Weeds and Weed Management on Arable Land

An Ecological Approach

Weeds and Weed Management onArable Land

An Ecological Approach

Sigurd Hkansson

Department of Ecology and Crop Production ScienceSwedish University of Agricultural Sciences

UppsalaSweden

CABI Publishing

CABI Publishing is a division of CAB International

CABI PublishingCAB InternationalWallingfordOxon OX10 8DEUK

Tel: +44 (0)1491 832111Fax: +44 (0)1491 833508E-mail: cabi@cabi.orgWebsite: www.cabi-publishing.org

CABI Publishing44 Brattle Street

4th FloorCambridge, MA 02138

USA

Tel: +1 617 395 4056Fax: +1 617 354 6875

E-mail: cabi-nao@cabi.org

CAB International 2003. All rights reserved. No part of this publication maybe reproduced in any form or by any means, electronically, mechanically, byphotocopying, recording or otherwise, without the prior permission of thecopyright owners.

A catalogue record for this book is available from the British Library, London, UK.

Library of Congress Cataloging-in-Publication Data

Hakansson, Sigurd.Weeds and weed management on arable land : an ecological approach /

Sigurd Hkansson.p. cm.

Includes bibliographical references (p. ).ISBN 0-85199-651-5 (alk. paper)1. Weeds. 2. Weeds--Ecology. 3. Weeds--Control. 4. Weeds--Sweden.

5. Weeds--Ecology--Sweden. 6. Weeds--Control--Sweden. I. Title.SB611 .H35 2003632.5--dc21 2002010487

ISBN 0 85199 651 5

Typeset by AMA DataSet Ltd, UK.Printed and bound in the UK by Biddles Ltd, Guildford and Kings Lynn.

Contents

Preface xBackground and Focus of the Book xGuide to Central Topics Covered in Chapters 113 xiAcknowledgements xiv

1. Introduction 1Comments on the book 1The concept of weeds 1Some definitions and descriptions 3

2. Classification of Plants Based on Traits of Ecological Significance 4Life Form, Growth Form, Lifespan 4Raunkirs Life-form System 5Classification Using the Term Life Form in a Wide Sense 6

Overview 6Seed banks and germination of weed seeds in the soil in different seasons 6Classification of weeds 7

Annuals 8Biennials, monocarpic perennials (pluriennials) 9Stationary perennials 9Creeping perennials 10Perennials with other or modified regenerative structures 12

Classification Problems with Species Distributed over Wide Geographical Areas 12

3. Annual and Perennial Crops 14

4. Weed Communities Looked Upon as Early Stages in Secondary VegetationSuccession 16

5. Weeds with Diverse Life Forms in Various Types of Crops 20Overview in a North-European Perspective 20

Comments on crops and cropping systems in the Nordic countries 20Relationships of weeds to crops largely depend on times and types of tillage 21Changes in tillage and other measures selectively affecting weeds 27

v

Scandinavia and Finland Field Surveys and General Observations 28Summer annuals 32Winter annuals (facultative) 33Biennials (monocarpic perennials) 34Stationary perennials 35Creeping perennials with stolons or prostrate above-ground stems 35Creeping perennials with rhizomes sensitive to soil cultivation 35Creeping perennials with rhizomes tolerant to soil cultivation 35Creeping perennials with plagiotropic thickened roots 36

Changes in weed flora after the 1940s 36Northern and Southern Europe 39Global Perspectives 44

Life forms of weeds 44C4 plants among arable weeds 49Plants of different families as arable weeds 52Parasitic weeds 54

Which Crops in a Crop Sequence Favour the Population of a Specified Weed? 55

6. Germination, Emergence and Establishment of Crop and Weed Plants 56Introduction 56Seed Dormancy and Germination and Soil Seed Banks 56From Seed Swelling to Plant Establishment 60Experiments Exemplifying Influences on Plant Emergence and Establishment 61

Experimental conditions 61Dry matter losses before plant emergence 61Plant establishment 62

Examples representing cultivated plants 62An example representing annual weeds 64

Plant emergence and early mortality among emerged plants 65Overview of Factors Influencing Plant Emergence and Establishment 67

Seed size and depth location in soil, light intensity and temperature 67Water conditions in the soil 69

Soil Moisture Content Seed Germination Plant Growth 71Effects of the relative time of emergence of neighbouring plants in a cerealstand 71Sowing depth and growth of crop and weed plants 72Sowing depth principles regarding crop and soil in the field 73

Seedbed Preparation and Sowing Considering Relative Times of Crop andWeed Emergence 75

Ordinary spring sowing examples regarding different soils 75Delayed spring sowing 76Autumn sowing 76Effects of rolling 77

Late Emergence of Weeds Consequences in Differently Competitive Crops 78Fertilizer Placement 78

7. Competition in Plant Stands of Short Duration 81The Concept of Competition 81

General 81Impacts of allelochemicals allelopathy 83

Competition and Phenotypic Plasticity 83Intraspecific and Interspecific Competition 85

vi Contents

Relative Competitiveness 85Competition as Influenced by Plant Spacing and Relative Time of Emergence 86Effects of Plant Density 87

General characteristics of productiondensity relationships 87Production in crop stands 90

Yield in different crops 90Plasticity of crop plants and quality of harvested products examples 92

Cropweed interactions 93Effects of the Spatial Distribution of Plants 100

Theory 100Row spacing and plant distribution in the row 101

Effects of the Relative Time of Emergence 106General 106Experiments with cereals and weeds 107

Competition Modifies the Response of Plants to Various Factors Examples 111Response to nitrogen 111Growth of plants from seeds of different sizes placed at various depthsin the soil 113Regrowth after breakage and burial of regenerative structures of perennialweeds with an example of nitrogen influences 114Response to temperature 115Response to infections example, discussion 117

8. Weed Flora and Plant Adaptation to Environment and CompetitiveConditions 119Extremely Complex Causalities 119Changes in the Weed Flora as Related to Increased Fertilizer Use Some Scandinavian Examples 119Competitive Conditions in Different Crops 121Competitive Ability of Weeds 123

The importance of plant traits 123Reasons for variation often hard to identify in the field 124

Phenotypic Adaptation to Environments Exemplified by Weeds in Lawns 125

9. Measurements of Competition and Competitiveness in Plant Stands ofShort Duration 128Introduction: Stands of Short and Long Duration 128Unit Production Ratio (UPR) 129UPR in Measuring the Relative Competitiveness of Plants 129Use of UPR and Percentage (or Relative) UPR Examples 132Selective Effects of Competition within Heterogeneous Components in MixedStands 136Determination of ProductionDensity Relationships in Mixed Stands 137Reference Models 138Reference Models in Evaluating Relationships and Indices Aimed atCharacterizing Relative Competitiveness Examples 139Plant Growth as Influenced by Density and Relative Time of Emergence 144Growth Reduction Rate (RR) 149Changes in UPR and Percentage UPR Over Time Comparing Growth Ratesand Competitive Conditions in Mixed Stands 151The Relative Crowding Coefficient (k) and the Relative Yield Total (RYT)Observed Over Time 153

Contents vii

Critical Periods of Weed Control in a Crop Stand 154Early Predictions of Yield Loss Due to Weeds 155

10. Soil Tillage Effects on Weeds 158Introduction 158

Uncultivated land brought into cultivation 158Selective effects of tillage 159No more tillage than necessary 159

Tillage Effects on Weed Seeds in the Soil Seed Bank 160Tillage Effects on Seedlings 162

Seedlings of annuals 162Seedlings of perennials 163

Tillage Effects on Perennial Plants with Fully Developed Vegetative Structures 164General statements 164

Plants with low tolerance to tillage 165Plants with higher tolerance to tillage 165

Factors affecting the response of growing plants to tillage an overview 166Perennials with rhizomes or plagiotropic roots 167

Undisturbed growth and response to a single soil operation 168Response to repeated tillage 178

A Cropping System Approach to Effects of Tillage 183Periods when the field is free of growing crops 183

Seedbed preparation effect on weed seeds in the soil 183Tillage considering the growth rhythm of Elymus repens 185Tillage between harvest and winter 186Tillage between harvest and autumn sowing 189

Harrowing and hoeing in growing crops 190Reduced Tillage 193

11. Chemical Weed Control as an Element in the Cropping System 197Introductory Comments 197Intended Effects of Herbicides 198

The importance of a uniform herbicide distribution 198Joint effects of herbicides and competition in a crop stand 199A reduced herbicide dose a proportionally less reduced effect 200Herbicide application techniques 203Compounds mixed with herbicides in commercial formulations oradded to the spray liquid 203Systemic and contact herbicides 204Pre-emergence soil application of herbicides 205The persistence of herbicides 206Measurements of herbicide effects 206

Unintended Effects of Herbicides 207General 207Toxicity 208Persistence movements 208Harm of pesticide pollution 209Problems in handling herbicides 210Resistance in response to recurrent use of similar herbicides 210

Herbicide and Tillage Problems in Efforts to Develop Sustainable CroppingSystems 211

viii Contents

12. Special Management Measures 214Various Methods of Mechanical or Physical Weed Control 214Cover Crops and Mulches 215Harvesting Timing and Methods 217

Combine harvesting 217Harvesting of root crops 218

Breeding for Increased Competitive Ability of Crops 218Biological Control 219

13. Important Points for Understanding the Occurrence and RationalManagement of Weeds 222General Views 222Illustrations: Case Studies in Sweden 222

Allium vineale a bulb-forming perennial 223Introductory remarks 223The study and its results 223

Morphology and reproduction (Fig. 66) 224Annual rhythm of growth and development (Fig. 67a) 224Population structure 225Sensitivity to tillage 225Occurrence in autumn-sown cereals 225Growth in years with and without spring tillage (Table 44) 227Occurrence on heavier and lighter soil and in winter rye andwinter wheat 227Interpretation of results 231Effects of herbicides 231Occurrence and management of A. vineale discussed on thebasis of the investigation 232

Sinapis arvensis a summer annual 233Avena fatua a summer annual in the Nordic countries 234Elymus repens and Agrostis gigantea perennial rhizomatous grasses 236Cirsium arvense a perennial dicot with plagiotropic roots 242

References 248

Species Index 269

Subject Index 272

Contents ix

Preface

Background and Focus of the Book

This book may in the first instance be of interest to teachers, postgraduate students andresearchers in weed science, but will, I hope, also prove useful to undergraduate studentswho are specializing in this and related subjects. Discussions and illustrations in the bookalso relate to issues that are of interest in other fields of agriculture, horticulture and plantecology. There are also concepts in the book that I hope will attract other categories ofreaders interested in biology and ecology.

