field. Ecology, 81, 887–92. (2001). Contrasting effects of a cladoceran
Wilson, E. O. and Petr, F. M. (1986). BioDiversity. (Daphnia galeata) and a calanoid copepod
Washington, DC: National Academy Press. (Eodiaptomus japonicus) on algal and microbial
Winkler, L. W. (1888). Die Bestimmung des im Wasser plankton in a Japanese Lake, Lake Biwa. Oecologia,
gelösten Sauerstoffs. Berichte der deutsches chemische 129, 602–10.
Gesellschaft, 21, 2843–54. Youngman, R. E. (1971). Algal Monitoring of Water Supply
Wiseman, S. W. and Reynolds, C. S. (1981). Sinking Reservoirs and Rivers, Technical Memorandum No.
rate and electrophoretic mobility of the TM63. Medmenham: Water Research Association.
freshwater diatom Asterionella formosa: an Yunes, J. S., Niencheski, L. F., Salomon, P. S. and
experimental investigation. British Phycological Parise, M. (1998). Toxicity of Cyanobacteria in the
Journal, 16, 357–61. Patos Lagoon Estuary, Brazil. Verhandlungen der
Wiseman, S. W., Jaworski, G. H. M. and Reynolds, C. S. internationalen Vereinigung für theoretische und
(1983). Variability in the sinking rate of the angewandte Limnologie, 26, 1796–2000.
freshwater diatom, Asterionella formosa Hass: the Zacharias, O. (1898) Das Potamoplankton. Zoologische
influence of the excess density of the colonies. Anzeiger, 21, 41–8.
British Phycological Journal, 18, 425–32. Zehr, J. P. (1995). Nitrogen fixation in the sea: why
Witte, F., Goldschidt, T., Wanink, J. et al. (1992). The only Trichodesmium? In Molecular Ecology of Aquatic
destruction of the endemic species flock: Microbes, ed I. Joint, pp. 335–64. Berlin:
quantitative data on the decline of the Springer-Verlag.
haplochromine cichlids of Lake Victoria. Zhuravleva, V. I. and Matsekevich, Ye.S. (1974).
Environmental Biology of Fishes, 34, 1–28. Electrokinetic properties of Microcystis aeruginosa.
Woese, C. R. (1987). Bacterial evolution. Microbiology Hydrobiological Journal, 10(1), 55–7. (English
Reviews, 51, 221–71. translation of original paper in Gidrobiologeskii
Woese, C. R., Kandler, O. and Wheelis, M. L. (1990). Zhurnal.)
Towards a natural system of organisms: proposal Ziegler, S. and Benner, R. (2000). Effects of solar
for the domains Archaea, Bacteria and Eucarya. radiation on dissolved organic matter cycling in a
Proceedings of the National Academy of Sciences of the tropical seagrass meadow. Limnology and
USA, 87, 4576–9. Oceanography, 45, 257–66.
Wolf-Gladrow, D. A. and Riebesell, U. (1997). Diffusion Zimmermann, U. (1969) Ökologische und
and reactions in the vicinity of plankton: a physiologische Untersuchungen an der
refined model for inorganic carbon transport. planktonischen Blaualge Oscillatoria rubescens D.C.
Marine Chemistry, 59, 17–34. unter besonderen Berücksichtigung von Licht
Wolk, C. P. (1982). Heterocysts. In The Biology of the und Temperatur. Schweizerische Zeitschrift für
Cyanobacteria, ed. N. G. Carr and B. A. Whitton, Hydrologie, 31, 1–58.
pp. 359–86. Oxford: Blackwell Scientific Zohary, T. and Pais-Madeira, A. M. (1990). Structural,
Publications. physical and chemical characteristics of
REFERENCES 507
Microcystis aeruginosa hyperscums from a elongatus at 3.8 resolution. Nature, 409,

hypertrophic lake. Freshwater Biology, 23, 339–52. 739–43.
Zohary, T. and Robarts, R. D. (1989). Diurnal mixed Zuniño, L. and Diaz, M. (2000). Autotrophic
layers and the long-term dominance of Microcystis picoplankton along a trophic gradient in
aeruginosa. Journal of Plankton Research, 11, 25–48. Andean–Patagonian lakes. Verhandlungen der
Zouni, A., Witt, H.-T., Kern, J., et al. (2001). Crystal internationalen Vereinigung für theoretische und
structure of photosystem II from Synechococcus angewandte Limnologie, 27, 1895–99.
Index to lakes, rivers and seas
LAKES AND Carrilaufquen Chica, Argentina Garda, Lago di, Italy 45◦ 35 N,
RESERVOIRS 41◦ 13 S, 69◦ 26 W 335 10◦ 25 E 330, 331 Table 7.4
Carrilaufquen Grande, Argentina George, Lake, Uganda 0◦ 0 N,
41◦ 7 S, 69◦ 28 W 335 30◦ 10 E 59, 105, 344
Akan-Panke, Japan 43◦ 27 N,
Castle Lake, California, USA Grasmere, UK 54◦ 28 N, 3◦ 1 W 241,
144◦ 06 E 326
Ammersee, Germany 48◦ N, 11◦ 6 E 41◦ 14 N, 122◦ 23 W 167 332 Table 7.5, 333, 352
Caviahue, Lago, Argentina 37◦ 54 S, Great Bear Lake, Canada 66◦ N,
329, 330 Table 7.3
70◦ 35 W 425 120◦ W 323
Aralskoye More,
Como, Lago di, Italy 45◦ 58 N, Great Slave Lake, Canada 62◦ N,
Kazakhstan/Uzbekistan 45◦ N,
60◦ E 318, 353 9◦ 15 E 330, 331 Table 7.4 114◦ W 324
Coniston Water, UK 54◦ 20 N, Hamilton Harbour, Lake Ontario,
Arresø, Denmark 55◦ 59 N, 12◦ 7 E
3◦ 4 W 331, 332 Table 7.5 Canada 43◦ 15 N, 79◦ 51 E 344
344
Attersee, Austria 47◦ 50 N, 13◦ 35 E Correntoso, Lago, Argentina Hartbeespoort Dam, South Africa
329, 330 Table 7.3 40◦ 24 S, 71◦ 35 W 336 25◦ 24 S, 27◦ 30 E 215
Balaton, Hungary 46◦ 50 N, 18◦ 42 E Crater Lake, Oregon, USA 42◦ 56 N, Hawes Water, UK 54◦ 31 N, 2◦ 48 W
89, 345, 354 122◦ 8 W 111, 117 Table 3.2 331, 332 Table 7.5
Balkhash, Ozero, Kazakhstan 46◦ N, Crose Mere, UK 52◦ 53 N, 2◦ 50 WE Huron, Lake, Canada/USA 44◦ 30 N,
76◦ E 343, 344 109, 117 Table 3.2, 245, 339, 82◦ W 323
368, 373 Iseo, Lago d’, Italy 45◦ 43 N, 10◦ 4 E
Bangweolo, Zambia 11◦ S, 30◦ E 343
Crummock Water, UK 54◦ 31 N, 330, 331 Table 7.4
Bassenthwaite Lake, UK 54◦ 39 N,
3◦ 18 W 332 Table 7.5 Issyk-kul, Ozero, Kyrgyzstan 43◦ N,
3◦ 12 W 332 Table 7.5
Dead Sea, Israel/Jordan 31◦ 25 N, 78◦ E 89, 325
Bautzen Reservoir, Germany
35◦ 29 W 172 Kaspiyskoye More, Azerbaijan/
51◦ 11 N, 14◦ 29 E 403
Derwent Water, UK 54◦ 34 N, 3◦ 8 W Iran/Kazakhstan/Russia/
Baykal, Ozero, Russia 53◦ N, 108◦ E
331, 332 Table 7.5 Turkmenistan 42◦ N,
89, 322, 324, 390
Drontermeer, Netherlands 52◦ 16 N, 51◦ E 322
Bayley-Willis, Lago, Argentina
5◦ 32 E 345 Kasumigaura-Ko, Japan 36◦ 0 N,
40◦ 39 S, 71◦ 43 W 196
Eglwys Nynydd, UK 51◦ 33 N, 140◦ 26 E 235, 345
Biwa-Ko, Japan 35◦ N, 136◦ W 142,
3◦ 44 W 82, 87 Kilotes, Lake, Ethiopia 7◦ 5 N,
219, 327
Blelham Tarn, UK 54◦ 24 N, 2◦ 58 W Ennerdale Water, UK 54◦ 32 N, 38◦ 25 E 113, 117 Table 3.2
3◦ 23 W 332 Table 7.5 Kinneret, Yam, Israel 32◦ 50 N,
82, 120, 245, 284, 332 Table 7.5,
333 Erie, Lake, Canada/USA 41◦ 45 N, 35◦ 30 E 233, 327, 366
Bodensee, Austria/Germany/ 81◦ W 290, 323, 324 Kivu, Lac, D. R. Congo/Rwanda 2◦ S,
Switzerland 47◦ 40 N, 9◦ 20 E Erken, Sjön, Sweden 59◦ 25 N, 29◦ E 339, 341
48 Table 2.2, 71, 329, 330 Table 18◦ 15 E 338 Königsee, Germany 47◦ 40 N,
7.3 Espejo, Lago, Argentina 40◦ 36 S, 12◦ 55 E 329, 330 Table 7.3
Brothers Water, UK 54◦ 21 N, 71◦ 48 W 336 Lacar, Lago, Argentina 40◦ 10 S,
Esrum Sø, Denmark 56◦ 00 N, 71◦ 30 W 336
2◦ 56 W 332 Table 7.5
12◦ 22 E 338 Ladozhskoye, Ozero, Russia 61◦ N,
Budworth Mere, UK 53◦ 17 N,
Esthwaite Water, UK 54◦ 21 N, 32◦ E 89, 324
2◦ 31 W 344
3◦ 0 W 88, 108, 126, 295, 332, Lanao, Philippines 7◦ 52 N,
Buenos Aires/General Carrera, Lago,
332 Table 7.5, 370, 371, 390 124◦ 13 E 233, 342, 354
Argentina/Chile 46◦ 35 S, 72◦ W
Eyre, Lake, Australia 29◦ S, 137◦ E Léman, Lac, France/Switzerland
322
343 46◦ 22 N, 6◦ 33 E 329, 330 Table
Buttermere, UK 54◦ 33 N, 3◦ 16 W
Fairmont Lakes Reservoir, 7.3
332 Table 7.5
Minnesota, USA 43◦ 16 N, Llanquihue, Lago, Chile 41◦ 6 S,
Cabora Bassa Dam, Mozambique
93◦ 47 W 416 72◦ 48 W 336
15◦ 38 S, 33◦ 11 E 205
Fonck, Lago, Argentina 41◦ 20 S, Loughrigg Tarn, UK 54◦ 26 N,
Carioca, Lagoa, Brasil 19◦ 10 S,
71◦ 44 W 196 3◦ 1 W 332 Table 7.5
42◦ 1 W 115
508
INDEX TO LAKES, RIVERS AND SEAS 509
Lowes Water, UK 54◦ 34 N, 3◦ 21 W Ontario , Lake, Canada/USA Trummen, Sjön, Sweden 56◦ 52 N,
332, 332 Table 7.5 43◦ 40 N, 78◦ W 324 14◦ 50 E 413
Lugano, Lago di, Italy 45◦ 57 N, P. K. le Roux Reservoir, South Africa Turkana (formerly Lake Rudolf),
8◦ 58 E 330, 331 Table 7.4 30◦ 0 S, 24◦ 44 E 113, 117 Kenya/Ethiopia 4◦ N, 36◦ E
Maggiore, Lago, Italy/Switzerland Table 3.2 339
46◦ N, 8◦ 40 E 330, 330 Table Ranco, Lago, Chile 40◦ 12 S, Ullswater, UK 54◦ 35 N, 2◦ 53 W
7.3, 331 Table 7.4 72◦ 25 W 336 332, 332 Table 7.5
Malham Tarn, UK 54◦ 6 N, 2◦ 4 W Rapel, Embalse de Chile 34◦ 2 S, Valencia, Lago, Venezuela 10◦ 11 N,
129 71◦ 35 S 233 67◦ 40 W 342
Mälaren, Sweden 59◦ 18 N, 17◦ 6 E Red Rock Tarn, Australia 38◦ 20 S, Veluwemeer, Netherlands 52◦ 55 N,
198 Fig. 5.6 143◦ 30 W 105 5◦ 45 E 345, 410
Malawi, Lake (formerly Lake Nyasa), Rostherne Mere, UK 53◦ 21 N, Victoria, Lake,
Malawi/Mozambique/Tanzania 2◦ 23 W 119, 339, 366 Kenya/Tanzania/Uganda 1◦ S,
12◦ S, 35◦ E 339, 340 Rotongaio, Lake, New Zealand 33◦ E 339, 341
Matano, Danau, Indonesia 2◦ 35 S, 38◦ 40 S, 176◦ 2 E 121 Vierwaldstättersee, Switzerland
121◦ 23 E 322 Rydal Water, UK 54◦ 27 N, 3◦ 0 W 47◦ N, 8◦ 20 E 59, 329, 330
Memphrémagog, Lac, Canada/USA 332 Table 7.5 Table 7.3
45◦ 5 N, 72◦ 16 W 89 Sagami-Ko, Japan 35◦ 24 N, 139◦ 6 E Villarrica, Lago, Chile 39◦ 15 S,
Mendota, Lake, USA 43◦ 21 N, 198 Fig. 5.6 72◦ 35 W 336
89◦ 25 W 289 Schlachtensee, Germany 52◦ 28 N, Volta Grande Reservoir, Brasil
Michigan, Lake, USA 43◦ 45 N, 13◦ 25 E 410 20◦ 4 S, 48◦ 13 W 415
86◦ 30 W 290, 323, 324, 390 Schöhsee, Germany 54◦ 10 N, Vostok, Lake, Antarctica approx.
Mikolajske, Jezioro, Poland 53◦ 10 N, 10◦ 25 E 338 85◦ S, 50◦ W 322
21◦ 33 E 298, 339 Śniardwy, Poland 53◦ 45 N, 21◦ 45 E Wahnbach Talsperre, Germany
Millstätter See, Austria 46◦ 48 N, 344 50◦ 50 N, 7◦ 8 E 409
13◦ 33 E 325, 350 Søbygaard, Denmark 56◦ 3 N, Walensee, Switzerland 47◦ 7 N,
Montezuma’s Well, Arizona, USA 9◦ 40 E 413 9◦ 16 E 330 Table 7.3
34◦ 37 N, 111◦ 51 W 233, 241 St James’s Park Lake, UK 51◦ 30 N, Washington, Lake, Washington, USA
Murtensee, Switzerland, 46◦ 55 N, 0◦ 9 E 345 47◦ 35 N, 122◦ 14 W 408
7◦ 5 E 115 Stechlinsee, Germany 53◦ 10 N, Wast Water, UK 54◦ 26 N, 3◦ 17 W
Mount Bold Reservoir, Australia, 13◦ 3 E 326, 337 332 Table 7.5
35◦ 0 S, 138◦ 48 E 113, 117 Superior, Lake, Canada/USA Wellington Reservoir, Australia
Table 3.2 47◦ 30 N, 89◦ W 117 Table 3.2, 33◦ 20 S, 116◦ 2 E Fig 29
Nauel Huapi, Lago, Argentina 323 Windermere, UK 54◦ 20 N, 2◦ 53 W
40◦ 50 S, 71◦ 50 W 336 Tahoe, Lake, California/Nevada, USA 80, 115, 117 Table 3.2, 119, 221,
Neagh, Lough, UK 54◦ 55 N, 6◦ 30 W 39◦ N, 120◦ W 198 Fig. 5.6 225, 245, 247, 284, 289, 292,
48 Table 2.2, 72, 345 Tai Hu, China 31◦ N, 120◦ E 344 332, 332 Table 7.5, 333, 354,
Negra, Laguna, Chile 37◦ 49 S, Tanganyika, Lac, Burundi/ D. R. 368, 401, 402
70◦ 2 W 113 Congo/Tanzania/Zambia 6◦ S, Winnipeg, Lake, Canada 53◦ S,
Ness, Loch, UK 57◦ 16 N, 4◦ 30 W 30◦ E 263, 339, 340, 392 98◦ W 343
126, 390 Tchad, Lac, Chad/Niger/Nigeria Wolderwijd, Netherlands 52◦ 29 N,
Neusiedlersee, Austria/Hungary 12◦ 30 N, 14◦ 30 E 105, 343, 5◦ 51 E 345
47◦ 47 N, 16◦ 44 E 377 Fig. 7.22 344 Wörthersee, Austria 46◦ 37 N,
Nyos, Lake, Cameroon 6◦ 40 N, Thames Valley Reservoirs, UK 14◦ 10 E 325
10◦ 13 E 126 51◦ 30 N, 0◦ 40 W 139, 405, 415, Zürichsee, Switzerland 47◦ 12 N,
N’zigi (formerly Lake Albert), D. R. 416 8◦ 42 E 59, 235, 330 Table 7.3,
Congo/Uganda 1◦ 50 N, 31◦ E Thirlmere, UK 54◦ 32 N, 3◦ 4 W 332 366
339 Table 7.5
Oak Mere, UK 53◦ 13 N, 2◦ 39 W Titicaca, Lago, Bolivia/Peru 15◦ 40 S,
339, 345, 425 69◦ 35 W 342
RIVERS AND
Okeechobee, Lake, Florida, USA Todos Los Santos, Lago, Chile ESTUARIES
27◦ N, 81◦ W 413 41◦ 5 S, 72◦ 15 W 336
Onezhskoye, Ozero, Russia 62◦ N, Traful, Lago, Argentina 40◦ 36 S, Angara, Russia 55◦ 50 N, 110◦ 0 E
36◦ E 89, 324, 325 71◦ 25 W 336 322
510 INDEX TO LAKES, RIVERS AND SEAS
Ashes Hollow, UK 52◦ 30 N, 2◦ 50 W Gulf of Bothnia 62◦ N, 20◦ E 112, Trondheimsfjord 64◦ N, 9◦ E 311
48 Table 2.2 312 Ullsfjord 71◦ N, 27◦ E 311
Danube, Romania/Russia 43◦ 15 N, Gulf of Finland 60◦ N, 28◦ E 312 Mediterranean Sea
29◦ 35 E 312 Gulf of Maine 43◦ N, 68◦ W 310, Black Sea 43◦ N, 35◦ E 259, 312
Dnepr, Ukraine 46◦ 30 N, 32◦ 0 E 407 Golfo di Napoli 40◦ 45 N, 14◦ 15 E
313 Gulf of Mexico 25◦ N, 90◦ W 313
Dnestr, Moldova/Ukraine 46◦ 3 N, Gulf Stream (Bahamas) 25◦ N, Tyrrhenian Sea 40◦ N, 12◦ E
30◦ 23 E 312 75◦ W 313
Don, Ukraine 47◦ 11 N, 39◦ 10 E 313 English Channel 50◦ N, 2◦ W 234, Indian Ocean 111, 134, 162
Guadiana, at Mourão, Portugal 310, 311 Andaman Sea 12◦ N, 96◦ E 309
38◦ 24 N, 7◦ 22 W 242 Irish Sea – Clyde Estuary Arabian Sea 15◦ N, 65◦ E 309
Mississippi, USA 29◦ 0 N, 89◦ 20 W 55◦ 50 N, 5◦ 00 W 112 Arafura Sea 10◦ S, 135◦ E 310
126 Irish Sea – Severn Estuary Red Sea 20◦ N, 38◦ E 40
Selenga, Russia 52◦ 15 N, 106◦ 40 E 51◦ 20 N, 4◦ 00 W 48 Table 2.2, Pacific Ocean
322 117 Table 3.2 ALOHA 22◦ 45 N, 158◦ W
Severn, UK 51◦ 20 N, 3◦ W 112, 242 Labrador Sea 60◦ N, 55◦ W 306 305
Thames, at Reading, UK 51◦ 27 N, Naragansett Bay (RI) 41◦ 23 N, Baja California 25◦ N, 110◦ W
0◦ 29 W 48 Table 2.2 71◦ 24 W 310 407
White Nile, Sudan/Uganda 15◦ 30 N, North Atlantic Drift Current California Current (San Francisco)
32◦ 50 E 339 45◦ N, 45◦ W 289, 306 35◦ N, 115◦ W 309
North Atlantic Subtropical Gyre East China Sea 30◦ N, 125◦ E
(Mauritania/Senegal) 15◦ N, 310
OCEANS AND SEAS 20◦ W 309 Kuroshio Current 40◦ N, 150◦ E
North Atlantic Tropical Gyre 5◦ N, 306, 307
Arctic Ocean 15◦ W 306 North Pacific Subtropical Gyre
Polar Basin 80◦ N, 180◦ W 306 North Sea 56◦ N, 4◦ E 1, 310, 407, 25◦ N, 160◦ W 133, 162, 304,
Atlantic Ocean 111, 134 427 354, 382
Baltic Sea 55◦ N, 20◦ E 134, 310, Norwegian Sea 70◦ N, 0◦ E 306, Peru Current (Galapagos Islands)
312, 426 307 5◦ S, 90◦ W 134, 309
Bay of Fundy 45◦ N, 66◦ W 42 Øresund 56◦ 00 N, 12◦ 50 E Polar Front ∼40◦ S 308
Benguela Current (Gabon) 0◦ N, Oslofjord 59◦ N, 11◦ E 101, 312 Sea of Okhotsk 55◦ N, 90◦ W 134,
0◦ E 134, 309 Ria de Vigo 42◦ N, 9◦ W 311, 313 310
Canaries Current 30◦ N, 15◦ W 309 Sargasso Sea 30◦ N, 60◦ W 111, 117 South Pacific Gyre 25◦ S, 140◦ W
Table 3.2, 162 306
Grand Banks of Newfoundland St Lawrence Estuary 49◦ N, 64◦ W Southern Ocean 60◦ S, 30◦ E 168,
45◦ N, 52◦ W 310, 427 134, 407 306, 307
Index to genera and species of phytoplankton
Genera mentioned in the text are listed together with their systematic position (Phylum ORDER) and authorities for
each species are cited. Synonyms are included where appropriate. Entries are suffixed ‘M’ or ‘F’ to indicate marine or
freshwater occurrence, or ‘M/F’ for genera appearing in both. The bold entry for freshwater species (A, B, etc.) refers
to the trait selected functional grouping of Reynolds et al. (2002) as summarised in Table 7.1.
Achnanthes (Bacillariophyta Anabaena solitaria Kleb. F, H2 233, Aphanizomenon flos-aquae Ralfs ex

