WESTERN CORN ROOTWORM Ecology and Management Prelims00.Qxd 15/11/04 2:16 PM Page Ii Prelims00.Qxd 15/11/04 2:16 PM Page Iii

prelims00.qxd 15/11/04 2:16 PM Page i

WESTERN CORN ROOTWORM Ecology and Management prelims00.qxd 15/11/04 2:16 PM Page ii prelims00.qxd 15/11/04 2:16 PM Page iii

WESTERN CORN ROOTWORM Ecology and Management

Edited by Stefan Vidal

Georg-August University, Göttingen, Germany Ulrich Kuhlmann

CABI Bioscience, Delémont, Switzerland

and C. Richard Edwards

Purdue University, Indiana, USA

CABI Publishing prelims00.qxd 15/11/04 2:16 PM Page iv

CABI Publishing is a division of CAB International

CABI Publishing CABI Publishing CAB International 875 Massachusetts Avenue Wallingford 7th Floor Oxfordshire OX10 8DE Cambridge, MA 02139 UK USA

Tel: +44 (0)1491 832111 Tel: +1 617 395 4056 Fax: +44 (0)1491 833508 Fax: +1 617 354 6875 E-mail: [email protected] E-mail: [email protected] Website: www.cabi-publishing.org

© CAB International 2005. All rights reserved. No part of this publication may be reproduced in any form or by any means, electronically, mechanically, by photocopying, recording or otherwise, without the prior permission of the copyright owners.

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

Library of Congress Cataloging-in-Publication Data Western corn rootworm : ecology and management / edited by Stefan Vidal, Ulrich Kuhlmann, C. Richard Edwards. p. cm Includes index ISBN 0-85199-817-8 (alk. paper) 1. Western corn rootworm--Congresses. I. Vidal, Stefan. II. Kuhlmann, Ulrich, 1964- III. Edwards, C. Richard.

SB945.W53W44 2005 633.1′5976′48--dc22 2004012843

ISBN 0 85199 817 8

Typeset by MRM Graphics Ltd, Winslow, Bucks Printed and bound in the UK by Biddles Ltd, King’s Lynn. prelims00.qxd 15/11/04 2:16 PM Page v

Contents

Contributors ix

Preface xiii

1 Invasive Alien Species – a Threat to Global Biodiversity and Opportunities to Prevent and Manage Them 1 Rüdiger Wittenberg

2 Monitoring of Western Corn Rootworm (Diabrotica virgifera virgifera LeConte) in Europe 1992–2003 29 József Kiss, C. Richard Edwards, Harald K. Berger, Peter Cate, Mirela Cean, Sharon Cheek, Jacques Derron, Husnija Festic, Lorenzo Furlan, Jasminka Igrc-Barc˘ic´, Ivanka Ivanova, Wiebe Lammers, Victor Omelyuta, Gabor Princzinger, Philippe Reynaud, Ivan Sivcev, Peter Sivicek, Gregor Urek, Otmar Vahala

3 A Synopsis of the Nutritional Ecology of Larvae and Adults of Diabrotica virgifera virgifera (LeConte) in the New and Old World – Nouvelle Cuisine for the Invasive Maize Pest Diabrotica virgifera virgifera in Europe? 41 Joachim Moeser and Bruce E. Hibbard

4 Western Corn Rootworm, Cucurbits and Curcurbitacins 67 Douglas W. Tallamy, Bruce E. Hibbard, Thomas L. Clark and Joseph J. Gillespie

5 Natural Mortality Factors Acting on Western Corn Rootworm Populations: a Comparison between the United States and Central Europe 95 Stefan Toepfer and Ulrich Kuhlmann

v prelims00.qxd 15/11/04 2:16 PM Page vi

vi Contents

6 Movement, Dispersal and Behaviour of Western Corn Rootworm Adults in Rotated Maize and Soybean Fields 121 Joseph L. Spencer, Eli Levine, Scott A. Isard and Timothy R. Mabry

7 Within-ﬁeld Spatial Variation of Northern Corn Rootworm Distributions 145 Michael M. Ellsbury, Sharon A. Clay, David E. Clay and Douglas D. Malo

8 Heterogeneous Landscapes and Variable Behaviour: Modelling Rootworm Evolution and Geographic Spread 155 David W. Onstad, Charles A. Guse and Dave W. Crowder

9 Sampling Devices and Decision Rule Development for Western Corn Rootworm (Diabrotica virgifera virgifera LeConte) Adults in Soybean to Predict Subsequent Damage to Maize in Indiana 169 Corey K. Gerber, C. Richard Edwards, Larry W. Bledsoe, John L. Obermeyer, Gyorgy Barna and Ricky E. Foster

10 Western Corn Rootworm (Diabrotica virgifera virgifera LeConte) and the Crop Rotation Systems in Europe 189 József Kiss, Judit Komáromi, Khosbayar Bayar, C. Richard Edwards and Ibolya Hatala-Zsellér

11 Application of the Areawide Concept Using Semiochemical- based Insecticide Baits for Managing the Western Corn Rootworm (Diabrotica virgifera virgifera LeConte) Variant in the Eastern Midwest 221 Corey K. Gerber, C. Richard Edwards, Larry W. Bledsoe, Michael E. Gray, Kevin L. Steffey and Laurence D. Chandler

12 Genetically Enhanced Maize as a Potential Management Option for Corn Rootworm: YieldGard® Rootworm Maize Case Study 239 Dennis P. Ward, Todd A. DeGooyer, Ty T. Vaughn, Graham P. Head, Michael J. McKee, James D. Astwood and Jay C. Pershing

13 Is Classical Biological Control against Western Corn Rootworm in Europe a Potential Sustainable Management Strategy? 263 Ulrich Kuhlmann, Stefan Toepfer and Feng Zhang prelims00.qxd 15/11/04 2:16 PM Page vii

Contents vii

14 Maize Growing, Maize High-risk Areas and Potential Yield Losses due to Western Corn Rootworm (Diabrotica virgifera virgifera LeConte) Damage in Selected European Countries 285 Peter Baufeld and Siegfried Enzian

Index 303 prelims00.qxd 15/11/04 2:16 PM Page viii prelims00.qxd 15/11/04 2:16 PM Page ix

Contributors

James D. Astwood, Monsanto Company, 800 N. Lindbergh Blvd, St Louis, MO 63167, USA Gyorgy Barna, Syngenta Seeds Kft., 1123 Budapest, Alkotas u. 41, Hungary Peter Baufeld, Federal Biological Research Centre for Agriculture and Forestry, Department for National and International Plant Health, Stahnsdorfer Damm 81, 14532 Kleinmachnow, Germany Khosbayar Bayar, Department of Plant Protection, Szent István University, 2100 Gödöllö, Hungary Harald K. Berger, Institute of Plant Health, Austrian Agency for Health and Food Safety, 1126 Vienna, Austria Larry W. Bledsoe, Department of Entomology, Purdue University, W. Lafayette, IN 47907-2089, USA Peter Cate, Institute of Plant Health, Austrian Agency for Health and Food Safety, 1126 Vienna, Austria Mirela Cean, Department of Entomology, Central Laboratory for Phytosanitary Quarantine, 077190 Bucharest, Romania Laurence D. Chandler, RRVARC, USDA-ARS, Fargo, ND 58105-5677, USA Sharon Cheek, Central Science Laboratory, Department for Food, Environment and Rural Affairs, Sand Hutton, York YO4 1LZ, UK Thomas L. Clark, USDA-ARS, 205 Curtis Hall, Department of Entomology, University of Missouri, Columbia, MO 65211-7020, USA David E. Clay, Plant Science Department, SD State University, Brookings, SD 57007, USA Sharon A. Clay, Plant Science Department, SD State University, Brookings, SD 57007, USA Dave W. Crowder, Department of Natural Resources and Environmental Sciences, University of Illinois, Urbana, IL 61801, USA

ix prelims00.qxd 15/11/04 2:16 PM Page x

x Contributors

Todd A. DeGooyer, Monsanto Company, 800 N. Lindbergh Blvd, St Louis, MO 63167, USA Jaques Derron, Swiss Federal Research Station for Plant Production, 1260 Nyon, Switzerland C. Richard Edwards, Department of Entomology, Purdue University, W. Lafayette, IN 47907-2089, USA Michael M. Ellsbury, Northern Grain Insects Research Laboratory, USDA- ARS, Brookings, SD 57006, USA Siegfried Enzian, Federal Biological Research Centre for Agriculture and Forestry, Institute for Technology Assessment in Plant Protection, Stahnsdorfer Damm 81, 14532 Kleinmachnow, Germany Husnija Festic, Faculty of Agriculture, University of Sarajevo, 71000 Sarajevo, Bosnia-Herzegovina Ricky E. Foster, Department of Entomology, Purdue University, W. Lafayette, IN 47907-2089, USA Lorenzo Furlan, Department of Agronomy, Entomology, University of Padua, Legnaro, 2-35122 Padova, Italy Corey K. Gerber, Department of Entomology, Purdue University, W. Lafayette, IN 47907-2089, USA Joseph J. Gillespie, Holistic Insect Systematics Laboratory, Department of Entomology, Texas A&M University (TAMU), Rm 519, Minnie Belle Heep Bldg, 2475 TAMU, College Station, TX 77843-2475, USA Michael E. Gray, Department of Crop Science, University of Illinois, Urbana, IL 61801, USA Charles A. Guse, Department of Natural Resources and Environmental Sciences, University of Illinois, Urbana, IL 61801, USA Ibolya Hatala-Zsellér, Csongrád County Plant and Soil Protection Service, PO Box 99, 6801 Hódmezövásárhely, Hungary Graham P. Head, Monsanto Company, 800 N. Lindbergh Blvd, St Louis, MO 63167, USA Bruce E. Hibbard, USDA-ARS, 205 Curtis Hall, University of Missouri, Columbia, MO 65211-7020, USA Jasminka Igrc-Barc˘ic´, Department of Agricultural Zoology, Faculty of Agriculture, University of Zagreb, 10000 Zagreb, Croatia Scott A. Isard, Department of Geography, 220 Davenport Hall, University of Illinois, Urbana, IL 61801, USA Ivanka Ivanova, Central Laboratory for Plant Quarantine, 1330 Soﬁa, Bulgaria József Kiss, Department of Plant Protection, Szent István University, 2100 Gödöllö, Hungary Judit Komáromi, Department of Plant Protection, Szent István University, 2100 Gödöllö, Hungary Ulrich Kuhlmann, CABI Bioscience Switzerland Centre, Rue des Grillons 1, 2800 Delémont, Switzerland Wiebe Lammers, Plant Protection Service, PO Box 9102, 6700 HC Wageningen, The Netherlands prelims00.qxd 15/11/04 2:16 PM Page xi

Contributors xi

Eli Levine, Center for Economic Entomology, Illinois Natural History Survey, 172 Natural Resources Building, 607 E. Peabody Drive, Champaign, IL 61820-6917, USA Timothy R. Mabry, Holden Foundation Seeds, 503 S. Maplewood Ave., Williamsburg, IA 52361, USA Douglas D. Malo, Plant Science Department, South Dakota State University, Brookings, SD 57007, USA Michael J. McKee, Monsanto Company, 800 N. Lindbergh Blvd, St Louis, MO 63167, USA Joachim Moeser, Institute for Plant Pathology and Plant Protection, Entomological Section, Georg-August University, Grisebachstrasse 6, 37077 Göttingen, Germany John L. Obermeyer, Department of Entomology, Purdue University, W. Lafayette, IN 47907-2089, USA Victor Omelyuta, Institute of Plant Protection of Ukraine Academy of Agrarian Sciences, 03022 Kiev, Ukraine David W. Onstad, Department of Natural Resources and Environmental Sciences, University of Illinois, Urbana, IL 61801, USA Jay C. Pershing, Monsanto Company, 800 N. Lindbergh Blvd, St Louis, MO 63167, USA Gabor Princzinger, Department of Plant and Soil Protection, Ministry of Agriculture and Rural Development, 1860 Budapest, Hungary Philippe Reynaud, Entomology Unit, INRA, 34060 Montpellier, France Ivan Sivcev, Institute for Plant Protection and Environment, 11080 Zemun, Serbia Peter Sivicek, Central Control and Testing Institute of Agriculture, 84429 Bratislava, Slovakia Joseph L. Spencer, Center for Economic Entomology, Illinois Natural History Survey, 172 Natural Resources Building, 607 E. Peabody Drive, Champaign, IL 61820-6917, USA Kevin L. Steffey, Department of Crop Science, University of Illinois, Urbana, IL 61801, USA Douglas W. Tallamy, University of Delaware, Department of Entomology and Wildlife Ecology, 250 Townsend Hall, Newark, DE 19717, USA Stefan Toepfer, CABI Bioscience Switzerland Centre, Rue des Grillons 1, 2800 Delémont, Switzerland Gregor Urek, Agricultural Institute of Slovenia, 1001 Ljubljana, Slovenia Otmar Vahala, Regional Division of the State Phytosanitary Administration, 62800 Brno, Czech Republic Ty T. Vaughn, Monsanto Company, 800 N. Lindbergh Blvd, St Louis, MO 63167, USA Stefan Vidal, Institute for Plant Pathology and Plant Protection, Entomological Section, Georg-August University, Grisebachstrasse 6, 37077 Göttingen, Germany Dennis P. Ward, Monsanto Company, 800 N. Lindbergh Blvd, St Louis, MO 63167, USA prelims00.qxd 15/11/04 2:16 PM Page xii

xii Contributors

Rüdiger Wittenberg, CABI Bioscience Switzerland Centre, Rue des Grillons 1, 2800 Delémont, Switzerland Feng Zhang, CABI Bioscience Switzerland Centre, Rue des Grillons 1, 2800 Delémont, Switzerland prelims00.qxd 15/11/04 2:16 PM Page xiii

Preface

Since its first discovery in Europe in 1992 near Belgrade, Serbia, the western corn rootworm (WCR), Diabrotica virgifera virgifera LeConte, which is often referred to as the most import pest insect of maize in the USA, has spread to more than 15 Eastern and Central European countries (Chapter 2). Increasing public and governmental concerns within the European Union regarding the impact of this pest on maize production has resulted in funding directed at determining potential management options for European countries. In 2000, the European Union funded a multi-country project on WCR Ecology and Management in Europe (QLRT-1999-01110). At the conclusion of this project, an international symposium was organized in Göttingen, Germany, in spring 2003. This symposium brought together WCR scientists and experts from Europe, and North and South America. Papers were presented on the current state of knowledge of the diabroticine beetles, including such topics as distribution, behaviour, natural enemies and management options. These papers and the discussions that followed formed the basis for this book. The 1986 J.L. Krysan and T.A. Miller book, Methods for the Study of Pest Diabrotica, is not necessarily considered comprehensive in regard to some of the recent issues in Diabrotica research. There is much to be learned from this invasive pest species, and recent research on WCR has added insight into evolutionary aspects of pest species, adaptation to changing environments, as well as to management options to counterbalance these adaptations. As such, this symposium-based book has been published with the aim of updating knowledge on the ecology and management of Diabrotica, which is equally relevant and applicable for management decisions in North America and Europe. Concepts and management options related to WCR are reviewed in this book to classify the invasion of Europe by WCR, and to place it into the context of best prevention and management practices of invasive alien species (Chapter 1). Specific management options for WCR should be adapted based on their behaviour in the field. The implications for

xiii prelims00.qxd 15/11/04 2:16 PM Page xiv

xiv Preface

alternative host plant use by WCR in Europe are numerous, and understanding the nutritional ecology of this species (Chapter 3) and the phylogenetic background of host plant use in different plant families (Chapter 4) may explain the rapid establishment of WCR in Europe. Detailed life table analyses (Chapter 5), carried out in Hungary, allow for the evaluation of the combined impact of mortality factors that occur naturally and that may serve as regulating factors of WCR in Europe. Since no comprehensive information was available on natural enemies occurring in the area of origin of diabroticine beetles in Central America, a review of classical biological control possibilities for WCR through the introduction of specialized natural enemies is provided (Chapter 13). Recently, WCR has changed its egg-laying behaviour in some regions in the USA due to intense selection pressure and adaptation to annual crop rotation systems, particularly in soybean and maize. On the other hand, the northern corn rootworm (NCR), Diabrotica barberi Smith and Lawrence, has adapted to this selection pressure by extending the diapause period in some individuals. The different adaptations of these two Diabrotica species, together with reports of insecticide-resistant populations in some areas of the Midwestern USA, point to the importance of additional or alternative management options for the diabroticine beetles. Background on the phenomenon of egg laying in a non-host crop, such as soybean, has been elaborated in detail in Chapter 6. To understand the adaptation of Diabrotica to crop rotation, a model was developed (Chapter 8) that explains such phenomena in areas of the Midwestern USA. A field experiment designed to evaluate the possibilities of adaptation of WCR to crop rotation systems in Europe and the conclusions drawn from this experiment are outlined in Chapter 10. Diabroticine beetles are not evenly distributed in fields. The aggregation of beetles makes sampling and quantification of populations within a field difficult. An evaluation of the spatial distribution of NCR larvae in the field (Chapter 7) and an evaluation of sampling devices and decision rules to predict subsequent maize damage (Chapter 9) both serve to clarify possible management strategies. The areawide management approach (Chapter 11) and the implementation of genetically modified maize cultivars expressing a specific toxin targeted against WCR are both additional management options which could be implemented in Europe to keep WCR populations below threshold levels. Finally, as maize is grown less intensively and is more scattered in Europe compared to the USA, a quantification of high-risk areas prone to WCR damage may help to focus management options within Europe. This book would not have been published without the continuous interest and help of Tim Hardwick, CABI Publishing. We therefore owe our thanks to him and his efforts.

Stefan Vidal, Ulrich Kuhlmann and C. Richard Edwards Göttingen, Germany; Delémont, Switzerland; and W. Lafayette, Indiana, USA – April 2004 chap01.qxd 12/11/04 10:56 AM Page 1

1 Invasive Alien Species – a Threat to Global Biodiversity and Opportunities to Prevent and Manage Them

Rüdiger Wittenberg CABI Bioscience Centre Switzerland, Delémont, Switzerland

Introduction

Changes in distribution of species are a natural phenomenon; ranges expand and retract and species colonize new areas outside their natural range by long-distance dispersal, for example reptiles on floating wood to new islands. However, these events are rare and restricted by natural barriers. The rather recent globalization of trade and travel has inadvertently led to the increased transport of organisms and introduction of alien species, tearing down these natural barriers. Alien species are not bad per se, in fact, many species are used for human consumption, e.g. most crop species are grown as aliens. Some of them may become harmful and pose threats to the environment and human populations, as will be discussed below. These so-called invasive alien species (IAS) are increasingly recognized as one of the major threats to biodiversity. This global problem needs global reaction or, even better, proactive measures and solutions. Thus the Global Invasive Species Programme (GISP) deals with IAS to find the best practices to prevent and manage IAS in response to the undertaking in the Convention on Biological Diversity (Article 8h: ‘Each Contracting Party shall, as far as possible and as appropriate: Prevent the introduction of, control or eradicate those alien species which threaten ecosystems, habitats or species’). This chapter will give a brief summary of the current knowledge on IAS, what IAS are, what their perceived impact is, how they travel to new places and what their characteristics are, and will focus on what can be done to prevent and manage IAS. The latter is based on best prevention and management practices derived from successful case studies. However, the chapter cannot cover all knowledge and options; this would not fit into an entire book, so only an overview and selected highlights can be covered. The complexity of the topic ‘IAS’ stems from the very different species involved, their diverse origin, the

2 R. Wittenberg

variety of pathways used, their varied impact on the new environment, their relationships with indigenous and other alien species, the ecosystem changes caused, their dependence on other factors such as global warming, their human dimensions, including the change in political and ethical views, and their ongoing evolution.

Historical Example

Besides habitat loss due to human destruction of natural ecosystems, IAS are the single most important threat to biodiversity. These negative effects are best documented in bird extinctions on islands where the majority of the bird extinctions since 1800 were caused by IAS (BirdLife International, 2000). Island birds often did not evolve with mammalian predators lacking natural defences to introduced predators. Therefore, it is of great concern that a quarter of all globally threatened bird species are currently affected by alien predators. The story of the Stephen Island wren (Xenicus lyalli (Rothschild)), which belonged to the family of the New Zealand wrens or Acanthosittidae, will serve as an example to these abstract figures. This is a very old bird family at the base of passerine radiation with only two remaining species (Chambers, 1989; Sibley and Monroe, 1990). The Stephen Island wren occurred only on Stephen Island, a small offshore island of New Zealand. Its demise started with the building of a lighthouse. One fine day the domestic cat (Felis catus (L.)) of the lighthouse keeper, which he kept as a companion, brought home a dead bird, which he did not recognize. So he gave it to ornithologists on the mainland and they confirmed that it was a species new to science (Quammen, 1996). While the lighthouse keeper, Mr Lyall, lent his name to the newly described species, his cat continued to feast on it until it was never found again. It is still unclear whether the Stephen Island wren was flightless or had some rudimentary flight ability – this knowledge was taken to the grave by the cat, the only one who would know. The lighthouse keeper saw the species only twice, running like a mouse in the evening in 1894 (Falla et al., 1993). Thus in this clear and special case it was only one specimen of an IAS that caused the extinction of a rare bird species. In most other cases it is not that obvious and it is more probably a combination of factors leading to extinction of a species rather than being monocausal. However, one subspecies of the bush wren (Xenicus longipes (Gmelin)), the other extinct species of the family, was restricted to Steward Island and some outliers. The demise of that subspecies is directly attributable to the arrival of the ship rat (Rattus rattus (Linnaeus)) and mustelids (Heather and Robertson, 1996). It is interesting to note that the four species of the New Zealand wren family were not adapted to mammalian predators in their natural environment and the two species extinct were ground dwellers with at most weak flight abilities, one in the forests and one on a small island, whereas the other two species live in more protected niches, one species in high alpine fields and the other chap01.qxd 12/11/04 10:56 AM Page 3

Invasive Alien Species 3

being a canopy species with good ﬂight ability. This little example of the fate of four closely related species indicates a more general phenomenon. Small island populations without adaptation to mammalian predators and breeding on or near the ground are most vulnerable to intentional or accidental introductions of alien opportunistic predators.

Who Is Who

Who and what are IAS? They are species introduced by human intervention, either intentionally or accidentally, into a new environment, a new area, where they are able to adapt and thrive, causing negative impacts on the native ecosystem, species or humans. It needs to be stressed, as mentioned above, that not all alien species cause problems. In fact, most of the crop species used worldwide are grown as non-indigenous species, for example the wide use of American maize (Zea mays L.) in Europe and Africa. Bearing in mind that creating lists of native and introduced species and ascertaining the impact of IAS are not flawless, as a rule of thumb the tens rule can be employed; this states that 10% of introduced species are able to establish in their new environment and 10% of those established become invasive (Williamson, 1996). IAS are found in virtually every taxonomic group. The following examples will attest this statement. The West Nile virus causing encephalitis hitched a ride to the New World in an infected bird, mos- quito or human (Enserink, 1999). The bacterium Vibrio cholerae, the causal organism of the human disease cholera, is a member of brackish water communities and is frequently found in ballast water of ships (McCarthy and Khambaty, 1994), by which means some new highly viru- lent strains have been redistributed leading to epidemic outbreaks of cholera. Some fungal pathogens are amongst the IAS with the most disruptive impact on ecosystems, a well-known example being fungi attacking trees, e.g. chestnut blight (Cryphonectria parasitica (Murrill) Barr) was introduced with alien chestnuts to North America, where it virtually eradicated the American chestnut (Castanea dentata (Marsh.) Borkh.), which was a dominant tree in eastern forests, thereby changing the entire ecosystem and composition of the forests (Hendrickson, 2002). Many marine phytoplankton species cause harmful algal blooms, in particular dinoflagellates, which can kill fish and especially sessile organisms and also produce potent toxins that can find their way to humans through consumption of seafood (Weidema, 2000). One of the worst invaders in the Mediterranean Sea is the macroalga Caulerpa taxifolia (Vahl) C. Agardh, which escaped from an aquarium and is replacing the native seaweed beds and thus altering large tracts of coastal ecosystems. Weeds are the predominant group of IAS known to cause economic problems as well as deleterious effects on the environment. The giant reed (Arundo donax L.) is used in many countries, for example, as wind-breaks and is chap01.qxd 12/11/04 10:56 AM Page 4

4 R. Wittenberg

readily invading natural areas, and the small herb common ragweed (Ambrosia artemisiifolia L.) is swiftly expanding its exotic range in Europe causing severe allergic problems for the human population. An almost global problem, disrupting, or at least altering, the natural succession is bushes such as broom (Cytisus scoparius (L.) Link) and gorse (Ulex europaeus L.), introduced as ornamentals and soil-enhancing plants; another successful invader of natural forest is the vine old man’s beard (Clematis vitalba L.), which smothers remnant forest tracts in New Zealand, and finally trees such as pines (Pinus spp.) and eucalypts (Eucalyptus spp.) escape from their plantations and may replace native vegetation and deplete water sources (Van Wilgen et al., 1997). Many different kinds of worms found their way to new areas with human assistance, especially parasitic worms from the Platyhelminthes and Nemathelminthes, and also in the marine environment, for example the polychaete Marenzelleria viridis (Verrill) introduced into the North and Baltic Seas (Zettler et al., 2002). A spectacular disaster caused by an introduced snail was the introduction of the carnivore rosy wolfsnail (Euglandina rosea (Férussac)) to many subtropical and tropical islands, destroying the diverse endemic snail faunas. Another example of an introduced mollusc in Europe is perhaps the most aggressive freshwater invader worldwide, the zebra mussel (Dreissena polymorpha (Pallas)), inflicting not only huge economic costs but also severe biotic changes as it functions as an ecosystem engineer species (Karateyev et al., 2002). Small introduced crustaceans dominate the fauna of many rivers and lakes worldwide due to the increased ship traffic transporting organisms in their ballast water to new areas and also due to the creation of canals connecting formerly insurmountable natural barriers between watersheds. Thus, alien species (mainly crustaceans and molluscs) dominate the Rhine in total abundance and biomass by more than 80% (Haas et al., 2002). Interestingly, introduced insects, despite their diversity, have not shown a high potential for causing environmental problems, albeit they can be devastating pests in agriculture and forestry. However, several ant species destroy the native faunas, especially on islands – but also, for example, the Argentine ant (Linepithema humile (Mayr)) in southern Europe. The infamous cane toad (Bufo marinus (L.)) is quickly spreading over Australia feeding on everything smaller than itself and poisoning the bigger predators, such as quolls (Dasyurus spp. E. Geoffroy St-Hilaire). One of the most devastating introduced reptiles is the brown tree snake (Boiga irregularis (Merrem)) on Pacific islands; it arrived on Guam with military help and brought the silent spring to the island by feasting on the bird species. Moreover, it is causing frequent power cuts and is a danger to babies because of its venom. The Nile perch (Lates niloticus (L.)), introduced into Lake Victoria to improve fisheries, caused the extinction of more than 100 fish species of the cichlid family, most of them endemic to the lake – before the predator came it was called an evolutionary laboratory, afterwards an ecological disaster. The American ruddy duck (Oxyura jamaicensis (Gmelin)) was introduced to England as an addition chap01.qxd 12/11/04 10:56 AM Page 5

Invasive Alien Species 5

to the wildfowl fauna, where it ostensibly does no harm, but then it spread to Spain where it readily endangers the native closely related white- headed duck (Oxyura leucocephala (Scopoli)) by hybridization. Feral mammals introduced to islands brought many bird species to the brink of extinction or beyond by feeding on their eggs and chicks (e.g. Long, 2003). However, some taxonomic groups seem to include more invasive species than others do. Mammals are a major threat to island faunas and floras. Whereas rats, mongooses, mustelids and feral cats devastate the local bird and reptile fauna of islands, feral goats (Capra hircus) can diminish the native flora drastically. Islands, due to their size and other characteristics, which cannot be summarized here (see, for example, Loope and Mueller- Dombois, 1989), are particularly vulnerable to these invaders. These introduced mammalian plagues are due chiefly opportunistic predators, with high abilities for adaptation to different circumstances and food items as a precondition for bringing native species to extinction. It is perhaps not surprising that the largest angiosperm families supply the largest percentages in the world of invading species (Heywood, 1989). The high number of invasive mammals and members of the Asteraceae, which are regarded as some of the most advanced groups (amongst classes of vertebrates and families of angiosperms, respectively) from the evolutionary point of view, is also an indication that these groups are currently in radiation, evolutionar- ily successful at these times, and have developed biological features which ensure both survival under extreme conditions and high reproductive rates. This suggests that, for example, in the era of the dinosaurs, the reptiles would have been the most damaging invaders. IAS are recorded from every part of the globe, emphasizing the worldwide impact of human disturbance, leaving almost no ecosystem untouched and pristine. All continents and habitats seem to be vulnerable to invasions, though islands are particularly at risk, as mentioned above, and some patterns between continents seem to arise. In the highly populated area of Central Europe IAS seem to be of less importance to biodiversity than on other continents with large tracts of more natural habitats. The smaller reserves in Central Europe are easier to manage and control of alien species in these places is often more practical. The long association between introduced species and the human population in Europe is a very different situation compared to other continents, as all habitats are highly altered and human-made habitats dominate. These human-made habitats are regarded as a valuable heritage of Central Europe and are often based on alien species introduced long ago. There are also broad differences between the numbers of introduced species. Kowarik (2002) states that in the German flora 22.4% are alien species, whereas New Zealand has 1579 established alien angiosperms to 2212 native ones (Clout and Lowe, 2000). In the case of New Zealand the numbers of native to alien land mammals is of even greater concern, with a ratio of two indigenous bats to 34 alien species. Comparing the numbers of introduced to established and invasive alien species, one has to bear in mind the long lag phases, which often chap01.qxd 12/11/04 10:56 AM Page 6

6 R. Wittenberg

occur. Most introduced species will need some time before they become invasive, i.e. enter the lag phase. Kowarik (2003) shows that for woody alien plants in a part of Germany the average time-lag between first introduction and their expansion is about 147 years. The reasons for the lag phase can be manifold and will be different from case to case. The founder population needs some time to adapt to the local environment, genetic changes could also play a role with a new mutation better adapted, ongoing releases could also be a crucial factor in the establishment, or a change in human disturbance can be of additional help for an alien species. Whatever the reason, the fact that often time-lags occur makes predictions on invasiveness of alien species very difficult. A species showing no harm today can still be an invasive of tomorrow, especially in combination with other global changes. Figure 1.1 summarizes some important factors influencing the invasiveness of species.