As a researcher and teacher in weed science and crop production ecology for almost40 years, I have had the opportunity to follow developments in these subjects during a periodof great change. In weed research, the interest has gradually shifted from a strong focus onchemical weed control in the 1950s and 1960s to an increasing accent on integrated weedmanagement during the 1980s and 1990s.

Investigations with a broader biological and ecological approach to weed managementreceived proportionally little financial support in most countries from the 1950s to the late1970s. This also applies to Sweden, although such investigations were awarded relativelymore money in Sweden than in many other countries.

In 1954, I had the opportunity to start research on wild garlic (Allium vineale L.)occurring as a weed on arable land in South-eastern Sweden. In discreet opposition todemands from above for experiments focusing on chemical control, I was able to give myinvestigation a broader ecological approach. My youthful ambition was to present a casestudy illustrating the minimum knowledge needed for understanding the possibilities anddifficulties of a plant growing and persisting as a weed in different cropping systems whichis, at the same time, the minimum knowledge needed for rational combinations of direct andindirect control of the weed.

When I became a university lecturer in the 1960s, the period of early, almost unreservedenthusiasm over modern herbicides had not yet passed. A discussion on weed managementwas largely synonymous with a discussion on chemical weed control. Like the bodiesfunding research, students were influenced by the prevailing circumstances, most of themexpecting weed science courses to be focused on chemical weed control.

In spite of this, I persisted with presenting weed science courses with a wider, moreecological approach. The courses were on the whole accepted and appreciated by themajority of my students, although protests also arose against wasting time on economicallyunimportant issues. However, general opinions gradually changed in favour of broader

x

ecological approaches to weed management. The proportions between different subjectareas in my courses may have varied more or less from year to year, but the overall structureof the courses has remained unchanged over the years. This structure is reflected in thepresent book.

The strong focus on chemical control in weed research and weed management practicesfor about three decades resulted in an overuse of herbicides, which indeed deservescriticism. However, my personal opinion is that this criticism must not lead to a completerejection of chemical weed control. Pluralism is always fruitful; at the same time as thedevelopment of cropping systems without any use of chemical weed control should besupported, a cautious and restricted use of herbicides will certainly do well in manycropping systems seriously structured to minimize negative environmental impacts.Verdicts on adverse side effects of herbicides should be based on comprehensive herbicidetests, which are now being increasingly performed and reported. Like other culturalmeasures in a cropping system, chemical weed control should be based on broad biologicaland ecological knowledge and considerations.

Many herbicides are positive tools in the control of vegetation since they allow areduction in tillage where soil erosion is a problem. This is particularly important in warmareas with high precipitation, on sloping ground and in dry, windy areas, but intensivetillage and passes with heavy machinery have negative impacts everywhere. Soil compac-tion is one of the persistent problems in modern agriculture and intensified soil tillage cannever be a satisfactory means of replacing chemical weed control. Minimum and zero tillageare being practised or attempted in many agricultural areas for the purpose of minimizingsoil tillage disadvantages.

A one-sided focus on the short-term effects of weed control measures, either chemicalor mechanical, should thus be avoided. In harmony with intentions that are generallyrecognized today, different control and cultural measures should be combined into systemsof integrated weed management. I hope that the facts and viewpoints in my book willcontribute to a balanced development of such systems.

Guide to Central Topics Covered in Chapters 113

Chapter 1

All plants occurring in an arable field in a given situation without the growers intentionare termed weeds in that situation; thus also volunteer crop plants.

Chapters 2 and 3

Classification of plant species into life forms on the basis of ecologically significant traits isdiscussed. For species occurring as arable weeds, the classification here considers traitswith primary influence on the appearance of the species in crop stands differing as totimes and means of establishment and harvesting, and with various geometry, annualdevelopmental rhythm and duration (1 or more years). The importance of dormancyof weed seeds, resulting in soil seed banks, is discussed. Seasonal variation in thegermination readiness of seeds in the weed seed banks is illustrated (e.g. Fig. 1) andits influence on the appearance of weeds in different situations is dealt with. Perennialweeds are classified according to properties of vegetative structures determining theirestablishment and persistence as arable weeds.

Preface xi

Chapters 4 and 5

Arable weeds (non-parasitic) can be looked upon as plants in early stages of secondaryvegetation succession (Tables 14). Relationships between plants with different life formsand their ability to establish as weeds in different crops (considering associated culturalmeasures) are therefore found to present essentially similar patterns in Scandinavian,European and global perspectives (Tables 315).

The occurrence of plants with C4 photosynthesis as arable weeds is illustrated(Tables 12, 14 and 15) and differences are discussed. Particular attention is devoted tothe geographical distribution of rhizomatous C4 species. The distribution is discussedcomparing the occurrence of the C3 species Elymus repens.

A weed with better possibilities to establish, grow and reproduce in a given crop than inthe others in a crop sequence when its growing plants are not subjected to direct control mayoften be more strongly restricted in that crop than in the others when its growing plants areeffectively killed by direct control and its reproduction thus prevented. This may apply evenif plants of the weed are effectively controlled when emerged in the other crops. Thoughimportant and easily explained, this is too often overlooked in general discussions on thegrowth and reproduction of weeds in crop sequences.

Chapter 6

The early growth of plants from seeds is illustrated, considering influences of soilconditions, seedbed preparation and means of sowing. Both crops and weeds are regarded(Tables 1821; Figs 312). The conditions for their germination (or bud activation) andtheir emergence and further early growth largely determine their competitive position inthe future growth of the cropweed stand (Tables 2224).

Chapters 7 and 8

The concept of plant competition and factors affecting the competitive outcome in plantstands of short duration are analysed regarding both cropweed stands and pure cropstands. Phenotypic plasticity enables plants to adapt to different environments, includingcompetitive conditions. Differing plasticity is one of the reasons for differences in therelative competitiveness of plants. Variation in properties (quality) of harvested cropproducts is, to a great extent, caused by the plastic responses of the crop plants todifferences in their environment, including their competitive situation.

Competition in a plant stand always modifies, sometimes very drastically, the responseof plants to external factors. This also means that the relative competitive abilities ofgenetically identical plants may be strongly affected by changes in the external environment.This applies to their competitive ability in terms of both effects on other plants andresponses to competition from them (e.g. Tables 31, 32 and 36; Fig. 34). The ranking orderof crops or weeds as to their relative competitiveness may thus differ strongly with theenvironment.

Weakening of plants caused by mechanical or chemical means, diseases, etc. leads toan increased suppressive effect of competition from surrounding plants, when these plantsare not weakened or are less weakened (Tables 26, 27, 34, 35 and 37). The importance ofinteractive effects of weed control measures and cropweed competition is emphasized hereand further stressed in Chapters 10 and 11.

xii Preface

Chapter 9

Methods, difficulties and pitfalls in measuring results of competition in plant stands ofshort duration are dealt with. Reference models are described. They have been derivedboth to deduce reasonable characters of various relationships between production anddensity of different plants in mixed stands, and to test the usefulness of various ratios orindices for characterizing the relative competitiveness of plants and to specify restrictionsin their usefulness. Thus, for instance, UPR, which is a central ratio in my own work,has been evaluated in this way. Reference models are based on production-densityequations of pure stands of identical plants. The pure stands are then assumed to representmixtures of different plant categories (A, B, etc.) uniformly distributed in the stands atvarying densities (see Eqns 46, p. 138).

Chapter 10

Soil tillage is studied considering both indirect effects on weed populations and directeffects on growing weed plants. Life history and annual developmental rhythm largelydetermine the response of the weeds to different kinds of tillage at different times of theyear. Principles are discussed and illustrated on the basis of experiments. Effects of tillagein interaction or combination with effects of other measures and situations in a croppingsystem are dealt with. Under the motto no more tillage than necessary, various forms ofreduced tillage are reviewed.

Chapter 11

Chemical weed control measures are discussed as elements in cropping systems, con-sidering intended and unintended effects. By weighing desired and adverse effects, thejustification for chemical weed control is debated. Considering both positive and negativeeffects, comparisons with tillage are of great interest.

Chapter 12

A number of management methods, applied or hypothetical, are surveyed regardingexperienced or possible effects. These methods include mechanical or physical measures,the use of cover crops and mulches, timing and methods of harvesting, breeding forincreased competitiveness of crops and biological weed control.

Chapter 13

Issues requiring knowledge for the guidance of rational integrated weed management in acropping system are listed.