BACILLARIALES) M/F 7 Table 1.1 332, 335, 336 Born. et Flah. F, H1 24, 26 Table
Achnanthes taeniata Grun. M 306, Anabaena spiroides Kleb. F, H1 338 1.2, 184 Table 5.0, 186, 219 Table
312 Anabaenopsis (Cyanobacteria 5.4, 312, 333, 338, 339, 426
Actinocyclus (Bacillariophyta NOSTOCALES) F 6 Table 1.1, 26 Aphanizomenon gracile Lemm. F,
BIDDULPHIALES) M/F 309, 312, Table 1.2, 81, 205, 341, 344, 401 probably H2 233
317, 353 Anacystis (Cyanobacteria Aphanocapsa (Cyanobacteria
Alexandrium (Dinophyta CHROOCOCCALES) F CHROOCOCCALES) F
GONYAULACALES) M 233, 312, Genus defunct, part of Several species, all colonial; F, K;
313 Synechococcus free cells are picoplanktic
Alexandrium tamarense (Lebour) Anacystis nidulans, see also Synechoccus, Z
Balech M 201, 311, 407 Synechococcus 183 Aphanothece (Cyanobacteria
Amphidinium (Dinophyta Ankistrodesmus (Chlorophyta CHROOCOCCALES) F, K 6 Table
GYMNODINIALES) M 7 Table 1.1 CHLOROCOCCALES) F 6 Table 1.1, 26 Table 1.2, 320 Table 7.1,
Amphisolenia (Dinophyta 1.1, 344, 425 344
DINOPHYSIALES) M 203 Table Once large genus, many separated Arthrodesmus (Chlorophyta
5.2, 234 into other genera ZYGNEMATALES) F 6 Table 1.1
Anabaena (Cyanobacteria Ankistrodesmus braunii 184 Table Arthrospira (Cyanobacteria
NOSTOCALES) mostly F 6 Table 5.0 OSCILLATORIALES) mostly F 6
1.1, 26 Table 1.2, 58, 59, 65, 67, Ankistrodesmus falcatus, see also Table 1.1, 26 Table 1.2, 320
81, 121, 129, 130, 165, 172, 205, Monoraphidium contortum 338 Table 7.1
206, 213, 216, 226, 228, 229, Ankyra (Chlorophyta Asterionella (Bacillariophyta
230, 232, 241, 249, 271, 293, CHLOROCOCCALES) F 6 Table BACILLARIALES), M/F 7 Table
297, 328, 329, 330, 333, 340, 1.1, 82, 212, 218, 226, 228, 229, 1.1, 23, 24, 33 Table 1.6, 54, 63,
341, 342, 344, 366, 401, 404, 405 230, 231, 232, 247, 274, 284, 64, 65, 66, 67, 80, 158, 172, 191,
Anabaena circinalis Rabenh. ex 285, 300, 320 Table 7.1, 333, 192, 195, 196, 197, 201, 202,
Born. et Flah. F, H1 24, 26 344, 359, 370 206, 207, 208, 213, 218, 220,
Table 1.2, 57 Table 2.4, 339 Ankyra judayi (G. M. Smith) Fott F, 221, 222, 224, 225, 226, 228,
Anabaena cylindrica Lemmermann X1 20 Table 1.1, 26 Table 1.2, 231, 233, 242, 245, 246, 248,
F, SN ? 187 219 Table 5.4, 277 Table 6.3, 284 249, 269, 270, 274, 280, 290,
Anabaena flos-aquae (Lyngb.) Bréb. Table 6.4, 285 Table 6.5, 338 292, 293, 294, 298, 299, 300,
ex Born. et Flah. F, H1 54 Table Anthosphaera (Haptophyta 325, 329, 333, 338, 339, 344,
2.3, 157 Table 4.2, 165, 184 COCCOLITHOPHORIDALES) M 365, 378
Table 5.0, 194, 195, 219 Table 312 Asterionella formosa Hassall. F, B
5.4, 248, 320 Table 7.1, 333, 338 Apedinella (Chrysophyta and C 20 Table 1.1, 26 Table 1.2,
Anabaena lemmermannii Richter F, PEDINELLALES) M 7 Table 1.1 29 Table 1.3, 31, 31 Table 1.5,
H2 58, 233, 312, 320 Table 7.1, Apedinella spinifera (Throndsen) 32, 52, 54 Table 2.3, 61 Table
322, 332, 426 Throndsen M 35 Table 1.7 2.5, 63, 66, 107, 113, 119, 129,
Anabaena minutissima Pridmore F, Aphanizomenon (Cyanobacteria 157 Table 4.2, 184 Table 5.1,
SN 320 Table 7.1 NOSTOCALES) mostly F 6 Table 195, 197, 202, 219 Table 5.4,
Anabaena ovalisporum Forti F, H2 1.1, 26 Table 1.2, 59, 67, 81, 129, 220, 221 Table 5.5, 221 Table
327 165, 205, 216, 230, 233, 320 5.6, 232, 245, 246, 247, 277
Anabaena planctonica Brunnth. F, Table 7.1, 325, 330, 341, 401, Table 6.3, 320 Table 7.1, 324,
H2? 338 404, 405 326, 328, 330, 331, 338
511
512 INDEX TO GENERA AND SPECIES OF PHYTOPLANKTON
Asterionella (Bacillariophyta) (cont.) Calciopappus (Haptophyta Chaetoceros diadema (Ehrenb.) Gran

Asterionella japonica Cleve M 234, COCCOLITHOPHORIDALES) M M 306, 310
311 312 Chaetoceros neglectus Karsten M 308
Aulacoseira (Bacillariophyta Carteria (Chlorophyta VOLVOCALES) Chaetoceros radicans Schütt M 311
BIDDULPHIALES) F 7 Table 1.1, F 234, 311 Chaetoceros socialis Lauder M 311,
23, 28, 54, 62, 63, 130, 177, 213, Cerataulina (Bacillariophyta 312, 313
214, 243, 245, 249, 250, 325, BIDDULPHIALES) M 7 Table 1.1, Chilomonas (Cryptophyta
329, 336, 341, 342, 344, 415 234, 312, 317 CRYPTOMONADALES) F 6 Table
Aulacoseira alpigena Cerataulina bergonii H. Perag. M 1.1
(Grun.) Krammer F, A (also 312 Chlamydocapsa (Chlorophyta
recorded as Melosira distans) 203 Cerataulina pelagica (Cleve) Hendey TETRASPORALES) F 293
Table 5.2 M 311, 312 Chlamydocapsa planctonica (W. and
Aulacoseira ambigua (Grun.) Ceratium (Dinophyta .G. S.West) Fott F, F (formerly
Simonsen 232, 320 Table 7.1, GONYAULACALES, M/F 7 Table known as Gloeocystis planctonica)
326 1.1, 13, 24, 60, 68, 88, 206, 213, 57 Table 2.4
Aulacoseira baicalensis (K. I. Mey.) 216, 218, 219, 228, 230, 231, Chlamydomonas (Chlorophyta
Simonsen F, A 322 232, 233, 234, 249, 293, 295, VOLVOCALES) M/F 6 Table 1.1,
Aulacoseira binderana Kütz. F, B 26 311, 320 Table 7.1, 329, 330, 35 Table 1.7, 49, 78, 212,
Table 1.2 331, 332, 333, 338, 339, 348, 232, 233, 312, 326, 333, 343, 352, 425
Aulacoseira granulata (Ehrenb.) 366, 370, 371, 381, 385 Chlamydomonas reinhardtii P. A.
Simonsen F, P 6 Table 1.1, 29 Ceratium arcticum (Ehrenb.) Cleve Dangeard F, X1 127, 157 Table
Table 1.3, 123, 219 Table 5.4, M 306, 307 4.2
232, 320 Table 7.1, 324, 327, Ceratium azoricum Cleve M 307 Chlorella (Chlorophyta
328, 330, 336, 339, 340, 341, Ceratium carriense Gourret M 307 CHLOROCOCCALES) F 32, 33
373 Ceratium furca (Ehrenb.) M 307, Table 1.6, 113, 127, 137, 148,
Aulacoseira islandica (O. Müller) 310, 311, 316 150, 182, 184 Table 5.0, 188,
Simonsen F, C 225, 233, 320 Ceratium fusus (Ehrenb.) M 234, 196, 207, 208, 209, 212, 230,
Table 7.1, 322, 323, 324, 330 306, 307, 310, 311, 312, 313, 231, 260, 298, 320 Table 7.1,
Aulacoseira solida (Eulenstein) 316 324, 333, 343, 344, 352, 357,
Krammer F, B 225, 233, 320 Ceratium hexacanthum Gourret M 365, 370, 378, 425
Table 7.1, 322, 323, 324, 330 307 Chlorella minutissima Fott et
Aulacoseira subarctica (O. Müller) Ceratium hirundinella (O. F. Müller) Nováková F, X3 203 Table 5.2,
Haworth F, B 29 Table 1.3, 54 Dujardin F, LM 6 Table 1.1, 20 332
Table 2.3, 61 Table 2.5, 62, 129, Table 1.1, 83, 129, 184 Table 5.0, Chlorella pyrenoidosa Chick F X1 20
202, 225, 233, 247, 320 Table 205, 216, 219 Table 5.4, 229, Table 1.1, 26 Table 1.2, 157
7.1, 324, 325, 331, 333 248, 325, 338, 339 Table 4.2, 424
Aureococcus (Bacillariophyta Ceratium lineatum (Ehrenb.) M 307 Chlorella vulgaris Beijerinck F, X1
BIDDULPHIALES) M 407 Ceratium longipes (Bail.) Gran M (includes strains formerly
306, 307, 310, 312 known as Chlorella pyrenoidosa)
Bacillaria (Bacillariophyta Ceratium tripos (O. F. Müller) 54 Table 2.3, 61 Table 2.5
BIDDULPHIALES) M Nitzsch M 305, 307, 310, 311, Chlorobium (Anoxyphotobacteria) F
Bacteriastrum (Bacillariophyta 312, 313, 316 321 Table 7.1
BIDDULPHIALES) M 203 Table Chaetoceros (Bacillariophyta Chlorobotrys (Eustigmatophyta) F 6
5.2, 311 BIDDULPHIALES) mostly M 7 Table 1.1, 424
Binuclearia (Chlorophyta Table 1.1, 60, 306, 307, 310, 311, Chlorococcum (Chlorophyta
ULOTRICHALES) F 233, 259 312, 313, 314, 317, 353 CHLOROCOCCALES) F 54 Table
Bitrichia (Chrysophyta Chaetoceros atlanticum Cleve M 308 2.3, 61 Table 2.5, 352
HIBBERDIALES) F 7 Table 1.1 Chaetoceros compressus Lauder M Chlorogonium (Chlorophyta
Botryococcus (Chlorophyta 234, 310, 311, 312 VOLVOCALES) F 326
CHLOROCOCCALES) F 6 Table Chaetoceros convolutum Castracene Choricystis (Chlorophyta
1.1, 54, 320 Table 7.1, 322, 339, M 307 CHLOROCOCCALES) F 6 Table
344 Chaetoceros curvisetum Cleve M 313 1.1, 328
Botryococcus braunii Kützing F, F Chaetoceros debilis Cleve M 309, Chromatium (Anoxyphotobacteria) F
331, 425 310, 311, 312 321 Table 7.1
INDEX TO GENERA AND SPECIES OF PHYTOPLANKTON 513
Chromulina (Chrysophyta Coelastrum (Chlorophyta Cryptomonas erosa Ehrenb. F, Y 334

CHROMULINALES) F 7 Table 1.1, CHLOROCOCCALES) F 320 Table Cryptomonas marssoni Skuja F, Y
20 Table 1.1, 247, 251, 284, 300, 7.1, 341, 344 334
328, 425 Coelastrum microporum Nägeli F, J Cryptomonas ovata Ehrenb. F, Y 6
Chroococcus (Cyanobacteria Coenochloris (Chlorophyta Table 1.1, 20 Table 1.2, 184
CHROOCOCCALES) F 6 Table CHLOROCOCCALES) F 24, Table 5.0, 219 Table 5.4, 226,
1.1, 26 Table 1.2, 171, 325, 327 56, 82, 129, 203 Table 5.2, 276, 277 Table 6.3, 334, 338,
Chroomonas (Cryptophyta 210, 228, 273, 293, 320 339
CRYPTOMONADALES) M/F 6 Table 7.1, 325, 326, 340, Cryptomonas profunda Butcher M
Table 1.1, 338 359 35 Table 1.7
Chroomonas salina (Wislouch) Coenochloris fottii (Hindák) Cyanobium (Cyanobacteria
Butcher M 35 Table 1.7 Tsarenko F, F (formerly known CHROOCOCCALES), F 213, 269,
Chrysochromulina (Haptophyta as Sphaerocystis schroeteri) 57 327
PRYMNESIALES) M/F 7 Table 1.1, Table 2.4, 219 Table 5.4, 230, Picoplanktic Z, free cells of
13, 233, 234, 251, 312, 320 Table 248, 331 Cyanodictyon or Synechocystis
7.1, 323, 324, 327, 330, 331, 336, Coenococcus (Chlorophyta Cyanodictyon (Cyanobacteria
338, 343, 402 CHLOROCOCCALES) F 231 CHROOCOCCALES) F 213, 269
Chrysochromulina herdlensis Coenocystis (Chlorophyta Two species, all colonial; F, K?;
Leadbeater M 35 Table 1.7 CHLOROCOCCALES) F 203 Table free cells are picoplanktic
Chrysochromulina parva Lackey F, 5.2, 212 Cyanobium, Z
X2 20 Table 1.1, 338 Genus includes some species Cyanophora (Glaucophyta) F, X1 6
Chrysochromulina polylepis Manton previously attributed to Table 1.1, 11
et Parke M 312, 407 Coccomyxa Cyanothece (Cyanobacteria
Chrysococcus (Chrysophyta Corethron (Bacillariophyta CHROOCOCCALES), F 269
CHROMULINALES) F 7 Table 1.1, BIDDULPHIALES) M 307, 311 Relatively new genus
20 Table 1.1, 251, 320 Table 7.1, Corethron criophilum Castracene M accommodating larger species
332 306, 308 of Synechococcus, X2?
Chrysolykos (Chrysophyta Corethron hystrix Hensen M 307 Cyathomonas (Cryptophyta
CHROMULINALES) F 7 Table 1.1, Coscinodiscus (Bacillariophyta CRYPTOMONADALES) M/F 159,
203 Table 5.2 BIDDULPHIALES) mostly M 307, 425
Chrysosphaerella (Chrysophyta 309, 310, 312, 317, 341 Cyclococcolithus (Haptophyta
CHROMULINALES) F 7 Table 1.1, Coscinodiscus subbulliens Jørgensen COCCOLITHOPHORIDALES) M
129, 213 M 306 234
Clathrocystis (Anoxyphotobacteria) F Coscinodiscus wailseii Gran et Angst. Cyclococcolithus fragilis (Lohmann)
6 Table 1.1 M 50, 53 Deflandre M 234
Closterium (Chlorophyta Cosmarium (Chlorophyta Cyclostephanos (Bacillariophyta
ZYGNEMATALES) F 6 Table 1.1, ZYGNEMATALES) F 6 Table 1.1, BIDDULPHIALES) F 340, 341
23, 192, 340, 341, 344 129, 202, 203 Table 5.2, 233, Cyclotella (Bacillariophyta
Closterium aciculare T.West 279, 320 Table 7.1, 331, 336, BIDDULPHIALES) M/F 7 Table
F, P 20 Table 1.1, 26 385, 424 1.1, 23, 63, 129, 195, 197, 201,
Table 1.2, 219 Table 5.4, Cosmarium abbreviatum Raciborski 202, 208, 225, 232, 249, 313,
232, 320 Table 7.1, 328, F, N 195, 201, 247, 332 325, 326, 353
330, 339 Cosmarium contractum Kirchner F, Cyclotella bodanica Eulenstein F, A
Closterium acutum Bréb. F, P 15 Fig N 332 323, 329, 385
1.1 Cosmarium depressum (Nägeli) Cyclotella caspia Grun. M 234, 312,
Coccolithus (Haptophyta Lundell F, N 20 Table 1.1 313, 317
COCCOLITHOPHORIDALES) M 7 Crucigena (Chlorophyta Cyclotella comensis Grunow F, A 203
Table 1.1 CHLOROCOCCALES) F 231 Table 5.2, 320 Table 7.1, 325,
Coccolithus pelagicus (Wallich) Cryptomonas (Cryptophyta 331
Schiller M 35 Table 1.7 CRYPTOMONADALES) M/F 6 Cyclotella glomerata Bachm. F, A
Coccomyxa (Chlorophyta Table 1.1, 228, 229, 230, 247, 203 Table 5.2, 323, 325,
CHLOROCOCCALES) F 352 260, 269, 274, 284, 300, 312, 320 329
Some planktic species are now in Table 7.1, 323, 324, 328, 330, Cyclotella litoralis Lange et
Coenocystis 335, 338, 359 Syvertsen M 309, 317
Cyclotella (cont.) Dinobryon sertularia Ehrenb. F, E Euglena mutabilis F. Schmitz F, W1