Human dimension Extrinsic factors

Abiotic factors, such as Attractiveness to climatic conditions and soil humans: pathways, use of species, number of introductions No. natural enemies

No. competing Variation: species: native and species traits exotic such as seed dispersal, Invasiveness of species number, size Interactions with native and exotic species: pollination, dispersal, food Intrinsic factors source, change of ecosystem

Ecosystem characteristics, e.g. disturbed (natural or human-induced) Control of this and Land use other IAS change Global climate change

Fig. 1.1. Factors affecting the bioinvasion process and the invasiveness of alien species. chap01.qxd 12/11/04 10:56 AM Page 7

Invasive Alien Species 7

There are three major factors which determine the ability to become invasive:

1. Intrinsic factors or species traits, such as the ability to adapt to different conditions, a wide amplitude to abiotic factors, pre-adapted to different climatic zones, and a high reproductive rate. 2. Extrinsic factors or relationships between the species and abiotic and biotic factors, such as the number of natural enemies, the number of competing species (native and alien), other interactions with native and alien species (pollination, dispersal, food source, ecosystem engineers), climatic conditions, soil conditions, degree of disturbance (natural and human-induced), global climate change, change in land use patterns, and control and eradication of other IAS. 3. Human dimension. The attractiveness and importance for humans inﬂuence the pathways, vectors, the numbers of specimens introduced, the numbers of introductions and the potential to eradicate or control the species.

Pathways and Vectors

The ways in which alien species are introduced are almost unlimited. Some important pathways, for example to an island, are shown in Fig. 1.2. The increasing boat trafﬁc also accelerated the introductions of

Aqua- culture

Fig. 1.2. Some important pathways into the terrestrial, freshwater and marine environments of a given island. chap01.qxd 12/11/04 10:56 AM Page 8

8 R. Wittenberg

species by transport both within and upon the vessels. One of the most important pathways for marine and freshwater species is transport in ballast tanks. The species are either in the ballast water or live in the ballast sediment. In former days when the ships took soil as ballast, even terrestrial species were carried, e.g. plants. Another important way for introductions of aquatic species is the hull-fouling organisms, which can reproduce and colonize new areas from the travelling ship. The cargo of ships and other transport vehicles also harbours species. Moreover, people carry hitchhiking species around on their dirty boots and tents. All these pathways transport species accidentally. Some other groups are deliberately introduced, such as species used in aquaculture, for ﬁsheries, as forest trees, for agricultural purposes, as species for hunting, plants for soil improvement and just for the pleasure of humans as ornamentals. It can be summarized that most aquatic species and invertebrates in general mainly hitchhike accidentally, whereas most plants and vertebrates are deliberately introduced. Minchin and Gollasch (2002) and Carlton and Ruiz (2004) give excel- lent overviews on pathways and vectors in more depth. The latter divide pathways into cause (why a species is transported), route (the geographical path) and vector (how a species is transported).

Consequences

The impacts of IAS are often considerable, as ecosystem functioning can be altered and species can be brought to extinction. There is no way back to the former state in the latter case. The environmental impacts can be divided into four major factors: (i) competition; (ii) predation (including herbivory); and more subtle interactions such as (iii) hybridization and (iv) transmission of diseases. All these factors alone or in concert with other factors can decrease biodiversity and cause extinction. The most obvious examples for competition are between introduced and native plants for nutrients and exposure to sunlight. Resource competition has led to the replacement of the native red squirrel (Sciurus vulgaris L.) by the introduced American grey squirrel (Sciurus carolinensis Gmelin) in almost all of Great Britain. The latter forages more efficiently for food and is stronger than the native species (Williamson, 1996). Impacts due to predation and herbivory are very extensive on island fauna and flora, as mentioned above. The brown tree snake eliminated most of the bird species on Guam, and feral goats are a menace to native island vegetation, where they were often released as a living food depot. A well-known hybridization example from Europe is the ruddy duck, which hybridizes with the native white-headed duck, as mentioned before. In some cases IAS can harbour diseases and be a vector for the diseases to native species. This is the case with American-introduced crayfish species to Europe, which are carriers of the crayfish plague (Aphanomyces astaci Schikora) without many symptoms, but the native noble crayfish (Astacus astacus (L.)) is chap01.qxd 12/11/04 10:56 AM Page 9

Invasive Alien Species 9

highly susceptible to the disease, and thus is struggling to coexist with populations of the American crayfish species. Introduced species can interact with natives in a variety of ways and indirect effects can be very difficult to demonstrate. Direct and indirect effects can lead to very complex interactions and a combination of effects can cause complex impacts. In addition to the biodiversity impacts, many IAS also impose enormous economic costs. These costs can arise through direct losses of agricultural and forestry products and through increased production costs associated with control measures (US Congress, 1993; Pimentel et al., 2000). One often-cited example is the costs caused by D. polymorpha to water plants by clogging water pipes and other structures in the Great Lakes in North America. Costs for environmental problems are more difficult to calculate than costs imposed in the agricultural sector. A North American study calculated costs of US$138 billion per annum to the USA caused by IAS (Pimentel et al., 2000). This takes into account all costs associated with the IAS, not only the direct costs such as loss of harvest, costs of control, etc. Some of the costs calculated in the paper are rather estimates; however, even giving or taking a number of such magnitude, it is a very impressive number and shows the importance of IAS. Some IAS also have implications for human health. Giant hogweed (Heracleum mantegazzianum Sommier et Levier) was introduced from the Caucasus to Europe as an ornamental plant. It produces copious amounts of a sap which is phototoxic and can lead to severe burns of the skin. Regularly, in particular children are hospitalized after contact with the plant, especially when they play with the hollow stems and petioles. The racoon dog (Nyctereutes procyonoides Gray), introduced as a fur animal, can, like the native fox, be the vector of the most dangerous parasitic disease vectored by mammals to humans in Central Europe, i.e. the fox tapeworm (Echinococcus multilocularis Leuckart) (Thiess et al., 2001). Although the racoon dog is only an additional vector, this can have effects on the population dynamics of the parasite and lead to an increase in the disease in humans. To address these impacts, populations of existing invasive species need to be managed. New introductions must be assessed as to the threat they may present and only introduced on the basis of a risk analysis, and new invasions must be minimized.

Prevention

The rapid global increase in trade, travel, transport and tourism is leading to an increase in introductions of non-indigenous species (Peck et al., 1998). Prevention is the ﬁrst and most cost-effective line of defence against IAS. An ounce of prevention is worth a pound of cure – this maxim of medicine, dictating such measures as quarantine and inocula- tion, is equally valid for dealing with biological invasions. It is essential chap01.qxd 12/11/04 10:56 AM Page 10

10 R. Wittenberg

to prevent as many alien species as possible from entering the country, in addition to following a planned risk assessment process for species and pathways. Although there are significant costs associated with prevention, failure to prevent a single highly invasive species from colonizing, such as the zebra mussel in North America or the western corn rootworm (Diabrotica virgifera virgifera LeConte) in Europe, might lead to enormous costs outweighing all prevention efforts. One cost of prevention, and the most obvious, is the expense of maintaining the exclusion apparatus (salary and training of interception personnel, and facilities such as fumigation chambers, inspection apparatus and quarantine quarters). A second cost is that affecting individuals who are not allowed to profit from bringing in alien species (which may or may not be intended for release to the environment), and equally the public that might have ben- efited from a planned introduction disallowed by the prevention procedures. These costs are offset by the benefits that accrue to society from potential invasions that are prevented. In many cases, the benefit from an introduced species lies with the importer, whereas control costs of IAS are borne by the public. However, prioritizing support for a costly prevention system is often difficult, because the impact of alien species and the potential costs of invasions cannot be reliably predicted because of the complexity and calculated against the actual costs of the apparatus. Most prevention measures are focused on certain species known to be pests elsewhere, since this is the most reliable feature of invasive species. However, most of these quarantine species are economically important species for the agriculture, forestry or human health sectors. Prevention of entry of species on these black lists is the rather conservative goal for quarantine and other measures taken at present. A more recent approach, in order to incorporate all potentially dangerous organisms, not only in an economic view but also in terms of saving the world’s biodiversity, is a move to using white lists (e.g. US Congress, 1993; Panetta et al., 1994). The approach is also often called ‘guilty until proven innocent’ (Ruesink et al., 1995). A proposed intermediate step is the use of ‘pied lists’, which are more realistic to implement when constrained by a lack of facilities, staff and funding. The ‘pied list’ would contain a section of known pest species (equivalent to black lists) with strict regulations and measures to ensure pest-free imports. Another section of the list would describe species cleared for introduction (white lists) – organisms declared as safe. All species not listed on either list would be regarded as potential threats to biodiversity, ecosystems or economic sectors. A stakeholder proposing an intentional introduction would have to demonstrate beyond reasonable doubt the safety of the proposed introduction in a risk assessment process. Species assessed for their likely invasiveness would be moved to the white or black list depending on the outcome of this investigation. There are three principal strategies to reduce further introductions:

1. Interception based on regulations and their enforcement with inspec- tions and fees. This approach involves inspection, decontamination and chap01.qxd 12/11/04 10:56 AM Page 11

Invasive Alien Species 11

potential constraints to specific trade commodities rated as high-risk. The illegal importation of prohibited items – smuggling – is also of high concern, since the items are not inspected and smugglers are unlikely to pay attention and take precautions against hitchhiking species. In order to meet the precautionary principle a risk assessment process should be the basis for every proposed intentional introduction unless the species is already on the white list. 2. Treatment of goods or their packaging material suspected to be con- taminated with alien organisms, including biocide applications (e.g. fumigation, immersion or spraying), heat and cold treatment, pressure, and irradiation (Sharp and Hallman, 1994). 3. Trade prohibition. Under the World Trade Organization Sanitary and Phytosanitary (WTO SPS) Agreement (WTO, 1994), member countries have the right to take sanitary and phytosanitary measures to the extent necessary to protect human, animal or plant life or health provided these measures are based on scientific principles and are not maintained without sufficient scientific evidence.

Public education is an essential part of prevention and management programmes. In fact, some scientifically well-devised projects have been interrupted or stopped because of public disapproval. Besides these extreme cases, public awareness and support can greatly increase the success of projects to protect and save biodiversity. Travellers are often unaware of laws and regulations to prevent introductions of alien species, and the reasons for them. Education should focus on raising the awareness of the reasons for the restrictions and regulatory actions, and the environmental and economic risks involved. In addition to printed material, e.g. posters and brochures, video presentations and announcements on aeroplanes are a promising approach. The public as well as industry should perceive prevention measures not as an arbitrary nuisance but rather as a necessity for travel and trade. The most common approach for prevention of invasive organisms in the past was to target individual pest species. However, a more comprehensive approach is to identify major pathways that lead to harmful invasions and manage the risks associated with these. Pathways such as ballast water can only be targeted on the entire pathway and should be analysed for risks involved. IAS are not just species brought into a country from another country, but also within-country movements must be considered. Political boundaries do not make sense in the definition of alien species, which should rather be based on eco-regions and natural boundaries, for example fish species transported between unconnected watersheds. Some species translocated within the same country can be as disruptive to ecological systems as a species from a different continent. Human-made structures may enhance subsequent spread of alien species formerly restricted to one area. The completion of the Welland Canal between Lake Ontario and Lake Erie enabled invasive chap01.qxd 12/11/04 10:56 AM Page 12

12 R. Wittenberg

organisms, such as the sea lamprey (Petromyzon marinus L.), to bypass the Niagara Falls and subsequently spread to other lakes and river systems (Simberloff, 1996). The opening of the Suez Canal initiated a remarkable influx of hundreds of Red Sea species into the oligotrophic Mediterranean Sea, outcompeting and replacing indigenous species (Galil, 1999). In conclusion, prevention is the most cost-effective control. However, it is difficult to mobilize a sufficient budget before an IAS causes problems and inflicts costs. Exclusion methods based on pathways rather than on individual species are a more efficient way to concentrate efforts. Three major possibilities to minimize further invasions are interception, treatment of suspect imported material and prohibition of particular commodities under international regulations. Deliberate introductions should all be subject to an import risk assessment.

Early Detection

The longer an alien species goes undetected, the higher will be the population (approaching the lag phase), the fewer options will remain for its control and in particular eradication, and the more expensive any intervention will become (Mack et al., 2000). Eradication options will swiftly fade with the building up of the alien population. However, during the lag phase, it can be difﬁcult to distinguish doomed populations from future invaders. Since not all alien species will necessarily become invasive, species known to be invasive elsewhere under similar conditions are priorities for early detection. The possibility of early eradication or getting a new colonizer under effective early control makes investment in early detection worthwhile. Surveys for early detection should be carefully designed and targeted to answer speciﬁc questions as economically as possible. Some invasive species are easily seen, while others are cryptic and require special efforts to locate or identify them, particularly when they are low in numbers. Traps can be very effective for the more cryptic species. Surveys by experts should be made for certain groups of pests to enable a rapid response before the invasive species becomes well established. A contingency plan is usually a carefully considered plan of the action that should be taken when a new invasive species is found or an invasion is suspected. The plan may be just a simple paper document that all staff, selected volunteers or relevant organizations have written, are aware of and will act on in a contingency situation. Alternatively the plan may be expanded to include comprehensive kits of tools that are stored in a ‘ready-to-use’ condition at appropriate locations. Contingency funding must also be immediately accessible to deal with species colonization in an early stage. chap01.qxd 12/11/04 10:56 AM Page 13

Invasive Alien Species 13

Management

Even the most effective prevention and early warning systems will have leaks and thus new introductions have to be anticipated. For those species and the ones already present in the country at the onset of prevention measures, control options have to be investigated. In a management project dealing with invasive species, several important issues have to be addressed, including planning, budgeting, monitoring, analysis, recording, reporting, follow-up and dissemination of the results. Adequate funding needs to be secured for all steps until the project goal, set prior to the beginning of the project, is met. The first step is to determine the management goal for any project for the management of invasives. The target area needs to be defined. It may be an entire country, all or part of an island or all or part of a reserve or conservation area. In some instances regional projects will include more than one country and need good coordination between countries. In particular in cases of management of IAS in national parks, which protect high biodiversity and an important habitat of the area, the management of IAS should be incorporated into a strategy for the national park. The strategy needs to state the goals of the protected area and the objectives of how to conserve the native biodiversity. The management of IAS is only one part of the bigger picture to conserve or restore the natural processes. In many cases IAS will need to be addressed. Thus, the final goal will be the preservation of the unique ecosystems and the development of sustainable use of ecosystem services. The management area, as defined in the management goal, has to be surveyed for alien as well as native species to assess the potential loss of natural habitat and to estimate the impacts. These surveys include literature search, collection records and actual surveys in the area. The docu- mentation has to include the best available knowledge about the abundance and distribution of alien species, their impact on the habitat and, when justified (e.g. based on experience in neighbouring areas), a prediction of future impact. If there are earlier data available, a comparison between past and current species composition and distribution of single alien (and native) species can reveal the status and spread of species in that area (and their impact). Past control actions and their success or failure should be summarized too. The next step would be to examine the management options for each target species, using local knowledge, information from databases and published and unpublished sources. Local circumstances, such as cultural, religious and socio-economic features may affect the suitability of different options. Options for eradication, containment or control and needs for further surveys, experimental investigations and other research should all be evaluated. Priorities should be set, with the highest priority given to existing infestations that are expanding most rapidly, are most disruptive and affect the most highly valued areas of the site. The priority-setting process can be difficult, partly because many factors need to be chap01.qxd 12/11/04 10:56 AM Page 14

14 R. Wittenberg

considered and the limited budget needs to be spent where it has the highest impact for the management goals. Important factors for the prior- itization process are the current extent of the species on or near the site, the current and potential impacts of the species, the value of the habitats that the species infests or may infest in the near future, and the practical- ity of control. In the long run, it is usually most efﬁcient to devote resources to preventing new problems and immediately addressing incip- ient infestations. All IAS populations need monitoring, because many species not yet regarded as invasive may in fact be ‘sleeping’ organisms passing through their lag phase of invasion and will become invasive later on when entering the lag phase. The three main strategies for dealing with alien species that have already established populations in the area under concern will be presented below, i.e. eradication, containment and control.

Eradication

Eradication is the elimination of the entire population of an alien species, including any resting stages, in the managed area. When prevention has failed, an eradication programme, as a rapid response to the early detection of an alien species, is often the key to a successful and cost-effective solution. However, eradication should only be attempted if it is feasible to eradicate the species with the budget and the methods available. A careful analysis of the costs (including indirect costs) and likelihood of success must be made (rapidly) and adequate resources mobilized before eradication is attempted. However, if eradication of the invasive species is achieved it is more cost-effective than any other measure of long-term control. Eradication programmes can involve several control methods on their own or a combination of these. The methods vary depending on the invasive species, the habitat and the circumstances. Successful eradication in the past has been based on mechanical control (e.g. hand-picking of snails or hand-pulling of weeds), chemical control (e.g. using toxic baits against vertebrates), spot spraying of plants, biopesticides (e.g. Bacillus thuringiensis (Bt) sprayed against insect pests), habitat management (e.g. grazing and prescribed burning), or hunting of invasive vertebrates. Some groups of organisms are more suitable for eradication efforts than others. The best option is to base an eradication programme on a successful case study, although the circumstances may vary slightly and lead to a different outcome. Each situation needs to be evaluated to ﬁnd the best method(s) in that area under the given circumstances. Plants can be eradicated best by a combination of mechanical and chemical treatments, e.g. cutting woody weeds and applying a herbicide to the cut stems. Many successful eradication programmes have been carried out against land mammals on islands. The methods most frequently used were bait sta- tions where toxic substances were offered to the invasive species, e.g. rat chap01.qxd 12/11/04 10:56 AM Page 15

Invasive Alien Species 15

control. Bigger animals can be hunted, provided the ecosystem is of an open kind with less cover in which to hide. A particular issue with eradication programmes against land vertebrates may be adverse public opinion, especially that of animal rights groups. It can be very difficult to try to eradicate a fluffy and cute animal. Amongst land invertebrates only snails and insects have been successfully eradicated on occasion. Snails can be hand-picked, whereas the commonest options to eradicate insects are based on the use of insecticides or biopesticides, usually by widespread application, or using baits or traps or a combination of methods. The use of sterile male releases, often in combination with insecticide control, has been effective on several occasions against insects, such as fruit flies and the screw-worm fly. There are two published successful eradications of invasive species in the marine environment to date. An infestation of a sabellid worm in a bay in the USA was eliminated by hand-picking of the host (Culver and Kuris, 1999); the other case was the eradication of a mussel species in Australia using pesticides (Bax, 1999). Foreign freshwater fish species have been eradicated in the past by using toxins specific to fish (Courtenay, 1997). Pathogens of humans and domesticated animals have been eradicated by vaccination of the respective host. In general, it seems more feasible to apply methods for eradication to the hosts (cf. mosquitoes) rather than directly to the pathogens. If an eradication programme is feasible, it is the preferred choice for action against an IAS. The advantage of eradication as opposed to long- term control is the opportunity for complete rehabilitation to the conditions prevailing prior to the invasion of the alien species. There are no long-term control costs involved (although precautionary monitoring for early warning may be appropriate) and the ecological impacts and economic losses diminish to zero immediately after eradication. This method is the only option that totally meets the management goal, because the invasive species is completely eliminated. The major drawbacks of eradication programmes are that they are very costly and their success cannot be guaranteed. The programme needs full commitment and attention until its successful completion; no eradication programme should be started unless an assessment of the available options and methods has shown that eradication is feasible. Thus, eradication should only be pursued when funding and commitment of all stakeholders are secured. Public awareness of the problems caused by the invasive species should be raised beforehand and public support sought. A well-designed and realistic eradication approach has to be developed to achieve the required goal. Many failed attempts were highly costly and had side-effects on non-target species, as in the case of the attempt to eradicate South American red fire ants (Solenopsis invicta Buren) in the southern states of the USA (Simberloff, 1996). The insecticide initially used proved disastrous to wildlife and cattle. The ant bait subsequently developed also had non-target effects, and proved to be more effective against native ant species than against the intruder. This in fact enhanced the populations of the alien species due to a decrease of interspecific com- chap01.qxd 12/11/04 10:56 AM Page 16

16 R. Wittenberg

petition with native ant species. Finally, the eradication efforts had to be abandoned. Eradication efforts have been especially successful in island situations. The term ‘island’ is in this case not restricted to real islands but can also include ecological islands isolated by physical or ecological barriers, e.g. forest remnants surrounded by agricultural fields. However, the target species may survive in small populations outside an ecological island and depending upon the degree of isolation could rapidly re-invade the ecological island after an eradication campaign. Although eradication methods should be as specific as possible, the rather rigorous nature of concentrated efforts for eradication will often inflict incidental casualties on non-target species. In most cases these losses can be seen as inevitable and acceptable costs in achieving the management goal and can be balanced against the long-term economic and biodiversity benefits. However, the potential non-target effects should be evaluated beforehand. When attempting eradication using toxins, it should be ensured that these are as specific as possible and that their persistence in the ecosystem is of short duration. How- ever, some toxins unacceptable for use in a long-term control programme might justifiably be used in an eradication campaign over a short period of time. Eradication (or control) of well-established non-indigenous species that have become a major element of the ecosystem will influence the entire ecosystem. Predicting the consequences of the successful elimination of such species will be difficult. The relationships of the invasive species to indigenous and alien species have to be considered. A strong carnivore–prey relationship between two invasive species points to the need to investigate the potential for combined methods to eliminate both species at the same time. Control of one species in isolation could have drastic effects on the population dynamics of the second species. Elimination of the normal prey may eliminate the carnivore or it may cause it to change its behaviour and feed on native species. Elimination of an introduced carnivore is likely to allow the introduced prey to increase greatly in numbers and may cause more damage than when both were present. By way of synthesis, basic criteria for a successful eradication programme are summarized as follows:

The programme needs to be scientifically based. Unfortunately, most traits rendering species invasive make eradication efforts more difficult, e.g. high reproduction rate and dispersal ability. That means that invasive species are likely to be difficult to eliminate due to their very nature. Eradication of all individuals must be achievable. It must be borne in mind that it becomes progressively more difficult and costly to locate and remove the final individuals at the end of the programme, when the population is dwindling away. chap01.qxd 12/11/04 10:56 AM Page 17

Invasive Alien Species 17

Support by the public and all stakeholders must be ensured beforehand and opposition not underestimated. Sufﬁcient funding must be secured for an intensive programme (allowing for contingencies) to make sure that eradication can be pursued until the last individual is removed. Small, geographically limited populations of alien species are easiest to eliminate. Thus, immediate eradication is the preferred option for most species found in early-detection surveys. Therefore it is crucial that the early warning programme has contingency funds available for these actions. Immigration of the alien species must be zero, i.e. the management area must be completely isolated from other infested areas, as is the case for islands. Potential pathways for the species between infested areas and the management area must be controlled to prevent new invasions. All individuals of the population must be susceptible to the eradication technique used. If individuals learn to avoid the technique (trap-shy), they would not be susceptible to the technique and would survive. Perhaps a combination of methods more successful at high and low densities respectively would be more successful under these circumstances. A technique needs to be designed to ensure detection of the last sur- vivors at very low densities at the end of the programme, e.g. pheromone traps installed at high densities in high-risk areas. Organisms that have less obvious stages that can survive for long periods, e.g. seed banks of weeds, need monitoring for an elongated period. A subsequent monitoring phase for several years should be part of the eradication programme to make sure that eradication has been achieved. Methods for prevention and early detection of the eradicated species should be put in place.

Containment

Containment of non-indigenous invasive species is a special form of control. The aim is to restrict the spread of an alien species and to contain the population in a defined geographical range. The methods used for containment are the same as those described for prevention, eradication and control. The invasive species population is suppressed using different methods along the border of the defined area of containment, individuals and colonies spreading beyond this are eradicated and introductions into areas outside the defined containment area are prevented. Species most likely to be successfully contained in a defined area are likely to be those that spread slowly over short distances. The nearest suitable habitat for the species should be separated by a natural barrier or an effective artificial barrier. The most suitable cases for containment are chap01.qxd 12/11/04 10:56 AM Page 18

18 R. Wittenberg

habitat islands without suitable connections that would allow the easy spread of invasive species. Containing a species in a deﬁned area will, however, need constant attention, control measures at the border and prevention measures against spread of the species. Thus, successful containment is difﬁcult to achieve and involves several different costly methods.