The importance of these issues is illustrated in case studies on the appearance of anumber of plants with different traits as arable weeds in Sweden (e.g. Tables 4244;Figs 6671). Differences in the occurrence of these plants in diverse agricultural situationsare described and changes over periods of years are discussed in relation to changes in thecropping systems. Difficulties in interpreting causalities expose gaps in current knowledge.

Preface xiii

Acknowledgements

This book would not have been written without positive encouragement and support frommy friends at the Department of Ecology and Crop Production Science in Uppsala. ThisDepartment, one of the major departments of the Swedish University of AgriculturalSciences, has changed names and fields of responsibility several times during the past fivedecades, but weed science and related crop production ecology have always been centralfields. My successor, Professor Hkan Fogelfors, has always encouraged me and stronglysupported my writing this book. I am sincerely indebted to him and to his co-workers forproviding me with excellent opportunities to continue working at the Department, not onlyaccording to the traditional rights of a professor emeritus but also as a member of the team.

There are a number of people to whom I am indebted and whom I would like to thankhere. First I want to mention Karl-Gustav Ursberg, who has been my skilled technicalassistant for more than 30 years and who has more recently given me invaluable help withmany computer problems. He has, among other things, scanned all my graphs and giventhem a uniform style. I am also very grateful to all those at the University who have readdifferent chapters in my manuscript and given me valuable criticism. I mention themin alphabetical order: Dr Lars Andersson, Dr Tommy Arvidsson, Dr Bengt Bodin,Dr Ullalena Bostrm, Professor Sten Ebbersten, Professor Inge Hkansson, Dr Erik Hallgren,Dr Margareta Hansson and Dr Lars Ohlander.

From colleagues in neighbouring countries, I have had many valuable viewpoints andstrong encouragement and support in writing my book. I express my sincere gratitude to allof them: Dr Angelija Buciene and Dr Virginijus Feiza in Lithuania, Professor Haldor Fykse inNorway, Dr Jukka Salonen in Finland and Professor Jens C. Streibig in Denmark.

As Swedish is my mother tongue, my text would have had too much of a Swedishflavour without competent revision. I am very indebted to Dr Mary McAfee, Wiltshire,England, for her many improvements of my language.

The Swedish Research Council for Environment, Agricultural Sciences and SpatialPlanning has financially supported the publishing of this book, which I gratefully acknowl-edge. I also acknowledge with sincere gratitude direct and indirect financial support fromthe Department of Ecology and Crop Production Science and the Royal Swedish Academy ofAgriculture and Forestry.

Last but not least, I want to thank my wife Kerstin for having supported my writingby her tolerance and constant encouragement. I also want to thank my daughters and theirfamilies for stimulating me by showing keen interest in the progress of the project. I dedicatethis book to the three generations of my family.

Sigurd HkanssonUppsala, October, 2002

xiv Preface

1Introduction

Comments on the book

This book deals with the appearanceand management of plants with differenttraits occurring as weeds on arable land.It describes and discusses matters andrelationships that are important as a basisfor understanding the varying occurrence ofweeds in different crops and cropping sys-tems and, at the same time, understandingthe response of different weeds to specifiedmanagement measures. This understandingis particularly crucial in planning systemsof weed management measures over longperiods of time; in other words, in planningintegrated weed management.

Principles are stressed before details.Although the matters discussed are oftenillustrated by experimental results repre-senting conditions in temperate humidareas, the illustrations exemplify principlesof general interest. Some illustrationsdirectly treat global questions (e.g. Tables 14and 15). Most of the illustrations are basedon previously published material, althoughnot always material published in English.Some previously unpublished illustrations,based on experiments devised for theauthors university lectures, are also pre-sented. The need for increased knowledgeand understanding of certain relationshipsis stressed and experimental and measure-ment methods are discussed in many cases.

The book focuses in particular onissues that have been emphasized in theauthors lecturing and experimental work.

New approaches to certain issues may makethe book a useful complement to otherbooks on weeds and weed managements,such as those by King (1966), Altieri andLiebman (1987), Radosevich et al. (1996)and Liebman et al. (2001). In addition, tablesand diagrams in this book are largely basedon material representing areas that arevery seldom discussed in international text-books. Principles emphasized in researchfrom these areas underlie many approachesto weed management problems in tropicalagriculture (e.g. Alstrm, 1990; fors, 1994).

Any procedure in a cropping systemhas some effect on weed populations, eitherdirectly on growing weed plants or indi-rectly through influences on weed seeds orother propagules in the soil, etc. Definitivedistinctions can therefore not be madebetween weed management procedures andother cropping procedures. This should, ofcourse, go without saying in discussions onintegrated weed management, when long-term effects of procedures and conditionsin a cropping system are to be considered.When, for instance, seedbed preparationand sowing are discussed in the following,effects on crops and weeds are evaluated atthe same time.

The concept of weeds

The term weed is given many meaningsin the literature, although it usually refersto plants occurring in situations where

CAB International 2003. Weeds and Weed Management on Arable Land:an Ecological Approach (S. Hkansson) 1

they are unwanted. The word is usuallyunderstood as a term limited to herbaceousplants, but it sometimes also includesshrubs and trees. According to The ShorterOxford English Dictionary, the wordusually represents a herbaceous plantnot valued for use or beauty, growing wildor rank, and regarded as cumbering theground or hindering the growth of superiorvegetation. Outside the scope of thisbook, weed is a word frequently usedfiguratively, characterizing creatures ofvarious kinds regarded as unprofitable,sickly, troublesome or noxious.

As this book focuses on arable land,weeds are discussed within the definitionof plants occurring on arable land withoutthe growers intention. The term weedtherefore includes volunteer plants of culti-vated species established in arable fields insituations where they are not intended.

Even when restricted to plants onarable land, numerous definitions of weedare found in the literature. However, theylargely fit into the descriptions above. (Forexamples of definitions, see Rademacher,1948; Salisbury, 1961; Holzner, 1982;Radosevich et al., 1996.)

Weeds on arable land are frequentlycalled arable weeds. This brief and conve-nient designation has been widely used inthe literature and is also used here. Arableweeds may be synonymously called agrestalweeds (Holzner, 1982).

A plant species, as such, is not a weed inthe meaning given above. Weeds are thoseplants of the species that occur in situationswhere they are unwanted. This should bekept in mind when weed plants representingdifferent species are named weed speciesfor the sake of brevity, in this book as inother literature. Characteristic features andvarious origins of arable weeds are welldescribed by Baker (1965).

Within many species, certain biotypeshave become cultivated plants whereasother biotypes occur as weeds. An exampleis Daucus carota. Even genetically rathersimilar plants within one and the samespecies occur as troublesome weeds under

certain conditions with respect to climate,soil and/or stages of agricultural develop-ment, but are, or have been, grown as culti-vated plants under other conditions. Avenastrigosa is an example.

Volunteer plants of cultivated speciesmay originate from seeds or vegetative plantparts (e.g. potato tubers) lost from previouscrops or sown with contaminated seed, etc.Plants of this kind may cause great harm,particularly when they resemble the presentcrop plants to the extent that they canneither be controlled nor separated fromthe harvested product. When they representa previously grown cultivar of the samespecies as the present crop, volunteers withunwanted properties may lower the qualityof the harvested product.

Arable land includes agricultural andhorticultural fields where tillage is a regularmeasure. The present book is structured pri-marily to consider agricultural conditions,but the fundamentals dealt with here arenaturally valid for parallel conditions inhorticulture.

Although weeds are, on the whole,understood to be unwanted plants, all plantsother than the intended crop plants areusually termed weeds. This even appliesto plants causing little harm. It should beemphasized that many plants termed weedscan in fact be favourable from variousaspects when they occur at low densities.

Extensive experience indicates thatonly a minority of the plant species inhabit-ing a geographical area as wild plantsare able to become persistent weeds inintensively cultivated agricultural or horti-cultural fields. It will be seen in thefollowing that there is no simple biologicaldefinition characterizing an arable weedwith universal validity. Not unexpectedly,the significance of a specified plant trait isconditional. Traits such as life form, annualrhythm of development and growth,dormancy pattern of seeds and/or otherpropagules, competitive ability of plants,etc., are discussed in the following withrespect to their influences on the ability of aplant to become a weed in a given situation.

2 Chapter 1

Some definitions and descriptions

A few words and concepts are listed inalphabetical order below, and defined andcommented on regarding their use in thisbook.

Control: see Weed control. Cropping system: The concept com-

prises a crop rotation or a periodof years embracing a typical cropsequence with associated measures forthe establishment, growth and harvest-ing of the crops. It thus also includesmeasures for managing weeds andpests. Typical cropping systems inScandinavia and Finland are discussedon pp. 2021.

Fallow: The word stands for an arablefield temporarily set aside from plantproduction in order to control weeds,to make the soil richer by growingplants for green manuring, etc.

Latin names of plants: Unlessotherwise stated, names are usedaccording to Flora Europaea (Tutinet al., 19641980).

Ley: Particularly in Scandinavian liter-ature, the word is used in the meaningof an arable field with a (fodder) cropconsisting of one or more grasses,mostly mixed with some leguminousplant, largely red clover (seldom cloveror lucerne solely). From the 19thcentury on, perennial (usually grassclover) leys with a duration of 23(15) years have been regularly alter-nated with annual crops, mainly cere-als, sometimes also other crops and/ortilled fallows, etc., in more or less fixedrotations. The leys are usually under-sown in cereals. From the followingyear, the first ley year, it is cut forfodder. In recent decades, however,an increasing number of farms have

discontinued livestock farming andgrowing of leys for fodder.