Cyclotella meneghiniana Kütz. F, B 331 424, 425
20 Table 1.1, 61 Table 2.5, 195, Dinobryon sociale Ehrenb. F, E 157 Eunotia (Bacillariophyta
197 Table 4.2 BACILLARIALES) M/F 425
Cyclotella nana Hustedt M 53 Dinophysis (Dinophyta Eutetramorus (Chlorophyta
Cyclotella praeterissima Lund F, B 20 DINOPHYSIDALES) M 203 Table CHLOROCOCCALES) F
Table 1.1, 202, 247, 331 5.2, 309, 311, 313 Redundant genus name – see
Cyclotella radiosa (Grun.) Dinophysis acuminata Clap. et Lach. Coenococcus
Lemmermann F, A (formerly M 310 Eutreptia (Euglenophyta
known as Cyclotella comta) 203 Ditylum (Bacillariophyta EUTREPTIALES) M 6 Table 1.1,
Table 5.2, 233, 323, 326, 331 BIDDULPHIALES) M 55, 130 234, 311, 312
Cylindrocystis (Chlorophyta Ditylum brightwellii (West) Grun. M Eutreptiella (Euglenophyta
ZYGNEMATALES) F 424 55, 130 EUTREPTIALES) M 313
Cylindrospermopsis (Cyanobacteria Dunaliella (Chlorophyta Exuviella (Dinophyta
NOSTOCALES) F, SN 6 Table 1.1, VOLVOCALES) M 6 Table 1.1, 32, PROROCENTRALES) M 7 Table
26 Table 1.2, 205, 320 Table 7.1, 78, 234, 311 1.1, 307, 353
340, 341, 342, 344, 366, 401 Dunaliella tertiolecta Butcher M 35 Exuviella baltica Lohm. M 312
Cylindrospermopsis raciborskii Table 1.7 Florisphaera (Haptophyta
(Woloszynska) Seenayya et COCCOLITHOPHORIDALES) M 7
Subba Raju F, SN 216, 384 Elakatothrix (Chlorophyta Table 1.1, 309
ULOTRICHALES) F 6 Table 1.1 Fragilaria (Bacillariophyta
Dactylococcopsis (Cyanobacteria Emiliana (Haptophyta BACILLARIALES) M/F 7 Table 1.1,
CHROOCOCCALES), F 325 COCCOLITHOPHORIDALES) M 7 24, 51, 63, 226, 228, 229, 232,
Detonula (Bacillariophyta Table 1.1, 13, 203 Table 5.2, 311, 233, 245, 249, 269, 300, 330,
BIDDULPHIALES) M 7 Table 1.1 312 338
Detonula pumila (Castracene) Gran Emiliana huxleyi (Lohm.) Hay et Fragilaria capucina Desmaz. F, P 6
M 35 Table 1.7 Mohl. M 35 Table 1.7, 130, 234, Table 1.1
Diacanthos (Chlorophyta 307, 310, 313 Fragilaria crotonensis Kitton F, P 6
CHLOROCOCCALES) F, X3 352 Ethmodiscus (Bacillariophyta Table 1.1, 20 Table 1.1, 23, 29
Diacronema (Haptophyta BIDDULPHIALES) M 50, 234 Table 1.3, 31 Table 1.5, 54 Table
PAVLOVALES) M 7 Table 1.1 Ethmodiscus rex (Rattray) Wiseman 2.3, 57 Table 2.4, 61 Table 2.5,
Diatoma (Bacillariophyta et Hendey M 50, 234 62, 63, 66, 129, 184 Table 5.0,
BACILLARIALES) F 7 Table 1.1, Euastrum (Chlorophyta 219 Table 5.4, 232, 247, 320
324 ZYGNEMATALES) F 6 Table 1.1 Table 7.1, 324, 333, 336, 339
Diatoma elongatum (Lyngb.) C. Eucampia (Bacillariophyta Fragilaria oceanica Cleve M 307,
Agardh F, C 172, 338 BIDDULPHIALES) M 311, 312 311
Dictyosphaerium (Chlorophyta Eucampia cornuta (Cleve) Grun. M Fragilariopsis (Bacillariophyta
CHLOROCOCCALES) F 6 Table 311 BACILLARIALES) M 7 Table 1.1,
1.1, 210, 336 Eucampia zodiacus Ehrenb. M 307, 308
Dictyosphaerium pulchellum 312 Fragilariopsis cylindrus (Grun.)
H. C.Wood F, F 20 Table 1.1, 331 Eudorina (Chlorophyta VOLVOCALES) Krieger M 311
Dinobryon (Chrysophyta F 6 Table 1.1, 24, 182, 211, 216, Fragilariopsis doliolus Wallich M
CHROMULINALES) F 7 Table 1.1, 228, 229, 230, 232, 247, 269, 309
89, 129, 130, 210, 219 Table 5.4, 270, 293, 299, 320 Table 7.1, Fragilariopsis nana (Steemann
228, 229, 231, 232, 320 Table 338, 343 Nielsen) Paasche M 307
7.1, 323, 325, 329, 330, 333, 334, Eudorina elegans Ehrenb. F, G? 26
336, 338, 385 Table 1.2, 157 Table 4.2 Geminella (Chlorophyta
Dinobryon bavaricum Imhof F, E Eudorina unicocca G. M. Smith F, G ULOTRICHALES) F, T 6 Table 1.1,
331 26 Table 1.2, 57 Table 2.4, 184 233, 294, 320 Table 7.1
Dinobryon cylindricum Imhof F, E Table 5.0, 219 Table 5.4, 226, Gephyrocapsa (Haptophyta
203 Table 5.2, 322, 326, 331 339 COCCOLITHOPHORIDALES) M 7
Dinobryon divergens Imhof F, E 20 Euglena (Euglenophyta Table 1.1, 203 Table 5.2
Table 1.1, 184 Table 5.0, 331, EUGLENALES) M/F 6 Table 1.1, Gephyrocapsa oceanica Kamptner M
336, 338 343, 425 234
Glaucocystis (Glaucophyta) F 6 Gyrodinium uncatenatum Hulbert M Kirchneriella (Chlorophyta

Table 1.1 35 Table 1.6 CHLOROCOCCALES) F 6 Table
Glenodinium (Dinophyta Gyrosigma (Bacillariophyta 1.1, 336
PERIDINIALES) F 7 Table 1.1, 13, BACILLARIALES) F, D 344 Koliella (Chlorophyta
324, 334 ULOTRICHALES) F 6 Table 1.1,
Gloeocapsa (Cyanobacteria Halosphaera (Prasinophyta, 320 Table 7.1
CHROOCOCCALES) F 325 HALOSPHAERALES) M 306 Koliella longiseta (Vischer) Hindák
Gloeotrichia (Cyanobacteria Hemiaulus (Bacillariophyta F, X3 332
NOSTOCALES) F 6 Table 1.1, 26 BIDDULPHIALES) M 166, 305,
Table 1.2, 67, 81, 129, 205, 213, 385 Lagerheimia (Chlorophyta
216, 338, 401 Hemiaulus hauckii Grun. M 311, CHLOROCOCCALES) F
Gloeotrichia echinulata (J. E. Smith) 316, 317 344
Richter F, H2 216, 320 Table 7.1, Hemidinium (Dinophyta Lagerheimia genevensis (Chodat)
338 PHYTODINIALES) F 7 Table 1.1 Chodat F, X1 425
Golenkinia (Chlorophyta Hemiselmis (Cryptophyta Lauderia (Bacillariophyta
CHLOROCOCCALES) F, X1 320 CRYPTOMONADALES) M 234 BIDDULPHIALES) M
Table 7.1, 352 Heterocapsa (Dinophyta Lauderia annulata Cleve M
Gomphosphaeria (Cyanobacteria PERIDINIALES) M 233, 234, 311
CHROOCOCCALES) F 6 Table 312 Lauderia borealis Gran M 312
1.1, 26 Table 1.2, 81, 166, 205, Heterocapsa triquetra (Ehrenb.) Lepocinclis (Euglenophyta
233, 330, 332, 338 Stein M 310, 311, 312, 313, 316 EUGLENALES) F, W1 6 Table 1.1,
Main species G. naegeliana Heterosigma (Raphidophyta 343, 425
transferred to Woronichinia CHLOROMONADALES) M Leptocylindrus (Bacillariophyta
Goniochloris (Xanthophyta Heterosigma carterae (Hulbert) BIDDULPHIALES) M 203 Table
MISCHOCOCCALES) F 6 Table Taylor M 35 Table 1.6 5.2, 312, 317
1.1 Histoneis (Dinophyta DINOPHYSALES) Leptocylindrus danicus Cleve M 309,
Gonium (Chlorophyta VOLVOCALES) M 203 Table 5.2 310, 311
F, W1 321 Table 7.1, 343 Limnothrix (Cyanobacteria
Gonyostomum (Raphidophyta Isochrysis (Haptophyta OSCILLATORIALES) F 6 Table
CHLOROMONADALES) F 6 PRYMNESIALES) M 7 Table 1.1, 1.1, 26 Table 1.2, 114, 191, 213,
Table 1.1, 12, 321 Table 7.1 234 401, 405
Gonyostomum semen (Ehrenb.) Isochrysis galbana Parke M 35 Limnothrix redekei (van Goor)
Diesing F, Q 344 Table 1.7 Meffert F, S1 206, 233, 320
Gymnodinium (Dinophyta Table 7.1, 345, 366, 416
GYMNODINIALES) M/F 7 Table Karenia (Dinophyta Lingulodinium (Dinophyta
1.1, 233, 311, 325, 326, 328, 336, GYMNODINIALES) M 233 GONIAULACALES) M 7 Table 1.1,
425 Karenia brevis (Davis) G. Hansen et 68, 311
Gymnodinium baicalense Authority Moestrup M 408 Lingulodinium polyedrum (Stein)
not found F 322 Karenia mikimotoi (Miyake et Dodge M 234, 309, 316, 408
Gymnodinium catenatum Graham M Kominami) G. Hansen et Lyngbya (Cyanobacteria
68, 233, 309, 408 Moestrup M 35 Table 1.6, 310, OSCILLATORIALES) F 6 Table
Gymnodinium helveticum Penard F 311, 312, 316, 408 1.1, 26 Table 1.2, 59, 325
131, 324 Katablepharis (Cryptophyta Lyngbya limnetica Lemm. F, R
Gymnodinium sanguineum Hirasaka CRYPTOMONADALES) M 159 211
M 35 Table 1.6 Katodinium (Dinophyta
Gymnodinium simplex Lohm. M 35 GYMNODINIALES) M/F 316 Mallomonas (Chrysophyta
Table 1.6 Kephyrion (Chrysophyta SYNURALES) F 7 Table 1.1, 23,
Gymnodinium uberrimum (Allman) CHROMULINALES) F 7 Table 1.1 129, 232, 293, 320 Table 7.1,
Kofoid et Swezey F 324 Kephyrion littorale Lund F, X1 20 323
Gymnodinium vitiligo Ballantine M Table 1.1 Mallomonas akrokomos Ruttner in
35 Table 1.6 Keratococcus (Chlorophyta Pascher F, X1 333
Gyrodinium (Dinophyta CHLOROCOCCALES) F Mallomonas caudata Iwanoff F, X2
PERIDINIALES) M 7 Table 1.1, Keratococcus raphidioides Hansgirg 20 Table 1.1, 130, 203 Table 5.2,
233, 234, 316 F, X1 425 331
Mantoniella (Prasinophyta Monoraphidium griffithii (Berkeley) Many planktic genera reclassified