Control

Control programmes against IAS should aim for long-term reduction in density and abundance to below an acceptable threshold. The harm caused by the species under this threshold is considered acceptable in regard to biodiversity and economy. The weakened state of the invasive species allows native species to regain ground and they may even further diminish the abundance of the alien species. In rare cases this might even lead to local extinction of the alien species (especially when combined with habitat restoration efforts to support native species and put intact natural systems back in place). If prevention methods have failed and eradication is not feasible, managers will have to live with the introduced species and can only try to mitigate the negative impacts to biodiversity and ecosystems. All control methods, with the exception of classical biological control, which, when successful, is self-sustaining, need long-term funding and commitment. If the funding ceases, the population and the corresponding negative impacts will increase in most cases, if the IAS is not outcom- peted by native organisms. Since in the short term control seems to be a cheaper option than eradication, it is often the preferred method. The investment in funding and commitment does not need to be as great as for eradication programmes, the success will be seen earlier on and those who beneﬁt often bear most of the costs. Also, funding can be varied between the years depending on the perceived importance of the problem, political pressure and public awareness. However, the lower recurring costs are deceptive, because in the long run effective control is more expensive in total than a successful eradication campaign. Mechanical, chemical and biological control, habitat management and integrated combinations of methods are all used successfully in controlling population levels of invasive species. In many cases, the best practice to manage an invasive species may involve a system of integrated management tailored for the species and the location. Thus, it is important to accumulate the available information, assess all potential methods and use the best method or combination of methods to achieve the target level of control. As mentioned above, managing an invasive species should not be the management goal but only one tool in the process to achieve a higher goal, such as habitat restoration, preservation of a ‘pristine’ ecosystem, re- chap01.qxd 12/11/04 10:56 AM Page 19

Invasive Alien Species 19

establishment of the natural succession process and establishment of sustainable use of ecosystem services for local people. The successful control of the population of an invasive species itself can have indirect effects on native species, the ecosystem and the entire local biodiversity. The potential effects of reducing or eradicating the invasive species in a habitat should be evaluated beforehand and measures taken for these effects to be solely positive or on balance strongly positive. For example, removal of an aggressive invasive plant from a site might need to be accompanied by planting of indigenous species to fill the gaps and to prevent these gaps being filled by other unwanted plants. There are a huge number of specific methods to control invasive species. Recognizing the highly complex nature of invasion ecology and the importance of local conditions, general statements about suitable control methods for groups of alien species, in specific habitats or world regions, should be approached with great caution. Precise predictions of the behaviour, spread and impacts of alien species introduced into new environments are not available, because too many of the parameters used to describe the situation are no more than informed guesses. In many cases even the taxonomic status of the invasive species is uncertain. However, descriptions of methods used to control certain species and their effectiveness under specific environmental factors are available. These experience-based reports are essential for management of invasives and need to be made readily available, for example in databases accessible through the Internet.

Mechanical control Mechanical control can be carried out by directly removing individuals of the target species either by hand or by the use of special tools. In many cases introduced pests can be controlled or even eradicated in small-scale infestations by mechanical control, for example hand-pulling weeds (Cronk and Fuller, 1995) or hand-picking animals. An advanced method of mechanical control is the removal of plants by speciﬁcally designed machines, such as harvesting vehicles for water hyacinth (Eichhornia crassipes (Mart.) Solms)-infested lakes and rivers. In some cases of very persistent plants and depending on the area, e.g. on large open areas like pastures, even bulldozing may be appropriate. Mechanical control is highly speciﬁc to the target, and non-target effects are mostly restricted to disturbance by human presence and direct mechanical impact on vegetation and soil. The downside of the method is the fact that it is always highly labour-intensive. In countries where human labour is costly, the use of physical methods is almost prohibitive or at best limited mainly to volunteer groups. Most manual work has to be repeated for several years to remove all individuals, and to keep the volunteers interested might present a problem. For weeds whose seeds can be dormant in the soil for a long period, monitoring through that potential dormancy period is necessary following local eradication. The chap01.qxd 12/11/04 10:56 AM Page 20

20 R. Wittenberg

method can be effective when the population of the invader is still small and the population is limited to a small area. Weeds that grow vigorously from cut plant parts or multiply vegetatively are more difﬁcult to control (Cronk and Fuller, 1995).

Chemical control Chemical pesticides, including herbicides and insecticides, developed to meet the markets for control of pests in agricultural production and elimination of vectors of diseases, can be used to decrease population levels of invasive organisms below a threshold of ecologically tolerable impact. In the past, extensively used broad-spectrum pesticides such as DDT had massive detrimental impacts on the environment as well as on human health, but today these are banned in most countries, and there are more specific products on the market with fewer negative non-target effects. Some insecticides, such as those based on chemical structures similar to insect hormones, can also be specific to target groups of insects. Nevertheless, major drawbacks are the high costs, the necessity for repeated applications and the potential impacts on other species. Moreover, an additional problem, very clearly demonstrated in agriculture and human disease vector control, is that repeated use of pesticides can provide selective pressure, which enables many target species to evolve increasingly effective resistance to these chemicals. In response either the dose has to be increased or a different group of pesticides has to be used, usually further increasing the control costs. In cases, where resistance is anticipated, the likelihood of its development can be reduced by using chemicals in combination, or in sequence, or rotated with other control measures. Selection of a pesticide for use in controlling an invasive species begins with a determination of effectiveness against the target and all appropriate non-target species that might come in contact with the chemical, either directly or through secondary sources. Additionally, the environmental half-life, method of delivery, means of reducing non-target species contact, demonstration of efficacy and collection of data to ensure environmentally safe use must be evaluated. Most countries require pesticides to be registered for specific uses. Once identified, tested and registered, a pesticide can allow the rapid control of a target species over large areas and, as a result, reduce the need for personnel and costs for the more traditional methods such as traps and barriers.

Biological control Biological control is the intentional use of populations of natural enemies or naturally synthesized substances against pest species to suppress pest populations. Biological control can be split into several approaches grouped under two headings: those that are self-sustaining and those that are not. Methods that are not self-sustaining include: chap01.qxd 12/11/04 10:56 AM Page 21

Invasive Alien Species 21

Mass release of sterile males to swamp the population with males which copulate with the ‘wild’ females without producing any offspring in the next generation. The very successful screw worm (Cochliomyia hominivorax (Coquerel)) eradication programme was carried out using this method. Given the high costs, this approach is perhaps best used in eradication and containment programmes. Inducing host resistance against the pest. This approach is particularly relevant to agriculture where plant breeders select (or create) varieties resistant to diseases and insects. Biological chemicals, i.e. chemicals synthesized by living organisms. This category overlaps with chemical control and whether to list a particular method in one or the other category is a question of deﬁnition, e.g. while applying living Bt is without doubt a biological control option, to which group the use of the toxins stored in Bt belong could be debatable. Other examples of chemicals in this group are rotenone, neem and pyrethrums, extracted from plants. Inundative biological control using pathogens, parasitoids or predators that will not reproduce and survive effectively in the ecosystem. Large- scale or mass releases are made to react quickly to control the pest population.

Self-sustaining biological control includes:

Classical biological control. At its simplest, this is the introduction of natural enemies from the original range of the target species into new areas where the organism is invasive. Invasive alien species are often controlled in their indigenous range by their natural enemies, but are usually introduced into new environments without these population- controlling natural enemies. Freed of their parasitoids, parasites and predators, alien species often grow and/or reproduce more vigorously in the country of introduction. Natural enemies for introduction are selected on the basis of their host speciﬁcity, to minimize risks of sig- niﬁcant effects on non-target species. The aim is not the eradication of the invasive alien, but to reduce its competitiveness with native species, hence reducing its density, and its impact on the environment. Augmentation of natural enemies can be used during critical periods in the season (e.g. early in the season, when natural populations are low), or under pest outbreak conditions to try and achieve immediate control. Once released, these natural enemies can reproduce in the new environment, and an important part of the impact may be produced by the progeny in following generations. The control agent is reared or cul- tured in large numbers and released. Conservation biological control, such as habitat management, can be used to encourage populations of native predators and parasitoids, e.g. release/replant native alternative hosts and food resources.

Probably the most important of these for management of invasive alien chap01.qxd 12/11/04 10:56 AM Page 22

22 R. Wittenberg

species is classical biological control, which has been used extensively against weeds and insect pests with many examples of successful control (Greathead and Greathead, 1992; Julien and Griffiths, 1998). Like eradication, successful classical biological control has quite high ‘up-front’ costs, but relatively low recurring costs, and so has an attractive benefit:cost ratio. Conservation managers are coming to realize that this method, if used following modern protocols for careful screening of potential biological control agents, provides a safe and cost-efficient approach that can be used to solve many IAS problems. In comparison with other methods, classical biological control is, when successful, highly cost-effective, permanent and self-sustaining (Tisdell, 1990; Fowler, 1996). The main disadvantages are the lack of certainty about the level of control that will be achieved, the delays until the established agents achieve their full impact and the potential impact on non-target species, especially indigenous species closely related to the target, since the agents cannot be eradicated after release (Wajnberg et al., 2001). A risk assessment exercise should be carried out for any proposed biological control introduction, as for any introduction. The International Plant Protection Convention’s Code of Conduct for the Introduction of Exotic Biological Control Agents (FAO, 1996) provides guidance on procedures for introduction of biological control agents for those countries which do not already have suitable national protocols in place. In the end, the decision as to whether to proceed with a biological control introduction is a national one, taking into consideration the concerns of neighbouring countries, and based on the values and priorities of society at that time. The main problem is rather with individuals or organizations who try to do their own biological control, after they have learned about its potential, without any science-based assessment.

Habitat management

PRESCRIBED BURNING. In certain environments the practice of prescribed burning can change the vegetation cover in favour of native plant species, thereby decreasing population levels of weeds. Fire has been used quite frequently to manage IAS in the USA, for example to eradicate Australian pine (Casuarina equisetifolia L.) in pine forests and other ﬁre-tolerant communities in the USA. However, only trained and experienced people should undertake prescribed burning due to the many health and safety risks involved.

GRAZING. Habitat management with grazing mammals can be a suitable option to obtain the desired plant cover. This method works best where the plants that are to be preserved are adapted to grazing, i.e. they are either adapted to high populations of large herbivorous mammals or are prevalent in human-made habitats such as pastures and heathland and the invasive species are more palatable than the natives. On the other hand, unmanaged grazing often favours alien plants, as grazing can pref- chap01.qxd 12/11/04 10:56 AM Page 23

Invasive Alien Species 23

erentially remove native vegetation, leaving alien plants, especially toxic species, to grow under reduced competition. This twofold enhancement leads sooner or later to a monotypic stand of an alien plant, e.g. leafy spurge infestations in the USA.

HUNTING. Continuous hunting can be used to control exotic species, such as deer, originally introduced for hunting purposes (Mack et al., 2000). There are two approaches: commercial hunting principally for meat and recreational hunting. Both approaches can generate income for the landowner and/or the state. Some exotic species are both comparatively easy to hunt and favoured species for hunters, and so should be straight- forward to manage by hunting, but conversely more wary species or those less preferred by hunters are less likely to be effectively managed. Many other invasive species can be eaten or have edible fruits, which can be exploited for human consumption or as fodder for domesticated animals. In many parts of the world with high human density, invasive plants are esteemed also for their production of highly valued firewood or other uses. A high percentage of introduced fish and crustacean species make a good meal; thus recreational as well as industrial fishing is certainly helping to control these invasive populations. However, in the promotion of an alien species as a food resource lurks the danger of providing an incentive for individuals to spread the alien species to as yet uninfested areas or to breed them in captivity, from where they may eventually escape. This issue has to be evaluated on a case-by-case basis in order to estimate the potential danger and benefits.

Integrated pest management (IPM) A combination of methods, such as those described above, will often provide the most effective and acceptable control. For example, the ben- eﬁts of a chemical herbicide application will be much greater if followed by vegetation management and restoration efforts. As a tool to achieve the overall management goal, the regularly monitored, coordinated integra- tion of methods based on ecological research will almost always achieve the best results in managing an invasive species population. Since these control projects are dependent on so many variables, no general recom- mendation can be given for any taxonomic group. The strategy has to evolve based on the knowledge available on the invasive organism, the ecosystem invaded, the climatic conditions and other native and alien species in the same habitat and is case-speciﬁc. Figure 1.3 summarizes the different steps in dealing with alien species (see Wittenberg and Cock, 2001, for more details). The most useful tools to prevent introductions of invasive organisms are national and international regulations and their enforcement. Intentional introductions need to be screened for their invasive potential in a risk assessment process. Accidental introductions need to be minimized by effective quarantine inspection services. After establishment of an alien species, chap01.qxd 12/11/04 10:56 AM Page 24

24 R. Wittenberg

Fig. 1.3. Summary of steps to be taken into consideration when addressing alien species. Black bars mark the potential ﬁnal stages of introduced alien species. Diamonds symbolize important bifurcations and decision points. chap01.qxd 12/11/04 10:56 AM Page 25

Invasive Alien Species 25

eradication is the best management option, if feasible. If eradication is not possible, or fails, control measures need to be employed, which are often most effective when used in an integrated way.

Conclusions

Are IAS a biodiversity threat or hysteria? Despite the fact that sometimes xenophobia is pushed by using IAS and that media need to sell their stories, the case studies of some invaders show the importance of IAS and their extreme negative impact, including direct biodiversity losses and species extinctions. Due to the complexity of the problem, predictions about the potential of species becoming invasive and their impact are limited. This understanding strengthens the need to manage introductions. The costs of only a few IAS are skyrocketing, but these figures clearly show that prevention of IAS is paying off if it keeps out the small number of very invasive species. Considering these impacts, it seems to be clear that action is needed to stop this process. National or regional strategies against IAS will be a crucial step forward to deal with alien species. Being proactive is the key to being most cost-effective, but all methods need to be employed. Public support of IAS management is imperative; its basis is raising awareness of the problems. A pathways analysis is also crucial for successful prevention. The management of IAS is only one tool towards the goal of preserving and restoring ecosystems and ecosystem processes and the sustainable use of these resources. Results, successes and failures need to be widely disseminated through databases and information exchange to accelerate the success rate of IAS management. The only generalization about IAS is the fact that generalizations are bound to provide a wrong security for predictions due to exceptions to the rules and their general complexity. Although the comparison between Central Europe and other continents, such as Australia and North America, illustrates some major differences in importance of impacts of IAS, alien species need resources and are resources for others; thus they will be part of the ecosystem and will have an impact on the environment. It is debatable whether they do not have any impact or whether it is only difficult to measure these impacts. If Himalayan balsam (Impatiens glandulifera Royle) is growing densely packed along river banks, then it has replaced something else and might have some indirect effects on the water and soil. Anyway, while globalization is an important feature of human progress, the global introductions of some opportunistic species should be closely monitored and some management options employed to stop these processes. Species are threatened with extinction due to IAS, and the uniqueness of the places, the differences between the places, e.g. the specific flora and fauna, are being lost and the places become more similar to each other. This should be incentive enough to act. Global changes will affect the distribution of IAS in ways difficult to predict. Global warming is likely to help global chap01.qxd 12/11/04 10:56 AM Page 26

26 R. Wittenberg

swarming in Central Europe, with more Mediterranean species expanding to the north.

References

Bax, N.J. (1999) Eradicating a dreissenid from Australia. Dreissena 10(3), 1–5. BirdLife International (2000) Threatened Birds of the World. Lynx Edicions and BirdLife International, Barcelona, Spain and Cambridge, UK, 852 pp. Carlton, J.T. and Ruiz, G.M. (2004) Vector science and integrated vector management in bioinvasion ecology: conceptual frameworks. In: Mooney, H.A., McNeely, J., Neville, L.E., Schei, P.J. and Waage, J.K. (eds) Invasive Alien Species: Searching for Solutions. Island Press, Covelo, California. Chambers, S. (1989) Birds of New Zealand. Locality Guide. Arun Books Hamilton, Hamilton, New Zealand, 511 pp. Clout, M.N. and Lowe, S.J. (2000) Invasive species and environmental changes in New Zealand. In: Mooney, H.A. and Hobbs, R.J. (eds) Invasive Species in a Changing World. Island Press, Washington, DC, pp. 369–384. Courtenay, W.R. (1997) Nonindigenous ﬁshes. In: Simberloff, D., Schmitz, D.C. and Brown, T.C. (eds) Strangers in Paradise. Island Press, Washington, DC, pp. 109–122. Cronk, Q.C.B. and Fuller, L. (1995) Plant Invaders. Chapman & Hall, London, UK. Culver, S.C. and Kuris, A.M. (1999) The sabellid pest of abalone: the ﬁrst eradication of an established introduced marine bioinvader? In: Marine Bioinvasions: Proceedings of the First National Conference. MIT Press, Cambridge, Massachusetts, pp. 100–101. Enserink, M. (1999) Biological invaders sweep in. Science 285, 1834–1836. Falla, R.A., Sibson, R.B. and Turbott, E.G. (1993) Birds of New Zealand. HarperCollins Publishers, Auckland, 247 pp. Food and Agriculture Organization of the United Nations (FAO) (1996) International Standards for Phytosanitary Measures (ISPM) Publication No. 3, FAO, Rome, 23 pp. Fowler, S.V. (1996) Saving the gumwoods in St Helena. Aliens 4, 9. Galil, B. (1999) The ‘silver lining’ – the economic impact of Red Sea species in the Mediterranean. In: Marine Bioinvasions: Proceedings of the First National Conference. MIT Press, Cambridge, Massachusetts, pp. 265–267. Greathead, D.J. and Greathead, A.H. (1992) Biological control in insect pests by insect parasitoids and predators: the BIOCAT database. Biocontrol, News and Information 13, 61–68. Haas, G., Brunke, M. and Streit, B. (2002) Fast turnover in dominance of exotic species in the river Rhine determines biodiversity and ecosystem function: an affair between amphipods and mussels. In: Leppäkoski, E., Gollasch, S. and Olenin, S. (eds) Invasive Aquatic Species of Europe. Distribution, Impacts and Management. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 426–432. Heather, B.D. and Robertson, H.A. (1996) The Field Guide to the Birds of New Zealand. Penguin Books, Auckland, New Zealand, 432 pp. Hendrickson, O. (2002) Invasive alien species in Canadian forests. In: Claudi, R., Nantel, P. and Muckle-Jeffs, E. (eds) Alien Invaders in Canada’s Waters, Wetlands, and Forests. Canadian Forest Service, Ottawa, Canada, 320 pp. Heywood, V.H. (1989) Patterns, extents and modes of invasions by terrestrial plants. In: Drake, J.A., Mooney, H.A., di Castri, F., Groves, R.H., Kruger, F.J., Rejmánek, M. and Williamson, M. (eds) Biological Invasions. A Global Perspective. John Wiley & Sons, Chichester, UK, pp. 31–60. chap01.qxd 12/11/04 10:56 AM Page 27

Invasive Alien Species 27

Julien, M.H. and Grifﬁths, M.W. (eds) (1998) Biological Control of Weeds. A World Catalogue of Agents and Their Target Weeds, 4th edn. CAB International, Wallingford, UK. Karateyev, A.Y., Burlakova, L.E. and Padilla, D.K. (2002) Impacts of zebra mussels on aquatic communities and their role as ecosystem engineers. In: Leppäkoski, E., Gollasch, S. and Olenin, S. (eds) Invasive Aquatic Species of Europe. Distribution, Impacts and Management. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 433–446. Kowarik, I. (2002) Biologische invasionen in Deutschland: zur rolle nichteinheimicher pflanzen. In: Kowarik, I. and Starfinger, U. (eds) Biologische Invasionen: Herausforderung zum Handeln? Lentz-Druck, Berlin, Germany, pp. 5–24. Kowarik, I. (2003) Biologische Invasionen: Neophyten und Neozoen in Mitteleuropa. Verlag Eugen Ulmer, Stuttgart, Germany, 380 pp. Long, J.L. (2003) Introduced mammals of the World. Their History, Distribution and Inﬂuence. CAB International, Wallingford, UK, 589 pp. Loope, L.L. and Mueller-Dombois, D. (1989) Characteristics of invaded islands, with special reference to Hawaii. In: Drake, J.A., Mooney, H.A., di Castri, F., Groves, R.H., Kruger, F.J., Rejmánek, M. and Williamson, M. (eds) Biological Invasions. A Global Perspective. John Wiley & Sons, Chichester, UK, pp. 257–280. McCarthy, S.A. and Khambaty, F.M. (1994) International dissemination of epidemic Vibrio cholerae by cargo ship ballast and other non-potable waters. Applied Environmental Microbiology 60, 2597–2601. Mack, R.N., Simberloff, D., Lonsdale, W.M., Evans, H., Clout, M. and Bazzaz, F.A. (2000) Biotic invasions: causes, epidemiology, global consequences, and control. Ecological Applications 10, 689–710. Minchin, D. and Gollasch, S. (2002) Vectors – how exotics get around. In: Leppäkoski, E., Gollasch, S. and Olenin, S. (eds) Invasive Aquatic Species of Europe. Distribution, Impacts and Management. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 183–192. Panetta, D., Pheloung, P., Lonsdale, W.M., Jacobs, S., Mulvaney, M. and Wright, W. (1994) Screening Plants for Weediness: a Procedure for Assessing Species Proposed for Importation to Australia. A report commissioned by the Australian Weeds Committee, Canberra, Australia. Peck, S.B., Heraty, J., Landry, B. and Sinclair, B.J. (1998) Introduced insect fauna of an oceanic archipelago: the Galapagos Islands, Ecuador. American Entomologist, Winter, 218–237. Pimentel, D., Lach, L., Zuniga, R. and Morrison, D. (2000) Environmental and economic costs of nonindigenous species in the United States. BioScience 50, 53–65. Quammen, D. (1996) The Song of the Dodo. Island Biogeography in an Age of Extinctions. Hutchinson, London, UK, 702 pp. Ruesink, J.L., Parker, I.M., Groom, M.J. and Kareiva, P.M. (1995) Reducing the risks of non- indigenous species introductions – guilty until proven innocent. Bioscience 45(7), 465–477. Sharp, J.L. and Hallman, G.J. (1994) Quarantine Treatments for Pests of Food Plants. Westview Press, Boulder, Colorado. Sibley, C.G. and Monroe, B.L. (1990) Distribution and Taxonomy of Birds of the World. Yale University Press, New Haven, Connecticut. Simberloff, D. (1996) Impacts of introduced species in the United States. Consequences 2(2). Thiess, A., Schuster, R., Nöckler, K. and Mix, H. (2001) Helminthenfunde beim ein- heimischen Marderhund (Nyctereutes procyonoides, Gray, 1834). Berliner und Münchner Tierärztliche Wochenschrift 114, 273–276. chap01.qxd 12/11/04 10:56 AM Page 28

28 R. Wittenberg

Tisdell, C. (1990) Economic impact of biological control of weeds and insects. In: Mackauer, M., Ehler, L.E. and Roland, J. (eds) Critical Issues in Biological Control. Intercept Ltd, Andover, UK. US Congress (1993) Harmful Nonindigenous Species in the United States. Office of Technology Assessment, OTA-F-565, US Congress, Government Printing Office, Washington, DC. Van Wilgen, B.W., Little, P.R., Chapman, R.A., Gorgens, A.H.M., Willems, T. and Marais, C. (1997) The sustainable development of water resources: history, financial costs, and benefits of alien plant control programmes. South African Journal of Science 93, 404–411. Wajnberg, E., Scott, J.K. and Quimby, P.C. (eds) (2001) Evaluating Indirect Ecological Effects of Biological Control. CAB International, Wallingford, UK. Weidema, I. (ed.) (2000) Introduced Species in the Nordic Countries. Nordic Council of Ministers, Copenhagen, 242 pp. Williamson, M. (1996) Biological Invasions. Chapman & Hall, London, 244 pp. Wittenberg, R. and Cock, M.J.W. (eds) (2001) Invasive Alien Species: a Toolkit of Best Prevention and Management Practices. CAB International, Wallingford, UK, 228 pp. World Trade Organization (WTO) (1994) Agreement on the Application of Sanitary and Phytosanitary Measures. WTO, Geneva, Switzerland. Zettler, M.L., Daunys, D., Kotta, J. and Bick, A. (2002) History and success of an invasion into the Baltic Sea: the polychaete Marenzelleria cf. viridis, development and strategies. In: Leppäkoski, E., Gollasch, S. and Olenin, S. (eds) Invasive Aquatic Species of Europe. Distribution, Impacts and Management. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 66–75. chap02.qxd 24/11/04 1:36 PM Page 29

2 Monitoring of Western Corn Rootworm (Diabrotica virgifera virgifera LeConte) in Europe 1992–2003

JOZSEF KISS,1 C. RICHARD EDWARDS,2 HARALD K. BERGER,3 PETER CATE,3 MIRELA CEAN,4 SHARON CHEEK,5 JAQUES DERRON,6 HUSNIJA FESTIC´,7 LORENZO FURLAN,8 JASMINKA IGRC-BARC˘ IC´,9 IVANKA IVANOVA,10 WIEBE LAMMERS,11 VICTOR OMELYUTA,12 GABOR PRINCZINGER,13 PHILIPPE REYNAUD,14 IVAN SIVCEV,15 PETER SIVICEK,16 GREGOR UREK17 AND OTMAR VAHALA18 1Department of Plant Protection, Szent István University, Gödöllö, Hungary; 2Department of Entomology, Purdue University, Indiana, USA; 3Institute of Plant Health, Austrian Agency for Health and Food Safety, Vienna, Austria; 4Department of Entomology, Central Laboratory for Phytosanitary Quarantine, Bucharest, Romania; 5Central Science Laboratory, Department for Food, Environment and Rural Affairs, York, UK; 6Swiss Federal Research Station for Plant Production, Nyon, Switzerland; 7Faculty of Agriculture, University of Sarajevo, Sarajevo, Bosnia-Herzegovina; 8Department of Agronomy, Entomology, University of Padua, Legnaro, Italy; 9Department of Agricultural Zoology, Faculty of Agriculture, University of Zagreb, Zagreb, Croatia; 10Central Laboratory for Plant Quarantine, Soﬁa, Bulgaria; 11Plant Protection Service, Wageningen, The Netherlands; 12Institute of Plant Protection of Ukraine Academy of Agrarian Sciences, Kiev, Ukraine; 13Department of Plant and Soil Protection, Ministry of Agriculture and Rural Development, Budapest, Hungary; 14Entomology Unit, INRA, Montpellier, France; 15Institute for Plant Protection and Environment, Zemun, Serbia; 16Central Control and Testing Institute of Agriculture, Bratislava, Slovakia; 17Agricultural Institute of Slovenia, Ljubljana, Slovenia; 18Regional Division of the State Phytosanitary Administration, Brno, Czech Republic

Introduction

The ﬁrst monitoring of the western corn rootworm (WCR), Diabrotica virgifera virgifera LeConte, in Europe took place in 1992 when WCR beetles

30 J. Kiss et al.

and lodged maize plants, due to WCR larval feeding, were first detected in a maize field at Surcin, near Belgrade Airport in the Federal Republic of Yugoslavia (FRY)1 (Baca, 1993). From the first detection of WCR in Europe a broad international cooperative network developed among individuals, institutes and nations. One aspect of this cooperation has been the monitoring of the spread of this invasive pest species. What follows is an overview of this monitoring activity, looking at WCR spread, area of economic activity and changes in monitoring purposes over time.