Management: See under Weedmanagement.

Names of plants: See under Latinnames of plants.

Rhizomes: Underground shoot (stem)branches, growing plagiotropically tovarious lengths until they change toorthotropic growth, forming aerialshoots.

Scientific names of plants: See underLatin names of plants.

Seed: The word is used here in a broadsense, comprising an entire dispersalunit containing a seed. It does notonly mean a seed in its strictlyorganographic meaning.

Seed bank: Here understood as thesoil seed bank, representing seeds sur-viving in the soil for longer or shorterperiods of years due to dormancy.

Stolons: Shoot (stem) branches creep-ing on the soil surface, rooting fromtheir nodes.

Weed: see under the heading Theconcept of weeds p. 1.

Weed control: Activities and modifica-tions of measures or conditions in thecropping system intended to reduceweed populations.

Weed management: A general term foractivities, procedures and modifica-tions of conditions in the croppingsystem that are intended to influenceweed populations. The term thusincludes weed control. The wordmanagement stresses the intentionof regulating weed populations toappropriate levels, considering bothshort-term economic and long-termecological aspects. Weed managementmay thus also include measures withthe aim of preserving weed populationsat some low or moderate level.

Introduction 3

2Classification of Plants Based on Traits ofEcological Significance

Plants have been classified in many waysand with various aims. We have the taxo-nomic groupings with taxa on differentlevels: families, genera, species, subspeciesand varieties. However, even species thatare closely related in a taxonomic classifi-cation may differ greatly in characteristicsof ecological importance. A woody plant, abush or a tree, may be found in the samegenus as a herb, which may, in turn, be ashort-lived annual or a long-lived perennialplant. Plants diverging in those ways differstrongly in respect of their ability to estab-lish and build up persistent populations indifferent environments. Plants have there-fore also been classified with regard totraits of ecological significance, such as lifeform, growth form and lifespan (for anoverview of definitions and classificationproposals, see Krumbiegel, 1998). Thefollowing discussions are restricted tovascular plants.

Life Form, Growth Form, Lifespan

The term life form and its well-known def-inition by Raunkir (1934) are discussedbelow as an introduction to the classifica-tion of arable weeds into life forms such asthis term is understood in this book.

Growth form is a term frequently usedfor characterizing genetically fixed morpho-logical structures of vegetative plant parts.The term considers structures in a wider

sense than life form according to the above.It has reference to the structure recognizedas the basis behind modifications into dif-ferent growth types, which describe typesof morphological adaptations of plant indi-viduals to the environment, enabled by thephenotypic plasticity of the individuals.Krumbiegel (1998) has discerned 20 growthforms among annual plants. Perennialweeds have been grouped by Korsmo(e.g. 1930) on the basis of the traits of theirperennating vegetative structures, withoutdistinguishing them in terms of life formand growth form as these terms are definedabove.

The lifespan of plants, from their earlygrowth through germination or sprouting ofindividuals, may be defined here as thelength of time of their survival by vegetativestructures adapted to enduring winters ordry seasons and their subsequent growth.For perennial plants, the lifespan of a givengenet (individual or clone: e.g. Harper, 1977)may be related either to separate individualsor to clones. The lifespan of a clone may bevery long even in cases when the life lengthof individuals or individual perennatingstructures is short, often as short as about1 year.

Different growth types of the samegenotype developing under different envi-ronmental conditions may be easily con-fused with genetically determined growthforms. Some weed species appear withmorphologically dissimilar plants. In manycases, we have an insufficient knowledge of

CAB International 2003. Weeds and Weed Management on Arable Land:4 an Ecological Approach (S. Hkansson)

what is due to genetic variation (growthforms) or to modifications in response to theenvironment (growth types). It is also oftendifficult or inconvenient to distinguishthe terms life form and growth form asdefined above. Even a growth type, repre-senting a common adaptation to environ-mental conditions in a larger area, is difficultto distinguish from growth form withoutthorough studies of genetic conditions. Theterm life form is therefore often used in awide sense, comprising growth form. Ofnecessity, life type frequently must also beconsidered a life form. Thus, life form isused in that wide sense in this book.

However, the classification according toRaunkirs (1934) life-form system will firstbe presented briefly, because it is often usedor referred to in literature on weeds.

Raunkirs Life-form System

Vascular plants are classified by Raunkir(1934) into life forms with regard to theirways of surviving unfavourable seasons, i.e.winters in temperate climates and dry sea-sons in warmer areas. The classification isbased on the position of surviving meri-stematic tissues in vegetative buds, includ-ing shoot apices, or in seeds. The meristemsin these structures are the initiators ofnew plant units or individuals growingin the subsequent vegetation period.The following main groups of plants aredistinguished.

Phanerophytes. Vegetative buds are sit-uated on stem structures reaching highlevels above ground; trees and bushesand, in warm climates, sometimes alsotall plants with non-woody stems.

Chamaephytes. Vegetative buds arepositioned above, but near, the groundsurface; plants with woody or non-woody stems.

Hemicryptophytes. Vegetative buds aresituated at the soil surface, fromslightly above to slightly below; plantswith taproots, with above-ground or

shallow below-ground runners, withsurviving stem bases or shallowrootstocks, etc.

Cryptophytes. Vegetative buds aresituated on underground structures.Geophytes and helophytes are dis-tinguished as subgroups. Geophytesare terrestrial plants with more deep-growing rhizomes or creeping roots,with stem or root tubers, bulbs, etc.Helophytes are plants in swamps, atshores, etc., with their perennatingbuds situated in waterlogged soil or inwater.

Therophytes. The only meristemsnormally surviving from one growingseason to the following are the seedembryos; annual plants, whose vegeta-tive parts do not survive adverse peri-ods (winters or dry periods) betweentwo growing seasons.

Comments regarding parallel or alter-native survival through vegetative andgenerative units. Most phanerogamsthat survive adverse periods betweengrowing seasons by vegetative organscan do this also by seeds. Individualsof facultative winter annuals (seebelow), which, after germination in latesummer or autumn, survive the winteras young plants, appear as hemicryp-tophytes, whereas those developedfrom seeds in the spring perform astherophytes.

The plant categories distinguished inRaunkirs system naturally differ in theirability to establish and persist under differ-ent abiotic (edaphic, climatic) and biotic(e.g. competitive) conditions (e.g. Crawley,1986). They therefore also have differentabilities to appear in different successionalstages of plant communities (see below).

Raunkirs classification system can beused for comparing the life-form composi-tion of weeds in various crops and/oragroecosystems. For this purpose, however,a classification where the term life form isused in a wider sense seems better suitedand more convenient and is therefore usedin this book, as described below.

Classification by Traits of Ecological Significance 5

Classification Using the Term Life Formin a Wide Sense

As stated above, it is frequently difficult,or inconvenient, to distinguish between theterms life form and growth form. Lifeform has therefore sometimes been used forboth of them (cf. Krumbiegel, 1998). Manyauthors (including Hkansson, 1995a,b,c)have used the term in a wider meaning,including both growth form and lifespan.Consequently, annuals, biennials, peren-nials, etc. are life forms in the meaning inwhich life form is used here.

Overview

Before more details of the classificationof arable weeds are presented, terms fora superordinate grouping level are definedbelow. Synonyms to the terms chosen asfirst-hand terms are added in parentheses.

1. Monocarpic plants (hapaxanthicplants, semelparous plants). Plant individu-als normally die entirely following the for-mation of generative reproduction organs,seeds in phanerogams. Seed formation isstarted and completed within one growingseason.

Annuals (monocyclic plants).Phanerogams with seed produc-tion. In temperate areas with realwinters, annuals are traditionallysubdivided into two categories: Summer annuals. Plant spe-

cies with individuals that nor-mally complete their growthand development, includingseed production, within onegrowing season.

Winter annuals. Plant specieswith individuals that can sur-vive winter, mostly in earlystages of development, aftergermination in late summeror autumn, and then continuegrowth and set seed in the fol-lowing growing season. Mostannuals of this category are

facultative winter annuals,germinating more or less fre-quently also in the spring, inwhich case the plants have asummer annual performance.

Biennials (bicyclic plants). Plantindividuals reach their generativephase, set seed and die in theirsecond growing season.

Monocarpic perennials (plurien-nials, polycyclic plants). Plantindividuals need more than twogrowing seasons to reach theirgenerative phase.

2. Polycarpic plants (pollacanthic plants,iteroparous plants). Genets (individuals orclones) survive winters or dry seasons bymeans of vegetative structures (perennatingorgans) and can repeat vegetative develop-ment and growth and form organs for genera-tive reproduction (seeds in phanerogams) inmore than 1 year.

Perennials (= polycarpic peren-nials) Stationary perennials. Slight,

or slow, horizontal extensionof individuals or clonesthrough shoots or roots. Theseare also known as simpleperennials.

Creeping perennials. Horizon-tal extension of individualsand clones through plagio-tropic (creeping) shoots orroots. These are also knownas running or wanderingperennials.

Perennials are usually understood tobe herbaceous plants, whereas woody plants(trees and bushes) are considered additionalgroups, although they could be seen asgroups among the perennials. Unless other-wise stressed, discussions in this bookconcern herbaceous vascular plants.