CHLORODENDRALES) M 6 Table Komárková-Legnerová F, X1 into Limnothrix, Planktothrix and
1.1 (taxon rationalises various Tychonema
Mantoniella squamata Manton et species and subspecies of
Parke M 35 Table 1.7 Ankistrodesmus) Pandorina (Chlorophyta
Melosira (Bacillariophyta Mougeotia (Chlorophyta VOLVOCALES) F 6 Table 1.1, 232,
BIDDULPHIALES) F 344 ZYGNEMATALES) F, T 233, 320 338, 343
Melosira varians Agardh F, D 344 Table 7.1, 330, 336 Pandorina morum (O. F. Müller)
Merismopedia (Cyanobacteria Nannochloris (Chlorophyta Bory F, G 219 Table 5.4, 329
CHROOCOCCALES) F, LO 6 VOLVOCALES) M 6 Table 1.1, Paulschulzia (Chlorophyta
Table 1.1, 26 Table 1.2, 165, 171, 234, 311, 425 TETRASPORALES) F 6 Table 1.1,
325, 326 Nannochloris ocelata Droop M 35 129
Micractinium (Chlorophyta Table 1.7 Paulschulzia pseudovolvox (Schulz)
CHLOROCOCCALES) F, X1 352 Nephrodiella (Xanthophyta Skuja F, F 331
Microcystis (Cyanobacteria MISCHOCOCCALES) F 6 Table Pavlova (Haptophyta PAVLOVALES) M
CHROOCOCCALES) 6 Table 1.1, 1.1 7 Table 1.1, 234
24, 26 Table 1.2, 50, 55, 59, 67, Nephroselmis (Prasinophyta Pavlova lutheri (Droop) Green M
81, 82, 87, 113, 123, 129, 165, CHLORODENDRALES) M/F 6 (formerly known as Monochrysis
166, 179, 184, 186, 190, 192, 205, Table 1.1 lutheri) 32, 35 Table 1.7
206, 208, 213, 214, 215, 226, Nitzschia (Bacillariophyta Pediastrum (Chlorophyta
228, 229, 231, 232, 247, 249, BACILLARIALES) M/F 7 Table 1.1, CHLOROCOCCALES) F 6 Table
267, 269, 271, 272, 275, 279, 194, 234, 307, 308, 313, 340, 1.1, 51, 192, 211, 320 Table 7.1,
285, 292, 293, 297, 298, 299, 424 324, 339, 341, 344
300, 320 Table 7.1, 327, 328, Nitzschia acicularis (Kütz.) W. Smith Pediastrum boryanum (Turpin)
339, 341, 344, 348, 353, 359, F 320 Table 7.1, 322, 342 Meneghin F, J 20 Table 1.1
366, 370, 381, 401, 403, 404, 405 Nitzschia bicapitata Lagerstedt M Pediastrum duplex Meyen F, J 67
Microcystis aeruginosa (Kütz.) 309 Pedinella (Chrysophyta
emend. Elenkin F, LM , M 26 Nitzschia closterium W. Smith M PEDINELLALES) F 7 Table 1.1
Table 1.2, 54 Table 2.3, 57 Table 308 Pedinomonas (Prasinophyta
2.4, 58, 81, 129, 157 Table 4.2, Nodularia (Cyanobacteria PEDINOMONADALES) F 6 Table
184 Table 5.0, 198, 209, 219, 219 NOSTOCALES) M/F 6 Table 1.1, 1.1
Table 5.4, 333, 339, 345 26 Table 1.2, 165, 205 Pelagococcus (Chrysophyta
Microcystis wesenbergii (Komárek) Nodularia spumigenea Mertens ex PEDINELLALES) M 7 Table 1.1,
Komárek in Kondrat’eva 195, Born et Flah. M/F H1 312, 426 35 Table 1.7
219, 271 Pelagomonas (Chrsophyta
Micromonas (Prasinophyta Ochromonas (Chrysophyta PEDINELLALES) M 7 Table 1.1,
CHLORODENDRALES) M 6 Table CHROMULINALES) F, X3 7 Table 305, 308
1.1, 234, 312 1.1, 131, 159, 251, 312, 323, 328, Pelodictyon (Anoxyphotobacteria) F 6
Micromonas pusilla (Butcher) 330, 331, 425 Table 1.1
Manton et Parke M 35 Table 1.7 Oocystis (Chlorophyta Peridinium (Dinophyta
Monochrysis (Chrysophyta CHLOROCOCCALES) F 6 Table PERIDINIALES) M/F 7 Table 1.1,
CHROMULINALES) M, F 332 1.1, 104, 233, 312, 324, 325, 326, 13, 68, 157 Table 4.2, 165, 213,
Monodus (Eustigmatophyta) F 6 340 233, 307, 311, 326, 327, 342,
Table 1.1, 20 Table 1.1 Oocystis lacustris Chodat F, F 129, 366, 385
Monodus subterraneus 184 Table 5.0 203 Table 5.2, 231, 320 Peridinium aciculiferum
Monoraphidium (Chlorophyta Table 7.1 Lemmermann F, W2 326
CHLOROCOCCALES) F 6 Table Ophiocytium (Xanthophyta Peridinium bipes Stein F, LM ?
1.1, 23, 312, 320 Table 7.1, 324, MISCHOCOCCALES) F 6 Table 336
333, 338, 343, 344, 352 1.1 Peridinium cinctum (O. F. Müller)
Monoraphidium contortum (Thuret) Ornithocercus (Dinophyta Ehrenb. F, LM 20 Table 1.1, 219
Komárková-Legnerová F, X1 DINOPHYSIALES) M 203 Table Table 5.4, 338
(taxon rationalises various 5.2, 234, 306, 314, 316 Peridinium depressum Bail. M 306,
species and subspecies of Oscillatoria (Cyanobacteria 307
Ankistrodesmus) 26 Table 1.2 OSCILLATORIALES) F 115 Peridinium faroense Paulsen M 310
Peridinium gatunense Nygaard F, LO 191, 207, 208, 209, 213, 225, Prorocentrum minimum (Pavillard)
205, 218, 327 228, 241, 259, 269, 271, 275, Schiller M 312
Peridinium inconspicuum 350, 358, 365, 370, 381, 401, Protoperidinium (Dinophyta
Lemmermann F, LO 332, 336, 405 PROROCENTRALES) M 313
338 Planktothrix agardhii (Gom.) Anagn. Prymnesium (Haptophyta
Peridinium lomnickii Wooszyska F, Y et Kom. F, S1 (formerly known PRYMNESIALES) M 7 Table 1.1,
320 Table 7.1, 334 as Oscillatoria agardhii) 54 Table 35 Table 1.7, 251, 270, 353, 402,
Peridinium ovatum Pouchet M 306 2.3, 115, 157 Table 4.2, 184 407
Peridinium pallidum Ostenfeld M Table 5.0, 186, 190, 191, 192, Prymnesium parvum N. Carter M
306 193, 206, 207, 231, 233, 320 313, 316
Peridinium umbonatum Stein F, LO Table 7.1, 324, 325, 333, 339, Pseudanabaena (Cyanobacteria
232, 338, 425 345, 366, 378, 408, 410, 416, OSCILLATORIALES) F 6 Table
Peridinium volzii Lemmermann F, 422 1.1, 26 Table 1.2, 233, 320 Table
LO 336 Planktothrix mougeotii (Bory ex 7.1, 345
Peridinium willei Huitfeldt-Kaas F, Gom.) Anagn. et Kom. F, R Pseudanabaena limnetica (Lemm.)
LO 232, 320 Table 7.1, 325, 331, (formerly known as Oscillatoria Komárek F, S1 416
332, 336 agardhii var. isothrix) 26 Table Pseudonitzschia (Bacillariophyta
Phacotus (Chlorophyta VOLVOCALES) 1.2, 33, 59, 219 Table 5.4, 226, BACILLARIALES) M 307, 311, 312
F 6 Table 1.1, 172, 343 233, 320 Table 7.1, 332 Pseudonitzschia delicatissima (P. T.
Phacus (Euglenophyta EUGLENALES) Planktothrix prolifica (Gom.) Anagn. Cleve) M (formerly known as
F 6 Table 1.1, 343 et Kom. F, R (formerly known Nitzschia delicatissima) 307, 311,
Phacus longicauda (Ehrenb.) as Oscillatoria prolifica) 59, 312
Dujardin F, W1 15 Fig 1.2 193 Pseudopedinella (Chrysophyta
Phaeocystis (Haptophyta, Planktothrix rubescens (Gom.) PEDINELLALES) M/F 7 Table 1.1,
PRYMNESIALES) M 7 Table 1.1, Anagn. et Kom. F, R (formerly 312, 332
13, 24, 234, 271, 306, 308, 311, known as Oscillatoria rubescens) Pseudopedinella pyriformis N. Carter
385, 407 59, 115, 193, 211, 233, 234, 320 M 35 Table 1.7
Phaeocystis antarctica Karsten M Table 7.1, 325, 328, 330, 346, Pseudosphaerocystis (Chlorophyta
308 366, 409 TETRASPORALES) F 6 Table 1.1,
Phaeocystis pouchetii (Hariot) Platymonas (Prasinophyta 56, 129, 211, 293, 320 Table 7.1,
Lagerheim M 35 Table 1.7, 312 CHLORODENDRALES) M 311 324
Phaeodactylum (Bacillariophyta Plectonema (Cyanobacteria Pseudosphaerocystis lacustris
BACILLARIALES) 177 OSCILLATORIALES) F 166 (Lemm.) Novák F, F
Phaeodactylum tricornutum Bohlin Porosira (Bacillariophyta (formerly known as Gemellicystis
M 28 BIDDULPHIALES) M 310, 311 neglecta) 57 Table 2.4, 231, 248,
Plagioselmis (Cryptophyta Porosira glacialis (Grun.) Jørgensen 331
CRYPTOMONADALES) F 6 Table M 310, 311 Pyramimonas (Prasinophyta
1.1, 218, 225, 230, 232, 233, 284, Prochlorococcus (Cyanobacteria PYRAMIMONADALES) M 308,
320 Table 7.1, 323, 324, 327, PROCHLORALES) M 6 Table 1.1, 313
330, 331, 336, 338, 343, 425 11, 26 Table 1.2, 203 Table 5.2, Pyrenomonas (Cryptophyta,
Plagioselmis nannoplanctica (Skuja) 213, 305, 306, 316, 322 CRYPTOMONADALES) M 6
Novarino et al. F, X2 (formerly Prochloron (Cyanobacteria Table 1.1
known as Rhodomonas minuta) PROCHLORALES) M 6 Table 1.1, Pyrenomonas salina (Wislouch)
20 Table 1.1, 172, 219 Table 5.4, 11, 26 Table 1.2 Santore M
277 Table 6.3, 338 Prochlorothrix (Cyanobacteria Pyrocystis (Dinophyta
Planktolyngbya (Cyanobacteria PROCHLORALES) F, T? 6 Table GONIAULACALES) M 13, 234,
OSCILLATORIALES) F 59, 115, 1.1, 11, 26 Table 1.2 305, 316
342 Prorocentrum (Dinophyta Pyrocystis noctiluca Murray M
Planktolyngbya limnetica (Lemm.) PROROCENTRALES) M 7 Table 55
Komárková-Legnerová et 1.1, 233, 234, 305, 311, 313 Pyrodinium (Dinophyta
Cronberg F, R 193 Prorocentrum balticum (Lohm.) GONIAULACALES) M 316, 408
Planktothrix (Cyanobacteria Loeblich M 311 Pyrodinium bahamense (Böhm)
OSCILLATORIALES) F 6 Table Prorocentrum micans Ehrenb. M Steidinger et al. M 316,
1.1, 23, 26 Table 1.2, 58, 83, 114, 312, 313 408
Radiococcus (Chlorophyta Scrippsiella trochoidea (Stein) Staurodesmus triangularis

CHLOROCOCCALES) F Loeblich M 310, 311, 312, 313, (Lagerheim) Teiling F, N 336
Radiococcus planctonicus Lund F, F 316 Stephanodiscus (Bacillariophyta
(formerly known as Coenococcus Skeletonema (Bacillariophyta BIDDULPIALES) F 7 Table 1.1,
planctonicus) 248, 331 BIDDULPIALES) M 7 Table 1.1, 23, 53, 54, 63, 148, 214, 233,
Rhizosolenia (Bacillariophyta 312, 317 243, 245, 249, 279, 329, 344
BIDDULPIALES) M 7 Table 1.1, Skeletonema costatum Grev. M 33, Stephanodiscus binderanus (Kütz.) W.
203 Table 5.2, 234, 305, 306, 130, 234, 306, 309, 311, 312, 313 Krieg. F, B (formerly known as
307, 313 Snowella (Cyanobacteria Melosira binderana) 198, 322,
Rhizosolenia alata Brightw. M 130, CHROOCOCCALES) F 6 Table 324
307, 308, 310, 311, 312 1.1, 24, 26 Table 1.2 Stephanodiscus hantzschii Grun. F, D
Rhizosolenia calcar-avis Schultze M Sphaerella (Chlorophyta 6 Table 1.1, 20 Table 1.1, 29
311, 313, 316, 317 VOLVOCALES) F Table 1.3, 31 Table 1.5, 184
Rhizosolenia delicatula Cleve M 309, (Most species transferred to Table 5.0, 320 Table 7.1, 338
310, 311, 312 Haematococcus) 425 Stephanodiscus minutulus Cleve et
Rhizosolenia fragilissima Bergon M Sphaerocauum (Cyanobacteria Moller F, B (formerly known as
312 CHROOCOCCALES) F, M 320 S. astraea var. minutula) 333,
Rhizosolenia hebetata Bail. M 306, Table 7.1 338
307, 308, 310 Sphaerocystis (Chlorophyta Stephanodiscus rotula (Kütz.)
Rhizosolenia setigera Brightw. M 60 CHLOROCOCCALES) F Hendey F, C (formerly known as
Rhizosolenia stolterforthii H. Perag. (some former species now in S. astraea) 6 Table 1.1, 20 Table
M 309 Coenochloris) 82, 203 Table 5.2 1.1, 29 Table 1.3, 31 Table 1.5,
Rhizosolenia styliformis Brightw. M Sphaerocystis schroeteri Chodat F, F 50, 52, 54 Table 2.3, 61 Table
305, 306, 307, 310, 311, 316, 317 194 2.5, 66, 232, 320 Table 7.1, 338,
Rhodomonas (Cryptophyta, Spirulina (Cyanobacteria 339
CRYPTOMONADALES) M/F 6 OSCILLATORIALES) M 6 Table Stichococcus (Chlorophyta
Table 1.1, 212, 234, 260, 311, 1.1, 26 Table 1.2, 59, 320 Table ULOTRICHALES) F, X1 6 Table
312, 325 7.1, 341, 404 1.1
Rhodomonas lens Pascher et Spondylosium (Chlorophyta Surirella (Bacillariophyta
Ruttner F 35 Table 1.7 ZYGMEMATALES) F, P 6 Table BACILLARIALES) F 341, 344
Richiella (Cyanobacteria, 1.1, 273 Synechococcus (Cyanobacteria
OSCILLATORIALES) M 166, 305 Staurastrum (Chlorophyta CHROOCOCCALES) M/F 6 Table
ZYGMEMATALES) F 6 Table 1.1, 1.1, 11, 20 Table 1.1, 26 Table
Scenedesmus (Chlorophyta, 24, 51, 129, 202, 232, 326, 336, 1.2, 97, 155, 165, 181, 183, 184
CHLOROCOCCALES) F 6 Table 338, 340, 415 Table 5.0, 186, 203 Table 5.2,
1.1, 67, 182, 195, 211, 269, 270, Staurastrum brevispinum West F, P 269, 277 Table 6.3, 305, 306,
320 Table 7.1, 324, 339, 341, 57 Table 2.4 320 Table 7.1, 326
344 Staurastrum chaetoceras (Schröder) Picoplanktic Z, free cells of
‘‘Scenedesmus quadricauda (Turpin) G. M. Sm. F, P 201 Aphanocapsa. Larger species in
Brébisson’ F, J Oft-recorded Staurastrum cf. cingulum (W. et Cyanothece
species is no longer regarded as G. S. West) F, P 247 Synechocystis (Cyanobacteria
an entity and is being Staurastrum dorsidentiferum W. et CHROOCOCCALES) M/F 6 Table
subdivided (John et al., 2002) 20 G. S. West F, P 328 1.1, 26 Table 1.2, 269
Table 1.1, 26 Table 1.2, 60, 157 Staurastrum longiradiatum W. et Several species, all colonial; F, K?;
Table 4.2, 195, 277 Table 6.3, G. S. West F, P? 219 free cells are picoplanktic
424 Staurastrum pingue Teiling F, P 20 Cyanobium, Z
Scherfellia (Prasinophyta Table 1.1, 26 Table 1.2, 195, 201, Synechocystis limnetica Popovskaya
CHLORODENDRALES) F 6 Table 219 Table 5.4, 228, 229, 320 322
1.1 Table 7.1, 333 Synedra (Bacillariophyta
Scourfieldia (Prasinophyta Staurodesmus (Chlorophyta BACILLARIALES) F 7 Table 1.1,
SCOURFIELDIALES) M 6 Table ZYGMEMATALES) F 6 62, 192, 233, 328, 344
1.1, 425 Table 1.1, 129, 203 Table 5.2, Synedra acus 54 Table 2.3, 320
Scrippsiella (Dinophyta 233, 293, 320 Table 7.1, 325, Table 7.1, 338
PERIDINIALES) M 312 331, 385 Synedra filiformis Grun. F, D
Synedra ulna (Nitzsch) Ehrenb. F, B Thalassiosira punctigera (Castracene) 213, 228, 229, 231, 292, 320
20 Table 1.1, 322 Hasle M 130 Table 7.1, 323, 325, 328, 330
Synura (Chrysophyta SYNURALES) F Thalassiosira subtilis Ostenf. M 309 Uroglena americana Calkins F, U
7 Table 1.1, 231, 232, 320 Table Thalassiosira weissflogii (Grun.) 326, 328
7.1, 321 Table 7.1, 343 Fryxell et Hasle M (formerly Uroglena lindii Bourrelly F, U 20
Synura petersenii Korshikov F, W1 known as Thalassiosira fluviatilis) Table 1.2
130 35 Table 1.7, 54 Table 2.3, 60, Urosolenia (Bacillariophyta
Synura uvella Ehrenb. F, E 331 130 BIDDULPHIALES) F 7 Table 1.1,
Thalassiothrix (Bacillariophyta 129, 232, 320 Table 7.1, 328,
Tabellaria (Bacillariophyta BIDDULPHIALES) M 307 336, 385
BACILLARIALES) F 7 Table 1.1, Thalassiothrix longissima Cleve et Uroslenia eriensis (H. L. Smith)
293, 320 Table 7.1, 325, 329, Grun. M 307 Round et Crawford F, A 203
330 Thiocystis (Anoxyphotobacteria) F 6 Table 5.2, 323, 331,
Tabellaria fenestrata (Lyngb.) Kütz. Table 1.1 335
F, P? 233, 323 Thiopedia (Anoxyphotobacteria) F 6
Tabellaria flocculosa (Roth.) Kütz. F, Table 1.1 Volvox (Chlorophyta VOLVOCALES) F
N 29 Table 1.3, 54 Table 2.3, Trachelomonas (Euglenophyta 6 Table 1.1, 12, 68, 83, 179, 205,
233, 323, 332, 333 EUGLENALES) F 6 Table 1.1 211, 216, 285, 292, 294, 320
Tabellaria flocculosa var. Trachelomonas hispida (Perty) Stein Table 7.1, 343
asterionelloides (Grun.) Knuds. F, F, X2 15 Fig 1.2 Volvox aureus Ehrenb. F, G 26 Table
N 6 Table 1.1, 20 Table 1.1, 61 Trachelomonas volvocina Ehrenb. F, 1.2, 157 Table 4.2, 182, 184
Table 2.5, 64, 184 Table 5.0, 247 W2 321 Table 7.1 Table 5.0
Tetraedron (Chlorophyta, Treubaria (Chlorophyta Volvox globator Linnaeus F, G 20
CHLOROCOCCALES) F 324 CHLOROCOCCALES) F 352 Table 1.2
Tetrastrum (Chlorophyta, Tribonema (Xanthophyta
CHLOROCOCCALES) F 6 Table TRIBONEMATALES) F 6 Table Westella (Chlorophyta
1.1, 51 1.1, 12, 320 Table 7.1, 324 CHLOROCOCCALES) F 352
Thalassionema (Bacillariophyta Trichodesmium (Cyanobacteria Willea (Chlorophyta
BIDDULPHIALES) M 307, 311 OSCILLATORIALES) M 6 Table CHLOROCOCCALES) F 203 Table
Thalassionema nitzschioides (Grun.) 1.1, 26 Table 1.2, 81, 166, 169, 5.2
Mereschkowsky M 307, 309, 203 Table 5.2, 234, 305, 306, Woloszynskia (Dinophyta
310, 311 316, 385 GYMNODINIALES) F, X2 7 Table
Thalassiosira (Bacillariophyta Trichodesmium thiebautii Gomont M 1.1
BIDDULPHIALES) M 7 58, 166 Woronichinia (Cyanobacteria
Table 1.1, 60, 67, 271, 307, 311, Tychonema (Cyanobacteria CHROOCOCCALES) F 6 Table
314, 317 OSCILLATORIALES) F 6 Table 1.1, 24, 26 Table 1.2, 81, 166,
Thalassiosira baltica Grun. M 312 1.1, 26 Table 1.2 233, 272, 320 Table 7.1, 325,
Thalassiosira decipiens (Grun.) Tychonema bourrellyi F, S1 115 332, 338, 401
Jørgensen M 311 Woronichinia naegeliana (Unger)
Thalassiosira fluviatilis (syn. T. Umbellosphaera (Haptophyta Elenkin F, LO (formerly
weissflogii) q.v. COCCOLITHOPHORIDALES) M 7 Gomphosphaeria naegeliana and
Thalassiosira hyalina Grun. M 311 Table 1.1, 203 Table 5.2, 305, Coelosphaerium naegelianum) 248
Thalassiosira nordenskioeldii Cleve M 306, 309, 333
234, 307, 310, 311, 312 Uroglena (Chrysophyta Xanthidium (Chlorophyta
Thalassiosira pseudonana Hasle et CHROMULINALES) F 7 Table 1.1, ZYGNEMATALES) F, N 6 Table
Heindal M 35 Table 1.7 24, 83, 129, 130, 203 Table 5.2, 1.1
Index to genera and species of organisms
other than phytoplankton
Abramis Actinopterygii, Blastocladiella Fungi, Blastocladiales Chydorus Crustacea, Cladocera 254