Overview of WCR Monitoring in Europe

1992–1994

The monitoring of WCR in Serbia was carried out using visual inspection techniques of ﬁelds within and around the WCR-infested area in 1993 and 1994. The yearly spread of the adult population was estimated to be 50–70 km covering about 10,000 km2 in 1993 and 20,000 km2 in 1994 (Sivcev et al., 1994).

1995

The International Working Group on Ostrinia and Other Maize Pests (IWGO) of the International Organization for Biological Control (IOBC) organized the first WCR international workshop in Europe in cooperation with the United Nations Food and Agriculture Organization (FAO) and European Plant Protection Organization (EPPO) in March of 1995 (Berger, 1996). The result of this meeting was the establishment of the IWGO Diabrotica Subgroup, which called for a collaborative effort among all countries and organizations to address the issues associated with the presence of this pest in Europe, including monitoring. To initiate this activity, C.R. Edwards (Purdue University, West Lafayette, Indiana, USA) and H.K. Berger (Institute for Phytomedicine, Federal Office and Research Centre for Agriculture, Vienna, Austria) provided cucurbitacin-based vial traps to interested countries for monitoring WCR in 1995. Parallel to the use of cucurbitacin traps, visual inspection of maize plants, Multigard® yellow sticky traps (Scentry, Billings, Montana, USA) and sex pheromone traps (Tóth et al., 1996) were also employed. By the end of 1995, the spread of the WCR population in the FRY reached the borders of Hungary and Croatia. This spread to the north and west occurred very quickly in 1995. As a result, the first detections of WCR adults in Croatia (Igrc-Barc˘ic´ and Maceljsky, 1996) and in Hungary (Princzinger, 1996) occurred in 1995. Although WCR adults were not trapped in Romania in 1995, it is assumed that the WCR population

1As of 4 February 2003 the FRY has become Serbia and Montenegro. In this chapter, however, Serbia will be used to indicate the Serbian portion of the union and Montenegro the other. chap02.qxd 24/11/04 1:36 PM Page 31

Monitoring of Western Corn Rootworm in Europe 31

reached the Serbian/Romanian borders by 1995. WCR monitoring was also conducted in 1995 in Bulgaria, but no WCR adults were captured.

1996

In 1996, WCR monitoring using the same techniques as described for 1995 continued. WCR adults were captured in Romania (Vonica, 1996) in 1996. However, although monitoring was under way in Bulgaria, Poland, Slovenia and Ukraine, no detections of WCR adults were made in these countries in 1996.

1997

An FAO Technical Cooperation Programme (TCP) activity, which had as a part of its mission the establishment of a permanent monitoring network for WCR, was initiated in 1997 in Bosnia-Herzegovina, Croatia, Hungary and Romania. The establishment of this network allowed for the following: Determinations as to the occurrence and spread of WCR into uninfested regions. Ability to gauge WCR population fluctuations within and between years in the permanent monitoring sites. Hungarian pheromone traps (Csalomon PAL, for capture of WCR males) were used for monitoring the population spread of WCR in the TCP activity. This type of trap is attractive enough to allow for early detection of WCR males, even at low WCR population levels. In order to catch WCR females, non-baited Multigard® yellow sticky traps were used. Csalomon PAL pheromone traps were placed at the previous year’s WCR spread line and beyond, while taking into account the natural barriers or pathways that might increase or decrease WCR spread. Multigard® traps were placed in maize fields within the infested area. As a part of the FAO TCP monitoring activity, the first detection of WCR in Bosnia-Herzegovina (Festic´ et al., 1998) was recorded. As of the end of the 1997 growing season, a total of approximately 100,000 km2 were infested in Serbia, Bosnia-Herzegovina, Croatia, Hungary and Romania, with economic populations present only in Serbia (Edwards et al., 1998). Bulgaria, Slovenia and Ukraine reported that no WCR trap- pings occurred in their countries.

1998

The monitoring activity under the FAO TCP project continued in 1998. The spread of WCR was monitored as in 1997 through the use of chap02.qxd 24/11/04 1:36 PM Page 32

32 J. Kiss et al.

Csalomon PAL and Multigard® traps. During the season, the first detection of WCR was reported from Bulgaria (Ivanova, 2002), Montenegro (Sivcev and Hrncic, 1998) and Italy (Furlan et al., 1998). The Italian captures were the first case of the so-called ‘jumping spread’ of WCR, the occurrence of adults far from the actual spread line. Also, Italy was the first member state of the European Union to be infested by WCR. Although the monitoring area increased to several new countries in 1998, there were no WCR adults detected in Austria, Germany, Slovak Republic, Slovenia and Ukraine (Edwards et al., 1999).

1999

In early 1999, the FAO TCP project terminated and the regional activity for WCR monitoring was continued under the WCR Network established in 1999 by the Plant Protection Department, Szent Istvan University, Hungary. The WCR spread continued in the infested countries and approached the borders of Austria, Slovak Republic and Slovenia. In Italy, an eradication programme was established in the infested area near Venice (Furlan et al., 2002). No WCR adults were detected in 1999 in Albania, Austria, Czech Republic, France, Germany, Poland, Slovenia, Slovak Republic and Ukraine, although monitoring was taking place.

2000

Multigard® traps were replaced in the monitoring network in 2000 by another yellow sticky trap, the Pherocon® AM trap (TREĆE´ Incorporated, PO Box 6278, Salinas, California, USA). This switch was made due to the fact that more data were available from the USA on what the actual numbers caught on the Pherocon® AM traps meant in regard to the possibility of the presence of WCR economic populations. In 2000, first detections of WCR were observed in the Slovak Republic (Sivicek, 2000) and Switzerland (Bürki, 2000), and a new infestation in Italy in the Milan region far from the 1998 detection near Venice was noted. No WCR adults were detected in 2000 in Albania, Austria, Czech Republic, France, Germany, Slovenia and Ukraine. In addition to Serbia, economic larval damage to maize roots was first observed in Croatia, Hungary and Romania in 2000.

2001

Beginning in 2001, FAO provided support for the WCR Network activity, including monitoring, in the region through a letter of agreement with the Plant Protection Department, Szent Istvan University, Hungary (Kiss and Edwards, 2002; Kiss, 2003). This support enabled the regional coopera- chap02.qxd 24/11/04 1:36 PM Page 33

Monitoring of Western Corn Rootworm in Europe 33

tive effort to continue. The spread of WCR in Europe continued, with the first detection of WCR in the Ukraine in the border triangle of Hungary/Ukraine/Romania (Movchan et al., 2001; Omelyuta and Filatova, 2001; Sadlyak et al., 2001). As evidence of the presence of economic populations in 2001, WCR larval damage in the form of visible plant lodging was noted in parts of Serbia, Croatia, Hungary and Romania. The largest area of larval damage was observed in Hungary, where root damage ratings in southern counties reached the economic level of 3 (Hills and Peters 1–6 root damage rating scale) on 3058 ha.

2002

In 2002, the greatest spread occurred toward the north and east in areas where the plains of Hungary, Slovak Republic and Romania or river valleys of the Mures River in Romania favoured beetle spread. In Serbia, Croatia and Bosnia-Herzegovina, the spread of WCR was moderate, approximately 20–30 km. WCR reached Austria in 2002 (Cate, 2002). The jumping spread of WCR resulted in the ﬁrst detection of WCR in the Czech Republic (Rasovsky and Vahala, 2002) and France, near Paris (Reynaud, 2002). The jumping spread ability of WCR poses the risk of allowing for the establishment of WCR in regions with suitable conditions for WCR far beyond the established spread line and could delay detection unless monitoring activities are utilized in vulnerable regions. Larval damage to maize was observed in Croatia, Serbia, Hungary, Romania and Italy in 2002. This was the ﬁrst year of observed larval damage in Italy. This damage occurred in the Lombardy region. The greatest area of larval damage observed in 2002, however, occurred in Hungary and Croatia.

2003

The spread of WCR continued in most of the infested countries in 2003. However, limited spread occurred in Bosnia-Herzegovina, Serbia and Romania. This was possibly due to barriers, such as mountains, and unfavourable weather, such as hot and dry conditions. Signiﬁcant spread was observed in the extensive maize-growing area of northern Italy. The jumping spread of WCR resulted in new occurrences of the pest in areas of France and Switzerland removed from previous infestation areas. Also in 2003, new WCR infestations were recorded for Belgium (Food Agency of Belgium, 2003), The Netherlands (Lammers et al., 2004) and UK (Cheek et al., 2004), all near airports. In 2003, WCR was not detected in the following countries which had monitoring activities: Albania (Cota, 2004), Germany (Baufeld, 2004) and Greece (Tsitsipis et al., 2004). chap02.qxd 24/11/04 1:36 PM Page 34

34 J. Kiss et al.

As of October 2003, the European Union introduced emergency measures to prevent the spread of WCR within the European Community (Commission Decision 2003/766/EC). Also, the Central and Eastern European countries of Bosnia-Herzegovina, Bulgaria, Croatia, Hungary, Romania, Serbia and Slovak Republic initiated a new regional activity for Integrated Pest Management for WCR in Central and Eastern Europe as an FAO Trust Fund Project (GTFS/RER/017/ITA; donor: Government of Italy). The activity is designed:

to protect maize production in Central Europe from losses caused by WCR through the development and implementation of IPM strategies by farmers, based on sound understanding of local agro-ecosystems and protection of biodiversity as the main element of sustainability of agricultural production.

Although not the major area of emphasis, the continuation of in-country monitoring is included in this project.

Monitoring as a Tool for Multiple Purposes

The monitoring of adult WCR populations by European countries has allowed for the rapid detection and determination as to the spread of this invasive pest species since the insect was ﬁrst observed in Serbia. Monitoring is still important in countries with no or isolated infestations, as well as in countries where WCR population is still spreading. Thus, in these areas, the monitoring has served as a means for detection, and in some instances, containment and control or, in the case of new appear- ances outside the regular spread line, for eradication purposes. These monitoring data can provide the necessary information for making quarantine and other regulatory decisions. Also, the monitoring of WCR populations in areas where the pest has been established for some time provides useful information on expected or potential economic damage by larvae and adults. Additionally, monitoring activities can contribute to raising the awareness of the dangers posed by WCR and stir individual interest in learning about pest biology and observing its developmental and behavioural characteristics under local conditions. If farmers are involved in monitoring, the method of learning by doing becomes an essential way to improve a farmer’s capacity to address issues related to WCR and develop integrated pest management (IPM) strategies to mitigate its impact on maize production. In areas with a well-established WCR population, the monitoring/ population check activity will be rather an information source for verify- ing the impact of farming practices and of IPM options on the WCR population level. chap02.qxd 24/11/04 1:36 PM Page 35

Monitoring of Western Corn Rootworm in Europe 35

Some Characteristics of the Spread of WCR in Europe

Although it has been reported that WCR probably arrived in Serbia in 1989 or 1990 (Sivcev et al., 1996), it is likely that the beetle was introduced into this area in the early to mid-1980s (Edwards et al., 1998). Based on the detection of the first WCR adults in nearby countries and on evidence of the first economic larval damage, it is assumed that a time period of about 5–6 years is needed for a population to build up to economically damaging levels. Therefore, it is estimated that the first beetles arrived in Serbia around the mid-1980s (Kiss et al., 2001), as had been surmized by Edwards et al. (1998). The total area infested in Europe by WCR reached approximately 311,000 km2 by the end of the 2003 maize growing season (Table 2.1), with about 70,000 km2 of land area with economic adult activity (Table 2.2) and approximately 97 km2 of larval damage (Table 2.3). The spread distance of the WCR population varies by year and from region to region. Within an area, a population may spread up to a distance of a few kilometres, as observed in Hungary in 1998, or 70–80 km, as seen in Hungary in 1999 and in Romania in 2001. The greatest spread of the WCR population has occurred in the Carpathian basin towards northern and eastern areas of the region, where the plains of Hungary, Slovak Republic and

Table 2.1. Spread of western corn rootworm, Diabrotica virgifera virgifera LeConte, in Europe, 1998–2003 (area infested in km2).

Country 1998 1999 2000 2001 2002 2003

Austria ––––9903,000 Belgium –––––+ Bosnia-Herzegovina 1,500 10,000 12,000 13,000 15,000 16,000 Bulgaria 200 1,000 3,000 7,000 8,500 12,000 Croatia 10,500 12,750 14,000 15,500 19,000 23,500 Czech Republic ––––++ France ––––++ Hungary 30,000 40,000 50,000 70,000 78,000 93,000 Italy 11 3 5 4,000 7,000 10,000 The Netherlands –––––+ Romania 12,000 14,000 35,000 60,000 65,000 65,000 Serbia and Montenegroa 61,400 63,000 67,500 72,400 72,500 73,000 Slovak Republic – – 500 6,300 9,531 10,062 Switzerland – – + 740 1,654 1,654 Ukraine – – – 60 575 3,000 UK + Total 115,600 140,750 182,000 249,000 277,750 310,216

aFormerly the Federal Republic of Yugoslavia. +, detected only or non-contiguous population. chap02.qxd 24/11/04 1:36 PM Page 36

36 J. Kiss et al.

Table 2.2. Area of western corn rootworm, Diabrotica virgifera virgifera LeConte, economic adult activitya in Europe, 1998–2003 (km2).

Country 1998 1999 2000 2001 2002 2003

Austria ––––PNEPNE Belgium –––––PNE Bosnia-Herzegovina PNE PNE PNE PNE PNE PNE Bulgaria PNE PNE PNE PNE PNE PNE Croatia PNE PNE 2 4,000 5,000 7,500 Czech Republic ––––PNEPNE France ––––PNEPNE Hungary PNE PNE 2 10,000 20,000 30,000 Italy PNE PNE PNE PNE 30 PNE The Netherlands –––––PNE Romania PNE PNE 5 11,000 11,000 13,000 Serbia and Montenegrob 14,000 16,000 26,500 26,500 28,000 20,000 Slovak Republic – – PNE PNE PNE PNE Switzerland – – PNE PNE PNE PNE Ukraine – – – PNE PNE PNE United Kingdom PNE Total 14,000 16,000 26,509 51,500 64,030 70,500

aEconomic adult activity refers to a WCR adult population that can result in economic larval damage to maize in the subsequent year. bFormerly the Federal Republic of Yugoslavia. PNE, present, but no economic damage reported.

Table 2.3. Total area of western corn rootworm, Diabrotica virgifera virgifera LeConte, economic larval damage (over 3 of Iowa 1–6 scale) in maize ﬁelds in Europe, 1998–2003 (km2).

Country 1998 1999 2000 2001 2002 2003

Croatia PNE PNE + + 1 6 Hungary PNE PNE 10 31 54 60 Italy PNE PNE PNE PNE + PNE Romania PNE PNE 1111 Serbia and Montenegroa 455 305 500 10 10 30 Total 455 305 511 42 66 97

aFormerly the Federal Republic of Yugoslavia. PNE, present, but no economic damage reported; +, economic larval damage on some ﬁelds.

Romania or river valleys, such as the Mures River in Romania, favour beetle spread. chap02.qxd 24/11/04 1:36 PM Page 37

Monitoring of Western Corn Rootworm in Europe 37

Experience in Europe has shown that the larger the size of the infested area the greater the possibility of a jumping-spread movement of WCR beyond the actual spread line. This has occurred in such areas as the Lombardy region in northern Italy, along with the canton Ticino in southern Switzerland in 2000, north-western Bosnia-Herzegovina in 2001, in the Czech Republic and near Paris, France, in 2002 and the UK, Belgium and The Netherlands in 2003. The jumping-spread ability of WCR poses the risk of establishment of WCR in regions with suitable conditions for WCR far beyond the established spread line and thus could result in detection delays anywhere in Europe where maize is grown and monitoring is not carried out. The permanent monitoring sites located in FAO Network partner countries have allowed for the measurement of population ﬂuctuations over the years. It has been shown that WCR populations at permanent monitoring sites may increase by two to three times when compared to the previous year, but may decrease as well. Decreases are most probably the effect of the cumulative consequences of unfavourable weather conditions for WCR, such as low precipitation and high temperatures, and agronomic practices, such as growing maize in crop rotation with non- host crops. Transportation means, packaging materials (especially of yellow colour) and light sources in areas where WCR is found may attract WCR adults and can contribute to their being moved from region to region.

References

Baca, F. (1993) New member of the harmful entomofauna of Yugoslavia Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae). IWGO News Letter 13(1–2), 21–22. Baufeld, P. (2004) Monitoring of western corn rootworm in Germany 2003. IWGO News Letter 25(1), 22. Berger, H.K. (1996) Multi-country coordination efforts to deal with the western corn rootworm (Diabrotica virgifera virgifera). IWGO News Letter 16(1), 26–29. Bürki, H.M. (2000) Maisschaedling in der Schweiz schon aufgetaucht. Mediendienst 2487, 19 October. http://www.lid.ch/portal/DesktopDefault.aspx?tabindex=5&tabid= 747&langid=1 Cate, P. (2002) The confirmation of WCR (Diabrotica virgifera virgifera LeConte) in Austria: occurrence, expansion and future prospects. IWGO News Letter 23(2), 18–19. Cheek, S., Baker, R.H.A., Cannon, R.J.C., Macleod, A., Agallou, E. and Bartlett, P. (2004) First finding of the western corn rootworm in the UK. IWGO News Letter 25(1), 21. Cota, E. (2004) Monitoring of western corn rootworm (Diabrotica virgifera virgifera LeConte) in Albania 2003. IWGO News Letter 25(1), 22. Edwards, C.R., Igrc-Barc˘ic´, J., Berger, H.K., Festic´, H., Kiss, J., Princzinger, G., Schulten, G. and Vonica, I. (1998) Overview of the FAO western corn rootworm management program for Central Europe. Pflanzenschutzberichte 57(2), 3–14. Edwards, C.R., Igrc-Barc˘ic´, J., Berberovic, H., Berger, H.K., Festic´, H., Kiss, J., Princzinger, G., Schulten, G.G.M. and Vonica, I. (1999) Results of the 1997–1998 Multi-Country chap02.qxd 24/11/04 1:36 PM Page 38

38 J. Kiss et al.

FAO Activity on Containment and Control of the Western Corn Rootworm, Diabrotica virgifera virgifera LeConte, in Central Europe. Acta Phytopathologica et Entomologica Hungarica 34(4), 373–386. Festic´, H., Faginovic, M., Berberovic, H. and Seratlic-Turkic, A. (1998) Die Ergebnisse des Warndienstes beim der Western Corn Rootworm Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae) in Bosnia-Hercegovina. Pflanzenschutzberichte 57(2), 28–32. Food Agency of Belgium (2003) Communiqué de presse 2003–09–05. Un coléoptère exo- tique menace nos champs de maïs. Chrysomèle des racines du maïs. http://www.favv- afsca.fgov.be Furlan, L., Vettorazzo, M., Ortez, A. and Frausin, C. (1998) Diabrotica virgifera virgifera has already arrived in Italy. Informatore Fitopatologico 12, 43–44. Furlan, L., Vettorazzo, M., Ortez, A. and Frausin, C. (2002) Diabrotica virgifera virgifera LeConte: what has been done and what will be done in Italy? Acta Phytopathologica et Entomologica Hungarica 37(1–3), 169–173. Igrc-Barc˘ic´, J. and Maceljsky, M. (1996) Monitoring Diabrotica virgifera virgifera LeConte in Croatia in 1995. IWGO News Letter 16(1), 11–13. Ivanova, I.E. (2002) The first occurrence of Diabrotica virgifera virgifera LeConte in Bulgaria. Acta Phytopathologica et Entomologica Hungarica 37(1–3), 155–157. Kiss, J. (2003) Final report on Western Corn Rootworm (WCR), Diabrotica virgifera virgifera LeConte, Network Activity. UN FAO Report No. PR 21261. Kiss, J. and Edwards, C.R. (2002) Final Report on Western Corn Rootworm (WCR), Diabrotica virgifera virgifera LeConte, Network Activity. UN FAO Report No. PR 19713. Kiss, J., Edwards, C.R., Allara, M., Sivcev, I., Igrc-Barc˘ic´, J., Festic´, H., Ivanova, I., Princzinger, G., Sivcev, I., Sivicek, P. and Rosca, I. (2001) A 2001 update on the western corn rootworm, Diabrotica virgifera virgifera LeConte, in Europe. In: Proceedings Book of the XXI IWGO Conference and VIII Diabrotica Subgroup Meeting. Veneto Agricoltura, Legnaro, Italy, pp. 83–87. Lammers, W., Meijer, A. and Stigter, H. (2003) First finding of the western corn rootworm in the Netherlands. In: Abstracts of the 10th IWGO Diabrotica Subgroup Meeting, 14–16 January 2004, Engelberg, Switzerland, p. 30. Movchan, O.M., Melnyk, P.O. and Konstantynova, N.A. (2001) The problem of corn rootworm – Diabrotica virgifera virgifera LeConte – in Ukraine. IWGO News Letter 22(1–2), 33–35. Omelyuta, V. and Filatova, N. (2001) Western corn rootworm (Diabrotica virgifera virgifera LeConte) in Ukraine: reality and outlook. IWGO News Letter 22(1–2), 35–37. Princzinger, G. (1996) Monitoring of western corn rootworm (Diabrotica virgifera virgifera LeConte) in Hungary 1995. IWGO News Letter 16(1), 7–11. Rasovsky, V. and Vahala, O. (2002) Monitoring and first record of Diabrotica virgifera virgifera LeConte (Coleoptera, Chrysomelidae) from the Czech Republic. Poster at 9th IWGO Diabrotica Subgroup Meeting, 2–4 November, 2002, Belgrade. Reynaud, P. (2002) First occurrence of Diabrotica virgifera virgifera in France. IWGO News Letter 23(2), 20–21. Sadlyak, A.M., Sikura, A.J. and Yakovets, P.I. (2001) Appearance of Diabrotica virgifera virgifera LeConte on a boundary of Ukraine. IWGO News Letter 22(1–2), 35. Sivcev, I. and Hrncic, S. (1998) Monitoring of Diabrotica virgifera virgifera LeConte in Yugoslavia in 1998. IWGO News Letter 18(2), 12–13. Sivcev, I., Manojlovic, B., Krnjajic, S., Dimic, N., Draganic, M., Baca, F., Kaitovic, Z., Sekulic, R. and Keresi, T. (1994) Distribution and harmful effect of Diabrotica virgifera LeConte (Coleoptera, Chrysomelidae), a new maize pest in Yugoslavia. Zastita Bilja 45(1), 19–26. chap02.qxd 24/11/04 1:36 PM Page 39

Monitoring of Western Corn Rootworm in Europe 39

Sivcev, I., Manojlovic, M., Baca, F., Sekulic, R., Camprag, D. and Keresi, T. (1996) Occurrence of Diabrotica virgifera virgifera LeConte in Yugoslavia in 1995. IWGO News Letter 16(1), 20–25. Sivicek, P. (2000) Report on survey of western corn rootworm (Diabrotica virgifera virgifera LeConte) in the Slovak Republic 2000. IWGO News Letter 21(1–2), 37–38. Tóth, M., Toth, V., Ujvary, I., Sivcev, I., Manojlovic, B. and Ilovai, Z. (1996) Sex pheromone trapping of Diabrotica virgifera LeConte in Central Europe. Növényvédelem 32(9), 447–452 Tsitsipis, J., Zarpas, K.D., Tóth, M., Polymerou, V., Lilis, K. and Mouroutoglou, C. (2004) Monitoring of Diabrotica virgifera virgifera LeConte by pheromone traps in Greece. IWGO News Letter 25(1), 23. Vonica, I. (1996) Monitoring for Diabrotica virgifera in Romania. IWGO News Letter 16(2), 15–16. chap02.qxd 24/11/04 1:36 PM Page 40 chap03.qxd 15/11/04 2:19 PM Page 41

3 A Synopsis of the Nutritional Ecology of Larvae and Adults of Diabrotica virgifera virgifera (Le Conte) in the New and Old World – Nouvelle Cuisine for the Invasive Maize Pest Diabrotica virgifera virgifera in Europe?

JOACHIM MOESER1 AND BRUCE E. HIBBARD2 1Institute for Plant Pathology and Plant Protection, Georg-August University, Göttingen, Germany; 2USDA-ARS, Plant Genetics Research Unit, University of Missouri, Columbia, Missouri, USA

Introduction

Basic biological facts set the frame for the management options of pest species. To manage a pest insect it is vital to understand its basic biology. Perhaps the most fundamental and also most crucial part is the nutritional ecology of an insect, because that is what makes it a pest species from the human point of view in the ﬁrst place. The necessity for this particular kind of knowledge is even more urgent when pest species are invading new areas threatening human production systems, especially when eradication is no longer feasible (Simberloff, 2003). Are their feeding habits similar to those in their area of origin? Does the new environment facilitate or impede the proliferation of the pest by supplying more or less suitable food? In the case of the invasion of Europe by Diabrotica virgifera virgifera LeConte (western corn rootworm), one could argue that by invading Europe it is merely catching up with the distribution of its main host plant, maize. While this is true to a certain extent, questions remain on what limits or promotes this invasion, which is taking place with an alarming speed (EPPO, 2003). This chapter will focus on the knowledge of D. v. virgifera nutritional ecology gained in its area of ﬁrst appearance as a major maize pest (USA)

42 J. Moeser and B.E. Hibbard

and will be supplemented with the latest research from its expansion area in Europe. Two major objectives of this chapter are: (i) To discuss the nutritional ecology of D. v. virgifera against the background of resistance management plans for genetically modified (GM) maize and the developing resistance to insecticides and crop rotation, which have become major issues in US agriculture during the last decade; and (ii) to provide an orientation for European researchers to understand the patterns behind the invasion process and for European maize producers to facilitate the implementation of integrated pest management (IPM) strategies by summarizing the knowledge gathered on D. v. virgifera until now. To understand the factors that influence the invasion process in Europe it is useful to consider how the invasion of the USA developed during the last century. The interactions with today’s main host plant, Zea mays L., were not always as simple and were not the only host–plant interaction in the life of D. v. virgifera as may appear in the light of the intensive agricultural systems of the 20th century. Unfortunately, it is not clear what species or group of species served as the ancestral host of D. v. virgifera. Without this information, host relationships today cannot, with certainty, be placed in their proper evolutionary context. Branson and Krysan (1981) suggested that ‘D. v. virgifera became a specialist on corn in the tropics or subtropics and “followed” the diffusion of corn into the temperate United States’. Krysan and Smith (1987) stated that ‘it is reasonable to conclude that the presence of D. v. virgifera in the USA does not predate the presence of corn’. These thoughts represent the current dogma. Its primary rationale is that maize is by far the best host found to date for D. v. virgifera (Branson and Ortman, 1967a,b, 1970; Clark and Hibbard, 2004; Oyediran et al., 2004). However, being the best host does not necessarily correlate to original host. In fact, Diabrotica longicornis (Say) does not feed on maize when feral, but actually its larvae develop on maize at a rate equal to or even faster than larvae of the northern corn rootworm, a sibling species that specializes on maize (Golden and Meinke, 1991). LeConte (1868) first collected D. v. virgifera in western Kansas. Branson and Krysan (1981) and Krysan and Smith (1987) ignored or were unaware of the fact that maize was apparently not grown in the area of western Kansas at that time. According to Goodman (1987), ‘At the time of European colonization of the New World, maize was being grown from southern Canada to central Chile, although little was grown in the grassy plains or savannas of the central USA and northern Argentina.’ The distribution map of Weatherwax (1954) indicates that indeed maize was not grown in western Kansas at that time. What plant species or group of species were maintaining D. v. virgifera larval populations in the 1860s in western Kansas? Are these and related species still of importance in the USA and Europe? These are important questions to which we do not yet know the answers. chap03.qxd 15/11/04 2:19 PM Page 43

Nutritional Ecology of Larvae and Adults 43

Adaptation of and Invasion of D. v. virgifera in US Maize Production

The first report of D. v. virgifera as a pest of maize was in 1909, not far from where LeConte had first collected the species approximately 50 years earlier (Gillette, 1912). From 1909 to 1948 the insect spread eastward at an average rate of about 19 km per year (Metcalf, 1986). Large- scale applications of soil insecticides were first made for D. v. virgifera control in 1949 and more than 700,000 ha were being treated with insecticides by 1954 (Ball and Weekman, 1962). However, resistance to the broadcast soil insecticides being used at the time was noted as early as 1959 (Ball and Weekman, 1962), and the insecticide-resistant strain spread eastward at the even faster rate of 112 km/year (Metcalf, 1983). Since maize was not present in the region of western Kansas where D. v. virgifera was first collected (Weatherwax, 1954), we believe that it is likely that the beetles first collected from maize in 1909 (Gillette, 1912) probably switched hosts from native grasses in the area. However, the large populations required to expand the range of this species from what was probably a relatively small area to much of the maize production areas of the USA today resulted from beetle production from maize. Physiological and behavioural adaptation to maize may have developed over the decades. D. v. virgifera adult production from non-maize plants from outside or within maize production fields has not been documented except in cases where non-maize hosts were artificially infested.