Seed banks and germination of weed seeds inthe soil in different seasons

Long survival of dormant seeds enables aplant population to build up a bank of seeds

6 Chapter 2

of different ages in the soil. Germination ofthese seeds in portions over a period ofyears is a prerequisite for the persistenceof populations of annual wild plants andweeds. Unless seeds are imported from out-side, these populations would otherwisebe eliminated in years when environmentalconditions obstruct plant growth and repro-duction. Some perennials largely rely oncontinuous vegetative survival (by budbanks), but sexual reproduction by seedsis usually of importance, particularly fortheir long-term survival, and, in fact, mostperennial weeds build up seed banks in thesoil. Tussilago farfara is an example of anexception. Its seeds germinate or die in afew weeks following maturation.

There is extensive literature on thedormancy of seeds, the build-up of soilseed banks, the conditions under whichseed dormancy can be broken and theprerequisites of germination (further, seeChapter 6). The seasonal variation in thegermination among seeds in soil seed banksis briefly commented on here.

In temperate areas, these seeds exhibita more or less distinct seasonal variationbetween dormant and non-dormant states,largely determined by the seasonal tempera-ture variation. There is thus a seasonalvariation in the intensity of germinationand seedling emergence. Different plantsperform differently. Figure 1 illustratesvariations typical of summer- and winter-annual weeds, looked upon as entire groups,in South and Central Sweden. Attentionshould be paid to differences exhibitedwhen the soil is undisturbed and disturbedby shallow tillage, respectively. Possibleflushes of germination and seedling emer-gence following heavy rain cannot be seen inFig. 1, as the curves there average differentyears, fields and species.

Classification of weeds

The following classification is largely basedon works by Korsmo (e.g. 1930, 1954) inNorway, though modified in some details(Hkansson, 1975a, 1982, 1983a, 1992).

Only phanerogams occur among themonocarpic plants, whereas vascularcryptogams, such as species of Equisetum,are represented among the polycarpicperennials.

A plant species sometimes appears inmore than one life form. The appearancemay differ within a species, even in a smallgeographical area. However, for agronomicpurposes, it seems reasonable to classifya species within a restricted climatic areaon the basis of its most typical appearancethere. Plant species presented in thischapter as examples of different life forms

Classification by Traits of Ecological Significance 7

Summer annuals

Winter annuals

L.Wi E.Sp L.Sp M.Su L.Su L.Au

Fig. 1. Seasonal variation in seedling emergencefrom weed seeds in arable soil in South and CentralSweden. Average appearance of summer- andwinter-annual weeds common in Sweden, as foundin two investigations (Hkansson, 1983a, 1992).Solid lines: Emergence in a short period (2 or 3weeks) following shallow soil tillage (harrowing orsimulated harrowing) carried out at different pointsin time as the only soil operation in the study year.Broken lines: Corresponding emergence fromsoil not tilled in that year. L.Wi, late winter; E.Sp,early spring; L.Sp, late spring; M.Su, mid-summer;L.Su, late summer; L.Au, late autumn. There weredifferences between species, sites and years, butthe general pattern was similar. A few species,considered summer annuals in Sweden, e.g.Chamomilla suaveolens, Spergula arvensis andUrtica urens, germinated more readily than othersummer annuals after soil tillage in summer andearly autumn. They are excluded here. Possiblesecondary emergence flushes are averaged out inthe curves.

are classified on the basis of their typical,or most common, appearance in North-European countries.

Annuals

The plants normally reproduce solelyby seeds. In temperate climates with coldwinters, summer and winter annuals aredistinguished.

1. Summer annuals. Seeds in their soilseed banks germinate mainly in spring orearly summer (Fig. 1). Plants then emerge inthe early part of the growing season, flowerand set seed in the year of emergence. Whenseeds germinate in the autumn which alsohappens the resulting plants normally diein the winter. Soil tillage stimulates germi-nation, the stimulation being quantitativelymost pronounced in the spring. Later in thegrowing season, the stimulation can alsobe relatively strong, but lower in absoluteterms.

Examples of weeds: Avena fatua,Chenopodium album, Polygonum per-sicaria, Sinapis arvensis.2. Winter annuals (facultative). Typicalof most winter annual weeds in NorthernEurope is the fact that seeds in the seedbanks may germinate in any season undersuitable temperature and moisture con-ditions. Germination is strongly favouredby soil tillage, especially in summer andautumn. In these seasons, extensive germi-nation mainly occurs after tillage (Fig. 1;Hkansson, 1983a, 1992). It may also betriggered by heavy rain (Roberts and Potter,1980). Plants having emerged in the earlypart of the growing season flower and setseed in the same year, like those of summerannual species. Plants from seeds that havegerminated later in the growing season sur-vive winter to a large extent in young stages.In typical cases, they flower and set seedin the following growing season. Annualspecies flowering both in their first and intheir second year of life, depending on thetime of germination, are here called faculta-tive winter annuals. Milberg et al. (2000)characterize plants with a flexible timeof germination as germination generalists.

Certain winter annuals (e.g. Apera spica-venti) germinate predominantly in late sum-mer or autumn, although to a lesser extentalso in spring (Avholm and Wallgren, 1976).No major weeds in Scandinavia are obligatewinter annuals by the definition that allplants develop from seeds germinated in thelater part of the summer or in the autumnand therefore do not set seed until the nextgrowing season. (On biotype diversity, seepp. 1213.)

As tillage strongly triggers seed germ-ination of winter-annual weeds in latesummer and early autumn (Fig. 1), it shouldbe stressed that discussions on the occur-rence of these weeds in autumn-sown cropsapply to situations where seedbed prepara-tion by tillage precedes autumn sowing unless otherwise stated. To what extent theoccurrence of winter-annual weeds will bechanged by exclusion of this tillage is notwell understood. One question is to whatextent such exclusion will be compensatedfor, or even counteracted, by the conse-quences of more seeds remaining in thesuperficial soil, if deep tillage by ploughingis excluded in the cropping system.

The strong effect of tillage on germina-tion in late summer and autumn applies toall winter annuals observed in the Swedishexperiments. The main reasons for thiseffect may vary with species and with theconditions in the soil closely surroundingthe seeds. Seed age and previous environ-mental influences are of great importance.Depending on all these factors, light willbe more or less influential. In any case, asuddenly improved gas exchange causedby tillage will facilitate germination (cf.Hkansson, 1983a). An example of theinfluence of seed age is that newly shed,fresh seeds of Matricaria perforata germi-nate much better in the presence of lightthan in darkness, whereas the response ofolder seeds to light is weaker (Kolk, 1962). Ithas also been often observed that fresh seedsof M. perforata germinate much more fre-quently on the soil surface than after shallowburial by tillage.

Based on certain observations of myown, it may be suspected that the emergenceof winter-annual plants in spring seedbeds

8 Chapter 2

(Fig. 1) to some extent or in some cases isa continuation of germination started in theautumn (see discussion on p. 34).

Koch (1969) grouped annual weedswith regard to their germination seasonalityon the basis of observations in Germany. Hepoints out that there is often a great variationwithin the same species even within a smallarea. None the less, his grouping seems tobe reasonably valid for temperate climatescharacteristic of Central and NorthernEurope and is presented as follows.

Germination mainly in early spring:e.g. Galeopsis speciosa, G. tetrahit,Polygonum aviculare, Bilderdykiaconvolvulus.

Germination mainly in spring: e.g.Chenopodium album, Polygonumlapathifolium, P. persicaria, Urticaurens (often rather late).

Germination mainly in late spring: e.g.Amaranthus retroflexus, Galinsogaparviflora, G. ciliata, Setaria viridis,Solanum nigrum, Sonchus asper, S.oleraceus.

Germination mainly in autumn andspring: e.g. Alopecurus myosuroides,Chamomilla recutita, Fumariaofficinalis, Galium aparine, Matricariaperforata, Viola arvensis.

Germination throughout the year:e.g. Capsella bursa-pastoris, Lamiumamplexicaule, L. purpureum, Myosotisarvensis, Poa annua, Senecio vulgaris,Spergula arvensis, Stellaria media,Thlaspi arvense, Veronica persica.

Germination mainly in autumn:e.g. Apera spica-venti, Centaureacyanus, Lapsana communis, Veronicahederifolia.

Germination mainly in winter: e.g.Ranunculus arvensis.

Biennials, monocarpic perennials(pluriennials)

Like annuals, these plants normally repro-duce only by seed. Although their seedsmainly germinate in spring or early sum-mer, they rarely flower and set seed in theirfirst year, unlike winter annuals that have

emerged in the spring. After an individualhas flowered and set seed, in the secondyear or later, it normally dies entirely. Theplants mostly remain in the vegetativephase for one or more years, strengtheningtheir vegetative structures and accumulat-ing food reserves before entering the genera-tive phase. Plants classified as biennialsusually flower in the second year in theirtypical habitats. Examples are Cirsiumvulgare, Arctium spp. and Daucus carota.However, depending on both genetic diver-sity and environmental conditions, theirvegetative phase may last over 2 years ormore, and they may even flower in the firstyear under conditions extremely favourableto their growth.

Many monocarpic perennials (orpluriennials) usually grow vegetativelyfor several years until they flower. Cirsiumpalustre and Angelica silvestris are exam-ples of species that may need a period of57 years of strengthening their vegetativestructures before flowering (Sjrs, 1971). Inthe genus Agave there are species whoseindividuals build up a very strong leafrosette for 2030 years before they flowerand set seed; after that they die completelywithin a few months.

Stationary perennials

The individuals of a stationary perennialplant extend only slightly or slowly fromthe spot where they originally established.With increasing age and size, their peren-nating parts may become fragmented due toattacks of herbivores or fungi or to death oftissues with age or for other reasons. Genetssuch as grass tufts may therefore graduallybecome more or less partitioned, formingcrowds of separate units. A big grass tus-sock may thus consist of several auto-nomous individuals. If not mechanicallydispersed by soil tillage, these individualscrowd in very dense clones. Their winter-ing buds are usually situated from slightlyabove to slightly below the ground surface.