Cypriniformes 422 293 Table 6.1, 261, 266 Table 6.2,
Acanthometra Rhizopoda, Radiolaria Bodo Zoomastigophora 252 Table 279, 287
252 Table 6.1 6.1, 259 Chydorus sphaericus 261, 266 Table
Acartia Crustacea, Calanoidea 261, Boeckella Crustacea, Calanoidea 255 6.2, 279
262 Table 6.1, 336 Chytridium Fungi, Chytridiales 293
Acartia clausi 262 Boeckella gracilipes 336 Ciona Urochorda, Ascidacea 258
Acartia longiremis 261 Bosmina Crustacea, Cladocera 254 Table 6.1
Acineta Ciliophora, Suctoria 252 Table 6.1, 266 Table 6.2, 282, Clavelina Urochorda, Ascidacea 258
Table 6.1 287 Table 6.1
Actinophrys Rhizopoda, Heliozoa 252 Bosmina longirostris 266 Table 6.2 Clio Mollusca, Gastropoda 256 Table
Table 6.1 Brachionus, Rotatoria, Monogononta 6.1
Aeromonas Bacteria, Vibrionaceae 253 Table 6.1, 259, 266 Table Clupea Actinopterygii, Clupeiformes
162 6.2, 281 261, 263, 427
Agrostis Angiospermae, Glumiflorae Brachionus calyciflorus 266 Table Clupea harengus 261, 427
335 6.2, 281 Coleps Ciliophora, Spirotricha 259
Alburnus Actinopterygii, Bythotrephes Crustacea, Cladocera Colpoda Ciliophora, Holotricha 252
Cypriniformes 422 254 Table 6.1, 261, 290 Table 6.1
Alosa Actinopterygii, Clupeiformes Calanus Crustacea, Calanoidea 255 Conochilus Rotatoria, Monogononta
290 Table 6.1, 261, 428 253 Table 6.1
Alosa pseudoharengus 290 Calanus finmarchius 261 Convoluta Platyhelminthes,
Apherusa Crustacea, Malacostraca Canthocamptus Harpacticoidea 255 Turbellaria 253 Table 6.1
255 Table 6.1 Table 6.1 Coregonus Actinopterygii,
Arcella Rhizopoda, Foraminifera 252 Carchesium Ciliophora, Peritricha Clupeiformes 289
Table 6.1 252 Table 6.1 Coregonus artedii 289
Argulus Crustacea, Branchiura 255 Carcinus Crustacea, Malacostraca 256 Cortaderia Angiospermae,
Table 6.1 Table 6.1 Glumiflorae 335
Artemia Crustacea, Anostraca 254 Centropages Crustacea, Calanoidea Craspedacusta Coelenterata,
Table 6.1, 269 255 Table 6.1, 261, 262, 282 Trachylina 253 Table 6.1
Asellus Crustacea, Isopoda 391 Centropages hamatus 261 Cyanea Coelenterata, Scyphozoa 253
Asplanchna Rotatoria, Monogononta Centropages typicus 262 Table 6.1
253 Table 6.1, 281 Cephalodella Rotatoria, Monogononta Cyclops Crustacea, Cyclopoidea 260
Asplanchna priodonta 281 292 Cyclops vicinus 260
Asterias Echinodermata, Asteroidea Ceratophyllum Angiospermae, Cyprinus Actinopterygii,
257 Table 6.1 Ranales 419 Cypriniformes 422
Asterocaelum Rhizopoda, Amoebina Ceriodaphnia Crustacea, Cladocera Cyprinus carpio 422
252 Table 6.1, 294 254 Table 6.1, 261, 277, 287 Cypris Crustacea, Ostracoda 254
Aurelia Coelenterata, Scyphozoa 253 Cetorhinus Selachii, Euselachii 263 Table 6.1
Table 6.1 Chaetonotus Gastrotricha 253 Cytophaga Bacteria, Cytophagales
Australocedrus Gymnospermae, Table 6.1 142
Coniferae 335 Chaoborus Arthropoda, Diptera 256
Table 6.1, 269 Daphnia Crustacea, Cladocera 86,
Bacillus Bacteria, Bacillales 162 Chara Chlorophyta, Charales 419 221, 254 Table 6.1, 261, 265,
Balanus Crustacea, Cirripedia 255 Chirocephalus Crustacea, Anostraca 266, 266 Table 6.2, 267, 269,
Table 6.1 254 Table 6.1 270, 274, 276, 277, 277 Table
Beroë Coelenterata, Nuda 253 Table Chironomus Arthropoda, Diptera 6.3, 279, 280, 282, 284, 285, 286,
6.1 422 287, 289, 290, 299, 359, 390,
Bicosoeca Zoomastigophora 252 Table Chironomus anthracinus 422 392, 404, 409, 418, 421, 422
6.1, 259 Chironomus plumosus 422 Daphnia cucullata 261, 287
520
INDEX TO GENERA AND SPECIES OF ORGANISMS OTHER THAN PHYTOPLANKTON 521
Daphnia galeata 221, 261, 266, 266 Euphausia tricantha 263 Keratella cochlearis 259, 281
Table 6.2, 267, 269, 276, 284, Euplotes Ciliophora, Spirotricha 252 Keratella quadrata 259, 281
287, 289 Table 6.1, 269
Daphnia hyalina 261, 266 Eurydice Crustacea, Malacostraca 255 Lates Actinopterygii, Perciformes
Table 6.2 Table 6.1 263, 264, 341
Daphnia lumholtzi 261 Eurytemora Crustacea, Calanoidea Lates niloticus 341
Daphnia magna 261, 266, 277, 421, 255 Table 6.1 Lepomis Actinopterygii, Perciformes
422 Evadne Crustacea, Cladocera 254 Lepomis cyanellus 422
Daphnia pulex 266, 270 Table 6.1, 262 Leptodiaptomus Crustacea,
Daphnia pulicaria 261, 277, 289, Calanoidea 409
290, 418 Festuca Angiospermae, Glumiflorae Leptodora Crustacea, Cladocera 254
Daphnia schoedleri 266 335 Table 6.1, 261
Diaphanosoma Crustacea, Cladocera Filinia Rotatoria, Monogononta 253 Leptomysis Crustacea, Malacostraca
254 Table 6.1, 409, 421 Table 6.1, 281 255 Table 6.1
Diaptomus Crustacea, Calanoidea Filinia longiseta 281 Limacina Mollusca, Gastropoda 256
266 Table 6.2, 389 Fitzroya Gymnospermae, Coniferae Table 6.1
Diaptomus oregonensis 266 Table 6.2 335 Limnocnida Coelenterata, Trachylina
Flavobacterium Bacteria 142, 162 253 Table 6.1
Diastylis Crustacea, Malacostraca 255 Flustra Ectoprocta 257 Limnothrissa Actinopterygii,
Table 6.1 Table 6.1 Clupeiformes 263, 264, 392
Difflugia Rhizopoda, Foraminifera Fulica Aves, Gruiformes Limnothrissa miodon 263, 264,
252 Table 6.1 Fulica atra 422 392
Dinophysis Dinophyta 252 Littorella Angiospermae, Campanales
Table 6.1 Gadus Actinopterygii, Gadiformes 419
Doliolum Urochorda, Thaliacea Gadus morhua 427 Lobelia Angiospermae, Campanales
258 Table 6.1 Gammarus Crustacea, Amphipoda 419
Dreissena Mollusca, 389, 391 Loligo Mollusca, Cephalopoda 256
Lamellibranchiata 256 Table Gastrosaccus Crustacea, Malacostraca Table 6.1, 264
6.1, 323 255 Table 6.1 Lolium perenne Angiospermae,
Dreissena polymorpha 323 Gigantocypris Crustacea, Ostracoda Glumiflorae 383
254 Table 6.1
Echinus Echinodermata, Echinoidea Globigerina Rhizopoda, Foraminifera Macrohectopus Crustacea,
257 Table 6.1 252 Table 6.1 Malacostraca 255 Table 6.1
Ensis Mollusca, Lamellibranchiata Megalogrammus Actinopterygii,
256 Table 6.1 Halteria Ciliophora, Spirotricha 252 Gadiformes 427
Eodiaptomus Crustacea, Calanoidea Table 6.1 Megalogrammus aeglifinus
287 Hertwigia Rotatoria, Monogononta 427
Eodiaptomus japonicus 287 292 Mesocyclops Crustacea, Cyclopoidea
Epistylis Ciliophora, Peritricha 252 Holopedium Crustacea, Cladocera 254 254 Table 6.1, 260
Table 6.1 Table 6.1 Mesocyclops leuckarti 260
Escherichia Bacteria, Entobacteriacae Holothuria Echinodermata, Mesodinium Ciliophora,
49, 155, 158, 181 Holothuroidea 258 Table 6.1 Rhabdophorina 68
Escherichia coli 49, 155, 158, 181 Hydra Coelenterata, Hydrida 253 Metopus Ciliophora, Spirotricha 252
Esox Actinopterygii, Mesichthyes 422 Table 6.1 Table 6.1
Hydractinia Coelenterata, Microstomum Platyhelminthes,
Esox lucius 422 Anthomedusae 252 Table 6.1 Turbellaria 253 Table 6.1
Eudiaptomus Crustacea, Calanoidea Moina Crustacea, Cladocera 254
255 Table 6.1, 266 Table 6.2, Isoetes Pteridophyta, Isoetales 419 Table 6.1
279, 280, 286, 390 Monas Zoomastigophora 252 Table
Eudiaptomus gracilis 266 Table 6.2, Kellicottia Rotatoria, Monogononta 6.1, 259
280, 390 253 Table 6.1, 281 Monosiga Zoomastigophora 252
Euphausia Crustacea, Malacostraca Kellicottia longispina 281 Table 6.1, 259
256 Table 6.1, 263 Keratella Rotatoria, Monogononta Morone Actinopterygii, Perciformes
Euphausia frigida 263 253 Table 6.1, 259, 281, 287 428
522 INDEX TO GENERA AND SPECIES OF ORGANISMS OTHER THAN PHYTOPLANKTON
Mulinum Angiospermae, Umbellales Penilia Crustacea, Cladocera 262 Salmo 389

335 Peranema Zoomastigophora 252 Salmo trutta 389
Myriophyllum Angiospermae, Table 6.1 Salpa Urochorda, Thaliacea 258
Myrtales 419, 422 Perca Actinopterygii, Perciformes Table 6.1
289, 290 Salpingooeca Zoomastigophora 252
Mysis Crustacea, Malacostraca 255 Perca flavescens 290 Table 6.1
Table 6.1 Perca fluviatilis 289 Salvelinus Actinopterygii,
Najas Angiospermae, Najadales Petromyzon Agnatha, Cyclostomata Clupeiformes 290
419 290 Salvelinus namaycush 290
Nassula Ciliophora, Holotricha 252 Petromyzon marinus 290 Schoenoplectus Angiospermae,
Table 6.1, 259 Phoronis Phoronidea 257 Table 6.1 Cyperales 419
Nebaliopsis Crustacea, Malacostraca Phragmites Angiospermae, Scomber Actinopterygii, Perciformes
255 Table 6.1 Glumiflorae 419 261
Nitella Chlorophyta, Charales Physalia Coelenterata, Siphonophora Scomber scombrus 261
419 253 Table 6.1 Sialis Arthropoda, Megaloptera 256
Noctiluca Dinophyta 251, 252 Pleurobrachia Coelenterata, Table 6.1
Table 6.1 Tentaculata 253 Table 6.1 Sida Crustacea, Cladocera 254 Table
Nothofagus Angiospermae, Fagales Pleuronema Ciliophora, Holotricha 6.1, 421
335 252 Table 6.1, 259 Simocephalus Crustacea, Cladocera
Notholca Rotatoria, Monogonta 253 Plumularia Coelenterata, 254 Table 6.1, 261, 421
Table 6.1 Leptomedusae 252 Table 6.1 Sphagnum Bryophyta, Sphagnales
Notonecta Arthropoda, Hemiptera Podochytrium Fungi, Chytridiales 424
263 293 Sphingopyxis Bacteria 142
Nyctiphanes Crustacea, Malacostraca Podon Crustacea, Cladocera 254 Sphyraena Actinopterygii,
256 Table 6.1 Table 6.1, 262 Perciformes 428
Polyarthra Rotatoria, Monogonta 259, Squilla Crustacea, Malacostraca 255
Obelia Coelenterata, Leptomedusae 281, 287 Table 6.1, 263
252 Table 6.1 Pontomyia Arthropoda, Diptera 256 Stentor Ciliophora, Spirotricha 252
Oikopleura Urochorda, Larvacea 258 Table 6.1 Table 6.1
Table 6.1, 259 Potamogeton Angiospermae, Stolothrissa Actinopterygii,
Oithona Crustacea, Cyclopoidea 254 Najadales 419, 420 Clupeiformes 263, 264, 392
Table 6.1 Prorodon Ciliophora, Holotricha 252 Stolothrissa tanganicae 263, 392
Ophiura Echinodermata, Table 6.1, 259 Strobilidium Ciliophora, Spirotricha
Ophiuroidea 257 Table 6.1 Protoperidinium Dinophyta 252 Table 252 Table 6.1, 259
Oreochromis Actinopterygii, 6.1 Strombidium Ciliophora, Spirotricha
Perciformes 341 Pseudopileum Fungi, Chytridiales 293 252 Table 6.1, 259, 340
Oreochromis niloticus 341 Pyrosoma Urochorda, Thaliacea 258 Synchaeta Rotatoria, Monogonta 253
Ostrea Mollusca, Lamellibranchiata Table 6.1 Table 6.1, 259
256 Table 6.1
Oxyrrhis Dinophyta 251, 252 Table Quercus Angiospermae, Fagales 331, Temora Crustacea, Calanoidea 255
6.1, 270 359 Table 6.1, 261, 282
Temora longicornis 261
Palinurus Crustacea, Malacostraca Rhizophydium Fungi, Chytridiales Thiobacillus Bacteria, Chromatiaceae
256 Table 6.1 293, 294, 295 162
Patella Mollusca, Gastropoda 256 Rhizophydium planktonicum emend Thiobacillus denitrificans
Table 6.1 293 162
Pelagia Coelenterata, Scyphozoa 253 Rhizosiphon Fungi, Chytridiales 293 Thunnus Actinopterygii, Perciformes
Table 6.1 Rozella Fungi, Chytridiales 294 428
Pelagonemertes Nemertea 253 Table Rutilus Actinopterygii, Tintinnidium Ciliophora, Spirotricha
6.1 Cypriniformes 278, 422 252 Table 6.1, 259
Pelagothuria Echinodermata, Rutilus rutilus 278, 422 Tomopteris Annelida, Polychaeta 253
Holothuroidea 258 Table 6.1 Table 6.1, 263
Pelomyxa Rhizopoda, Amoebina 252 Sagitta Chaetognatha 256 Trichocerca Rotatoria, Monogonta 253
Table 6.1 Table 6.1 Table 6.1
INDEX TO GENERA AND SPECIES OF ORGANISMS OTHER THAN PHYTOPLANKTON 523
Tropocyclops Crustacea, Cyclopoidea Velella Coelenterata, Siphonophora Zostera Angiospermae, Najadales

254 Table 6.1 253 Table 6.1 420
Typha Angiospermae, Typhales 419 Vibrio Bacteria, Vibrionaceae 158, Zygorhizidium Fungi, Chytridiales
162 293, 294
Vampyrella Rhizopoda, Amoebina Vorticella Ciliophora, Peritricha 252 Zygorhizidium affluens 294
294 Table 6.1 Zygorhizidium planktonicum 293
General index
absolute viscosity, 44 Arrhenius coefficient, 107 bacteriochlorophylls, 5, 6 Table 1.1