Host Plant Phenology

D. v. virgifera adults do not lay their eggs on a host plant. The eggs are laid in the soil in late autumn, and the larvae emerge the following spring. Host location is then carried out underground by the neonate larvae. The time between egg hatch and larval establishment is one of the more vulnerable times in the life cycle of these insects (see Toepfer and Kuhlmann, Chapter 5, this volume). If host establishment is delayed for as little as 24 h, survival to the adult stage is significantly reduced (Branson, 1989). Timing of egg hatch and root availability is critical. Delayed planting results in reduced root damage and adult emergence, and peak emergence time is also delayed (Musick et al., 1980), presumably because early hatching larvae do not become established. Root toughness may also be a factor. Stavisky and Davis (1997) evaluated the effects of maize maturity class on larval development and adult emergence in a field evaluation in New York. They found that adult emergence peaked later, was more spread out and was higher on later maturing hybrids, probably because new nodes of roots developed later into the season on later maturing hybrids. Even more interestingly, they found that occurrence of first-instar larvae ended earlier in the early maturing hybrid. Their data suggest that western corn rootworm larvae not only prefer newly developed roots, as was already known (Apple and Patel, chap03.qxd 15/11/04 2:19 PM Page 44

44 J. Moeser and B.E. Hibbard

1963; Strnad and Bergman, 1987), but that D. v. virgifera larvae may require ‘young’ roots for establishment. Y.M. Schweikert and B.E. Hibbard (unpublished data) conducted greenhouse and field trials to determine which root phenologies were optimal for the establishment and development of D. v. virgifera larvae. In the field, plants of serial phenologies, beginning at planting and continuing weekly from V11 to V13 were infested with 0, 200 and 600 viable D. v. virgifera eggs, respectively. In greenhouse trials, 30 neonate D. v. virgifera larvae were allowed to feed on maize roots every week for 8 (2001 greenhouse) to 10 (2002 greenhouse) weeks, starting at the V2 stage. For all trials except the 2001 greenhouse (root damage omitted), two larval recovery samples, root damage and adult emergence data were collected. Plant damage gradually lessened towards later infestation dates, as expected when larger root systems are attacked. Larval recovery did not change significantly between infestation dates, appearing to contradict the data of Stavitski and Davis (1997). However, adult emergence was greatly reduced in the last infestation dates to the point where no adults were recovered on the last one or two infestation dates for several of the trials, despite a large number of larvae being recovered. The Missouri data indicate that D. v. virgifera can establish on plants with a later phenology, but that nutrition is not sufficient for adult production. P.G. Chege, T.L. Clark, and B.E. Hibbard (unpublished data) collected similar data from several of the better alternative hosts from Clark and Hibbard (2004), Digitaria sanguinalis, Setaria faberi, Panicum capillare, Eriochloa gracilis and Setaria viridis, with similar conclusions.

Breeding for Plant Research

Traditional plant resistance breeding programmes (Painter, 1951) divided plant resistance into three mechanisms: tolerance, non-preference and antibiosis. Insect resistance has been deﬁned as any inherited characteristic of a host plant that lessens the effects of insect feeding (Pedigo, 1989). These inherited characteristics may involve an increase or decrease in the expression of morphological and/or biochemical constituents. Resistant plants, by Painter’s (1951) deﬁnition, must inhibit the host-selection process of an insect, deter feeding, be toxic or otherwise be capable of tolerating insect feeding by maintaining growth and production despite damage done by insect feeding. The techniques used in plant resistance work with corn rootworms was thoroughly reviewed by Branson (1986) and a number of these techniques were directly compared by Knutson et al. (1999). Recent advances in techniques used in plant resistance work include the implement developed by Praiswater et al. (1997) for dislodging root systems from the soil, a new, more intuitive rating system developed by Oleson (1999), and a system for bar coding plants developed by B.E. Hibbard (unpublished). There have been several serious research programmes aimed at iden- chap03.qxd 15/11/04 2:19 PM Page 45

Nutritional Ecology of Larvae and Adults 45

tifying and developing sources of maize with resistance to corn rootworm larval feeding. As far back as the 1930s and 1940s, J.H. Bigger was evaluating maize strains in Illinois for resistance to feeding by larvae of the southern corn rootworm, Diabrotica undecimpunctata howardi (Bigger et al., 1938, 1941). They documented that most open-pollinated varieties were much more susceptible to damage than hybrids and that certain hybrids were more susceptible to damage than others. Iowa State University has had a series of researchers working on corn rootworm resistance. Although most plant resistance publications were from the 1970s, plant resistance work continues there today. Hills and Peters (1971) developed a system for rating maize roots for damage, which is still routinely used today. Rogers et al. (1975, 1976a,b, 1977) evaluated commercial germplasm, a root-pulling technique and yield relationships and calculated expected gains based upon different selection techniques. According to Rogers et al. (1977), selection for reduced damage results in little or no improvement, but selection for root size and secondary root development would result in improved populations, especially with increased replication and locations. Russell et al. (1971) released lines B64, B67 and B69, which all had tolerance to corn rootworm larval feeding. Owens et al. (1974) evaluated 221 inbred lines developed from an Iowa Stiff Stalk Synthetic background for root damage, size, regrowth and lodging. Genotypic, phenotypic and error correlations indicated that selection for root size would probably result in reduced damage and lodging as well as increased root regrowth. Finally, Wilson and Peters (1973) and Wilson et al. (1995) evaluated exotic germplasm for resistance to D. v. virgifera larval feeding. One of the most productive groups to date working on corn rootworm plant resistance has been the US Department of Agriculture Agricultural Research Service (USDA-ARS) Northern Grain Insects Research Laboratory in Brookings, South Dakota. Research has included evaluation of species and genera related to maize in a search for resistance sources (Branson et al., 1969; Branson, 1971; Branson and Guss, 1972; Branson and Reyes Rueda, 1983), evaluation of exotic maize germplasm (Branson et al., 1986), evaluation of Corn Belt germplasm (Ortman and Fitzgerald, 1964; Fitzgerald and Ortman, 1965; Riedell and Evenson, 1993), technique development (Fitzgerald and Ortman, 1964; Ortman and Gerloff, 1970; Ortman et al., 1974; Ortman and Branson, 1976; Branson et al., 1981; Riedell, 1989), evaluation of the mechanism of resistance (Branson et al., 1982a; Kahler et al., 1985a), identiﬁcation of (Branson et al., 1983) and release of (Kahler et al., 1985b) resistant germplasm sources. More recently, a research group from the University of Ottawa demonstrated that maize lines with extremely high levels of hydroxamic acids have low levels of antibiosis resistance (Xie et al., 1990, 1992a,b; Assabgui et al., 1993, 1995a,b; Arnason et al., 1997). However, some of the lines that showed resistance in Canada were quite susceptible in Missouri (D.J. Moellenbeck, personal communication). Since the establishment of D. v. virgifera in Europe, breeders in Croatia have collaborated with US scientists to evaluate and develop chap03.qxd 15/11/04 2:19 PM Page 46

46 J. Moeser and B.E. Hibbard

maize lines with resistance (Ivezic´ et al., 2003). This ongoing research programme appears to be making good progress in comparison to numerous breeding programmes over the years in the USA. Finally a USDA-ARS group based at the University of Missouri has been focusing heavily on native plant resistance since 1992. Initially, the focus was on maize relatives and technique development (Moellenbeck et al., 1994, 1995; Praisewater et al., 1997; Hibbard et al., 1999a; Knutson et al., 1999). The focus then shifted more to the identification and improvement of resistance sources (Hibbard et al., 1999b; Abel et al., 2000). More recently, the focus has been on the release of resistant germplasm (probably early in 2004) and identifying regions of the maize genome associated with corn rootworm resistance using quantitative trait loci (QTL) techniques (D.B. Willmot, B.E. Hibbard and others, unpublished data). Branson et al. (1983) and Hibbard et al. (1999b) documented reduced damage in certain maize germplasm lines. B.E. Hibbard, Willmot and L.L. Darrah (unpublished data) have documented even greater levels of reduced damage more recently. However, the mechanism of resistance in all of these lines remains unclear at this time. The best sources of resistance definitely have reduced damage, but adult emergence is similar between susceptible and resistant lines. When hybrids from the 1960s era were compared to those from the 1970s and 1980s in the same field, the 1980s era hybrids suffered less lodging and had greater vertical pull resistance than 1960s era hybrids (Riedell and Evenson, 1993). Despite significant research efforts over several decades, outlined above, élite modern hybrids are still susceptible to corn rootworm larval feeding and there was no difference in actual root damage ratings between hybrids of different eras, indicating that modern hybrids were more tolerant of damage, but did not have antibiosis resistance resulting in reduced larval feeding (Riedell and Evanson, 1993).

Food Conversion Efﬁciency: Plant-Insect Interactions Revisited

The actual effect of certain food items on D. v. virgifera larval performance has rarely been studied in detail due to methodological problems. Feeding trials with larvae have mainly been devoted to the search for tolerance or antibiosis resistance. The practice of taking larval survival as a measure for host plant suitability (Branson and Ortman, 1967a, 1970) has proved insufﬁcient in the past to comprehend the interactions between D. v. virgifera and its host plants in detail. Recent studies (Clark and Hibbard, 2004; Oyediran et al., 2004) have included parameters such as larval growth but have neglected the plant portion of this interaction, by measuring resistance as damage done to the root system (= amount of food eaten by D. v. virgifera). To measure both sides of the interaction between different maize varieties, land races or alternative host plants and D. v. virgifera larvae, the development of a new methodology was necessary. Previously existing methods for measuring food conversion efﬁciency chap03.qxd 15/11/04 2:19 PM Page 47

Nutritional Ecology of Larvae and Adults 47

(Waldbauer, 1968) were modified to suit subterranean insect larvae (Moeser, 2003). A bioassay consisted of a single second-instar larva and a cut portion of secondary root from a given plant, which were tested in a no-choice trial for 6 days. The initial and final weights were recorded for the larva as well as for the food item. The weight gain of the larva and the amount of ingested food were measured and the resulting index for efficiency of conversion of ingested food (ECI) was calculated by dividing the amount of ingested food by the weight gain of the larva. Because the ECI is known to be problematic due to its biased data (Raubenheimer and Simpson, 1992; Horton and Redak, 1993), an additional analysis of covariance (ANCOVA) was performed using the initial fresh weight of the larva as the covariate to correct for an eventual bias due to differences in larval weight at the beginning of the experiments. Significant differences in larval weight gain could be shown, as well as differences with regard to the amount of ingested food when fed with maize roots. This no-choice design showed that feeding on the same spatial portion of maize roots from different varieties resulted in highly variable food conversion efficiencies between maize lines. Some varieties showed little feeding but good larval weight gain while others were fed heavily upon with even a decrease in larval weight. This is another indicator that maize roots are a more heterogeneous food source than assumed previously (Strnad and Bergman, 1987). The future search for resistant varieties may be supplemented by deploying these techniques of measuring direct effects of different varieties on D. v. virgifera larval performance. If the amount of beetles hatching from susceptible and resistant varieties is of the same magnitude (Branson et al., 1983; B.E. Hibbard et al., unpublished data), the resistant varieties may provide food of higher quality, which allows the larvae to feed less but with a high food conversion efficiency, thus enabling them to gain as much weight as when feeding more on low-quality food. The identification of possible resistance mechanisms is ongoing. As Moeser and Vidal (2004a) showed, basic plant compounds such as phytosterols have significant effects on host plant suitability for D. v. virgifera larvae. Phytosterols are a heterogeneous group of isoprenoid derivates commonly found in plant membranes. They are essential for herbivorous insects because they cannot be synthesized de novo by most herbivores (Svoboda and Thompson, 1985). They serve as precursors for a variety of insect hormones including the moulting hormone ecdysone (Svoboda, 1984). A decreased total amount of sterols or an altered composition of individual sterols may therefore be one factor contributing to an increased resistance level (Bodnaryk et al., 1997).

Larval Feeding on Alternative Hosts

Of 41 grasses and 27 broadleaf species evaluated in early studies, larvae survived in Petri dishes for 10 days on only 18 grass species, giving insight into ‘grasses only’ as larval hosts (Branson and Ortman, 1967a, chap03.qxd 15/11/04 2:19 PM Page 48

48 J. Moeser and B.E. Hibbard

1970). Developmental information was absent and many maize field weeds and native prairie grass species were not evaluated. Recently, Oyediran et al. (2004) evaluated a series of 21 prairie grass species thought to be among those dominant in the western Great Plains in the 1860s for D. v. virgifera growth and development, along with maize and sorghum as controls. Twenty pots of each species were planted. Five weeks after planting, pots were infested with 20 neonate D. v. virgifera larvae using a moistened camel’s-hair paintbrush. At 5, 10, 15 and 20 days after infestation, four pots of each species were placed in Tullgren funnels for larval recovery. The remaining four pots of each species were used for adult emergence. The percentage of larvae recovered, larval head capsule width and adult emergence varied significantly between the grass species. The percentage of larvae recovered from western couch grass, Pascopyrum smithii (Rydb.), pubescent couch grass, Elytrigia intermedia (Host), and side-oats grama, Bouteloua curtipendula (Michx.), were not significantly different from the percentage of larvae recovered from maize. The number of adults produced by pubescent couch grass was not significantly different from the number produced from maize. The average dry weight and head capsule width of adults produced from maize were not significantly different from the head capsule widths and dry weights of those adults from any grass species. Overall, adults were produced from 14 of the 23 species evaluated. In summary, several of the couch grasses were quite good hosts for D. v. virgifera, perhaps even their ancestral host. The dogma that D. v. virgifera followed maize up from Mexico may be correct, but hosts other than maize were maintaining the D. v. virgifera population in western Kansas in the 1860s and a number of the species evaluated are capable of producing D. v. virgifera adults today (though not documented in the field). Although D. v. virgifera has not yet been documented as being produced naturally from hosts other than maize, its subspecies, the Mexican corn rootworm, Diabrotica v. zea Krysan and Smith, has. Mexican corn rootworm beetles were collected in emergence traps placed over a mixture of four grass species: Brachiaria plantaginea (Link), Eleusine indica (L.), Eragrostis indica (Hornem.) and Digitaria ciliaris (Retz.) (Branson et al., 1982b). Larvae and pupae of the Mexican corn rootworm were also collected from B. plantaginea and Panicum hallii (Vasey) and the sedge Cyperus macrocephalus (Liebm.) in Mexico (Branson et al., 1982b). Other examples suggest that this was not an isolated incident. In Sutton County, Texas, Mexican corn rootworm adults were found at a density of two beetles per plant in a maize field in its third year of production (the second generation of potential beetle production from maize) despite more than 250 km of isolation from any other maize production area (Krysan and Smith, 1987). The establishment of an economically damaging population (Witkowski et al., 1986) in the second possible year of adult production, despite isolation of more than 250 km, strongly suggested that this population of the Mexican corn rootworm was maintained on hosts other than maize in the area. To evaluate the food conversion efficiency for alternative host plants, chap03.qxd 15/11/04 2:19 PM Page 49

Nutritional Ecology of Larvae and Adults 49

2 0 –2 –4 ECI Index –6

–8

–10

–12

–14 PAN DK 440 Marano S. glauca OSSK 617 C. dactylon Greenﬁelds S. verticilata P. miliaceum S. halepense Host plants/maize varieties

Fig. 3.1. Food conversion efﬁciency of different alternative hosts (left) and different maize varieties (right). Same letters above bars indicate no signiﬁcant differences between host plants (ANOVA, Bonferroni adjustment). These data are an excerpt from a bigger data set from Moeser (2003).

Moeser and Vidal (2004b) applied the same technique as for maize roots described above. Larval weight gain, the amount of ingested food and the ECI were measured for several monocot and dicot weeds and crops which are common in or near European maize fields. The same quantity of food as in the ECI experiments with maize roots was used. It was shown that larval growth and performance were comparable to maize root feeding when they were given sufficient root mass of alternative hosts (Fig. 3.1). There were significant differences between the different host plants and varieties (analysis of variance (ANOVA): F9.276 = 3.5; P = 0.001). These differences were more pronounced within the varieties and within the alternative host plants than between maize varieties and alternative hosts. There were no significant differences between the most suitable alternative hosts (Setaria verticillata (L.), Setaria glauca (Poiret.) and Panicum miliaceum (L.)) and the most suitable maize varieties (OSSK 617, Greenfields and Panama), as well as between the worst host plants from both categories (weeds: Cynodon dactylon (L.) and Sorghum halepense (L.); maize: DK440 and Marano). These findings suggest that root morphology and root mass of suitable alternative hosts are more important for larval development than anticipated before. Providing insufficient food for the larvae in earlier studies (Branson and Ortman, chap03.qxd 15/11/04 2:19 PM Page 50

50 J. Moeser and B.E. Hibbard

1967a, 1970) may have led to the assumption that alternative host plants are in general less suitable than maize. On several host species that were not suitable for larval development, a considerable amount of feeding was still recorded. This gives the ﬁrst evidence that antifeedant substances are not responsible for this incompatible insect–plant interaction (no antibiosis). It is more likely that the presence or absence of certain compounds of regular plant biochemistry render these host plants unsuitable for D. v. virgifera.

Implications of Alternative Hosts for Quarantine Measures

Quarantine and eradication measures have the goal of exterminating a locally introduced small population. They are using the absence of maize from the region of invasion to deplete the larvae of their food source. The effect of monocot weeds, crops or native grasslands that could serve as hosts on the population dynamics of D. v. virgifera is unknown at this time. However, given that almost all hosts other than maize produce fewer and inferior adults, it is not likely that alternative hosts would have a signiﬁcant impact on quarantine measures to control D. v. virgifera, especially if it is a local introduction with a limited number of adults in a small area. The oviposition of eggs near suitable optimal hosts other than maize seems unlikely.

Implications of Alternative Hosts for IPM Strategies (i.e. Crop Rotation)

Crop rotation was a very cost-effective and environmental friendly tool against D. v. virgifera in the US Corn Belt until a crop rotation-resistant variant was detected. It continues to be so in most areas. European crop rotation systems are more diverse than the maize/soybean rotation in the US Corn Belt and incorporate more monocot crops. While European crop rotation systems are an advantage because they reduce the selection pressure on D. v. virgifera (see Chapter 6, this volume), US maize/soybean rotation is rather problematic. The efﬁcient use of winter wheat roots (and other monocot crops) has far-reaching implications for crop rotation. The adaptability of D. v. virgifera to changing environments has been demonstrated several times (as developing resistance to different selective pressures has clearly shown). Thus, crop rotation with monocot crops should be avoided when practical and when the phenology of the monocot crop is such that it could serve as a host. Crop rotation is a heavy selection pressure and, if applied on a large scale (see Chapter 6, this volume), could select for D. v. virgifera as a more general cereal pest in the long run. Thus, similarly to resistance management with insecticides and GM maize, an adaptation management to alternative hosts/crops should be considered in Europe. chap03.qxd 15/11/04 2:19 PM Page 51

Nutritional Ecology of Larvae and Adults 51

Implications of Alternative Hosts for the Invasion Process

The invasion process is affected by alternative hosts only to a limited extent, as the majority of the eggs will be laid in maize fields where weeds do not play a major role. As has been shown in the invasion of the USA during the early 20th century, the vast majority of the beetles will be produced in maize fields. These are the only places where the large population densities necessary for a range expansion can be achieved. The European monocot weed flora may increase the number of potential alternative host plants, although detailed screening is still missing. Of more importance will be the monocot crops that are grown in close spatial range to maize fields, because they may be capable of acting as a reservoir, even in the absence of maize. When migration flights end in areas without maize, it seems likely that D. v. virgifera females would use monocot weeds or crops as places for oviposition. Females are known to prefer clumps of Setaria spp. to bare soil and maize stalks for oviposition (Johnson and Turpin, 1985). Although the presence of monocots during oviposition does not necessarily mean that monocots will be there during egg hatch, this is the case if winter wheat is planted early enough for oviposition.

Implications of Alternative Hosts for Resistance Management

As part of the registration process for Bacillus thuringiensis (Bt) crops in the USA, all registrants must submit an insect resistance management (IRM) plan to the Environmental Protection Agency (EPA). The goal of the IRM plans is to delay the development of resistance to the toxins by producing susceptible insects and planting crops in such a way that the resistant insects are more likely to mate with susceptible insects than other resistant insects (it is generally assumed that resistant insects initially are rare). Development of an appropriate IRM plan for transgenic maize hybrids that control corn rootworms must include, among other things, an understanding of important biological parameters, because the interactions of many factors could affect the duration of the product (Onstad et al., 2001). For instance, movement of larvae from susceptible to transgenic plants, and vice versa, has been hypothesized to adversely affect IRM in several ways (Mallet and Porter, 1992; Davis and Onstad, 2000). In the case of corn rootworms, initial development would need to be relatively close because larval movement is limited (Hibbard et al., 2003). Since a mixed seed (transgenic and non-transgenic seed sold as mixed seed to the grower) has not been registered for sale by the EPA in the USA, this means that the susceptible host would need to be within the maize production ﬁeld. It has been known for some time that a number of grasses are hosts for larval development of D. v. virgifera (Branson and Ortman, 1967a,b, 1970), but the majority of the species evaluated were tested only in Petri dishes without soil. chap03.qxd 15/11/04 2:19 PM Page 52

52 J. Moeser and B.E. Hibbard

Clark and Hibbard (2004) evaluated larval survivorship and growth parameters of D. v. virgifera larvae on the roots of 29 plant species, mostly maize field grassy weeds. Larval recovery and growth (measured as increases in head capsule width and accumulation of dry weight) data were recorded at five sample dates (6, 10, 14, 20 and 24 days after infestation) after initial infestation of the 29 species. Larvae survived at least 6 days after infestation on 27 species and 24 days on 23 plant species (Table 3.1). Larval recovery and growth was affected by both species and time after infestation. Growth and development of larvae on plant species were significantly slower on most species than growth and development on maize. However, 18 of the species had larvae develop to the second instar while larvae on 14 species had development to the third instar. Adults were recovered from five plant species in addition to maize. Although adults were produced from a smaller percentage of species by Clark and Hibbard (2004) than in a similar study focused on dominant prairie grasses of the western Great Plains (Oyediran et al., 2004), larvae survived for at least 24 days on 23 of the 29 species evaluated. Because later rootworm instars are more tolerant to transgenic endotoxins (EPA Scientific Advisory Panel Meeting, 2002), initial development on a susceptible plant (a grassy weed or maize plant) followed by subsequent migration to a nearby transgenic plant could accelerate the rate of adaptation if heterozygotes with the resistance gene survived exposure to the endotoxin at higher rates. Likewise, if larvae briefly fed on a transgenic root and then migrated to a nearby susceptible root, this too could accelerate the rate of resistance development if heterozygotes with the resistance gene were preferentially selected. However, if a low-dose product produced susceptible beetles, movement of larger larvae onto transgenic roots from less suitable alternative hosts could actually increase product durability by producing additional susceptible insects from within the transgenic field. Storer (2003) modelled adaptation of corn rootworms to rootworm-resistant Bt maize. The model predicts that, if as few as 0.5% of the adults come from spatially well-distributed non- maize hosts, the onset of resistance would be significantly delayed in a system with a poorly distributed 5% fixed location refuge, although this delay is not significant under more conservative refuge deployment scenarios, such as the 20% refuge being required for the product that is currently registered (Nick Storer, personal communication). Both Clark and Hibbard (2004) and Oyediran et al. (2004) have documented significant adult production from hosts other than maize. However, what percentage, if any, of the adults commonly found in maize agroecosystems developed as larvae on hosts other than maize is unknown. Initial development on alternative hosts followed by movement to transgenic roots (perhaps after herbicide sprays) will probably have a greater impact on resistance management than adult production from alternative hosts. However, at this time, we do not know whether grassy weeds present in transgenic fields will speed or slow the development of resistance. The response may be specific to the transgenic product and the dose of toxin associated with it. chap03.qxd 15/11/04 2:19 PM Page 53