In most cases, the stationary perennialscan be characterized as hemicryptophytes inRaunkirs life-form system. They comprisespecies with very different morphological

Classification by Traits of Ecological Significance 9

habits. These may be distributed into twomain subgroups depending on whethertheir perennating regenerative organs are:(1) short stems or (2) taproots.

1. Perennating organs are short stems orstem parts, branched or unbranched, fromwhich adventitious roots and aerial shootsdevelop. New orthotropic aerial shootsannually develop from meristems (buds) onthe perennating stems. Adventitious rootsdevelop from these stem parts and mostlyalso from lower parts of the new orthotropicshoots. In many cases, for instance amonggrasses, the primary roots rapidly becomereplaced or complemented by adventitiousroots from the stem bases. The size of theperennating system and the number of aerialshoots usually increase from year to year.Examples:

Tufted grasses and sedges.Perennating stems are the basalparts of aerial shoots, verticalor almost vertical: e.g. Agrostiscapillaris, Arrhenatherum elatius,Deschampsia flexuosa, Holcuslanatus.

Dicotyledonous plants with shortvertical or almost vertical peren-nating stems close to the soil sur-face: e.g. Leontodon autumnalis,Plantago major, Ranunculus acris.

Dicotyledonous plants with ausually oblique or horizontalperennating stem close to the soilsurface, in some species connectedwith persistent primary roots dev-eloping as more or less branchedtaproots: e.g. Alchemilla spp.,Artemisia vulgaris, Centaureajacea, Leucanthemum vulgare,Plantago lanceolata, Seneciojacobaea.

2. Taproots, more or less branched. Inundisturbed plants, new aerial shootsnormally develop from buds situated nearthe soil surface on the upper part of thetaproot. This part is a stem, becomingmore or less vertically split or branchedwith age. Examples of species: Anchusaofficinalis, Bunias orientalis, Centaureascabiosa, Rumex crispus, R. longifolius,

R. obtusifolius, Taraxacum officinale,Symphytum officinale.

The regenerative ability of the variousparts of a fragmented taproot differs betweenspecies. In Taraxacum officinale, forinstance, all parts of the taproot and itsbranches exceeding a few millimetres inthickness are regenerative (Korsmo, 1930;Kvist and Hkansson, 1985). Healy (1953)and Hudson (1955) report that, in Rumexcrispus and R. obtusifolius, only the upper57 cm of the taproot can develop newshoots. Fykse (1986) reports similar observa-tions in these species and in R. longifolius.However, Cavers and Harper (1964) notedplant development from lower parts of thetaproot in Rumex in early spring, otherwisenot. In laboratory experiments with R.crispus, new shoots only developed frombuds on the upper stem part of the taproot,the crown, but in the spring, shoots devel-oping from lower parts of the taproot wereseen in the field, on plants with dead crowns(Kvist and Hkansson, 1985).

Creeping perennials

The plants extend horizontally by means ofcreeping vegetative organs. These are eitherplagiotropic lateral stems or plagiotropicthickened roots. The plagiotropic stems areeither: (1) above-ground prostrate stems,rooting from their nodes, sometimesmorphologically specialized stolons, or (2)underground stems, rhizomes. (3) Creepingroots, originally thin, become regenerativeafter secondary growth in thickness.

By means of the creeping stems or roots,new plants gradually develop at variousdistances from the position of an originalplant. Fragmentation, caused by ageing,fungi, animals, soil cultivation, etc., resultsin clones with increasing numbers ofseparate individuals. Plants may disperseover wide areas by means of cultivationimplements or other agents.

1. Dispersal and regeneration by stolonsor prostrate above-ground stems. Stolonsare plagiotropic shoots developed from budsat lower nodes of orthotropic shoots. Thestolons may branch both plagiotropically

10 Chapter 2

and orthotropically (forming assimilatinggreen shoots). It seems that they do not growout until orthotropic primary shoots havereached a stage when they produce a surplusof photosynthates (S. Hkansson, unpub-lished data). The stems of the stolons oftenmarkedly differ from those of the orthotropicshoots. They have often long distinctinternodes. In many species, on the otherhand, longer or shorter parts of the basalstems of the aerial shoots lie on the groundbut are otherwise morphologically rathersimilar to the vertical stems.

New individuals arise through thedevelopment of roots and shoots from thenodes of the creeping stems. They becomeseparate individuals with the death ofthe joining stolon internodes. Lower partsof these individuals become perennatingorgans, enabling a continued growth andvegetative spread and multiplication. Stoloninternodes usually do not survive thewinter. Examples of North-Europeanspecies with stolons or prostrate stems areAgrostis stolonifera, Glechoma hederacea,Poa trivialis, Potentilla anserina, P. reptans,Prunella vulgaris, Ranunculus repens,Veronica serpyllifolia.2. Dispersal and regeneration by under-ground plagiotropic stems, rhizomes. Thefollowing description of rhizomes appliesmore directly to those with distinct nodes,well separated by comparatively long,slender internodes. These are typical ofrhizomatous grasses such as Elymus repens(Fig. 50) but are also represented in manyother plant families. Principles also apply tocorresponding underground stem structureswith short and stout internodes (accordingto the definition of a rhizome in the above).Rhizomes originally grow out from basalunderground stem bases of aerial shootswhen these have developed a foliagelarge enough for producing a surplus ofphotosynthates (Hkansson, 1967, 1982;Hkansson and Wallgren, 1976). Lateralbuds develop from superficial tissues inthe axils of scale leaves at the nodes ofthe rhizomes. Rhizome branches of variousorders may develop within a season andpenetrate the soil to different depths.

New aerial shoots develop both byformation of orthotropic shoots from nodesnear the soil surface of earlier aerial shoots(tillering) and/or by formation of orthotropicshoots from the rhizomes. In the latter case,the new aerial shoots are mostly a result ofrhizome apices changing from plagiotropicto orthotropic growth. In undisturbed plants,they usually only develop to a minor extentfrom lateral rhizome buds. When these areactivated, they mostly grow plagiotropicallyforming rhizome branches. However, theapices of these sometimes rather soon bendupwards forming new orthotropic shoots.

Not only the plagiotropic rhizome partsare regenerative but also the undergroundvertical stem bases of aerial shoots. Frag-ments of these have proved to develop newplants (Hkansson, 1969a). The proportionof buds developing aerial shoots is enhancedwith an increased degree of fragmentation,until the fragments have become too smallto contain enough food reserves and/or toomany of them lack viable buds.

Rhizomes sensitive to soilcultivation. Rhizome system shallow:

e.g. Achillea millefolium, A.ptarmica, Cerastium arvense,Galium mollugo, Lamiumalbum, Urtica dioica.

Rhizomes tolerant to soilcultivation. Rhizome system shallow;

branches with spool-shapedswellings: e.g. Menthaarvensis, Stachys palustris.

Rhizome system shallow; bran-ches without spool-shapedswellings: e.g. Agrostis gig-antea, Elymus repens, Holcusmollis, Polygonum amphibium.

Rhizome system reachinggreater depths: e.g. Equisetumarvense (lateral tubers fre-quently formed at thenodes), Phragmites australis,Tussilago farfara.

Comments: Rhizome system shallowmeans that rhizomes normally grow withinthe topsoil layer; the great majority of themmostly in the upper 10-cm soil layer at

Classification by Traits of Ecological Significance 11

undisturbed growth. For species withrhizome branches reaching greater depths,the rhizome system may penetrate the soilfrom shallow layers to depths far below thetopsoil layer. The maximum depth dependson species and soil properties. The degreeof tolerance of the rhizome structures tosoil cultivation is far from being strictlycorrelated with the morphological robust-ness (when rhizomes representing differentspecies are compared).3. Dispersal and regeneration by plagio-tropic thickened roots. Perennating thick-ened roots develop among slender roots asa result of secondary growth in thickness.They grow at various inclinations, horizon-tal to vertical. After a period of growth inthickness, new aerial shoots develop frombuds in the thickened roots. The horizontalroots enable an effective dispersal of theplant, even without intervention by soilcultivation implements or other agents.Aerial shoots develop from buds at irregulardistances from each other on the root. Thesebuds are differentiated from tissues belowcortex. In undisturbed plants, only a fewbuds grow out forming aerial shoots,whereas most of them stop growing in earlystages of development owing to correlativeinhibition.

Fragmented thickened roots are regen-erative. The proportion of buds developingaerial shoots is enhanced with an increaseddegree of fragmentation, until the fragmentshave become too small to contain enoughfood reserves. New roots develop both fromthe thickened roots and from the under-ground stem parts of the shoots. Fragmentsof the stem bases of aerial shoots are regener-ative (if the shoots are not very young). Theydevelop shoots and new roots from tissues atnodes (e.g. Korsmo, 1930, 1954; Hkansson,1969d).

Perennating root system shallow:e.g. Rumex acetosella, Sonchusarvensis.

Perennating root system reachinggreater depths: e.g. Cirsiumarvense, Convolvulus arvensis(in combination with rhizomatousstructures).