accessory pigments, 10, 12, 13, 25, Arrhenius temperature scale, 186 bacteriophages, 295
95, 115, 193 arrow worms (chaetognaths), 256 bacterioplankton, 2, 140, 141;
acetylene-reduction assay of Table 6.1, 263 concentrations and daily
nitrogenase activity, 164 artificial neural networks (ANN), production rates, 141 Table 3.4;
‘acid rain’, 423 235 definition of, 2
acidification of natural waters, 423; ascendency, 303, 355, 357 bacterivorous phytoplankton, 131,
reversal by adding phosphorus, ascidaceans, 258 Table 6.1 141
425; reversal by liming, 425 ‘ascidian tadpole’, 258 Table 6.1 bacterivory in algae, 159
Actinopterygii, 258 Table 6.1 ascorbate, 123 barium, 167
adaptive traits in phytoplankton, ash content of phytoplankton, 25, barnacles, 255 Table 6.1
207 26 Table 1.2, 27 barracuda (Sphyraena), 428
ADP (adenosine diphosphate), 99 assembly rules for phyoplankton basking shark (Cetorhinus), 263
advective patchiness, 87 communities, 362, 362 Table 7.8 bicarbonate transport, 128
aestivation, 59 atelomictic lakes, 74, 342 bicosoecids, 251
aeolian deposition, 170 atomic-absorption spectroscopy, 167 Biddulphiales (centric diatoms), 7
affinity adaptation of nutrient atmospheric carbon-dioxide Table 1.1, 13
uptake, 157, 194, 202, 203 invasion in, 127 bioassays, 169
airborne remote sensing, 134 ATP (adenosine triphosphate), 5, 95, biochemical oxygen demand (BOD),
akinetes, 216, 249 99; phosphorylation by, 148, 149 297
alder flies (Megaloptera), 256 Table Auftrieb, 3 biodiversity, 368
6.1 Aufwuchs, 419, 420 biologically available phosphorus
alewife (Alosa pseudoharengus), 290 autopoesis; (see also (BAP), 153, 154 Table 4.1, 155, 398
alkaline phosphatase, 155, 159 ‘self-organisation’ of biomanipulation, 263, 416
alkalinity, 124 communities), 355 biomineral reinforcement of cell
allogenic vs. autogenic drivers of auxospore, 60, 64 structures, 27
change in species composition, Avogadro number, 95 bioturbation, 250, 413
360 avoidance reactions, 122 birds, 1, 76, 84
alternative steady states in shallow bivalves (lamellibranchs), 243, 256
lakes, 343, 382, 417, 421; forward Bacillariales (pennate diatoms), 7 Table 6.1, 290
and reverse switching, 421, 423; Table 1.1, 13 bleak (Alburnus), 422
influence of fish, 421; role of bacillariophytes; (see also diatoms); Blelham Enclosures, 217, 220, 226,
zooplankton, 421 density of, 53, 55; deposition of 245, 274, 284, 284 Table 6.4, 299,
aluminium, 424 silica, 182, 245; evolution of, 13; 357, 366, 370, 373, 374
amino acids, 123 growth limitation through silicon bloom, 56; in the Kattegat, 1988,
amoebae, 252 Table 6.1, 259 deficiency, 196; intervention in 312
amphipods, 263 silicon cycling by, 174; ‘bloom-forming’ Cyanobacteria, 401
anatoxins, 403 mixed-depth threshold for blue-green algal blooms, 401
annelids, 253 Table 6.1 growth, 80, 244; pigmentation bodonids, 251
anoxygenic photosynthesis, 5 variability in, 115; primary boron, 28
anoxyphotobacteria, 5, 6 Table 1.1 evolutionary strategies of, 234; bosminids, 261, 279
anthropogenic impacts, on food silicon content of, 28, 29 Table bottom deposits, sampling the
webs, 290, 426; on pelagic 1.3, 31 Table 1.5, 53, 220; silicon superficial layer of, 220
systems, 395 requirements of, 27, 173, 197; bottom–up and top–down processes,
antioxidants, in photosynthesis, 123 silicon uptake by, 28, 174; 250, 287, 288, 416
apatite, 151 structure of, 13 boundary layer, around a plankter,
apoptosis, 297 Bacteria, 2, 5, 140, 158, 164, 170, 147, 148
aquo polymers, 39 264, 277 Table 6.3, 388, 392 branchiopods, 254 Table 6.1, 260
archaeans, 5, 93 bacterial pathogens, 292, 295 bream (Abramis), 422, 423
524
GENERAL INDEX 525
brown tides, 407 limitation of photosynthesis by, chaetognaths, 256 Table 6.1, 263
bryophytes, 11 127, 130; limitation of Chaetophorales, 12
bryozoans, 353 phytoplankton growth by, 30; chain formation, 60
buoyancy regulation, through ionic metabolic turnover of, 127; chaoborines, 256 Table 6.1
balance, 55; in Cyanobacteria, 56, pelagic animal content of, 281; chaotic behaviour, 396
58 recycling of, in aquatic systems, Charales, 12
buoyant forces in water masses, 72 391; requirement of for algal cell chelates, 168, 169
burgundy blood alga, 115 doubling, 146, 188; sources of, 3 chemiluminescence, 167
carbon dioxide; atmospheric chitin, 60
C4 carbon fixers, 98 content of, 14; concentrations in chloride, algal requirement for, 172
C-, S- and R- primary evolutionary water, 124; depletion in chlorine, 28
strategies, 209, 210 Table 5.3, 212, photosynthesis, 96, 98, 126; as a Chlorobiaceae, 6 Table 1.1, 193
231, 233, 234, 298, 315; ecological factor regulating species Chlorococcales, 6 Table 1.1, 12
traits of C strategists, 352, 357, composition, 13; flux from Chlorodendrales, 6 Table 1.1, 11
370; ecological traits of S atmosphere to water, 127, 132; Chloromonadales, 6 Table 1.1
strategists, 359 supersaturating concentrations chlorophyll a, occurrence in algae,
CS, CR, RS intermediate strategies, of, 126; uptake by phytoplankton, 5, 6 Table 1.1, 7 Table 1.1, 12, 13,
210 127 25, 26 Table 1.2, 95; algal content
CSR triangle, 210, 232, 315, 363, carbonic anhydrase, 130 of, 33, 35 Table 1.7, 36, 37, 113; as
364, 396, 408 carboxylation, 94, 99, 127 a proxy of algal biomass, 34;
calanoids, 254 Table 6.1, 260, 261, carboxylic acid, 100 relative to cell carbon, 114, 125
279 carotenoids, 25, 95, 123 chlorophyll b, occurrence in algae, 6
calcium, 28; algal requirement for, carp (Cyprinus carpio), 422, 423 Table 1.1, 11, 95
171; ionic concentrations in the cell growth cycle, 179 chlorophyll c, occurrence in algae, 6
sea, 171; in lakes, 171 cell assembly, regulation of, 181 Table 1.1, 7 Table 1.1, 12, 13
calcium bicarbonate and DIC, 171 cell division, 180; in chlorophytes, chlorophyll concentration
calcium carbonate in phytoplankton 182; in diatoms, 180, 182; in supported in relation to TP, 399
exoskeletons, 13, 25, 53 dinoflagellates, 180; in chlorophyll synthesis, 168
calcium carbonate scales, 25 prokaryotes, 181; light efficiency Chlorophyta; classification of, 6
calcium hardness, 171 of (r/I), 190, 206 Table 1.1; DIC requirements of,
Calvin cycle, 94, 95, 96, 98, 99, 100 cell division rates; in culture, 183; 123, 129; evolution of, 11;
CAMP receptor protein (CRP), 128, as a function of algal osmotrophy in, 131
181 morphology, 183; as a function of chloroplast, 146
capacity, environmental carrying; of temperature, 186; in relation to choanoflagellates, 13, 251
atmospheric carbon flux, 136; of fluctuating light intensity, 193; in Chromatiaceae, 6 Table 1.1, 193
available silicon, 202; of the relation to light income, 206; in chromatic adaptation, 115
nutrient resources, 152, 194; of relation to persistent low light chromatographic analysis, 167
PAR income, 138; of phosphorus intensity, 192, 207; in relation to chromophores, 123
availability, 159; of primary nitrogen deficiency, 195; in chromosomes, 179
production, 131, 134, 136, 138; as relation to nutrient deficiency, Chromulinales, 6 Table 1.1
set by mixed depth, 138; as set by 194, 200, 203, 204; in relation to Chroococcales, 6 Table 1.1, 26 Table
transparency, 117 Table 3.2 nutrient stoichiometry, 200; in 1.2
carapace gape of cladocerans, 260, situ, 219 Table 5.4; in relation to chrysolaminarin, 7 Table 1.1
261, 280 phosphorus deficiency, 194; in Chrysophyta; classification of, 6
carbamylation, 99 relation to photoperiod, 189; in Table 1.1, 13; DIC requirements
carbohydrate, 53 relation to resource interaction, of, 129; distribution and calcium
carbon; algal content of, 28, 30, 32, 197; in relation to resource hardness, 171; evolution of, 11, 12;
33, 33 Table 1.6, 35 Table 1.7, 36, supply, 188; maxima at 20 o C, mixotrophy in, 131; pigmentation
146; availability to photosynthetic 184 Table 5.1 variability in, 115; silicon
organisms, 124, 125; bacterial cell quota, definition, 31; of requirements, 27; storage
content of; concentration phosphorus, 31 products in, 25; vitamin
mechanism, 128; external cell wall, 24, 27 requirements of, 170
subsidies of, to pelagic systems, centropomids, 263 chrysose, 7 Table 1.1
390; fixation in the plankton, 93; cephalopods, 1, 18, 256 Table 6.1 chydorids, 261, 279
526 GENERAL INDEX
chytrids, 66, 292 critical patch size, 88, 89 decomposition rates of algae, 297
ciliates, 252 Table 6.1, 259, 260, 261, crustaceans, 259; classification of, deep ‘chlorophyll maxima’ (DCM),
353, 354 254 Table 6.1 59, 83, 115, 192, 328
ciliophorans, 252 Table 6.1, 259 Cryptomonadales, 6 Table 1.1; density; of air, 46; of carbohydrates,
cirripedes, 255 Table 6.1 osmotrophy in, 131; 53; of diatoms, 53, 55; of
cisco (Coregonus artedii), 289 photadaptation in, 115 metabolic oils, 53; of mucilage, 55
cladocerans, 254 Table 6.1, 260, 261, cryptophytes, evolution of, 12; ; of nucleic acids, 53; of
354 morphology of, 6 Table 1.1, 25 phytoplankton, 51, 53, 54 Table
Cladophorales, 12 ctenophores (sea combs and sea 2.3, 55; of polyphosphate bodies,
climate change, 431 gooseberries), 253 Table 6.1, 263 53; of proteins, 53; salinity
clupeoids, 263, 389 cultural eutrophication, definition dependence of, 40; of silica, 53;
coastal waters, phytoplankton of, of, 397 temperature dependence of, 40; of
233, 310 cultures, 167 water, 39 Table 2.1, 40
cobalt, 28, 167 cyanelles, 6 Table 1.1, 11 density gradients in water columns,
coccoliths, 7 Table 1.1, 13, 25, 130 Cyanobacteria; buoyancy in, 56, 58, 72, 74, 75, 79, 204
coccolithophorids, evolution of, 13, 67, 81, 124; cytology of, 25; density stratification, 73, 75
14; morphology of, 7 Table 1.1, 25 classification of, 6 Table 1.1, 10, desmids, 6 Table 1.1, 12, 25; DIC
coelenterates, 252 Table 6.1 26 Table 1.2; in deep chlorophyll requirements of, 129
Coleochaetales, 12 layers, 193; DIC requirements of, detergents as a source of P, 398
community assembly in the 130; distribution and calcium detritus, 419
plankton, 302 hardness, 171; enhanced diadinoxanthin–diatoxanthin
community ecology of abundance with eutrophication, reaction, 123
phytoplankton, 302, 350 402; evolution of, 10, 11; gas diatoms; (see also bacillariophytes),
compensation point (where vacuoles of, 24, 56; habits of, 10; 13; deposition of silica, 182, 245;
photosynthetic gains and in relation to low DIN, 196; DIC requirements of, 129;
maintenance losses are balanced), nitrogen fixation in, 164, 169, 398; evolution of, 27, 234; growth
16, 116, 118, 120 in relation to N : P ratio, 198; limitation though silicon
competition, definition of, 152 overwintering of, 405; pH deficiency, 196; intervention in
competitive exclusion principle (of tolerance in, 124, 425; silicon cycling, 174; mixed-depth
Hardin), 203, 369 photoprotection in, 123; threshold for growth, 80, 244;
compositional change in pico–microplanktic morphology of, 27;
communities rate of, 367 transformation, 269; production photoprotection in, 123; settling
condensates of assimilation, 6 Table of siderophores, 169; storage velocities of, 52, 66, 67, 245;
1.1, 7 Table 1.1, 25, 26 Table 1.2, products in, 25; structure of silicon content of, 28, 29 Table
189 photosynthetic apparatus, 96; 1.3, 31 Table 1.5, 32, 53, 220;
constancy, 304 toxicity in, 56, 402, 403; ‘warning’ silicon requirements of, 27, 173,
Continuous Plankton Recorder concentrations, 1229 197; uptake of silicon, 174;
(CPR), 307 cyanobacterial blooms, 401 vitamin requirements of, 170
contractile vacuoles, 25 cyanophages, 295 diatoms, centric, 6 Table 1.1, 25
convection, 52 cyanophycin,85 diatoms, pennate, 7 Table 1.1
coot (Fulica atra), 422, 423 cyclopoids, 254 Table 6.1, 260, 278 diatoxanthin, 6 Table 1.1, 12
copepods, 254 Table 6.1, 259, 260 cylindrical curves, 81 dilution functions, 240
copper, 28; algal requirement for, cysts, of dinoflagellates, 215 dimethyl sulphide (DMS), 173
167; toxicity of, 167 cytochrome, 96, 167, 168 dimethyl sulphonoproponiate
copper sulphate, as an algicide, 167, cytoplasm, 25 (DMSP), 173, 307; algal
416 osmoregulation by, 173;
coregonids, 389 DAPI (DNA-specific stain), 220 DMS–DMSP metabolism,
corethrines, 256 Table 6.1 DCMU, 66 173
Coriolis force, 17, 42 DNA, 140, 146 dinitrogen reductase, 164
crabs (decapods), 256 Table 6.1, 263 DNA : cell carbon ratio, as an index dinoflagellates; adaptive radiation
Craspedophyceae, 13 of DNA replication, 220 in, 13, 14; DIC requirements of,
crepuscular habitats, 192 daphniids, features of, 261; habitats 129, 130; evolution of, 12, 13, 16;
critical depth model (of Sverdrup), of, 261 mixotrophy in, 131;
119, 120 decapods, 256 Table 6.1, 263 photosynthesis in, 123;
GENERAL INDEX 527
‘swimming’ velocities, 68; vitamin eddy spectrum, 17, 18, 38, 42, 48, exergy, 303, 356; fluxes of, 357,
requirements of, 170 251 375
Dinophyta, 7 Table 1.1; dinophyte El Niño, 305, 309, 382 exoskeletons, 25, 27
zooplankters, 252 Table 6.1 electivity of predators, 278 exponential phase of population
dipterans, 263 electrophoretic mobility, 67 growth, 183
disentrainment of phytoplankton, endemism among phytoplankton, extracellular production, 100, 140,
75, 123 353 239
disequilibrium explanations of endoplasmic reticulum, 25 eyespots, 12
species diversity, 369; consumer endosymbiosis, 11, 13, 15, 305
effects, 371; intermediate energy flow in pelagic systems, 387, factor limitation of capacity, 169
disturbances, 372, 374, 384; 388; relevance to structure, 393, faecal pellets, 288
pathogenic effects, 371; 427 ferredoxin, 96, 165, 168
resource-based competition, 371 engineer species, 382, 419 Fick’s diffusion laws, 127, 147
dispersal of plankton, 353 entrainability, 17 Ficoll, 51
dissipation of turbulence, 47 entrainment of phytoplankton, 38, filter-feeders, 260, 261, 262
dissipative ecological unit (DEU), 39, 67, 69, 74, 75, 204, 243 filter-feeding; on detrital POC, 277;
351, 355, 382 entropy, 355 impacts of, 264, 270, 274;
dissolution of diatom frustules, 174 environmental grain, 47 intervention of planktivores, 277,
dissolved humic matter (DHM), 142, environmental heterogeneity, 289; limiting food concentrations
390, 424 78 for, 265, 275, 277; measurement
dissolved inorganic carbon (DIC), epilimnion, 75 of, 265; nutritional aspects, 270;
125; uptake in phytoplankton, 128 epilithic algae, 342 rates of, 265, 266, 274; in relation
epiphytes, 285 to food availability, 267; in
dissolved inorganic nitrogen (DIN), equitability (evenness), 366 relation to temperature, 274;
163 ethylene diamine tetra-acetic acid saturating food concentrations
dissolved organic carbon (DOC), 93, (EDTA), 168 for, 274, 276; size selection, 267;
139, 140, 141, 142, 261, 264, 388; Eubacteria, 5 in turbid water, 280
DOC produced by phytoplankton, Euglenales, 6 Table 1.1 filtering setae of cladocerans, 260,
100, 124; DOC as a source of DIC, euglenoids, 12, 25 261
126 euglenophytes; classification, 6 fish, 1, 18
dissolved organic matter (DOM), Table 1.1; evolution of, 11, 12; fishing industry, 427
142 osmotrophy in, 131; storage fish-rearing ponds, 418
dissolved organic nitrogen (DON), products in, 25 fish-supportive capacity of pelagic
163 eukaryotes, 11; early evolution of, production, 389
disturbance, 304, 372 11, 12, 13 flagellates, 6 Table 1.1; vertical
diversity indices, 366 euphausids, 256 Table 6.1, 263 distribution of, 83
doliolids, 258 Table 6.1, 262 euphotic zone, 116 flagellum, propulsion, 49
droop model of nutrient uptake, euplankton, 2 Flexibacteria, 5
150 Eustigmatophyta, 6 Table 1.1, 12 flood-plain lakes, 352
dry masses of phytoplankton, 25, 26 Eutreptiales, 6 Table 1.1 flow cytometry, 140
Table 1.2; in relation to volume, eutrophic food webs, 140 fluorescence, 122; as a surrogate of
25 eutrophic lakes, phytoplankton of, biomass, 122; as a surrogate of
ducks, 423 232 phyletic composition, 122
Dugdale model of nutrient uptake, eutrophication; of coastal waters, fluorometry, 52, 132
150 426; of lakes, 397; of seas, 425; of flushing, 240
dugongs, 427 Windermere, 333; reversal of, fluvial ‘dead zones’, 242
dust deposition (aeolian deposition), 408 fluvioglacial deposits, 337
170 evaporative heat losses from water foamlines, 85, 86
bodies, 41 food chains of the pelagic, 261;
echinoderms, 257 Table 6.1 evenness (or equitability), 366 energetic dissipation in, 291;
ecological efficiency of energy excretion, of excess photosynthate, interactive linkage strength, 291;
transfer, 261 123 length of, 291
ecological stoichiometry, 199, 262 excretory products of zooplankton, food location by chemoreception,
ectoprocts, 257 Table 6.1 287 280
528 GENERAL INDEX
food requirements, of calanoids, grazing losses of phytoplankton, high-nutrient, iron-deficient areas of