Nutritional Ecology of Larvae and Adults 53 0.0 n 9.5 jkl 9.3 jkl 8.7 kl 6.0 lm 2.7 mn 2.3 mn 2.0 mn 1.7 mn 0.7 n 0.3 n 0.0 n 0.0 n 39.0 a 33.0 b 32.7 bc 29.0 bcd 28.7 bcd 28.3 cde 24.7 def 24.0 ef 22.7 f 21.0 fg 21.0 fg 20.7 fg 16.7 gh 15.7 h 14.3 h 13.3 hij 10.6 jk = 0.05. Means P Time after initial infestation 0.0 k A 0.0 m A 0.0 l A 0.0 m A 0.0 h A 0.0 k A0.0 k A0.0 k A 1.7 ± 1.7 l A0.0 k A 0.0 m A 3.3 ± 3.3 kl A 0.0 m A 0.0 m A 0.0 m A 0.0 l A 0.0 l A 0.0 l A 6.7 ± 3.9 e–g A 1.7 ± 1.7 lm A 0.0 m A 0.0 m A 1.7 ± 1.7 gh A 0.0 h A 0.0 h A 6.7 ± 4.7 h–j C 23.3 ± 12.9 f–h AB 31.7 ± 1.7 cd A 26.7 ± 2.7 d–f AB 16.7 ± 5.8 b–d BC 5.0 ± 3.2 ij B2.2 ± 1.9 jk A5.0 ± 5.0 i–k BC 6.7 ±3.3 ± 4.7 j–l B 20.0 ± 1.9 ij B 11.9 g–i A6.7 ± 23.3 ± 6.7 ij A 12.3 f–h A 10.0 ± 4.3 h–j A 13.3 ± 9.4 ij AB 10.0 ±6.7 ± 5.8 ij AB 3.9 h–j A 3.3 ±1.7 ± 1.9 k–m B 1.7 jk A 10.0 ± 7.9 jk A 0.0 m A 15.6 ± 5.1 g–i A 20.0 ± 9.3 ef A 0.0 m C1.7 ± 0.0 m A 1.7 jk A 0.0 m A 8.3 ± 8.3 e–g A 8.3 ± 3.3 ± 5.0 ef AB 3.3 k–m B 0.0 l A 0.0 m A 0.0 h B 0.0 l A 5.0 ± 5.0 f–h BC 0.0 l A 0.0 l A 5.0 ± 5.0 k–m A 1.7 ± 1.7 lm A 6.7 ± 6.7 kl A 1.7 ± 1.7 gh A 0.0 m A 1.7 ± 1.7 gh A 0.0 h A 0.0 h A 42.2 ± 5.1 a A21.7 ± 9.6 c–e BC38.3 ± 14.5 ab A 43.3 ±30.0 ± 6.9 ab A 11.4 cd A 46.7 ± 7.7 a A23.3 ± 11.4 c–f B 26.7 ± 4.7 d–f B36.7 ± 45.0 ± 6.9 ab A 40.0 ± 11.7 ab A 7.2 a–c A31.7 ± 56.7 ± 43.3 ± 15.5 b–d A 10.4 a A 5.8 ab A 31.7 ± 9.6 cd AB20.0 ± 8.6 d–f BC 36.7 ± 40.0 ± 10.1 b–d AB 17.9 bc A 10.0 ± 16.7 ±18.3 ± 6.4 i–k B 12.6 h–j B 9.6 d–f BC 23.3 ± 46.7 ± 42.2 ± 12.9 ef B 15.2 ab A 13.9 a–c A 28.3 ± 18.3 ± 7.4 c–f AB 10.0 b–d C 30.0 ± 16.7 ± 10.4 de A 1.9 e–g AB16.7 ± 30.0 ± 35.0 ± 16.7 g–i B 10.0 c–f AB 3.2 bc A 33.3 ± 36.7 ± 8.2 bc A16.7 ± 13.5 c–e AB 6.9 e–g BC 35.0 ± 31.7 ± 20.0 ± 8.8 bc A 5.0 ± 8.8 c–e A 4.7 ab B 3.2 e–g B 16.7 ± 38.3 ± 10.0 f–h BC 8.3 a–d A 53.3 ± 16.7 ± 33.3 ± 5.0 ± 11.2 a A28.3 ± 5.8 ab B 11.6 b–e A 3.2 ef B 4.2 bc A 21.7 ± 6.3 e–g A–C 26.7 ± 9.4 ef A 21.7 ± 12.0 g–i BC 26.7 ± 15.0 ± 21.7 ± 9.8 a A 8.8 f–j B 7.4 de AB 16.7 ± 8.8 b–d C 23.3 ± 8.4 fg AB 16.7 ± 21.7 ± 8.3 ± 6.7 ± 5.8 e–h AB 11.3 f–h A–C 4.2 d–f C 6.7 fg B 20.0 ± 14.1 h–j B 16.7 ± 10.0 ± 15.0 ± 11.1 h–j BC 5.8 c–e C 7.4 e–i B 15.0 ± 5.0 bc AB 6.7 ± 10.0 ± 6.7 e–g C 4.3 ij B 1.7 ± 1.7 gh C 30.0 ± 8.4 bc A 20.0 ±21.7 ± 11.9 g–i AB 7.4 c–e A 8.3 ± 6.3 jk B 11.7 ± 9.6 i–k AB 0.0 l B 3.3 ± 3.3 k–m B 5.0 ± 5.0 f–h B 13.3 ± 7.2 h–j A 0.0 h B 11.67 ± 7.9 g–i A 10.0 ± 4.3 ij A 11.7 ± 5.0 g–j A 18.3 ± 8.3 g–i A 20.0 ± 11.9 bc A Z. mays Percentage of western corn rootworm larval recovery from Tullgren funnels (from Clark and Hibbard, 2004). Means between between sampling dates (rows) followed by the same uppercase letter are not significantly different. plant species within a sampling date (columns) followed by the same lowercase letter are not significantly different at Table 3.1. Host speciesZea mays Day 6 Day 10 Day 14 Day 20Non-infested Day 24 Cumulative Pascopyrum smithii Chloris gayana Phalaris arundinacea Agrostis gigantean Eragrostis trichodes Setaria viridis Setaria vericillatta Digitaria sanquinalis Panicum capillare Echinochloa crus–galli Eriochloa villosa Urochloa texana Setaria faberi Eleusine indica Setaria pumila Eragrostis curvula Panicum miliaceum Bromus tectorum Panicum italicum Triticum aestivum Cenchrus tribuloides Sorghum halepense Dactylis glomerata Panicum virgatum Sorghum drummondii Amaranthus retroflexus Avena sativa Sorghum bicolor chap03.qxd 15/11/04 2:19 PM Page 54

54 J. Moeser and B.E. Hibbard

Density-dependent Survival

A discussion of the nutritional ecology of D. v. virgifera larvae is not complete without a discussion of density-dependent survival/mortality. An understanding of this phenomenon is critical to understanding adult emergence data from control tactics involving toxins. For instance, an untreated check may produce fewer adults than an insecticide or transgenic treatment under high infestation levels and yet the amount of damage to the untreated check would be much greater. Under these circumstances, density-dependent mortality would result in fewer adults being produced from the untreated check, but this would not be known if damage were not also accessed along with adult emergence. Stressed larvae, in general, will produce inferior adults (Peters and Barbosa, 1977). Branson and Sutter (1985) evaluated the performance of D. v. virgifera adults from field plots with varying densities of D. v. virgifera larvae. They documented that high densities of eggs resulted not only in greater plant damage but also in a lower percentage of adults being recovered (density-dependent mortality). The adults they did recover had a smaller head-capsule width, died earlier and laid fewer eggs per female. In a similar set of studies conducted in the greenhouse, Weiss et al. (1985) documented the fact that higher infestation rates resulted in longer developmental times and an altered sex ratio of adults to produce a higher percentage of males. Hibbard et al. (2003) evaluated plant-to-plant movement by D. v. virgifera larvae under different plant densities and row spacings. One central plant was infested and plant damage and larval recovery were evaluated for the infested plant, three plants down the row on each side of the infested plant and the four closest plants in the two rows adjacent to the infested plant. The field was planted to soybeans the previous year, so no feral D. v. virgifera larvae were present. Post-establishment movement by D. v. virgifera larvae was clearly documented in two out of four treatment combinations in 1999 where larvae moved up to three plants down the row and across a 0.46 m row. Larvae did not significantly cross a 0.91 m row after initial host establishment in 1998 or 1999, whether or not the soil had been compacted by a tractor and planter. Larvae moved from highly damaged, infested plants to nearby plants with little to no previous root damage, but this movement did not occur until significant damage was found on the infested plant. When stressed, larvae moved to superior food sources on adjacent plants. In his model, Storer (2003) included an assumption that ‘most density-dependence occurs after the larvae have become established but before they have reached adulthood’. Hibbard et al. (2004) evaluated infestation levels of 100, 200, 400, 800, 1600 and 3200 viable eggs on a single plant and evaluated larval recovery and plant damage from the infested plant, three plants down the row, the nearest plant across the row and a control plant at least 1.5 m away from the infested plant. Density- dependent mortality was negligible during establishment, as a similar percentage of larvae was recovered from all infestation levels, supporting chap03.qxd 15/11/04 2:19 PM Page 55

Nutritional Ecology of Larvae and Adults 55

Storer’s (2003) assumption. Although egg density appeared not to be an important factor in larval establishment, it was an important factor in plant damage and, secondarily, subsequent larval movement. In general, damage was highest in the plots infested with the most viable eggs and decreased as distance from the infested plant decreased. Post-establishment movement generally occurred about the time that significant damage began to appear, rather than at the time of establishment. These data imply that plant-to-plant movement was motivated by a search for food and was density-dependent only because damage was density- dependent. If crowding caused larval movement rather than a lack of food, movement would be expected to have occurred earlier than was found. Density-independent factors such as unsuccessful host-finding, flooding, dry soil and/or unknown factors caused mortality that was considerable in the Hibbard et al. (2004) study, but comparable to that in other studies (Branson and Sutter, 1985; Spike and Tollefson, 1989; Riedell and Sutter, 1995; Hoback et al., 2002). Storer (2003) set density- independent survival at 5% in his model after citing data on egg predation, survival after establishment and per cent adult emergence. In the Hibbard et al. (2004) study, density-independent establishment was between 2.5 and 5.7% when plants were sampled on the optimal date as estimated by larval recovery using their sampling technique. This figure did not include any overwintering or pupal mortality, but the 5% figure used by Storer (2003) is quite close to the data given by Hibbard et al. (2004) and other studies cited by Storer (2003). Interestingly, density- independent mortality in greenhouse situations is nearly tenfold less (Weiss et al., 1985; B.E. Hibbard, unpublished data). An understanding of the differences in density-independent mortality between field and greenhouse conditions is an important gap in our understanding of corn rootworm biology. Regardless, density-independent mortality in larval establishment is an important factor that must be considered when selection intensity (an important component in resistance management models) of Bt maize or insecticides is calculated.

Adult Feeding on Various Food Resources

The economic impact of adult D. v. virgifera feeding is negligible with the exception of seed maize production (Culy et al., 1992). Here an increased number of adults that are silk feeding may lead to reduced ear ﬁlling and to seed shapes that cannot be used in modern planting equipment (C.R. Edwards, personal communication). Understanding the nutritional ecology of adult D. v. virgifera is important for several other reasons: the spreading and actual invasion process is carried out by the adult females, it’s the females that are responsible for the increase in population density and their nutritional status is therefore of utmost importance. D. v. virgifera is mainly a pollen feeder that also uses other above-ground plant organs of maize (Chiang, chap03.qxd 15/11/04 2:19 PM Page 56

56 J. Moeser and B.E. Hibbard

1973; Ludwig and Hill, 1975). Maize pollen and silk were shown to be the best for egg production in female D. v. virgifera (Elliot et al., 1990), but are only available during flowering, which is a relatively short time period. Protein-rich food is scarce after this time. Limited food supply within maize fields may lead to dispersion flight or even to large-scale migration towards the end of the maize-growing season (see Chapter 6, this volume). Movement at the maize field boundaries increases where females move frequently between maize fields and outlying pollen sources (J. Moeser, personal observation). Still, pollen feeding by D. v. virgifera is poorly documented for the USA, while data for D. v. zeae (the Mexican corn rootworm) are available. Pollen from plants from 63 genera was found in the gut of feral D. v. zeae in Texas (Jones and Coppedge, 2000). While D. v. zeae is generally believed to turn its attention to alternative pollen sources after the flowering period of maize, D. v. virgifera was expected to prefer feeding on other maize tissues (Naranjo, 1991). The European invasion was in need of some more specific data on the food sources used by D. v. virgifera, so Moeser and Vidal (2004c) performed gut content and pollen analysis with 1200 beetles from southern Hungary. He compared fields with high and low abundance of weeds to estimate the impact of alternative pollen sources on D. v. virgifera feeding behaviour. Moeser and Vidal (2004c) showed that D. v. virgifera fed on pollen from 73% of the flowering weeds or crops present in maize fields in southern Hungary (= 19 species). Ludwig and Hill (1975) mention only two weed species whose pollen was detected in D. v. virgifera guts. Whether this is due to the single sample date, the small sample size or the agroecosystems with a decreased number of alternative pollen sources remains to be investigated. The high plasticity in the adult feeding behaviour was also clearly demonstrated by Spencer et al. (Chapter 6, this volume), where an increase in soybean feeding was shown after the flowering period of maize. While McKone et al. (2001) demonstrated that Diabrotica barberi Smith and Lawrence fed on sunflower, but not D. v. virgifera, Moeser and Vidal (2004c) presented data which even suggest a certain preference in European D. v. virgifera for sunflower pollen. Similarly Mullin et al. (1991) showed data indicating antifeedant attributes for sunflower and Solidago canadensis L. pollen, concluding that Asteraceae are unsuitable pollen sources for D. v. virgifera. Moeser and Vidal (2004c) found pollen from five Asteraceae species in European D. v. virgifera, with a preference for Ambrosia artemisiifolia L. towards the later vegetation period. This preference was found in beetles which contained high numbers of pollen from this species but were sampled in weed-free fields not exhibiting this weed. Three major factors were identified as influencing the nutritional ecology of adult D. v. virgifera (Moeser and Vidal, 2004c):

1. Time and changes in maize phenology: The changing phenology of maize and time dictates which food is provided within the maize ﬁeld. During the ﬁrst weeks of adult life, maize provides enough high-quality chap03.qxd 15/11/04 2:19 PM Page 57

Nutritional Ecology of Larvae and Adults 57

food as pollen and silk are generally readily available at this time. Only after these food items became scarce did leaf and kernel feeding begin significantly. At this time, the number of individuals with empty guts also increased. At the same time, the use of alternative pollen sources increased as well. These pollen sources were located in and outside the maize fields. A similar increase in non-maize pollen over time was also noted for D. barberi (Cinereski and Chiang, 1968). The influence of changing maize phenology with regard to soybean foliage feeding was investigated thoroughly by O’Neal et al. (2002) and Spencer et al. (Chapter 6, this volume). 2. Habitat and abundance of alternative pollen sources: The amount of alternative pollen resources within the maize field influences the choice of adult D. v. virgifera. During the first weeks, beetles from weedy fields fed to a certain extent on weed pollen, while beetles from non-weedy fields began to use this resource only at a later time, when maize was not an attractive food source, especially during the later vegetation period. Weeds which provide pollen in large quantities during the whole vegetation period and especially towards the end of Diabrotica adult life are considered extremely valuable as alternative pollen sources. Fields with flowering crops like sunflower close to maize fields supply the beetles with an additional pollen source. Whether weeds keep the beetles inside maize fields when maize is becoming unsuitable and thus impede spreading remains to be investigated. The amount of pollen of each plant species does not reflect the plant’s abundance in the fields. Ambrosia and sunflower were clearly over-represented in the pollen analysis, while the abundance was minimal. A preference for late-flowering weeds (A. artemisiifolia) and flowers that offer a large quantity of pollen and are largely visible (sunflower) has to be considered. 3. Sex of the beetles: Female D. v. virgifera use alternative pollen sources to a greater extent than male beetles. Females from weedy fields feed more on alternative pollen than females from non-weedy fields. Males from weedy fields also feed more on alternative pollen than males from non-weedy fields. Males use a larger number of alternative host plants but less pollen of alternative hosts is found on average in males.

Adult D. v. virgifera do not use specific alternative pollen sources according to their abundance. Pollen of weeds that were not present in weed- free fields (like A. artemisiifolia) was detected in large quantities in the guts of D. v. virgifera beetles sampled in these weed-free fields. Short excursions outside their natal field, as documented for D. barberi (Naranjo, 1991), or flights from fields containing a larger array of alternative pollen sources are likely to explain these findings. The nutritional status of the adults influences oviposition, longevity, flight activity and migration behaviour (Naranjo, 1991). Feeding assays with pollen of Asteraceae led to the assumption that they are of lesser value for D. v. virgifera adult nutrition (Mullin et al., 1991). These no-choice experiments neglect the possibility that they may chap03.qxd 15/11/04 2:19 PM Page 58

58 J. Moeser and B.E. Hibbard

supply the beetles with essential protein when fed upon in conjunction with maize tissue like kernel or leaves. The no-choice experiments of the past (Ludwig and Hill, 1975) are not likely to explain the situation of adult D. v. virgifera in the field, where a mix of maize and non-maize diets is most often encountered (J. Moeser, unpublished data). Latest research suggests that adult D. v. virgifera profit greatly by using all pollen resources available, although Pavuk and Skinner (1994) concluded that weeds in maize fields have no influence on D. v. virgifera populations. These studies from the US Corn Belt have only limited applicability for Europe because European agroecosystems may provide a different and a wider array of possible pollen sources because of the reduced land use intensity. The use of alternative pollen may enlarge the amount of time available to the females for oviposition, increase fecundity, increase longevity in areas even when maize is already harvested and facilitate spreading by providing food in areas without maize.

Rearing of D. v. virgifera

The ability to rear a species being studied is a requirement for certain types of work and a great advantage in most types of studies. Methods for rearing D. v. virgifera have been available for about 40 years (George and Ortman, 1965) and several fairly comprehensive methods are described in papers (Howe and George, 1966; Branson et al., 1975; Jackson, 1985, 1986). Jackson (1986) is particularly useful because several different techniques applicable for differing reasons are presented. None of these techniques uses an artificial diet for rearing D. v. virgifera larvae. Since screening transgenic crops for production of proteins toxic to D. v. virgifera larvae requires an artificial diet, several seed companies have developed their own proprietary diets. Representatives from Monsanto have now published an artificial diet for D. v. virgifera larvae (Pleau et al., 2002). However, maize is still superior to the diet in producing large numbers of viable adults, so all of the ingredients for an optimal diet are not known at this time.

Implications of the Results on Nutritional Ecology for Management Options and Outlook

The results of research on D. v. virgifera nutritional ecology have far- reaching implications for the implementation of IPM strategies for maize producers, plant breeders and crop protection agencies. The invasion of Europe, the increasing resistance to crop rotation and insecticides in the USA and the requirement of resistance management plans in Bt ﬁelds have led to a new wave in research interested in D. v. virgifera host–plant interactions. Larval host ranges are better understood and the implications for the management practices like crop rotation are evident for chap03.qxd 15/11/04 2:19 PM Page 59

Nutritional Ecology of Larvae and Adults 59

Europe. How far the invasion process in Europe is affected by the use of alternative hosts remains to be seen. The focus of European agriculture on monocot crops other than maize in Central Europe may result in a completely new field of research if D. v. virgifera undertakes another host switch such as happened in the past. The potential threat posed by alternative hosts to resistance management plans in the use of GM maize is an area that has received more attention since registration by the EPA. The past has shown that relying on a single counter-strategy led to resistance within a few decades. In the case of GM maize another valuable tool would then be lost quickly if applied without the necessary caution. Finally, newly developed methods may contribute in screening for resistant maize varieties and deepen our understanding of this complex insect–plant interaction. The patterns behind the invasion process are becoming more clear as our understanding of resource usage by D. v. virgifera increases. While alternative pollen sources are widely used by D. v. virgifera in Europe, recommendations are difficult with regard to any implications of IPM strategies. On the one hand, the removal of weeds from maize fields may lead to an early start in the migration flight of a larger number of females, thus increasing invasion pressure. On the other hand, a high abundance of alternative pollen sources may provide additional nitrogen sources, which will possibly lead to an increase in successful oviposition and female fitness. Whether landscape diversity, for example, has any significant impact on the invasion process, besides the amount of maize coverage, is still unknown. The high adaptability of D. v. virgifera with regard to the use of alternative host plants as larvae and adults shows that Europe may indeed provide new areas for occupation, not only spatially (range expansion) but also by expansion of its ecological niche.

References

Abel, C.A., Berhow, M.A., Wilson, R.L., Binder, B.F. and Hibbard, B.E. (2000) Evaluation of conventional resistance to European corn borer (Lepidoptera: Pyralidae) and western corn rootworm (Coleoptera: Chrysomelidae) in experimental maize lines developed from a backcross breeding program. Journal of Economic Entomology 93, 1814–1821. Apple, J.W. and Patel, K.K. (1963) Sequence of attack by northen corn rootworms on the crown roots of corn. Proceedings North Central Branch ESA 18, 80–81. Arnason, J.T., Larsen, J., Assabgui, R., Xie, Y., Atkinson, J., Philogene, B.J.R. and Hamilton, R.I. (1997) Mechanisms of resistance in maize to western corn rootworm. In: CIMMYT (ed.) Insect Resistant Maize: Recent Advances and Utilization. Proceedings of an International Symposium Held at the International Maize and Wheat Improvement Center (CIMMYT), 27 November–3 December 1994, pp. 96–100. Assabgui, R.A., Arnason, J.T. and Hamilton, R.I. (1993) Hydroxamic acid content in maize (Zea mays) roots of 18 Ontario recommended hybrids and prediction of antibiosis to the western corn rootworm. Canadian Journal of Plant Science 73, 359–363. chap03.qxd 15/11/04 2:19 PM Page 60

60 J. Moeser and B.E. Hibbard

Assabgui, R.A., Hamilton R.I. and Arnason, J.T. (1995a) Hydroxamic acid content and plant development of maize (Zea mays L.) in relation to damage by the western corn rootworm, Diabrotica virgifera virgifera LeConte. Canadian Journal of Plant Science 75, 851–856. Assabgui, R.A., Arnason, J.T. and Hamilton, R.I. (1995b) Field evaluations of hydroxamic acids as antibiosis factors in elite maize inbreds to the western corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 88, 1482– 1493. Ball, H.J. and Weekman, G.T. (1962) Insecticide resistance in the adult western corn rootworm in Nebraska. Journal of Economic Entomology 55, 439–441. Bigger, J.H., Holbert, J.R., Flint, W.P. and Lang, A.L. (1938) Resistance of certain corn hybrids to attack of southern corn rootworm. Journal of Economic Entomology 31, 102–107. Bigger, J.H., Snelling, R.O. and Blanchard, R.A. (1941) Resistance of corn strains to the southern corn rootworm, Diabrotica duodecimpunctata F. Journal of Economic Entomology 34, 605–613. Bodnaryk, R.P., Luo, M. and Kudryk, L. (1997) Effects of modifying the phytosterol profile of canola, Brassica napus L., on growth, development, and survival of the bertha armyworm, Mamestra configurata Walker (Lepidoptera: Noctuidae), the flea beetle, Phyllotreta cruciferae (Goeze) (Coleoptera: Chrysomelidae) and the aphids Lipaphis erysimi (Kaltenbach) and Myzus persicae (Sulzer) (Homoptera: Aphididae). Canadian Journal of Plant Science 77, 677–683. Branson, T.F. (1971) Resistance in the grass tribe Maydeae to larvae of the western corn rootworm. Annals of the Entomological Society of America 64, 861–863. Branson, T.F. (1986) Larval feeding behavior and host-plant resistance in maize. In: Krysan, J.L. and Miller, T.A. (eds) Methods for the Study of Pest Diabrotica. Springer- Verlag, New York, pp.159–182. Branson, T.F. (1989) Survival of starved neonate larvae of Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae). Journal of the Kansas Entomological Society 62, 521–523. Branson, T.F. and Guss, P.L. (1972) Potential for utilizing resistance from relatives of cultivated crops. Proceedings North Central Branch ESA 27, 91–95. Branson, T.F. and Krysan, J.L. (1981) Feeding and oviposition behavior and life cycle strategies of Diabrotica: an evolutionary view with implications for pest management. Environmental Entomology 10, 826–831. Branson, T.F. and Ortman, E.E. (1967a) Host range of larvae of the western corn rootworm. Journal of Economic Entomology 60, 201–203. Branson, T.F. and Ortman, E.E. (1967b) Fertility of western maize rootworm as larvae on alternate hosts. Journal of Economic Entomology 60, 595. Branson, T.F. and Ortman, E.E. (1970) The host range of the larvae of the western corn rootworm: further studies. Journal of Economic Entomology 63, 800–803. Branson, T.F. and Reyes Rueda, J. (1983) The association of Diabrotica spp. with Zea diploperennis. Journal of the Kansas Entomological Society 56, 97–99. Branson, T.F. and Sutter, G.R. (1985) Influence of population density of immatures on size, longevity, and fecundity of adult Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae). Environmental Entomology 14, 687–690. Branson, T.F., Guss, P.L. and Ortman, E.E. (1969) Toxicity of sorghum roots to larvae of the western corn rootworm. Journal of Economic Entomology 62, 1375–1378. Branson, T.F., Guss, P.L., Krysan, J.L. and Sutter, G.R. (1975) Corn Rootworm: Laboratory Rearing and Manipulation. ARS-NC-28, USDA, 18 pp. Branson, T.F., Fisher, J.R., Kahler, A.L. and Sutter, G.R. (1981) Host plant resistance to chap03.qxd 15/11/04 2:19 PM Page 61

Nutritional Ecology of Larvae and Adults 61

corn rootworms. In: Seventeenth Annual Illinois Corn Breeders School, Allerton House, University of Illinois at Urbana-Champaign, 12–13 March 1981. University of Illinois at Urbana-Champaign. Branson, T.F., Sutter, G.R. and Fisher, J.R. (1982a) Comparison of a tolerant and a susceptible maize inbred under artificial infestations of Diabrotica virgifera virgifera: yield and adult emergence. Environmental Entomology 11, 371–372. Branson, T.F., Reyes Rueda, J. and Valdéz Martínez, H. (1982b) Field biology of Mexican corn rootworm, Diabrotica virgifera zeae (Coleoptera: Chrysomelidae), in Mexico. Environmental Entomology 11, 1078–1083. Branson, T.F., Welch, V.A., Sutter, G.R. and Fisher, J.R. (1983) Resistance to larvae of Diabrotica virgifera virgifera in three experimental maize hybrids. Environmental Entomology 12, 1509–1512. Branson, T.F., Valdéz Martínez, H. and Kahler, A.L. (1986) Evaluation of Mexican populations of maize for resistance to larvae of Diabrotica virgifera (Coleoptera: Chrysomelidae). Maydica 31, 173–177. Chiang, H.C. (1973) Bionomics of the northern and western corn rootworm. Annual Review of Entomology 18, 47–72. Cinereski, J.E. and Chiang, H.C. (1968) The pattern of movement of adults of the northern corn rootworm inside and outside of corn fields. Journal of Economic Entomology 61, 1531–1536. Clark, T.L. and Hibbard, B.E. (2004) A comparison of non-maize hosts to support western corn rootworm (Coleoptera: Chrysomelidae) larval biology. Environmental Entomology 33, 681–689. Culy, M.D., Edwards, C.R. and Cornelius, J.R. (1992) Effect of silk feeding by western corn rootworm (Coleoptera; Chrysomelidae) on yield and quality of inbred corn in seed corn production fields. Journal of Economic Entomology 85, 2440–2446. Davis, P.M. and Onstad, D.W. (2000) Seed mixtures as a resistance management strategy for European corn borers (Lepidoptera: Crambidae) infesting transgenic corn expressing Cry1ab protein. Journal of Economic Entomology 93, 937–948. Elliot, N.C., Gustin, R.D. and Hanson, S.L. (1990) Influence of adult diet on the reproductive biology and survival of the western corn rootworm, Diabrotica virgifera virgifera. Entomologia Experimentalis et Applicata 56, 15–21. EPA Scientific Advisory Panel Meeting (2002) Corn rootworm plant-incorporated protectant non-target insect and insect resistance management issues, part B. http://www.epa.gov/scipoly/sap/2002/august/august2002final.pdf EPPO (2003) http://www.eppo.org/QUARANTINE/Diabrotica_virgifera/diabrotica_ virgifera.html Fitzgerald, P.J. and Ortman, E.E. (1964) Breeding for resistance to the western corn rootworm. In: Heckendorn, W. and Sutherland, J.I. (eds) Proceedings of the Annual Hybrid Corn Industry Research Conference, 9–10 December Vol. 19, 1964, Chicago. American Seed Trade Association, Washington, DC, pp. 46–60. Fitzgerald, P.J. and Ortman, E.E. (1965) Two-year performance of inbreds and their single crosses grown under corn rootworm infestation. Proceedings North Central Branch ESA 20, 46–47. George, B.W. and Ortman, E.E. (1965) Rearing the western corn rootworm in the laboratory. Journal of Economic Entomology 58, 375–377. Gillette, C.P. (1912) Diabrotica virgifera as a corn rootworm. Journal of Economic Entomology 5, 364–366. Golden, K.L. and Meinke, L.J. (1991) Immature development, fecundity, longevity, and egg diapause of Diabrotica longicornis (Coleoptera: Chrysomelidae). Journal of the Kansas Entomological Society 64, 251–256. chap03.qxd 15/11/04 2:19 PM Page 62