Perennials with other or modified regenerativestructures

Perennials with regenerative structuresother than those presented above also occuras arable weeds. Allium vineale can bementioned as an example. This species,perennating and multiplying by bulbs,can become a noxious arable weed underspecial conditions, e.g. locally in Sweden(Hkansson, 1963a). Some species havetubers or bulbs that can be regarded as partsof rhizomes. Examples are Cyperusrotundus with tubers, i.e. swollen regenera-tive stem parts. These alternate with thininternodes in rhizome branches and/or theyterminate the branches (see Tables 12 and14). Oxalis latifolia forms bulb-like organsat the tips of its rhizome branches (e.g.Holm et al., 1997). The two latter speciesare important weeds in a global perspective,particularly C. rotundus, but none of themin cold-temperate regions. Species such asMentha arvensis and Stachys palustris formspool-shaped swellings near the tips ofrhizome branches.

Structures with tubers or bulbs asspecial formations on rhizome branchesmay be regarded as variants of rhizomatousstructures; the plants classified here amongthe rhizomatous perennials. Besides themain perennating structures of a plant, thereare often additional structures with a moreor less pronounced regenerative capacity.As understood from the above, undergroundstem parts are thus more or less regenerativebesides the thickened roots in plants such asCirsium arvense and Sonchus arvensis and,particularly, Convolvulus arvensis.

Classification Problems withSpecies Distributed overWide Geographical Areas

As previously stated, the life-form classi-fication of a plant species is often condi-tional. A species may comprise genotypesrepresenting diverse life forms occurringwithin the same or in different areas. It mayalso differ phenotypically in response to the

12 Chapter 2

environment. For example, its ability tosurvive winters as a winter annual or as aperennial differs, not only between climaticareas but also from year to year in the samefield.

The grouping of annual plants intosummer and winter annuals seems to beparticularly conditional. In species classi-fied as winter annuals under Swedish andNorth-European conditions, spring germi-nation is followed by a summer-annualperformance of the established plants,whereas plants established after autumngermination largely survive the winter asyoung plants. Spring and autumn germ-ination may sometimes be represented bydifferent biotypes, both in the same areaand in different geographical regions. Ifa population of weed species includessuch biotypes and these are not easily dist-inguished by morphological characters, itmay be practically reasonable, or inevitable,to consider it a taxonomic unit characterizedas a facultative winter annual.

Assume now that a geographicallywidely-distributed annual species is charac-terized as a (facultative) winter annual innorthern Europe. This characterization maybe irrelevant, or have a different meaning, inwarmer areas. It may be useful throughoutEurope but have somewhat different mean-ings in the North and the South. In southernEurope, the performance of a species as asummer or a (facultative) winter annual willmainly depend on its germination seasonal-

ity and, in a Mediterranean climate, on theresistance of its growing plants to summerdrought. It may only slightly, or not at all,depend on its ability to survive winters,which are moist and mild. A plant being atypical summer annual in a North-Europeanclimate does not germinate until spring inthat climate, though the dormancy of manyseeds in its soil seed bank is broken as earlyas in late autumn after a period of low tem-peratures. There, however, the temperaturesin late autumn and winter are too low forgermination. In southern Europe, even seedsof a plant with (possibly) identical geneticcharacters might germinate in the winter.The temperatures are low enough to breakdormancy, but, at the same time, frequentlyhigh enough to allow germination.

In the hot tropics, a grouping of annualplants into summer and winter annualsis inapplicable. However, there might bedifferences in the germination seasonalityamong the annuals motivating divisions intoother subgroups. Thorough investigationsare needed to clarify this issue (cf. Baskinand Baskin, 1998).

With the classification problems nowbeing discussed still in mind, relationshipsbetween different groups of weeds and theiroccurrence in crops of different types areexamined in Chapter 5. The question here isto what extent relationships valid in widergeographical perspectives can be defined.First, a simple grouping of crops based ontheir duration is presented.

Classification by Traits of Ecological Significance 13

3Annual and Perennial Crops

The arable crops are grouped here consider-ing their lifespan, annual life cycle anddegree of stand closure. A grouping intoannual and perennial crops is illustratedwith special regard to crops in temperateareas. The annual crops are subdividedwith respect to the season in which they aresown or planted. The grouping is indicatedin Tables 3 and 5. At the same time, thesetables present grading of the relativeabundance of various weeds in differentcategories of crop based on broad experi-ences from Sweden. The cropweed rela-tions seen there are analysed in Chapter 5.However, it should be stressed here thatthe grading applies to cropping systemswith ordinary soil tillage, where sowing orplanting is immediately preceded by soilcultivation. In a non-tillage situation withdirect sowing, the conditions may differvery strongly.

In its position as a crop, a species orvariety may differ completely from itsposition in the life-form system presentedabove. Thus, for instance, the potato plant isbasically a perennial, but is a spring-plantedannual crop. Sugar beet is a biennial plant,but is an annual spring-sown crop whengrown for sugar production. Spring-sownvarieties of small-grain cereals, rape andturnip rape are both summer-annual plantsand spring-sown annual crops. Autumn-sown varieties of these species may on thewhole be characterized as biennials. Theyshould not be considered winter annuals ifthey do not normally flower before winter

when sown in the spring. However, thereare differences, between both species andvarieties of winter crops, in their dispositionto flowering as annuals. Low springtemperatures sometimes enhance thisdisposition.

1. Annual crops, e.g. cereals, oilseedcrops, peas, beans, sugar beet, potatoes andvarious vegetables. In accordance with theabove, they are subdivided into spring crops(spring-sown or spring-planted annualcrops) and winter crops (autumn-sownannual crops). The spring crops are heresubdivided into crops with stands that closelate in the growing season, resulting in aweak competitive effect on weeds from earlystages of development, e.g. row crops, suchas potatoes, sugar beet and most vegetables,and crops whose stands close early, suchas small-grain cereals and (largely) oil-seed crops, which become comparativelycompetitive from early stages.2. Perennial crops. In the Nordic coun-tries, the principal perennial crops in a cropsequence on arable land are leys for cutting,mostly leys with grass plus clover or grassonly, more seldom clover in pure stands andrarely with lucerne alone or in mixtures.These leys, the plants of which are mostlyundersown in spring cereals, usually have aduration of 24 years. Pastures for grazingmostly have a much longer duration evenwhere established on arable land. The weedflora in leys alters with the age of theley (Table 4) following principles of vegeta-tion successions on non-tilled land, with

CAB International 2003. Weeds and Weed Management on Arable Land:14 an Ecological Approach (S. Hkansson)

modifications caused by the annual cutting.When surveying the weed occurrence in leys(Tables 3 and 5), young leys (chiefly in themeaning of first-year leys) and older leys aredistinguished in order to briefly characterizetypical weed flora changes with the ageing ofthe ley.

Conditions in perennial crops or planta-tions, such as orchards and vineyards, arediscussed later on (Tables 12, 13). Thesecrops represent much more heterogeneouscrop categories than the grassclover or

grass leys. In extreme cases, the entire fieldsaround trees or bushes are, each year, eithertilled or treated chemically to manageweeds, or covered with grass and/or othervegetation, which may be cut as needed. Theaverage field with such crops could bedescribed as a field with mosaics or stripsresembling areas with annual crops (or blackfallows) and areas with dense stands ofcrops such as leys for cutting. The two typesof area in the mosaics or strips thus favourdifferent life forms of weeds.

Annual and Perennial Crops 15

4Weed Communities Looked Upon as Early Stages inSecondary Vegetation Succession

Vegetation succession is a concept encom-passing those changes in the compositionof plant communities that normally takeplace with the progress of time. Followingthe classic work of Clements (1916), furtherstudies have illustrated and analysedimportant patterns in vegetation changes.Crawley (1986) describes vegetation succes-sions as community dynamics occurringon a time scale of the order of the life-spansof the dominant plants (in contrast to muchslower, evolutionary changes, occurringover hundreds or thousands of generations,or the much more rapid seasonal or annualfluctuations in species abundances).Primary and secondary successions aredistinguished in the literature (forreviews, see Tilman, 1986, 1990; Brownand Southwood, 1987; Miles, 1987;Mortimer, 1987; Bazzaz, 1990; Hansson andFogelfors, 1998).

Primary succession concerns changesin the composition of plant communitiesstarting with the establishment of plantson bare ground exposed after, for instance,glacial recession and volcanic eruption.

Secondary succession refers tochanges in the composition of communitiesof plants repopulating an area of bare soilafter destructive disturbance of a previousvegetation cover. The following solelyconcerns secondary successions.

The first plants repopulating bareground predominantly originate from seedsand vegetative plant parts remaining in thesoil. Mostly, they also come from dispersal

units entering from outside, but usually to aminor extent. Densities and proportions ofthe taxa and life forms of these first plantsthus largely depend on the previousvegetation and its treatments. Plants thatare superior competitors in a short-termperspective become dominant, e.g. plantswith high growth rates and/or originatingfrom propagules with large amounts offood reserves. For reasons discussed below,more short-lived plants occur among thedominants in the first year than in latersuccessional stages (Table 1). Rapidly grow-ing monocarpic plants (mostly annuals)and polycarpic plants (perennials) capableof fast and strong vegetative renewal afterdisturbance are usually represented amongthese dominants. See, for instance, Numata(1982) and Crawley (1986).

The changes in the vegetation on aban-doned arable land are prominent examplesof secondary vegetation successions. Thesechanges, summed up in Table 1, strikinglyillustrate that arable weed communities canbe characterized as early stages in such suc-cessions. An understanding of the dynamicsin these changes means an understandingof relations between weed communities onarable land and many plant communities inthe surrounding landscape.