262 220 the ocean, 170
of cladocerans,262; of pelagic green tides, 407 Hill–Bendall model of
fish, 389 greenhouse gases, oceanic photosynthesis, 94
food, zooplankton-supportive absorption of, 428 holopedids, 261
capacity of, 262 gross primary production (GPP), 141 holotrichs, 252 Table 6.1, 259
food-web control nodes, 394 growth rate; (see also replication horizontal patchiness of
foraging in calanoids, 279; in rate), 178 phytoplankton, 84, 88
relation to pelagic energy flows, growth-regulating nucleotides, 158 human metabolites, as a source of
388; saturating food availability, growth and reproductive strategies P, 398
280; superiority over of freshwater phytoplankton, 208; humic acids, 168
filter-feeding, 279, 286, 394 correspondence with morphology, humic lakes, plankton of, 12, 326
foraminiferans, 252 Table 6.1, 211; C-, S- and R- primary hydrogen, algal content of, 28, 32,
259 evolutionary strategies, 209, 233, 33 Table 1.6
form resistance of phytoplankton 315; CS, CR, RS intermediate hydroxyapatite, 129
(to sinking), 50, 60; coefficient of, strategies, 210; SS strategy, 211, hydrozoans, 252 Table 6.1
51; as contributed by chain 234; violent, patient and hyperparasitism, 294
formation, 60; as provided by explerent strategies, 211 hyperscum, 215
protuberances and spines, 60; of guanosine bipyrophosphate (ppGpp), hypersensitiviy of hosts to parasitic
stellate colonies, 63 158, 181, 214 attack, 294
friction velocity; (see also turbulent Gulf Stream (North Atlantic Drift hypnozygotes, 215
velocity), 46, 48 Table 2.2 Current), 289, 306 hypolimnion, 75
frictional drag coefficients, 41 Gymnoceratia, 216, 229
frustules; (see also diatoms, siliceous Gymnodiniales, 7 Table 1.1 ice cover, 306, 307, 308, 322
cell walls), 27, 32, 174 gymnodinioids, 13 information theory, 366
fucoxanthin, 6 Table 1.1, 7 Table 1.1, intermediate disturbance hypothesis
12, 13 HNLC oceans (high nitrogen, low (IDH), 373, 377
fulvic acids, 168 chlorophyll), 308 invasion by species ‘new to the
fungal epidemics, 293 habitat filtration, 360 locality’, 384
fungi, 2; parasitic on algae, 292 habitat templates, 315, 348, 394 internal phosphorus loading, 412
haddock (Megalogrammus aeglifinus), ionic regulation, 55
Gaia principle, 173 427 iron, 167, 335, 398, 424; algal
gas vacuoles, 56 Haeckel, E., 3 requirement for, 168; cell
gas vesicles, 57, 58 Halobacteria, 5 contents of, 28, 168
gastropods, 256 Table 6.1 Haptonema, 13 complexes with organic carbon
gastrotrichs, 253 Table 6.1 haptophytes; classification of, 7 (DOFE), 170; limitation of
Gelbstoff, 111 Table 1.1; evolution of, 11, 13; nitrogen fixation, 169; limiting
generation time, 178, 188 morphology of, 24; pigmentation concentrations, 168; sources of,
germination of resting stages, 214, variability in, 115; vitamin 170; symptoms of cell deficiency,
229, 250 requirements of, 170 168; uptake by algae, 168
gill rakers, 263, 389 harmful algal blooms (HAB), 312, iron-binding chelates, 168
Gilvin, 111 401 IRONEX oceanic fertilisation, 170,
glaciations, 336 heat flux, solar, 72 308
Glaucophyta, 6 Table 1.1 heleoplankton, definition of, 2 island biogeography, 352
global warming, 14 hemipterans, 263 isopycny, 19, 72, 82
glucose, 100 Henry’s law, 125, 126
glutathione, 123 Hensen, V., 3 jellyfish, 2
glycerol, 52 herbivory on phytoplankton, 250 Jenkin mud sampler, 249
glycogen, 6 Table 1.1, 25, 26 Table herring (Clupea harengus), 261, 263,
1.2, 59, 98, 189 427 kairomones, 269
glycolate excretion, 100, 239 heterocysts (heterocytes), 6 Table karyokinesis, 179
glycolic acid, 100, 123 1.1, 26 Table 1.2, 164, 196, 230 kataglacial lakes, phytoplankton of,
Gonyaulacales, 7 Table 1.1 heterotrophy, 12, 126, 141 232
gonyaulacoids, 13 Hibberdiales, 7 Table 1.1 Kelvin–Helmholtz instability, 79
GENERAL INDEX 529
keystone species, 382, 394, 427 light-saturation of photosynthesis, manganese, 28, 167, 424; algal
kinematic viscosity, 44, 48 103, 106 Table 3.1, 122, 136 requirement for, 167
KISS model of critical patch size, 88 limiting capacity, 152 mannitol, 6 Table 1.1
Kolmogorov eddy scale, 42, 45 limiting factors, 151 mantis shrimps, 263
krill (euphausids), 256 Table 6.1, 263 limnoplankton,, 2 marine snow, 66, 166, 248, 249, 430
lipid, 6 Table 1.1, 7 Table 1.1, 54; marl lakes, 129
accumulation, 54 mass mortalities of phytoplankton,
lake trout (Salvelinus namaycush), 290 lipo-polysaccharides, 403 214
lakes as a source of CO2 , 390 Lloyd’s crowding index, 82 match–mismatch hypothesis, 290
lamellibranchs, 256 Table 6.1 lobsters (decapods), 256 Table 6.1, maturation-promoting factors (MPF),
laminar flow, 44 263 181
Langmuir circulations, 85, 85 Lorenzian attractors, 396 mechanical energy loss in water
larvaceans, 258 Table 6.1, 259 loss rates of phytoplankton, 139, columns, 47
larvae, planktic: actinotrocha, 257 239; aggregated, 297; due to cell medusae, 252 Table 6.1
Table 6.1; amphiblastulae, 252 death, 240, 296; due to megalopterans, 256 Table 6.1, 263
Table 6.1; appendicularia, 258 consumption by animals, 240, megaplankton, 263
Table 6.1; auricularia, 257 Table 243, 278, 280, 285, 286; due to Mehler reaction, 100, 123
6.1; bipinnaria, 285 Table 6.5; downstream transport, 239, 242; membrane transport systems, 148
cyphonautes, 257 Table 6.1; due to parasitism, 240, 292, 295; memory in community assembly,
cypris, 255 Table 6.1; megalopa, due to sedimentation, 240, 243; 366, 380, 384
256 Table 6.1; nauplii larvae, 259; due to washout, 239, 240, 241; meromictic lakes, 74, 263
paralarvae, 264; phyllosoma, 256 from suspension, 70, 76; seasonal meromixis, 74, 340
Table 6.1; pilidium, 253 Table 6.1; variations, 297, 299 meroplankton, 2, 243
pluteus, 257 Table 6.1; losses, distinction between mesophytoplankton, size definition
trochophore, 256 Table 6.1; physiological and demographic of, 5
trochosphere, 253 Table 6.1; processes, 239 mesotrophic lakes, phytoplankton
veliger, 256 Table 6.1; zoea, 256 luxury uptake of nutrients, 31, 151, of, 233
Table 6.1 161, 194 mesozooplankto: of freshwaters,
latent heat of evaporation, 41 Lycopodium spores, as a model of 260; of the sea, 261
Laurentian great lakes, 290 suspended phytoplankton, 76, 80, metalimnion, 75
leptodorids, 261 245 methanogens, 5
leucosin, 6 Table 1.1 Michaelis–Menten kinetics, 150
licensing factors, 181 mackerel (Scomber scombrus), 261 microalgae (µ-algae), 2, 4
Liebig’s law of the minimum, macroecology, 380 microbial food web, 140, 259, 261,
152 macroinvertebrates (of macrophyte 318
light (see also photosynthetically beds), 419 microbial ‘loop’, 140
active radiation), 95; absorption macrophytes, 342, 391, 395; microcystins, 403, 404
by water, 138; attenuation architecture of, 418; effects of micronutrients, inorganic, 166;
(extinction) of, underwater, 103, eutrophication on, 420; leaf area organic, 170
109, 110, 113, 116; effect of cloud index in, 420; as refuges for microphytoplankton, size definition
on, 108; light underwater, 108, potamoplankton, 243 of, 5
110; surface incidence of, 108; macrophytoplankton, size definition microplanktic protists, 251
surface reflectance, 108 of, 5 microplanktic metazoans, 259
light harvesting by photosynthetic macroplanktic herbivores, 262 microstratification, 81
organisms, 94, 95, 123 macroplanktic predators, 261 microzooplankton, 170, 251, 259,
light-harvesting complex (LHC); areal macrothricids, 261, 279 261, 287, 392
concentrations of, 137; mechanics magnesium, 28; algal requirement migration of phytoplankton,
of, 96, 97, 192; numbers of, 113; for, 172; availability of, 172 vertical, 83
structure of, 95, 96 maintenance requirement of Milankovitch cycle, 431
light adaptation in phytoplankton, resource, 189 minimal communities, 351
113, 114, 120 mammals, 2, 18 minimum cell quota, of a given
light-dependent growth in malacostracans, 255 Table 6.1 nutrient, 150
phytoplankton, 206 manatees, 427 Mischococcales, 6 Table 1.1
light : nutrient ratio in lakes, 199 Mandala, of Margalef, 314, 408 mitochondria, 25
530 GENERAL INDEX
mitosis, 180 nanophytoplankton size definition nutrient-induced fluorescent

mixing times, 75, 76, 117 of, 5 transients (NIFT) indicative of P
mixolimnion, 340 nekton, 1, 2, 18, 263 deficiency, 159
mixotrophy in phytoplankton, 131, nematodes, 253 Table 6.1 nutrient ratios, in algal tissue, 28
159 net primary production (NPP), 134; nutrient regeneration, 287
models simulating phytoplankton by global domains, 135 Table 3.3
growth and performance, 234 netplankton, 4 ocean; heat exchanges, 41; renewal
Mollusca, 256 Table 6.1 niche, 303, 354 time, 39; total area of, 39; total
molybdate-reactive phosphorus Nile perch (Lates), 341 volume of, 39; water circulation
(MRP), 155, 399 Nile tilapia (Oreochromis), 341 in the, 41, 42
molybdenum, 28, 167, 335, 398; nitrate reductase, 163; synthesis of, oceanic fronts, 307, 308
algal requirement for, 167 168 oceanic provinces, 318
monimolimnion, 340 nitrilotriacetic acid (NTA), 168 Oedogoniales, 12
Monin–Obukhov length, 72, 79, 119, nitrogen; algal content of, 28, 30, oils (as storage product), 6 Table 1.1,
121, 243 32, 33, 33 Table 1.6, 35 Table 1.7, 7 Table 1.1, 25; effect on cell
Monod equation, 150 36, 161; availability to densities, 53, 54
monosccharides, 100 phytoplankton, 162, 163; cell oligotrophic lakes, 388;
Monosigales, 13 quotas of, 164; external phytoplankton of, 232
monosilicic acid, 28, 173, 182 concentrations favouring oligotrophic food webs, 140
monovalent : divalent cation ratio, dinitrogen fixers, 196; external open oceans, phytoplankton of,
172 concentrations half-saturating 233
morphological adaptations of phytoplankton growth, 196; operons, 149, 181
phytoplankton, 19, 23, 24, 48, 53, N-regulated capacity of opossum shrimps (mysids), 255
120 Patagonian lakes, 162, 196, 335; Table 6.1
motility, advantages of for potential chlorophyll yield, 164; organic composition of
phytoplankters, 205, 206 requirement of for cell doubling, phytoplankton, 28
mucilage, presence of in plankters, 146, 188, 195; sources of, 162; orthophosphoric acid, 153
271; as a defence against uptake rates of algal cells, 163 Oscillatoriales, 6 Table 1.1, 26 Table
digestion, 273; as a defence nitrogen fixation, 164; in 1.2
against grazers, 270, 273; as a Cyanobacteria, 164, 398; DIN osmoregulation, 25
defence against heavy metals, sensitivity of, 165, 196; osmotrophy, 131
273; as a defence against oxygen, dependence upon energy, 165; outgassing of CO2 from lakes and
273; for nutrient sequestration, dependence upon phosphorus, rivers, 126
272; production as a buoyancy 165; redox sensitivity of, 164; overwintering of Cyanobacteria,
aid, 55, 67; properties of, 24, 55, requirement for trace metals, 214, 405
271; relative volume of, 57 Table 165 oxygen, algal content of, 28, 32, 33
2.4; as a response to nutrient nitrogenase, 168 Table 1.6
deficiency, 271; for North American Great Lakes, 89
self-regulation, 272; for North Atlantic Oscillation (NAO), pH, 124, 423
streamlining, 272 289 pH–CO2 –bicarbonate system in
mucilaginous phytoplankters, 24 northern cod (Gadus morhua), 427 water, 125
mucilaginous threads, 67 Nostocales, 6 Table 1.1, 26 Table 1.2 pH sensitivity of phosphate
Müller, J., 3 nucleus, 25; nuclear division solubility, 153
mycoplankton, 2, 2 (karyokinesis), 179 packaging effect on pigment
mycosporine-like amino acids, 123 nucleic acids, 27, 53, 179 distribution, 192
mysids, 255 Table 6.1, 263 nutrients, 145; availability to paradox of the plankton (of
myxomycetes, 295 phytoplankton, 145, 146; cell Hutchinson), 368, 379
quotas, 150, 199; demand versus paralarvae, 256 Table 6.1
NADP (nicotinamide adenine supply, 145, 188; flux rates to paramylon (or paramylum), 6 Table
dinucleotide phosphate), 94, 96; cells, 147; intracellular sensitivity 1.1, 25, 98, 189
reduction of, 165 and control, 149; limitation, 151, parasites of phytoplankton, 292, 294
nannoplankton, 2 152; uptake by phytoplankton, parasitic fungi, 66
nanocytes, 215 145, 146, 148, 150; uptake, role of particulate organic carbon (POC),
nanoflagellates, 259, 280, 287 motion, 147 126, 279, 390, 393
GENERAL INDEX 531
particulate organic matter (POM), photoheterotrophy, 5 phototrophic bacteria, 5

390 photoinhibition, 104, 117, 121, 122, phycobilins, 6 Table 1.1, 10, 12, 25,
patchiness of phytoplankton, 78, 84, 124, 194, 246 26 Table 1.2, 168
88 photons, 95 phycobiliproteins, 115
Pavlovales, 7 Table 1.1 photon flux density (PFD), 95, 123 phycobilisomes, 10, 95
Péclet number (Pe), 147 photon interception and absorption, phycocyanin, 6 Table 1.1, 26 Table
Pedinellales, 7 Table 1.1 95 1.2, 115, 193
Pedinomonadales, 6 Table 1.1, 11 photooxidation, 121 phycoerythrin, 6 Table 1.1, 26 Table
penetrative convection, 72 photoprotection, 104, 121, 122, 194 1.2, 115, 193
perch (Perca fluviatilis), 289 photorespiration, 99, 123, 139, 239 phycomycetes, 292
perennation of phytoplankton, 249 photosynthesis, as a basis of aquatic phycoviruses, 295
Peridiniales, 7 Table 1.1 production,, 5, 93; biochemistry phyllopods, 260, 279
peridinians, 32 of, 94; carbon limitation of, 127, Phytodiniales, 7 Table 1.1
peridinin, 13 130; carbon oxidation, 100; carbon phytoplankters, 3
peritrichs, 252 Table 6.1 sourcing in, 124, 125; condensates phytoplankton, definition of, 2, 3,
Permian extinctions, 14 of, 25; electron transport, 94, 95; 36; functional classification of, 4;
phaeophytes, 12 hexose generation, 94, 95; phyletic classification of, 4, 5, 36;
phaeophytin, 96 Hill–Bendall model of, 94; general characteristics of, 16;
phagotrophy, 12; in algae, as a integration through depth, 104, numbers of species of, 4; size, 19,
means of nutrient sequestration, 119, 132; integration through 20 Table 1.1, 34, 35 Table 1.7, 36,
159 time, 132; light efficiency of (P/ I), 53; shape, 19, 20 Table 1.1, 22, 23,
phoronids, 257 Table 6.1 103, 105; light harvesting in, 94, 35 Table 1.7, 50; form resistance,
phosphatase production in algae, 95; light saturation of, 103, 106 50, 53, 60, 243; morphological
159 Table 3.1, 122, 136; measurement, adaptation, 19, 23, 24, 48, 53, 120;
phosphate binding in soils, 398 101, 105, 107; models of, 104, 110; turbulent embedding of, 48, 49;
phosphoglycolate, 99 net of respiration, 107, 132; ‘swimming’, 49; effect of viscosity,
phosphoglycolic acid, 123 oxygen generation, 94; productive 49; settling velocities of, 39, 49,
phosphorus, algal content of, 28, 31, yields, 133, 139; quantum 50, 52, 61 Table 2.5, 67, 68, 75, 77;
32, 33, 33 Table 1.6, 36, 151; efficiency of, 95, 98, 100; as a function of size, 53; as a
affinity adaptation, 157; reduction of CO2 , 94, 96, 98, 146; function of density, 53; as a
availability to phytoplankton, 398, reduction of water, 94; regulation function of chain formation, 62;
399; deficiency in cells, 158; by light, 101, 102, 108, 120, 121; in as a function of cylindrical
external concentrations relation to depth, 103; surface elongation, 62; as a function of
half-saturating phytoplankton depression of, 104 colony formation, 63; stellate
growth, 195; as a factor limiting photosynthetic inhibitors, 66 colonies, 63; settling velocities as
growth rate of phytoplankton, photosynthetic pigments, 5, 6 Table a function of cell vitality, 65, 66;
158; fractions in water, 154 Table 1.1, 7 Table 1.1, 11, 12, 25, 26 entrainment of, 38, 39, 67, 69, 74,
4.1; minimum requirements of Table 1.2, 123 75, 77; vertical distribution of, 80,
algae for, 151; potential photosynthetic quotient (PQ), 100, 83; horizontal patchiness of, 84,
chlorophyll yield, 160; quotas 105, 107, 163, 190 88; dry masses of, 25, 26 Table 1.2;
supporting replication, 158; photosynthetic rates, Tab 3.1, 104, densities of, 51, 53, 54 Table 2.3;
requirement of for cell doubling, 106 Table 3.1; temperature organic composition of, 28;
146, 188, 194; sources of, 151; dependence of, 106 elemental composition, 24, 25,
storage adaptation, 157; uptake photosynthetic yield to planktic 28, 35 Table 1.7, 36, 146;
rates of algal cells, 150, 155, 157 food webs, 139 photosynthesis in, 93; light
Table 4.2; velocity adaptation, 157 photosynthetically active radiation adaptation in, 113, 114, 120;
phosphorus, recycling of, 412 (PAR); (see also light), 16, 95, 108; uptake of carbon in, 127, 128;
phosphorus ‘release’ from carbon yield, 100; energy bacterivory in, 131; excretion in,
sediments, 413 equivalence, 95, 100 123; nutrient requirements of,
photadaptation, to low light doses, photosystem I, 94, 96, 122, 146; nutrient uptake, 19;
120, 193, 194 165 availability of nitrogen to, 164;
photoautotrophy, 5, 94, 141 photosystem II, 94, 95, 122, nitrogen uptake rates of, 163;
photodamage, 124 123 availability of phosphorus to, 153,
photodegradation (of DOM), 143 phototrophy, 12 154, 154 Table 4.1, 398, 399
532 GENERAL INDEX
phytoplankton, definition of (cont.) oligotrophic lakes, 319; of Planck’s constant, 95