62 J. Moeser and B.E. Hibbard

Goodman, M.M. (1987) The history and evolution of maize. Critical Reviews in Plant Science 7, 197–220. Hibbard, B.E., Barry, B.D., Darrah, L.L., Jackson, J.J., Chandler, L.D., French, L.K. and Mihm, J.A. (1999a) Controlled field infestations with western corn rootworm (Coleoptera: Chrysomelidae) eggs in Missouri: effects of egg strains, infestation dates, and infestation levels on corn root damage. Journal of the Kansas Entomological Society 72, 214–221. Hibbard, B.E., Darrah, L.L. and Barry, B.D. (1999b) Combining ability of resistance leads and identiﬁcation of a new resistance source for western corn rootworm (Coleoptera: Chrysomelidae) larvae in corn. Maydica 44, 133–139. Hibbard, B.E., Duran, D.P., Ellersieck, M.M. and Ellsbury, M.M. (2003) Post-establishment movement of western corn rootworm larvae (Coleoptera: Chrysomelidae) in Central Missouri corn. Journal of Economic Entomology 96, 599–608. Hibbard, B.E., Higdon, M., Duran, D.P., Schweikert, Y.M. and Ellersieck, M.R. (2004) Role of egg density on establishment and plant-to-plant movement by western corn rootworm larvae (Coleoptera: Chrysomelidae). Journal of Economic Entomology 97, 871–882. Hills, T.M. and Peters, D.C. (1971) A method of evaluating postplanting insecticide treatments for control of western corn rootworm larvae. Journal of Economic Entomology 64, 764–765. Hoback, W.W., Clark, T.L., Meinke, L.J., Higley, L.G. and Scalzitti, J.M. (2002) Immersion survival differs among three Diabrotica species. Entomologia Experimentalis et Applicata 105, 29–34. Horton, D.R. and Redak, R.A. (1993) Further comments on analysis of covariance in insect dietary studies. Entomologia Experimentalis et Applicata 69, 263–275. Howe, W.L. and George, B.W. (1966) Corn rootworms. In: Smith, C.N. (ed.) Insect Colonization and Mass Production. Academic Press, New York, pp. 367–383. Ivezic´, M., Tollefson, J.J., Raspudic-, E., Hibbard, B.E. and Brkic´, I. (2003) Evaluation of Croatian corn hybrids for tolerance to corn rootworm (Diabrotica virgifera virgifera LeConte) larval feeding. In: Proceedings of the XXIst IWGO Conference and VIII Diabrotica Subgroup Meeting at Venice, Italy, 27 October–3 November, 2001, Veneto Agricoltura, Legnaro, Italy, pp. 205–212. Jackson, J.J. (1985) Diabrotica. In: Singh, P. and Moore, R.F. (eds) Handbook of Insect Rearing. Elsevier Science Publishers, Amsterdam, pp. 237–254. Jackson, J.J. (1986) Rearing and handling of Diabrotica virgifera and Diabrotica undecimpunctata howardi. In: Krysan, J.L. and Miller, T.A. (eds) Methods for the Study of Pest Diabrotica. Springer-Verlag, New York, pp. 25–48. Johnson, T.B. and Turpin, F.T. (1985) Northern and western corn rootworm (Coleoptera: Chrysomelidae) oviposition in corn as inﬂuenced by foxtail populations and tillage systems. Journal of Economic Entomology 78, 57–60. Jones, G.D. and Coppedge, J.R. (2000) Foraging resources of adult Mexican corn rootworm (Coleoptera: Chrysomelidae) in Bell County, Texas. Journal of Economic Entomology 93, 636–643. Kahler, A.L., Olness, A.E., Sutter, G.R., Dybing, C.D. and Devine, O.J. (1985a) Root damage by western corn rootworm and nutrient content in maize. Agronomy Journal 77, 769–774. Kahler, A.L., Telkamp, R.E., Penny, L.H., Branson, T.F. and Fitzgerald, P.J. (1985b). Registration of NGSDCRW1(S2)C4 maize germplasm. Crop Science 25, 202. Knutson, R.J., Hibbard, B.E., Barry, B.D., Smith, V.A. and Darrah, L.L. (1999) Comparison of screening techniques for western corn rootworm (Coleoptera: Chrysomelidae) host- plant resistance. Journal of Economic Entomology 92, 714–722. Krysan, J.L. and Smith, R.F. (1987) Systematics of the virgifera species group of Diabrotica (Coleoptera: Chrysomelidae: Galerucinae). Entomography 5, 375–484. chap03.qxd 15/11/04 2:19 PM Page 63

Nutritional Ecology of Larvae and Adults 63

LeConte, J.L. (1868) New Coleoptera collected on the survey for the extension of the Union Pacific Railway, E.D. from Kansas to Fort Craig, New Mexico. Transactions of the American Entomological Society 2, 49–59. Ludwig, K.A. and Hill, R.E. (1975) Comparison of gut content of adult western and northern corn rootworm in northeast Nebraska. Environmental Entomology 4, 435–438. McKone, M.J., McLauchlan, K.K., Lebrun, E.G. and McCalls, A.C. (2001) An edge effect caused by adult corn-rootworm beetles on sunflower in tallgrass prairie remnants. Conservation Biology 15, 1315–1324. Mallet, J. and Porter, P. (1992) Preventing insect adaptation to insect-resistant crops: are seed mixtures or refugia the best strategy? Proceedings of the Royal Society of London B 250, 165–169. Metcalf, R.J. (1983) Implications and prognosis of resistance to insecticides. In: Georghiou, G.P. and Saito, T. (eds) Pest Resistance to Pesticides. John Wiley & Sons, New York, pp. 703–733. Metcalf, R.J. (1986) Foreword. In: Krysan, J.L. and Miller, T.A. (eds) Methods for the Study of Pest Diabrotica. Springer-Verlag, New York, pp. vii–xv. Moellenbeck, D.J., Barry, B.D. and Darrah, L.L. (1994) The use of artificial infestations and vertical root pulling evaluations to screen for resistance to the western corn rootworm (Coleoptera: Chrysomelidae). Journal of the Kansas Entomological Society 67, 46–52. Moellenbeck, D.J., Barry, B.D. and Darrah, L.L. (1995) Tripsacum dactyloides (Graminae) seedlings for host plant resistance to the western corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 88, 1801–1803. Moeser, J. (2003) Nutritional ecology of the invasive maize pest Diabrotica virgifera virgifera LeConte in Europe. PhD dissertation. http://webdoc.sub.gwdg.de/diss/2003/ moeser/ Moeser, J. and Vidal, S. (2004a) Response of larvae of the invasive maize pest Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae) to carbon/nitrogen ratio and phytosterol content of European maize varieties. Journal of Economic Entomology 97, 1335–1341. Moeser, J. and Vidal, S. (2004b) Do alternative hosts enhance the invasion of the maize pest Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae, Galerucinae) in Europe? Environmental Entomology (in press). Moeser, J. and Vidal, S. (2004c) Nutritional resources used by the invasive maize pest Diabrotica virgifera virgifera in its new south-east-European distribution range. Entomologia Experimentalis et Applicata (in press). Mullin, C.A., Alfatafta, A.A., Harman, J.L., Serino, A.A. and Everett, S.L. (1991) Corn rootworm feeding on sunflower and other Compositae. In: Hedin, P.A. (ed.) Naturally Occurring Pest Bioregulators. Symposium Series 449, American Chemical Society, Washington, DC, pp. 278–292. Musick, G.J., Chiang, H.C., Luckmann, W.H., Mayo, Z.B. and Turpin, F.T. (1980) Impact of planting dates of field corn on beetle emergence and damage by the western and the northern corn rootworms in the Corn Belt. Annals of the Entomological Society of America 73, 207–215. Naranjo, S.E. (1991) Movement of corn rootworm beetles, Diabrotica spp. (Coleoptera: Chrysomelidae), at cornfield boundaries in relation to sex, reproductive status, and crop phenology. Environmental Entomology 20, 230–240. Oleson, J.D. (1999) Linear scale for evaluating corn rootworm larval injury. 1999 North Central Branch, ESA Annual Meeting, Des Moines, Iowa, March. http://www.ent.iastate.edu/pest/rootworm/nodeinjury/nodeinjury.html Onstad, D.W., Guse, C.A., Spencer, J.L., Levine, E. and Gray, M.E. (2001) Modeling the dynamics of adaptation to transgenic corn by western corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 94, 529–540. chap03.qxd 15/11/04 2:19 PM Page 64

64 J. Moeser and B.E. Hibbard

Ortman, E.E. and Branson, T.F. (1976) Growth pouches for studies of host plant resistance to larvae of corn rootworms. Journal of Economic Entomology 69, 380–382. Ortman, E.E. and Fitzgerald, P.J. (1964) Evaluation of corn inbreds for resistance to corn rootworm. Proceedings North Central Branch ESA 19. Ortman, E.E. and Gerloff, E.D. (1970) Rootworm resistance: problems in measuring and its relationship to performance. In: Sutherland, J.I. and Falasca, R.J. (eds) Proceedings of the 25th Annual Corn and Sorghum Research Conference, Vol. 25, 8–10 December 1970, Chicago. American Seed Trade Association, Washington, DC, pp. 161–174. Ortman, E.E., Branson, T.F. and Gerloff, E.D. (1974) Techniques, accomplishments, and future potential of host plant resistance to Diabrotica. In: Maxwell, F.G. and Harris, F.A. (eds) Proceedings of the Summer Institute on Biological Control of Plant Insects and Diseases. University Press of Mississippi, Jackson, Mississippi, pp. 344–358. Owens, J.C., Peters, D.C. and Hallauer, A.R. (1974) Corn rootworm tolerance in maize. Environmental Entomology 3, 767–772. Oyediran, I.O., Hibbard, B.E. and Clark, T.L. (2004) Prairie grasses as hosts of the western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 33, 740–747. Painter, R.H. (1951) Insect Resistance in Crop Plants. Macmillan, New York. Pavuk, D.M. and Stinner, B.R. (1994) Influence of weeds within Zea mays crop plantings on populations of adult Diabrotica barberi and Diabrotica virgifera virgifera. Agriculture Ecosystems and Environment 50, 165–175. Pedigo, L.P. (1989) Entomology and Pest Management. Macmillan, New York, 646 pp. Peters, T.M. and Barbosa, P. (1977) Influence of population density on size, fecundity, and developmental rate of insects in culture. Annual Review of Entomology 22, 431–450. Pleau, M.J., Huesing, J.E., Head, G.P. and Feir, D.J. (2002) Development of an artificial diet for the western corn rootworm. Entomologia Experimentalis et Applicata 105, 1–11. Praiswater, T.W., Hibbard, B.E., Barry, B.D., Darrah, L.L. and Smith, V.A. (1997) An implement for dislodging maize roots from soil for corn rootworm (Coleoptera: Chrysomelidae) damage evaluations. Journal of the Kansas Entomological Society 70, 335–338. Raubenheimer, D. and Simpson, S.J. (1992) Analysis of covariance: an alternative to nutritional indices. Entomologia Experimentalis et Applicata 62, 221–231 Riedell, W.E. (1989) Western corn rootworm damage in maize: greenhouse technique and plant response. Crop Science 29, 412–415. Riedell, W.E. and Evenson, P.D. (1993) Rootworm feeding tolerance in single-cross maize hybrids from different eras. Crop Science 33, 951–955. Riedell, W.E. and Sutter, G.R. (1995) Soil moisture and survival of western corn rootworm larvae in field plots. Journal of the Kansas Entomological Society 68, 80–84. Rogers, R.R., Owens, J.C., Tollefson, J.J. and Witkowski, J.F. (1975) Evaluation of commercial corn hybrids for tolerance to corn rootworms. Environmental Entomology 4, 920–922. Rogers, R.R., Russell, W.A. and Owens, J.C. (1976a) Relationship of corn rootworm [Diabrotica] tolerance to yield in the Isss [Iowa stiff stalk synthetic] maize population. Iowa State Journal of Research 51, 125–129. Rogers, R.R., Russell, W.A. and Owens, J.C. (1976b) Evaluation of a vertical-pull technique in population improvement of maize for corn rootworm [Diabrotica virgifera, Diabrotica longicornis, Diabrotica undecimpunctata howardi] tolerance. Crop Science 16, 591–594. Rogers, R.R., Russell, W.A. and Owens, J.C. (1977) Expected gains from selection in maize for resistance to corn rootworms. Maydica 22, 27–36. Russell, W.A., Penny, L.H., Guthrie, W.D. and Dicke, F.F. (1971) Registration of corn germplasm inbreds (Reg. nos. GP 1 to 5). Crop Science 11, 140. chap03.qxd 15/11/04 2:19 PM Page 65

Nutritional Ecology of Larvae and Adults 65

Simberloff, D. (2003) How much information on population biology is needed to manage introduced species? Conservation Biology 17, 83–92. Spike, B.P. and Tollefson, J.J. (1989) Relationship of plant phenology to corn yield loss resulting from western corn rootworm (Coleoptera: Chrysomelidae) larval injury, nitrogen deﬁciency, and high plant density. Journal of Economic Entomology 82, 226–231. Stavisky, J. and Davis, P.M. (1997) The effects of corn maturity class on western corn rootworm (Coleoptera: Chrysomelidae) phenology. Journal of the Kansas Entomological Society 70, 261–271. Storer, N.P. (2003) A spatially explicit model simulating western corn rootworm (Coleoptera: Chrysomelidae) adaptation to insect-resistant maize. Journal of Economic Entomology 96, 1530–1547. Strnad, S.P. and Bergman, M.K. (1987) Distribution and orientation of western corn rootworm (Coleoptera: Chrysomelidae) larvae in corn roots. Environmental Entomology 16, 1193–1198. Svoboda, J.A. (1984) Insect steroids: metabolism and function. In: Nes, W.D., Fuller, G. and Tsai, L.S. (eds) Isoterpenoids in Plants: Biochemistry and Function. Marcel Dekker, New York, pp. 367–388. Svoboda, J.A. and Thompson, M.J. (1985) Steroids. In: Kerkut, G.A. and Gilbert, L.I. (eds) Comprehensive Insect Physiology, Biochemistry and Pharmacology, Vol. 4. Pergamon Press, New York, pp. 137–175. Waldbauer, G.P. (1968) The consumption and utilization of food by insects. Advances in Insect Physiology 5, 229–288. Weatherwax, P. (1954) Indian Corn in Old America. Macmillan, New York, 253 pp. Weiss, M.J., Seevers, K.P. and Mayo, Z.B. (1985) Inﬂuence of western corn rootworm larval densities and damage on corn rootworm survival, developmental time, size and sex ratio (Coleoptera: Chrysomelidae). Journal of the Kansas Entomological Society 58, 397–402. Wilson, R.L. and Peters, D.C. (1973) Plant introductions of Zea mays as sources of corn rootworm tolerance. Journal of Economic Entomology 66, 101–104. Wilson, R.L., Abel, C.A., Wiseman, B.R., Davis, F.M., Williams, W.P., Barry, B.D. and White, W.H. (1995) Evaluation for multiple pest resistance in European corn borer, Ostrinia nubilalis, resistant maize accessions from Peru. Journal of the Kansas Entomological Society 68, 326–331. Witkowski, J.F., Keith, D.L. and Mayo, Z.B. (1986) Western Corn Rootworm Soil Insecticide Treatment Decisions Based on Beetle Numbers. University of Nebraska Cooperation Extension G86–777A. Xie, Y.S., Arnason, J.T., Philogene, B.J.R. and Lambert, J.D.H. (1990) Role of 2,4-dihydroxy- 7-methoxy-1,4-benzoxazin-3-one (DIMBOA) in the resistance of maize to western corn rootworm, Diabrotica virgifera virgifera (LeConte) (Coleoptera: Chrysomelidae). Canadian Entomologist 122, 1177–1186. Xie, Y.S., Arnason, J.T., Philogene, B.J.R., Olechowski, H.T. and Hamilton, R.I. (1992a) Variation of hydroxamic acid content in maize roots in relation to geographic origin of maize germplasm and resistance to western corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 85, 2478–2485. Xie, Y.S., Arnason, J.T., Philogene, B.J.R., Atkinson, J. and Morand, P. (1992b) Behavioral responses of western corn rootworm larvae to naturally occurring and synthetic hydroxamic acids. Journal of Chemical Ecology 18, 945–957. chap03.qxd 15/11/04 2:19 PM Page 66 chap04.qxd 24/11/04 1:38 PM Page 67

4 Western Corn Rootworm, Cucurbits and Cucurbitacins

DOUGLAS W. TALLAMY,1 BRUCE E. HIBBARD,2 THOMAS L. CLARK2 AND JOSEPH J. GILLESPIE3 1University of Delaware, Department of Entomology and Wildlife Ecology, Newark, USA; 2USDA-ARS, University of Missouri, Columbia, Missouri, USA; 3Texas A&M University, Department of Entomology, College Station, Texas, USA

For more than a century, researchers have noted the curious attraction of adult luperine chrysomelids in the subtribes Diabroticina and Aulacophorina to cucurbit species rich in the bitter compounds collectively called cucurbitacins (Webster, 1895; Contardi, 1939; Metcalf et al., 1980). The attraction is curious, not because these beetles can locate cucurbits over long distances by tracking ﬂower and wound volatiles, and not because cucurbitacins are phagostimulants for Diabroticites that, despite their noxious effects on other insects (Nielsen et al., 1977; Tallamy et al., 1997a), cause them to eat anything containing these compounds (Sinha and Krishna, 1970; Metcalf et al., 1980). Such behaviours charac- terize most, if not all, phytophagous insects that specialize on a particular group of plants (Rosenthal and Janzen, 1979). What is curious is that most luperines that are attracted to cucurbits are not cucurbit specialists. Instead, they develop to maturity on plants from several families, such as Poaceae, Solanaceae, Convolvulaceae, Fabaceae and Asclepiadaceae. The western corn rootworm (WCR), Diabrotica virgifera virgifera, is no exception. As a member of the virgifera subdivision of the large genus Diabrotica (Wilcox, 1972a), it is a specialist on the Poaceae and ances- trally probably reached maturity only on grasses, such as Zea spp., Setaria spp. and couch grasses in several genera (Smith, 1966; Clark and Hibbard, 2004; Oyediran et al., 2004; Moeser and Hibbard, Chapter 3, this volume). It is no surprise, then, that WCR has so successfully adopted the comparatively nutritious and toxin-free Zea mays for growth and reproduction. Yet WCR is the quintessential example of a luperine that, upon reaching adulthood, leaves its nutritionally complete grass host in search of cucurbits embittered with cucurbitacins. This chapter will explore the phylogenetic, evolutionary, ecological and applied implications of WCR’s intriguing afﬁnity for cucurbits.

68 D.W. Tallamy et al.

Taxonomy of Western Corn Rootworm

WCR (Coleoptera: Chrysomelidae), D. v. virgifera LeConte, belongs to the largest tribe within the Galerucinae, the Luperini (Wilcox, 1965, 1972a). Historically, the division of Galerucinae into tribes was in a state of flux. Several authors devised numerous classification systems, resulting in taxonomic chaos until Wilcox (1965) provided a revision based upon male characters that stabilized classification at this level. Wilcox (1965) and Seeno and Wilcox (1982) divided the Luperini into three subtribes: Aulacophorina (formerly known as Monleptina), Luperina and Diabroticina, with the latter containing D. virgifera and being strictly New World. The Diabroticina are divided further into four sections: Diabroticites, Ceratomites, Phyllecthrites and Trachyscelidites, with the genus Diabrotica being placed within the Diabroticites (Wilcox, 1972a). Diabrotica is divided into three species subgroups: fucata, signifera and virgifera, and consists of 333 valid species (Wilcox, 1972a). The division of Diabrotica into subgroups was devised by Smith and Lawrence (1967) using external characters and has since been confirmed using allozyme markers (Krysan et al., 1989) and phylogenetic analysis of nuclear and mitochondrial genes (Clark et al., 2001). D. virgifera is classified within the virgifera species group, with LeConte providing the first formal description of the species in 1868 from beetles collected near Fort Wallace, Kansas, USA, in 1867 (Smith and Lawrence, 1967). While specimens in LeConte’s collection are considered to be the original type series, Smith and Lawrence (1967) contend that D. virgifera was also part of Say’s original Diabrotica longicornis (Say) mixed type series collected 21 km west of present-day Pueblo, Colorado, USA, in 1820. Their con- tention was based upon the original description and collection records from the same area in 1965. To prevent further taxonomic confusion within this group (D. v. virgifera would have had to become D. longicornis), Smith and Lawrence chose to keep the original names. Until 1980, WCR, as described by LeConte and known as a pest throughout northern central USA, was considered to be the only variant of the species; however, evidence based on mating compatibility (Krysan and Branson, 1977), egg diapause intensity (Krysan and Branson, 1977; Krysan et al., 1977), behaviour (Krysan et al., 1980) and morphology (Krysan et al., 1980) confirmed that D. virgifera has two subspecies, with WCR becoming D. v. virgifera and Mexican corn rootworm becoming D. v. zeae Krysan and Smith (Krysan et al., 1980).

Purported Origin of Western Corn Rootworm

The geographical distribution of WCR is primarily constrained by the presence of suitable larval hosts (Branson and Krysan, 1981). For populations to exist, there must be phenological synchrony between egg hatch and the availability of host roots for larvae to feed upon. One mechanism chap04.qxd 24/11/04 1:38 PM Page 69

Western Corn Rootworm, Cucurbits and Curcubitacins 69

that allows this synchrony to occur for many virgifera group taxa is an egg diapause that breaks as conditions become favourable (Branson and Krysan, 1981; Krysan, 1982). Branson and Krysan (1981) argued that this adaptation allows virgifera species to specialize on perennial grasses that become available during the rainy season in the tropics or during the summer months in more temperate regions. Their argument is supported by the presence of virgifera group taxa such as Diabrotica lemniscata LeConte, Diabrotica cristata (Harris) and D. longicornis (Say) in temperate regions of the USA where perennial grasses are a permanent feature in the landscape; however, most of the larval host plants are unknown, especially for non-economically important species. The planting of maize on an annual basis in temperate regions within the USA provides a perennial resource for WCR to exploit. Thus, the distribution of species in the virgifera group apparently is not directly related to climate, as most species in the group are capable of withstanding environmental extremes such as freezing and drought in the egg stage (Krysan and Smith, 1987). Instead, species in this group are limited to the distribution of their obligatory larval host plants. Furthermore, Krysan and Smith (1987) hypothesized that the progenitor of the virgifera group was probably a specialist on grasses in a region that typically had alternating wet and dry seasons, much like what occurs in parts of Mexico and Guatemala, the region that contains the greatest diversity of taxa for the virgifera species group. Several studies provide evidence that many grasses other than maize fall within the larval host range of WCR (Branson and Ortman, 1967, 1970; Clark and Hibbard, 2004; Oyediran et al., 2004). Furthermore, there is evidence that WCR and its subspecies, Mexican corn rootworm, continue to reproduce on plants other than maize in environments near maize production (Hill and Mayo, 1980; Branson et al., 1982). It is difﬁ- cult to infer origin of the species based on these works. Branson and Krysan (1981) provided the most plausible explanation regarding the origin of WCR. They hypothesized that D. virgifera (both subspecies) initially included progenitorial maize as an incidental host prior to its transformation to a food crop by indigenous peoples. As maize was developed in what is now present-day Guatemala and Mexico, it is thought that D. virgifera remained associated with it as its range expanded northward through trade and the expansion of prehistoric agriculture (Smith, 1966; Galinat, 1977; Branson and Krysan, 1981). While both rootworm subspecies were most probably associated with maize as it was cultivated by these early indigenous peoples, it is likely that neither subspecies reached pest status because the type of agriculture practised by those ancient peoples did not promote a build-up of rootworm populations (Smith, 1966; Mangelsdorf, 1974). Population expansion did not occur until the Spanish introduction of European-style maize monocultures, which created favourable environments for massive build-ups of D. virgifera populations in regions where these monocultures exist. A classic example of this occurred when the westward expansion of maize monoculture in northern central USA reached the eastern limits of the WCR chap04.qxd 24/11/04 1:38 PM Page 70

70 D.W. Tallamy et al.

Fig. 4.1. Cucurbitacin B.

geographical range, presumably some time in the 1940s in western Kansas or Nebraska, and ﬁnally the east coast of the USA by 1983 (Chiang, 1973; Krysan and Smith, 1987). While this evidence suggests a strong WCR association with maize, it should be noted that LeConte’s original type series collected in 1868 and Say’s original mixed type series of D. longicornis collected in 1820 (which most probably contained WCR) were both collected from Cucurbita foetidissima Humboldt, Bonpland and Kunth in geographical regions that had a high probability of being devoid of maize at the time (Smith and Lawrence, 1967). Therefore, it is likely that WCR had an association with grasses beyond maize prior to its well-documented eastward expansion in northern central USA.

WCR–Cucurbitacin Interactions

The cucurbitacins are a group of non-volatile, highly oxygenated tetracyclic triterpenes with a unique 19(10→9β)abeo-10-α-lanostane (cucurbitane) skeleton (Fig. 4.1). The biological activity of this group of compounds has been recognized for centuries and cucurbitacins have been used as a laxative and emetic and in the treatment of malaria, dysen- tery and dysmenorrhoea (Lavie and Glotter, 1971; Halaweish, 1987; Miro, 1995). More recently, cucurbitacins have received a great deal of attention because of their antitumour properties, differential cytotoxicity towards renal, brain tumour and melanoma cell lines (Cardellina et al., 1990; Fuller et al., 1994), their inhibition of cell adhesion (Musza et al., 1994) and possible antifungal effects (Bar-Nun and Mayer, 1989). Cucurbitacins are produced in at least some tissues of all members of chap04.qxd 24/11/04 1:38 PM Page 71

Western Corn Rootworm, Cucurbits and Curcubitacins 71

the Cucurbitaceae (Gibbs, 1974; Guha and Sen, 1975; Jeffrey, 1980) and a few species in other plant families (Curtis and Meade, 1971; Pohlman, 1975; Dryer and Trousdale, 1978; Thorne, 1981). In most species they are concentrated in roots and fruits, with lesser amounts in stems and leaves. Because of their extreme bitterness, cucurbitacins are thought to be involved in plant protection against herbivores (Metcalf, 1985; Tallamy and Krischik, 1989). Nevertheless, cucurbitacins are phagostimulants for both adults (Metcalf et al., 1980) and larvae (DeHeer and Tallamy, 1991) of several luperine species in the subtribes Aulacophorina and Diabroticina (Table 4.1) and can have important ecological consequences for plants that possess them (Tallamy and Krischik, 1989). Adult luperines can detect cucurbitacins in nanogram quantities and readily devour bitter plant material (Metcalf, 1994; Tallamy et al., 1998). In addition to WCR, cucurbitacins inﬂuence the behaviour of several important crop pests, including Diabrotica balteata LeConte, the banded cucumber beetle, Diabrotica barberi Smith and Lawrence, the northern corn rootworm, Diabrotica undecimpunctata howardi Barber, the southern corn rootworm, and Diabrotica speciosa, a crop pest in Central and South America. Studies have shown that, when WCR eat crystalline cucurbitacins for 2 days, they excrete 85% of the material and permanently sequester the

Table 4.1. Effects of cucurbitacins on the feeding of luperine leaf beetles (modiﬁed from Matsuda, 1988). Phagostimulation by cucurbitacin analogues is represented by a +. Apart from D.W.T. Tallamy (unpublished), data are from Chambliss and Jones (1966); Sinha and Krishna (1970); Metcalf et al. (1980); Ferguson et al. (1983); Nishida and Fukami (1990); Mehta and Sandhu (1992); Metcalf and Metcalf (1992); Lewis and Metcalf (1996); Eben et al. (1997); Tallamy et al. (1997a); and Abe et al., 2000).