Secondary vegetation successions, asbriefly characterized in Table 1, largelyresult from plant competition. It is easilyunderstood that late emergence of a plantin relation to neighbouring plants is a com-petitive disadvantage. Thus, it is hard, or

CAB International 2003. Weeds and Weed Management on Arable Land:16 an Ecological Approach (S. Hkansson)

impossible, for new plants to establish inan already closed canopy of plants. Thisparticularly applies to seedlings (Table 2).

Seedlings frequently use up their foodreserves and die in the shade without reach-ing stages of positive net photosynthesis

Weed Communities and Secondary Vegetation Succession 17

TimeVegetation succession(biomass proportions) Comments

Year 1: Firstgrowing seasonafter disturbance

More short-lived plants (largelyannuals on abandoned arableland) among the dominants thanin subsequent years

Weed communities in annual crops correspondto the first stage of succession. Weed controlcan be seen as an obstacle to the earlysuccessional course

Year 2 andonwards

Persistent perennials increasingin proportion to short-lived plants

Comparable to changes in leys (grassland)sown in arable fields

After varyingtime

Woody plants increasinglyestablished, influencing theherbaceous vegetation moreand more

Landscape gradually less open in areas withabandoned arable fields. Afforestation, now orearlier, is a way of governing and speeding upforest establishment. To keep landscape open,spontaneous growth of trees must be activelyprevented

Finally Forests of various types More stable plant communities although thecomposition of species is not constant

aThe vegetation on headlands surrounding arable fields, on verges and ditch-banks, etc., where the soilis not tilled, but cutting sometimes occurs, has many similarities to leys. Although varying greatly, suchvegetation largely represents intermediate stages in secondary succession. Perennial plants usuallydominate the vegetation. Perennial arable weeds often occur, and many of them, particularly creepingplants, can easily invade adjacent arable fields from there. If preventive measures are not taken, woodyvegetation often forms curtains of trees on such ground. Translated from Hkansson (1995d).

Table 1. Secondary vegetation succession: development from the start on bare soil followingdisturbance, e.g. by soil tillage. Vegetation changes typical of the successional course on abandonedarable land in humid climate.a

Ley plants: sown 0, 18 or 45 d.b.s.(= days before sowing of weeds)

Weed plants 53 days after their sowing

No. (m2) Dry weight (g m2)No ley plants sownStands even, not cut

0 d.b.s.18 d.b.s.45 d.b.s.

Stands uneven and cut45 d.b.s. (A)a45 d.b.s. (B)a

878

70763741

182171

337.00

236.009.50.04

0.280.32

aGreenhouse experiment in boxes, 80 80 cm2, at Uppsala 1973. Seeds of the annuals Chenopodiumalbum, Chamomilla recutita, Thlaspi arvense and Stellaria media, and the perennial Rumex crispus weremixed in soil that was spread as thin superficial layers, afterwards watered repeatedly. This was doneeither on previously unsown soil or on soil sown with seeds of ley species (Phleum pratense, Festucapratensis and Trifolium pratense, 8 + 8 + 4 kg ha1 of seed in even or uneven stands). Ley plants sownin even stands were not cut. Uneven stands had either four gaps of 15 15 cm2 (A), or one gap of30 30 cm2 (B) per box. They were cut at a height of 7 cm 27 days after weed sowing. (FromHkansson, 1979.)

Table 2. Number and biomass of above-ground shoots of weed plants established from seeds, eitherin pure stands or mixed with ley plants sown at different times in relation to the weeds.a

(Table 20), or they die due to droughtor mineral deficiency resulting from thewater and nutrient uptake by the olderplants.

In the experiment that lies behind Table2, concerning establishment of weed seed-lings in stands of ley plants (descriptionwith the table), water stress was eliminatedby repeated watering. In spite of this, mostweed plants originating from seeds sown 45days after the ley plants had been sown weredead, and the others were largely dying, onthe 53rd day after their sowing, particularlyin even and uncut ley stands. The unevenley stands were at least as even as good leystands in ordinary field situations. Moreplants per unit area were alive in the gapsthan in the other parts of the ley stand, butthe plant weights did not differ noticeably.Cutting, which was done in the uneven leystands (at a height of 7 cm 27 days after thesowing of weeds), had probably influencedthe weed seedlings more than the arrangedgaps. In case (B) as well as in case (A), thegaps were obviously too small to favour theseedlings appreciably (cf. Fig. 2).

In free vegetation development, peren-nials with long-lived individuals or clonesbecome increasingly dominant because,each year, they can start growth from vigor-ous vegetative units. Annuals then meetwith difficulties since they start growthas seedlings, which are competitively weakrelative to these perennial plants. Perennialswith short-lived genets soon meet withdifficulties too, as they frequently have tore-establish as seedlings.

The biomass in communities of undis-turbed plants in late successional stagesis usually dominated by a few species. Atthe same time, these communities can havea high species diversity. Even annualsmay establish, but mainly in spots withdestroyed or weakened perennials. Estab-lishment of seedlings, representing peren-nials as well as annuals, is favoured, ormaybe only possible, in temporary gaps orpatches with weakened plants. New peren-nial species, although weakly competitive injuvenile stages, can establish in this wayand then increase their proportion in theplant community, if they are sufficiently

competitive in more advanced stages ofplant development.

Grazing in permanent pastures reducesthe competition and dominance of certain

18 Chapter 4

400

200

0

g m1

400

200

0400

200

0400

200

0

ClovergrassAnnual weeds

One rowomitted:Gap 23 cm

Two rowsomitted:Gap 35 cm

Four rowsomitted:Gap 58 cm

Eight rowsomitted:Gap 104 cm

Fig. 2. Annual plants have difficulties in producingvigorous growth in already well-established plantstands. This was illustrated in a field experimentwith a first-year ley stand of red clover and grass,where gaps were arranged. Spontaneously emergingannual weeds (mainly winter annuals) could growvigorously only in gaps exceeding a certain width.The ley plants were sown in NS aligned rowsspaced at 11.5 cm. Gaps were arranged throughomission of ley plant rows. Bars show fresh weightsof aerial plant shoots, the bases of which were situ-ated within strips of a width of 11.5 cm and alength of 1 m; actual or omitted ley plant rows weresituated in the centres of the strips. Weights aremeans from strips with the same position relative togaps, on both sides of these. Weights were deter-mined at the normal time of hay-making, inmid-June. Weeds weighed less than 1 g in stripswith ley plants. In gaps with only one or two rowsomitted, they weighed 1.3 and 2.2 g, respectively.Matricaria perforata and Chamomilla recutitarepresented 7080% of weed biomasses, Capsellabursa-pastoris and Myosotis arvensis 2025%.(From Hkansson and Jendeberg, 1972.)

Weed Communities and Secondary Vegetation Succession 19

species. Moderate grazing therefore mostlyfacilitates diversity.

In humid climates, successions lead tothe formation of forests, providing thatground and temperature conditions enablestands of trees to establish and persist (Table1). Here, arable fields are largely situated onprevious forest land and free successionalchanges in the vegetation of abandonedarable fields sooner or later result in somekind of forest.

However, the establishment of trees isoften a slow process. Seedlings of trees often

have difficulties in establishing amongperennial grasses and herbs, which mayrapidly form dense covers (e.g. Olsson,1987). It therefore often takes a long timeuntil trees establish spontaneously, unlessthese dense covers are broken or weakenedby some disturbance. It is thus often intemporary gaps or in spots with weakenedvegetation that seedlings resulting inthe first trees establish. Trees of certainspecies, e.g. aspen, spread vegetatively bysuckers from creeping roots of olderindividuals.

5Weeds with Diverse Life Forms in Various Types of Crops

Overview in a North-EuropeanPerspective

Comments on crops and cropping systems inthe Nordic countries

The types of rotations or sequences inwhich crops are alternated in the fields varyextremely between agricultural areas andfarms and have changed with time. An ideaof crop sequences in the Nordic countriesmay be obtained from the following acreageproportions of the three main types of cropin the last five decades of the 20th century:annual spring crops 3060%, annualautumn-sown crops 535%, leys for cutting(hay, silage) 2040%.

Among annual crops, cereals havedominated. Spring-sown cereals have beenbarley, oats or wheat in various proportionsand autumn-sown cereals have been wheatand, to different extents, rye and triticale.Oilseed crops have been sown to variousextents. Particularly in Denmark and SouthSweden, sugar beet has represented con-siderable acreages, in certain areas alsopotatoes. Further north, row crops aremainly potatoes in minor acreages.

Leys for cutting are largely composed ofgrasses (mainly timothy, Phleum pratense)and clovers (mainly red clover, Trifoliumpratense) undersown in spring cereals in theyear preceding the first year of ley and aremostly maintained for 2, up to 4, years (butsometimes longer, particularly in northernScandinavia and Finland in the earlier

decades under discussion). The leys areusually broken by ploughing, sometimesafter shallower, sward-fragmenting tillageor, in recent decades, a herbicide treatment.

The acreage of leys in the areas as awhole has decreased since the early 1950s,largely resulting from the loss of leys onfarms giving up stock keeping. Behind thestatistics of mean acreages of leys there arethus stock-keeping farms with 3050% leysin alternation with annual crops, mainlycereals, and farms with solely annual crops.The proportions of different kinds of farmsvary between countries and districts.

The percentage acreage of leys forcutting has generally been higher in Finlandand Norway and lower in Denmark than inSweden, but changes over time have beenconsiderable in all areas. The percentageacreage of autumn-sown cereals has, on thewhole, been lower in Denmark, and muchlower in Finland and Norway, than inSweden.

In the first two or three decades after1950, levels of fertilizer use increased dra-matically. In the last two decades they havestabilized and even decreased