phosphorus
; uptake rates of, 150, subarctic lakes, 328; of alpine plasmalemma, 24, 25, 146
155, 157 Table 4.2; lakes, 329; of the English Lakes, planktivory, 392, 395
phosphorus-constrained growth 330; of Auracanian lakes, 335; of plankton, definition of, 1, 2
of, 158; cell phosphorus quotas, kataglacial lakes, 336; of the plastids, 11, 12, 13, 25, 123
161; major-ion requirements of, African Great Lakes, 339; of Lake plastoquinone (PQ), 96, 122
171; as trophic-level indicators, Titicaca, 342; of shallow lakes, plate tectonics, 309
129; growth and replication of, 342; in relation to supportive polder lakes, 366
178; growth rates of, (see also cell capacity, 346; overview of polychaetes, 253 Table 6.1, 263
division rates), 183; as a function mechanisms, 383 polymerase chain reaction (PCR),
of nutrient availability, 151; phytoplankton population dynamics 140
light-dependent growth in, 206; in natural environments, 217; polymictic lakes, 74
areas projected by, 192; adaptive estimating in-situ rates of growth, polynia, 306
traits of, 207; growth and 217; episodic outbursts of rapid polyphemids, 261
reproductive strategies, 208; increase, 217; observed increase polyphosphate bodies, 53
strategies for survival of resource rates, 217; frequency of dividing polysaccharides, 98
exhaustion, 211; loss from cells (FDC), 218; frequency of Porifera (sponges), 27
suspension, 70, 76; losses to nuclear division, 220; from post-upwelling relaxation, 309
grazers, 139, 141, 220, 250; resource depletion, 220; spring potamoplankton, 2, 242
counter-grazing adaptations, 269; increase in a temperate lake, 221; potassium, 28; algal requirement
effects of zooplankton selection effect of inoculum size, 223, 225, for, 172
on, 287; beneficial effects of 229; effect of nutrient availability, power, 303, 355
zooplankton, 287; selection by 223; effect of temperature, 223, Prasinophyta, 6 Table 1.1, 11
performance, 225; seasonality of 226; effect of underwater light Prasiolales, 12
selective mechanisms, 231; field, 223, 226; effect of resource prawns (decapods), 256 Table 6.1
eutrophic species of, 232; diminution, 223; effect of lake Prochlorales, 6 Table 1.1, 26 Table
oligotrophic species, 203 Table enrichment on timing of 1.2
5.2, 232, 233; of shallow lakes, maxima, 224; species selection by prochlorobacteria, 6 Table 1.1, 11,
233; of coastal waters, 233; of performance, 225; exploitative 26 Table 1.2
open oceans, 233; of rivers, 242; efficiency, 225; comparison of programmed cell death, 297
entrainment criteria for, 243; species-specific performances, 226; projected area of algal shape, 192
sedimentary fluxes of, 247, 249, effect of light-exposure prokaryotes, 25
318; parasites of, 292, 294; thresholds, 226; in relation to Prorocentrales, 7 Table 1.1
parasitic infections of, 66; phosphorus availability, 228; in PROTECH phytoplankton growth
perennation of, 249; regenerative relation to pH, 228; in relation to model, 236, 377, 383, 415
strategies of, 249; trait separation N:P ratio, 230; in relation to proteins, 27, 53
of, 319; community assembly in, transparency, 230; in relation to protistans, 2, 11, 27, 251
350; succession in the sea, 313; carbon dioxide, 231; in relation to Protobacteria (α-), 142
biomass levels in the sea, 313, seasonality, 231, 232, 234; Protobacteria (γ -), 142
388; biomass levels in inland quantification in a Blelham protomonadids, 251
waters, 345; dispersal Enclosure, 221 Table 5.5, 221 proton motive forces, 148
mechanisms, 353; of acidic Table 5.6; as influenced by protoplasm, 25, 30
waters, 423, 424; of humic lakes, simultaneous loss rates, 298; protozoans, parasitic on algae, 294
326; methods to control species selection by performance, Prymnesiales, 7 Table 1.1
abundance, 414 298; effect of inoculum size, 299; purple non-sulphur bacteria, 5
phytoplankton assemblages, sinking losses., 299; simulation pycnocline, 75, 79, 245, 340
communities and structure, 302; models of, 234 Pyramimonadales, 6 Table 1.1, 11
of high-latitude seas, 306, 308; of phytosociology, 318; application to pyrosomans, 262
the North Pacific, 304; of the the pelagic, 319
South Pacific, 306; of oceanic Phytotelmata, 292, 352 Q10 , of photosynthetic rate, 107; of
upwellings, 308; of continental picophytoplankton, size definition cell division, 186
shelves, 309, 310; of inshore of, 5; settling velocities of, 68; in quanta, 95
waters, 317; of large oligotrophic oligotrophic oceans, 234 quantum efficiency of
lakes, 319; of small and medium plagioclimax, 370 photosynthesis, 95
GENERAL INDEX 533
quantum yield of photosynthesis, Richardson numbers, 73 sedimented material, resuspension

98, 100 Rift Valley lakes, 339; food webs of, of, 250
quinones, 96 263 seiching, 340
river bed roughness, 46 selective planktic feeding, 278;
r-, K- and w- selection, 209, 212, 231, rivet theory (of Ehrlich and Ehrlich), impacts of, 280
314 382 ‘self-organisation’ of communities,
radiolarians, 3, 27, 252 Table 6.1, RNA, 140, 146, 199, 262 355
259 roach (Rutilus rutilus), 278, 392, 422 seston, definition of, 2
rails, 423 rotifers, 253 Table 6.1, 259, 292, settlement of diatom frustules, 175
Raphidomonadales, 6 Table 1.1 354 settling flux, 220
Raphidophyta, 6 Table 1.1, 12 RUBISCO (ribulose 1, 5-biphosphate shallow lakes, characteristics of,
raptorial feeders, planktic, 260, 278 carboxylase), 94, 95, 98, 99, 127, 319, 342, 391; alternative steady
recruitment of larval fish, 289 128; as an oxidase, 99, 123, 140 states in, 343, 417; phytoplankton
recycling of phosphorus, 412 RuBP (ribulose1, 5-biphosphate), 94, of, 233
recycling of resources in the 98, 99, 123 Shannon–Weaver index, 366, 378
pelagic, 288, 296 sharks, 427
Redfield stoichiometric ratio, 32, 33, salmonids, 389 shear, 148, 250, 413
33 Table 1.6, 37, 188, 198 salps, 258 Table 6.1, 261, 262 shear velocity; (see also turbulent
red tides, 407 sampling strategies for velocity), 46, 48 Table 2.2
redox-sensitivity of phosphate phytoplankton, 89 shelf waters, 117 Table 3.2, 135, 309,
solubility, 153; of nitrogen saprophytes, 297 430
speciation, 162; of nitrogen sarcodines, 259 Sherwood Number (Sh), 148
fixation, 164; to toxicity of metal satellite-borne remote sensing, 131, shortwave electromagnetic
micronutrients, 167 134 radiation, 72
relocation of phosphorus capacity, saxitoxins, 403, 407 siderophores, 169
away from plankton, 411 scales, calcareous, 7 Table 1.1, 25; sidids, 260
replication rate, 178 siliceous, 25 sieve effect, 112
reptiles, 1, 18 scum formation, 402, 404 silica, 28; opaline, 174; skeletal, 174
resilience (elasticity) of Scourfieldiales, 6 Table 1.1 siliceous cell walls, 25, 27, 53
communities to recover from sea bass (Morone), 428 silicification of diatoms, 13, 174
forcing, 304, 380, 383 sea birds, 428 silicon, 28; algal content of, 33
resistance of communities to sea combs (ctenophores), 263 Table 1.6, 173; cycling by diatoms,
forcing, 304, 380 sea gooseberries (ctenophores), 253 174; deposition, in diatoms, 182,
resource-based competition, 197, Table 6.1, 263 245; external SRSi concentrations
371 sea lamprey (Petromyzon marinus), half-saturating diatom growth,
resource segregation, 205 290 197; relative to carbon, 174;
respiration; basal rate, 108, 115, sea otters, 427 requirements of diatoms, 173,
132, 190; measurement of, 102; sea sawdust, 81, 166 197; silicon limitation of diatoms,
‘dark’ and ‘light’, 102, 189; as a sea urchin barrens, 426 196; silicon requirements of
balance to excess photosynthesis, seals, 427, 428 Synurophyceae, 173; solution from
139, 239, 388 seasonality in phytoplankton sinking of diatoms, 174; sources,
resting stages, 214, 249, 250 performances, 231, 232, 234 27, 173; uptake by diatoms, 173
restoration of lakes by seaweeds, 12 sinking behaviour, of
phosphorus-load reduction, 409; Secchi disk, 116, 118 phytoplankton, 39, 49, 50, 52,
by sediment removal,, 413; using secondary sewage treatment, 399 243; of colonial aggregates, 63; of
clay minerals, 413; sensitivity of sediment deposition, 249 cylindrical shapes, 50, 62; of
systems to treatment, 413 sediment diagenesis, 250 oblate spheroids, 50; of prolate
resuspension of sedimented sediment focusing, 250 spheroids, 50; in relation to
material, 79, 250 sediment semi-fluid surface layer, mixed-layer depth, 70
retention time (hydraulic), 240, 241 249 relevance to nutrient uptake,
Reynolds number, of water flow, 45, sediment traps, 220, 246 147; of sculpted models of algae,
50, 243; of particle motion, 49, 50 sedimentary flux, 284, 388, 390, 50, 51; spherical shapes, 50;
rhizopods, 252 Table 6.1, 292 430; of phytoplankton, 247, 249, of stellate colonies, 63
rhodophytes, 11, 12 318
534 GENERAL INDEX
sinking behaviour (cont.); of storage products of phytoplankton, trace elements, 167

teardrop shapes, 51; vital 6 Table 1.1, 7 Table 1.1, 25, 26 trait selection, 362
regulation of, 52, 65, 66, 123, 246 Table 1.2 trait-separated functional groups of
sinking-rate determination, 51 stratification of the water column, freshwater phyoplankton, 319,
size-efficiency hypothesis, 278 73, 75, 79, 204 320 Table 7.1, 346, 347 Table 7.6,
size spectra of phytoplankton straw, anti-algal effects of, 405 394
assemblages, 348 structural viscosity, 67 Tribonematales, 6 Table 1.1
slime moulds, 295 structured granules, 25 tripton, definition of, 2
sodium, 28; algal requirement for, succession, in terrestrial plant trishydroxymethyl-aminomethane
172; cyanobacterial requirement communities, 204; in the (tris), 168
for; , 172 plankton, 303, 359, 370, 384 trophic cascades, 263, 288
soil chemistry, 398 succession rate, 366 trophic states, 400; separation
SOIREE oceanic iron fertilisation, successional climax, 304, 359, 365 criteria, 401 Table 8.1
170, 308 sulphate, algal requirement for, 172 tuna (Thunnus), 428
solar energy income, 41; as a sulphur, algal content of, 28, 32, 33 tunicates, 258 Table 6.1, 261,
function of latitude, 41 Table 1.6 262
solar constant, 41, 95 sulphur-reducing bacteria, 5, 193 turbellarians, 253 Table 6.1
solar energy income, 108 supportive capacity, 346 turbulence, 42, 46; in confined
solar heat flux, 72 surface avoidance by flagellates, 83 channels, 46; associated with
soluble reactive silicon (SRSi), 173 surface-area-to-volume ratio in moving particles, 49
species composition in marine phytoplankton, 185 turbulent dissipation, 47, 71; rate
plankton, 302; in limnetic surface mixed layer, 47, 204; of, 47, 48, 48 Table 2.2
plankton, 318; in relation to light mixed-layer depth, 48 Table 2.2, turbulent embedding of
deficiency, 345; in relation to 71, 74, 75, 79, 80, 204, 244; mixed phytoplankton, 48, 49
nutrient deficiencies, 345 depth, in relation to light turbulent extent, 69
species richness of phytoplankton penetration, 117, 119, 138, 244; turbulent intensity, 74, 75
assemblages, 354, 364, 366, 367; relative to phytoplankton sinking turbulent kinetic energy, 47
a-, β-, ?-diversity, 380, 394; rates, 244; artificial enhancement turbulent velocity, 39, 45, 46, 48
relevance to community of, 414 Table 2.2, 69; fluctuations of, 45,
functioning, 382; in relation to suspension of planktic organisms, 1, 46
successional maturation, 384 17, 38 turgor pressure, 57, 59
species selection, in oligotrophic Synurales, 7 Table 1.1 tychoplankton, 1, 2
systems, 203; in rivers, 242 synurophycaens, 25
specific heat, of water, 16, 41 ‘Swimming’ velocities in ubiquity of phytoplankton, 353
spirotrichs, 252 Table 6.1, 259 phytoplankton, 49, 68, 83, 205 Ulotrichales, 6 Table 1.1, 12
squid (cephalopods), 256 Table 6.1, ‘Telescoping’ of algae at the surface, ultraplankton, 2, 4
264 402, 404 Ulvales, 12
stability in phytoplankton ultraviolet radiation, protection
communities, 304, 381 tench (Tinca), 423 from, 123
standing crop, 223 terrestrial sources of POM, POC, 390 underwater light fields, 108, 110,
starch, 6 Table 1.1, 7 Table 1.1, 25, Tetrasporales, 6 Table 1.1, 12 111, 119
98, 189 thalliaceans, 258 Table 6.1, 261, 262 upwellings, 308
statoblasts, 257 Table 6.1, 353 thermal bar, 89, 324 urea, as a source of nitrogen
stoichiometric lake metabolism thermal stratification, 73, 75, 290 available to phytoplankton, 163
model, 401 thermocline, 74, 75 Utermöhl counting technique, 179
stoichiometric oxygen demand of thylakoids, 10, 12, 25, 96, 146;
decomposition, 296 assembly, 168 vacuoles, 25
stoichiometry, as an ecological tides, 42 vanadium, 28, 167, 335, 398
driver, 199, 287 tidal mixing, 42, 48 Várzeas (flood-plain lakes), 352
Stokes’ Equation, 49, 51, 243 total iron (TFe), 169 vascular plants, 11
stolothrissids, 389 total phosphorus (TP), 155, 399 velocity adaptation of nutrient
stoneworts, 419 toxic algae, 312, 401, 407 uptake, 157, 194, 202
storage adaptation of nutrient toxic metals, 167; redox sensitivity velocity gradients through water
uptake, 157, 194, 202 of, 167 columns, 47, 80, 83
GENERAL INDEX 535
vertical migration of viscous behaviour of, 44, 49; young-of-the-year (YOY) fish, 289
phytoplankton, 205 specific heat of, 41
viruses, 2, 5, 292, 295 water, expansion coefficient of, 72 zeaxanthin–violoxanthin reaction,
viscosity of water, 39 Table 2.1, 40 water movements, 17, 38, 39, 41, 47, 123
vitamins, 170; microalgal benefits 77 zebra mussel (Dreissenia polymorpha),
from, 170; B12 requirements of water blooms, 56, 81, 401 290, 323
phytoplankton, 170 Wedderburn number, 74, 75, 243 zeta potential, 67
Vollenweider–OECD model, 399, WETSTEM electron microscopy, 67 zinc, 28, 167
400, 408, 410 whales, 263, 428 zoobenthos, 392, 395
volume; of the ice caps, 39; of ‘white water’ events, 13 zoomastigophorans, 251, 252 Table
inland waters, 39; of the sea, wind action, 42; stress applied to 6.1
39 water, 46, 69; velocity, 46 zooplankton, 2, 2, 251; phyletic
Volvocales, 6 Table 1.1, 12, 25; DIC classification of, 252 Table 6.1; to,
requirements of, 130 X-ray microanalysis, 167 258 Table 6.1; anti-predator
vorticellids, 259 xanthophylls, 12, 13, 25, 95;, 115, defences, 269; growth rates of,
123 274; competitive interactions
Water, physical properties of, 16, 39; Xanthophyta, 12, 25; classification among calanoids and cladocerans,
et seq., 39 Table 2.1; molecular of, 6 Table 1.1 286; zooplankton and optimal
structure, 39 foraging by fish, 392, 394
density of, 39 Table 2.1, 40; yellow perch (Perca flavescens), zoospores, 66
viscosity of, 39 Table 2.1, 40; 290 Zygnematales, 6 Table 1.1, 12

Ecology of Phytoplankton

Uploaded by

Copyright:

Available Formats

Ecology of Phytoplankton

Uploaded by

Document Information

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

Ecology of Phytoplankton

Uploaded by

Copyright:

Available Formats

The Ecology of Phytoplankton

Cambridge University Press

© Cambridge University Press 2006

This publication is in copyright. Subject to statutory exception and to the provision of

First published in print format 2006

isbn-13 978-0-511-19094-0 eBook (EBL)

isbn-13 978-0-521-84413-0 hardback

isbn-13 978-0-521-60519-9 paperback

Chapter 2. Entrainment and distribution in the pelagic 38

Chapter 3. Photosynthesis and carbon acquisition in

Chapter 4. Nutrient uptake and assimilation in

4.4 Nitrogen: requirements, sources, uptake and

Chapter 5. Growth and replication of phytoplankton 178

Chapter 6. Mortality and loss processes in phytoplankton 239

Chapter 7. Community assembly in the plankton: pattern,

Chapter 8. Phytoplankton ecology and aquatic ecosystems:

and are not necessarily contested. Thus, ‘plank-

Box 1.1 Some definitions used in the literaure

seston the totality of particulate matter in water; all material not

Some more, now redundant, terms

Box 1.2 The classification of phytoplankton according to

Maximum linear dimension Namea

Table 1.1 Survey of the organisms in the phytoplankton

Table 1.1 (cont.)

Table 1.1 (cont.)

Table 1.1 (cont.)

Table 1.1 (cont.)

of producing populations of common occurrence cells, there is a striking variety of planktic

Figure 1.1 Non-motile

Figure 1.2 Planktic unicellular

whether they occurred among other precursors

Figure 1.3 Coenobial

Figure 1.4 Filamentous

Figure 1.5 Colonial

Table 1.2 (cont.)

Notes to Table 1.2 (cont.)

independent but rather constitute a multicellu-

units, more surface bounds the volume than

s/v entered in Table 1.2B and C show several exam-

planktic cells will be equivalent to between 0.41

Wc (cell dry weight / pg)

Table 1.4 The silicon content of some freshwater planktic diatoms

Species Si (pg cell−1 ) Si nSiO2 Referencesa

dry weight (Ketchum and Redﬁeld, 1949). A

dry weight) between its deep-stratiﬁed and free-

ods of time point to 0.003–0.007 pg µm−3 , or

terms, the weaker is the photon ﬂux, and the

Table 1.7 Some cell measurements of cultured marine phytoplankton

Species shapea MLD v C N Chla

and a statistically predictable pattern of the rela-

The chapter provides an introduction to phy-

Entrainment and distribution in the pelagic

the persistence of the clone that residence is con-

Air Pure water Sea water

Figure 2.1 Plots showing (a) the

slippage of another across it, for the same given

2.2.2 Generating oceanic circulation

of turbulence have gradually given way to a

10 Figure 2.6 The onset of

numbers between R e = 500 and

te /t = log e 0.05/log 0.368 = 3.0

NP ≈ P max × z(0.5I k) (3.8)