Cucurbitacin analogues Species A B C D E I L

Diabrotica undecimpunctata undecimpunctata –+–+++– Diabrotica undecimpunctata howardi +++++++ Diabrotica virgifera virgifera –+–++++ Diabrotica longicornis –+–+++– Diabrotica cristata –+–+++– Diabrotica balteata ++–+++– Diabrotica speciosa –+–+––– Cerotoma arcuata –+–+––– Acalymma vittatum ++–+++– Acalymma trivittatum –+–+++– Aulacophora foveicollis ++++––– Aulacophora femoralis –+––++– Aulacophora nigripennis –+––––– Aulacophora lewisii –+–+++– Trachyscelida sp. – – chap04.qxd 24/11/04 1:38 PM Page 72

72 D.W. Tallamy et al.

remainder in their fat bodies, cuticles, haemolymph, spermatophores and developing eggs (Ferguson and Metcalf, 1985; Andersen et al., 1988; Tallamy et al., 2000). Regardless of the cucurbitacin configuration eaten, beetles transform it through glycosylation, hydrogenation, desaturation and acetylation into 23,24-dihydro-cucurbitacin D (Andersen et al., 1988; Nishida et al., 1992). There are decided defensive benefits to cucurbitacin sequestration. Beetles that have eaten cucurbitacins become highly distasteful and are readily rejected by predators such as mantids, mice and finches (Ferguson and Metcalf, 1985; Nishida and Fukami, 1990; D.W. Tallamy, unpublished data). Sequestered cucurbitacins may also discourage parasitoids such as tachinid flies in the genus Celatoria, although this has never been tested. Moreover, when cucurbitacins have been sequestered in eggs and larvae, both of which are denizens of pathogen-rich damp soil, survival after exposure to the entomopathogen Metarhizium anisopliae is significantly improved (Tallamy et al., 1998). This may explain why females shunt 79% of the cucurbitacins that are not excreted into their eggs or the mucus coating of the eggs (Tallamy et al., 2000). Despite the benefits to female WCR from eating bitter cucurbit tissues, it is males rather than females which actively seek these compounds in nature. In a field trial quantifying the sex ratio of beetles that came to cucurbitacin-rich fruits of Cucurbita andreana, Tallamy et al. (2002) found that 99% of the 224 WCR found at the fruits over a 5-day period were males. This result concurs with the male-biased sex ratios frequently found in cucurbitacin traps (Shaw et al., 1984; Fielding and Ruesink, 1985). Apparently females rely on males for their primary source of cucurbitacins (Tallamy et al., 2000). Males sequester 89% of the cucurbitacins not excreted after ingestion in their spermatophores and pass them to females during copulation. Whether such behaviour imparts a mating advantage to WCR males has not been investigated.

Pharmacophagy, Luperines and Cucurbits

One of the most interesting relationships between insect herbivores and their host plants occurs when a herbivore seeks a phytochemical that is toxic to other animals. Such behaviour has focused attention on luperine chrysomelids for decades (Webster, 1895; Metcalf et al., 1980; Nishida and Fukami, 1990; Nishida et al., 1992; Tallamy et al., 1999). As discussed above, cucurbitacins are phagostimulants for many luperine adults (Metcalf et al., 1980; Nishida and Fukami, 1990; Tallamy et al., 1997b) and larvae (DeHeer and Tallamy, 1991), despite their extreme bitterness and their ability to kill or repel most invertebrate and vertebrate herbivores (David and Vallance, 1955; Watt and Breyer-Brandwijk, 1962; Nielsen et al., 1977; Tallamy et al., 1997a). Of greatest interest are the species of luperines that do not (and apparently cannot) feed on cucurbit roots as larvae, but consume pure crystalline cucurbitacins as adults when given the chance. chap04.qxd 24/11/04 1:38 PM Page 73

Western Corn Rootworm, Cucurbits and Curcubitacins 73

The relationship between non-cucurbit specialists and cucurbitacins differs from that of most other insect herbivores that feed upon biotoxins. Many specialist herbivores use toxic phytochemicals as host recognition cues (Fraenkel, 1959; Feeny et al., 1970), while others encounter defensive compounds incidentally and unavoidably (Brower et al., 1972). These organisms, however, often exhibit elaborate behaviours to minimize exposure to their hosts’ toxins (Dussourd, 1993). Luperines such as WCR, in contrast, have been described as pharmacophagous insects (Nishida and Fukami, 1990) because they search for particular phytochemicals for purposes other than primary metabolism or host recognition (Boppré, 1990). Pharmacophagy was first described for certain Blattaria (Blattellidae, Blattidae), Orthoptera (Tettigoniidae, Gryllidae, Pyrgomorphidae), Coleoptera (Chrysomelidae, Cerambycidae), Lepidoptera (Danaidae, Ctenuchidae, Riodinidae, Pericopidae, Ithomiidae, Arctiidae, Noctuidae) and Diptera (Tephritidae, Chloropidae) which leave nutritionally suitable host plants to seek noxious pyrrolizidine alkaloids (Meinwald et al., 1969; Pliske, 1975; Schneider et al., 1982; Boppré et al., 1984; Krasnoff and Dussourd, 1989; Boppré, 1990). As with insects pharmacophagous toward pyrrolizidine alkaloids, there is no evidence that pharmacophagous luperines seek cucurbitacins to satisfy a nutritional requirement (although, once ingested, cucurbitacins may substitute for cholesterol under some conditions (Halaweish et al., 1999)). Nor can cucurbitacins serve as a cue for host plant recognition. By definition, there are no cucurbitacins in the hosts of pharmacophagous luperines! The most frequently cited benefit of pharmacophagy is defence (Boppré, 1984). Whether obtained through pharmacophagy or specialization on cucurbits, cucurbitacins persist in the cuticle, fat bodies and haemolymph (Ferguson et al., 1985; Andersen et al., 1988) and provide protection against predators (Ferguson and Metcalf, 1985; Nishida and Fukami, 1990) and/or pathogens (Tallamy et al., 1998). Perhaps because of their defensive benefits, both cucurbitacins and pyrrolizidine alkaloids have also become an integral component of the reproductive behaviour of participating species (Dussourd et al., 1991; LaMunyon and Eisner, 1993; Tallamy et al., 2000). In both cases, the pharmacophagous agent is consumed directly by females and/or is sequestered by males and passed within spermatophores to females. Females, in turn, shunt the majority of these materials to developing eggs.

Origins of Cucurbitacin Pharmacophagy

Ancestral host hypothesis

If defence and mating advantages are beneﬁts imparted to all pharmacophagous insects, the selective maintenance of such behaviour is no mystery. The origins of insect pharmacophagy, however, are controver- chap04.qxd 24/11/04 1:38 PM Page 74

74 D.W. Tallamy et al.

sial and poorly understood. Although larval host plants of pharmacophagous taxa, by definition, do not produce the chemical that is subsequently sought by adults (Boppré, 1984), larvae of a few ithomiid and danaid relatives of species pharmacophagous to pyrrolizidine alkaloids do, in fact, develop on plants containing pyrrolizidine alkaloids (Edgar et al., 1974). Thus, students of pyrrolizidine alkaloid pharmacophagy have cautiously concluded that, at some point in the evolutionary history of the diverse group of insects exhibiting this behaviour, immatures must have developed on plants that contained the pharmacophagous compound (Pliske, 1975; Boppré, 1978; Edgar, 1982; Trigo and Molta, 1990). This concept was modified by Dussourd (1986) to suggest that it is variation in the host’s production of the compound selected for receptors that enables adults with inadequate supplies of alkaloids to supplement their needs through pharmacophagous forays to non-host sources. The evolution of cucurbitacin pharmacophagy in the Luperini has been viewed similarly (Metcalf, 1979, 1994; Metcalf et al., 1980), because: (i) cucurbitacins are phagostimulants for some species of both Old World Aulacophorina and New World Diabroticina (Table 4.1); (ii) at least two genera of Luperini (Acalymma and Aulacophora) are larval host specialists on cucurbits (Wilcox, 1972a; Monroe and Smith, 1980); and (iii) all pharmacophagous species tested to date are stimulated most by cucurbitacin B, the most ubiquitous of the 46-plus known cucurbitacin configurations (Hill et al., 1991). Luperine responses to cucurbitacins are thought to be derived from plesiomorphic traits that arose from an ancestral host relationship with the Cucurbitaceae. Under this hypothesis, the phagostimulatory response to cucurbitacins in species that currently develop only on non-cucurbitaceous host plants is a relic of a long-lost need for host recognition and is a trait that is currently maintained through secondary selection from protection benefits associated with cucurbitacin consumption (Ferguson and Metcalf, 1985; Tallamy et al., 1998). The majority of luperine species, then, are assumed to have under- gone host shifts away from cucurbits to plants in other families. It is important to note here that pharmacophagous visits to cucurbits by adult luperines that have developed on other plants have vastly inflated luperine host records for Cucurbitaceae (e.g. Metcalf, 1994; Jolivet and Hawkeswood, 1995). Unfortunately, our knowledge of true larval hosts in the Luperini is extremely limited because the larvae of all species develop underground on roots and are seldom collected or correctly identified with their respective adults.

Loose receptor hypothesis

There is intuitive appeal in assuming an ancient relationship between luperine ancestors and early cucurbits, but it is not the most parsimonious explanation for why so many extant luperine lineages do not use cucurbits as larval hosts. Recent evidence from other pharmacophagous chap04.qxd 24/11/04 1:38 PM Page 75

Western Corn Rootworm, Cucurbits and Curcubitacins 75

insects suggests that an ancestral association with a particular compound may not be necessary to promote the evolution of pharmacophagy (Tallamy et al., 1999). Studies of cantharidin, a noxious, volatile monoter- pene produced de novo as a nuptial gift exclusively by male meloid and oedomerid beetles, provide a number of examples whereby exposure to a novel and potentially deleterious compound results in phagostimulation rather than deterrence. Cantharidin attracts and stimulates feeding in several diverse taxa, including pyrochroid, endomychid, anthicid, staphylinid and chrysomelid beetles, ceratopogonid, sciarid and anthomyiid flies, braconid wasps and mirid and tingid bugs, which may have had no evolutionary exposure to the compound (Young, 1984; Frenzel et al., 1992; Frenzel and Dettner, 1994; Mafra-Neto and Jolivet, 1994; Eisner et al., 1996). It is possible, however, that canthariphagy arose because of the coincidental acceptance of cantharidin as a novel agonist by peripheral receptors with less than perfect specificity rather than from adaptive responses to a historically familiar compound. The neurophysiological basis of peripheral perception is extraordi- narily complex in insect gustatory systems (Frazier, 1986; Simmonds et al., 1990; Städler, 1991; Schoonhoven et al., 1992; Mullin et al., 1994). In the simplest terms, feeding behaviour is stimulated if the chemoreception of phagostimulants exceeds the chemoreception of feeding deterrents (Dethier, 1980). In caterpillars and possibly all insects, taste sensilla contain cells specialized for the production of either deterrent and stimulatory inputs, or, more probably, neurons capable of producing both deterrent and stimulatory inputs (Frazier, 1986). Receptor sites on these cells can be highly specific (tight) or less specific (loose). Strychnine, for example, is a compound novel to phytophagous insects, but it readily depolarizes activation channels, leading to deterrent input in most insects. The binding requirements at these sites are sufficiently loose for a variety of molecular structures to meet the polarity and configuration specifications for binding there. The loose characteristics of receptor sites with deterrent capabilities may be adaptive because they protect the central nervous system from exposure to damaging novel compounds (Frazier, 1992). Critical to the loose receptor hypothesis is the fact that relatively loose binding properties of receptor sites can also enable novel and sometimes deleterious compounds to trigger feeding behaviour. There are several mechanisms by which this can happen (Frazier, 1986, 1992). Some molecules bind at receptor sites leading to deterrent inputs, but, rather than depolarizing the activation channels, they simply block them. Without inhibitory inputs, even small amounts of phagostimulants, including amino acids present in the insects’ saliva, are sufficient to acti- vate the stimulatory inputs at the sensillum and elicit feeding. Activation leading to deterrent inputs can also be inhibited when particular molecules block the stimulus removal system. Finally, loose stimulatory receptor sites themselves can encourage phagostimulation by novel compounds with the appropriate configuration and polarity at binding sites. chap04.qxd 24/11/04 1:38 PM Page 76

76 D.W. Tallamy et al.

This is apparently the mechanism by which the peptide aspartame mimics the carbohydrate sucrose at vertebrate receptors, a mimicry upon which much of the sweetener industry is based. Another example of a loose receptor exists within WCR as well. WCR larvae are highly attracted to carbon dioxide (Strnad et al., 1986; Hibbard and Bjostad, 1988) and dichloromethane (Jewett and Bjostad, 1996). Apparently, the size and polarity of the chlorine atoms in dichloromethane are similar to those of carbonic acid, the water-soluble form of carbon dioxide (Jewett and Bjostad, 1996). How might the loose receptor hypothesis explain cucurbitacin pharmacophagy in luperines? Although data are few, a review of the adult feeding habits of luperines suggests that there is a perfect correlation between adult afﬁnity for pollen and cucurbitacin phagostimulation: pollen is a substantial component of the adult diet in species stimulated to eat by cucurbitacins (Diabrotica, Acalymma, Aulacophora), while species in which the adult diet is largely conﬁned to foliage are repelled by cucurbitacins (Cerotoma trifurcata, Trachyscelida spp., all Luperina) (Sinha and Krishna, 1970; Metcalf et al., 1980; D.W. Tallamy, unpublished data). One scenario, then, is that the phagostimulatory response to cucurbitacins arose through adult feeding behaviours rather than through larval host dependencies. The link between pollen feeding and cucurbitacin phagostimulation may be based on similarities in structures of common pollen constituents and cucurbitacins (Tallamy et al., 1999). Amino acid neuroreceptors exhibiting γ-aminobutyric acid (GABA)/glycine pharmacology are located on the maxillary galeae of Diabrotica and have been implicated in the perception of both antifeedants and phagostimulants (Mullin et al., 1992; Chyb et al., 1995; Hollister and Mullin, 1998; Kim and Mullin, 1998). Cucurbitacins do not occur in pollen (Andersen and Metcalf, 1987), but pollen and meristematic tissues favoured by adult pharmacophagous luperines are enriched with similar mid-polar (mildly lipophilic) compounds such as brassinosteroids, ω-3-linoleic acid-containing lipids and hydroxycinnamic acid-polyamine amides, along with polar low- molecular-weight neutral amino acids, including GABA (Barber, 1971; Stanley and Linskens, 1974; Erhardt and Baker, 1990; Marquardt and Adam, 1991; Feldlaufer et al., 1993; Mullin et al., 1993; Lin and Mullin, 1999). Many of these compounds elicit phagostimulatory responses from Diabrotica amino acid receptors (Mullin et al., 1994; Hollister and Mullin, 1998; Lin and Mullin, 1999), either alone or while interacting at multiple sites. In particular, some pollen sterols potentiate the amino acid agonists in pollen (Chyb et al., 1995; Hollister and Mullin, 1998; Kim and Mullin, 1998). Current data suggest that it is these same taste neurons that are depolarized by cucurbitacins (Mullin et al., 1994). Cucurbitacins are structurally similar to many sterols, sharing both their hydroxylation and their stereochemistry (Dinan et al., 1997a,b). The unusual 4,4-dimethyl-5-ene structure of cucurbitacin B, for example, orients the 3-one group below the ring plane (Mullin et al., 1994) and thus closely resembles GABA-acting 3α-ol pregnane steroids (Purdy et al., chap04.qxd 24/11/04 1:38 PM Page 77

Western Corn Rootworm, Cucurbits and Curcubitacins 77

1990). Like pregnane steroids, cucurbitacins potentiate amino acids such as alanine, serine, praline and GABA (compounds present in pollen and saliva alike) and trigger a feeding response in WCR even at very low doses (Mullin et al., 1994). The structural similarity between cucurbitacins and many sterols also makes these molecules into powerful antagonists at insect ecdysteroid receptors (Dinan et al., 1997a,b). It is possible that selection on luperine amino acid receptors for the loose perception of pollen constituents simultaneously rendered these receptors susceptible to cucurbitacin depolarization. Since ancestral chrysomeloid beetles were associated with polliniferous food sources even before the division of the cerambycid and chrysomelid lineages in the early Cretaceous (Samuelson, 1994), it is likely that in certain lineages pollen feeding and the evolution of taste neurones associated with pollen feeding preceded luperine interactions with cucurbits and their cucurbitacins. Thus, the response of Luperini to cucurbitacins may not reflect an ancestral larval association with cucurbitacin-producing plants as suggested for luperines in particular (Metcalf, 1979) and implied by pharmacophagous theory in general (Edgar, 1982). Rather, it may be the product of a physiological coincidence mediated by the molecular configuration of cucurbitacins that promotes binding at a steroidal site on amino acid receptors. Under this scenario, beetles that evolved to exploit cucurbit pollen because of its abundance and nutritional richness (Reddi and Aluri, 1993) were, regardless of their larval host, serendipitously placed in contact with cucurbitacins, which are components of cucurbit anthers and flower petals, but not pollen (Andersen and Metcalf, 1987). In the early stages of the interaction, beetles that were stimulated to feed on cucurbitacins would have suffered reduced fitness. Cucurbitacin avoidance, however, could not have evolved without a significant tightening of the pollen- adapted amino acid receptors, compromising the detection of an evolu- tionarily entrenched adult food source. Repeated exposure to cucurbitacins would have favoured physiological tolerance, while advantages gained from their defensive properties would have encouraged beetle associations with cucurbitacin sources.

Phylogenetic Resolution

Whether or not cucurbitacin pharmacophagy in the luperine arose through ancestral host specialization on cucurbits or a stochastic similarity between the molecular structure of cucurbitacins and pollen sterols may never be known deﬁnitively, but mapping larval host range, cucurbitacin phagostimulation and pollen use on a reconstructed luperine phylogeny will infer much about the evolution of this behaviour. We have learned a great deal about phylogenetic relationships among the Luperini in recent years (Gillespie et al., 2003, 2004), but our knowledge of the distribution of cucurbitacin phagostimulation, pollen use and larval host range remains scanty. Nevertheless, Gillespie et al. (2003, 2004) provide chap04.qxd 24/11/04 1:38 PM Page 78

78 D.W. Tallamy et al.

a phylogenetic analysis of host range and cucurbitacin phagostimulation that is sufficiently robust to test the primary predictions of the ancestral host hypothesis (Fig. 4.2). Using partial DNA sequences from two gene regions, phylogenetic trees have been reconstructed for over 100 luperine taxa under several optimality criteria (equally weighted parsimony, dif- ferentially weighted parsimony and maximum likelihood), all of which unambiguously suggest that the rootworms are a paraphyletic assemblage of beetles. Interestingly, the Old World cucurbit specialists (Aulacophora and possibly related genera) and the New World cucurbit specialists (Acalymma) do not share a common ancestor at the tribal level within the subfamily Galerucinae. The luperine subtribe Diabroticina, which contains the cucurbit specialists Acalymma spp. and several other genera that are pharmacophagous towards cucurbitacins, shares affinities with two other galerucine tribes, the Metacyclini and the Galerucini. The subtribe Aulacophorina, which contains the cucurbit specialists in the genus Aulacophora, is a derived lineage within the third and largest rootworm subtribe, the Luperina. The remaining tribes of the Galerucinae, the Oidini and the Hylaspini, as well as the rogue diabroticine genus Trachyscelida, fluctuate in their placement in these rootworm lineages (Gillespie et al., 2004); however, the authors point out that the under- sampling of these lineages has clearly affected the stability of their phylogenetic position throughout the analyses. Recent analyses with thorough samplings of the Oidini and Hylaspini are adding more stability to the entire galerucine tree, and the mode of rootworm paraphyly discussed above is still strongly supported (Gillespie et al., 2004). If cucurbitacin pharmacophagy developed through an ancestral host association with Cucurbitaceae, then genera specializing on cucurbits (Acalymma and Aulacaphora) are predicted to be basal to other lineages within the Luperini. Moreover, if the behaviour is very old and arose before the breakup of Gondwana, as implied by the worldwide distribution of species specializing on cucurbits (Acalymma and Aulacaphora), and derived pharmacophagous species are expected to belong to the same or related lineages; that is, pharmacophagy should be a monophyletic trait. This evolutionary scenario is examined in a reconstructed phylogeny of the Galerucinae (Fig. 4.2A). When a single origin for a luperine–cucurbit association in an ancestor to this hypothetical Aulacophora/Acalymma lineage is assumed, it must be placed at the root to the Galerucinae sensu stricto (minus the flea beetles). Assuming that cucurbit specialization preceded cucurbitacin pharmacophagy in the hypothetical galerucine ancestor, this explanation for the evolution of pharmacophagy suggests that the majority of galerucine lineages, totalling over 4000 described species (Wilcox, 1972a,b), have shifted to other host plants outside the Cucurbitaceae, with only two genera retaining cucurbit specialization. This is clearly not the most parsimonious explanation for the evolution of cucurbitacin–luperine associations. To adopt this hypothesis, one must explain why the majority of species in such a successful beetle radiation have aborted a host plant affiliation that provides chap04.qxd 24/11/04 1:38 PM Page 79

Western Corn Rootworm, Cucurbits and Curcubitacins 79

A B Outgroups Ancestral Loose Host Receptor Hypothesis Alticini Hypothesis

Luperina + Oidini

* Old World cucurbit specialists * * * * ((AulacophoraAulacophora spp.spp)) * Aulacophorina

GalerucinaeGalerucinae Luperina GalerucinaeGalerucinae sensu lato + sensusensu lato lato sensu lato Hylaspini + Trachyscelida sp.

Metacyclini

GalerucinaeGalerucinae GalerucinaeGalerucinae sensusensu strictostricto sensusensu stricto stricto

Galerucini

Phyllecthrites

* Cerotomites * + Gynandrobrotica spp.

* * * * * New World * * * * cucurbit specialists * * * * (Acalymma spp)spp.) * * * Diabroticina Diabroticina ( - Trachyscelida sp.) ( - Trachyscelida sp.) * *

Diabroticites

* * ** **

Fig. 4.2. Mirror image phylogram showing the loss and gain of cucurbitacin feeding under two evolutionary scenarios (redrawn from Gillespie et al., 2004). Bold horizontal bars denote the gain of cucurbitacin feeding, and asterisks represent lineages known to utilize cucurbitacins. (A) The phylogram on the left (ancestral host hypothesis) illustrates cucurbitacin feeding in aulacophorine and diabroticine lineages as a plesiomorphic remnant of an ancestral host association between the Cucurbitaceae and Luperini, Metacyclini, Oidini, Hylaspini and Galerucini. (B) The phylogram on the right illustrates cucurbitacin feeding in some aulacophorine and diabroticine lineages as a product of convergent evolution in these geographically isolated lineages. chap04.qxd 24/11/04 1:38 PM Page 80

80 D.W. Tallamy et al.

highly selective advantages in the form of chemical defence and/or beneficial mating strategies. Moreover, it is difficult to support this hypothesis if an ancestral association between Cucurbitaceae and an immediate aulacophorine/diabroticine ancestor is perceived as the origin for pharmacophagy. In contrast, the loose receptor hypothesis carries no assumptions about the position of cucurbit specialists in the phylogeny of the Luperini. It does, however, assume that the evolution of pollen specialization preceded or is directly correlated with cucurbitacin pharmacophagy. Under this evolutionary scenario, pharmacophagy by convergent evolution unrelated to larval host use explains the independent origin of cucurbitacin utilization in Old and New World rootworm lineages (Fig. 4.2B). Given the strongly supported paraphyletic relationships between the Aulacophorina and Diabroticina, the loose receptor hypothesis is currently the more parsimonious model explaining the evolution of Cucurbitaceae/Luperini associations. It is important to note that cucurbitacin feeding may have arisen independently in basal lineages of both the Aulacophorina and the Diabroticina and then proceeded to cucurbitacin pharmacophagy via host switching as the ancestral host hypothesis predicts. Whether or not the ancestral host hypothesis supports the evolution of cucurbitacin pharmacophagy in Old World cucurbit feeders will remain an open question until more taxa are sampled and more is known about the host plants used by the basal lineages within the Aulacophorina. Within the Diabroticina, however, a well-supported ancestor is beginning to emerge (Fig. 4.2). A monophyletic group of bean-feeding species, comprised of taxa sampled from the Phyllecthrites, Cerotomites and the genus Gynandrobrotica (Diabroticites), is basally separated from the remaining Diabroticina (– Trachyscelida sp.). It is possible that the association between Acalymma spp. (Diabroticina) and cucurbits occurred in an ancestor to the Diabroticites, with pharmacophagy arising in other genera (such as Diabrotica, Paratriarius and Isotes) as a means of retaining the benefits imparted by cucurbitacin sequestration that are not provided by other host plants. This scenario, however, does not explain the affinity Cerotoma arcuata, a legume feeder, has with cucurbitacins (Nishida and Fukami, 1990). Because this species is grouped within the basal bean- feeding lineage (Cerotomites), we cannot account for its cucurbitacin affinity by an ancester that once specialized on cucurbits. Thus, it is easier to explain cucurbitacin associations as independently arising through convergence caused by the similarities of cucurbitacins and pollen compounds that favour pharmacophagy. While the loose receptor hypothesis is consistent with Gillespie’s phylogenetic reconstructions of the Luperini, more data from cucurbitacin sensitivity bioassays are needed to differentiate between cucurbit pollen feeders and species that incorporate cucurbitacins into their life cycles from non-pollen cucurbit tissues (Gillespie et al., 2004). chap04.qxd 24/11/04 1:38 PM Page 81

Western Corn Rootworm, Cucurbits and Curcubitacins 81

WCR Adult Attraction to Volatile Semiochemicals from Cucurbits

LeConte (1868) based the species, D. virgifera, on two specimens found on ‘wild gourd’, so the association of WCR with members of the Cucurbitaceae dates back to the earliest published record for the species. Diabrotica spp., in general, have been associated with blossoms of varying Cucurbita spp. (Fronk and Slater, 1956; Howe and Rhodes, 1976; Bach, 1977; Fisher et al., 1984). Andersen and Metcalf (1987) examined selected Cucurbita spp. and cultivars for differences in floral volatile release, blossom cucurbitacin content and pollen content of male blossoms, and correlated these data to preference by Diabrotica beetles. D. undecimpunctata howardi were primarily found on Cucurbita maxima, which were the only blossoms to contain detectable levels of cucurbitacins and also released the greatest quantity of volatiles. WCR most preferred C. maxima, but were also found in significant quantities on one of the cultivars of Cucurbita pepo. Pollen content did not vary significantly between the cultivars and was actually highest on Cucurbita moschata, the least attractive species to WCR beetles, indicating that pollen was not responsible for differences in beetle preference. Andersen (1987) identified 22 of the 31 major components of C. maxima floral aroma. Metcalf and Lampman (1991, and references therein) evaluated these compounds by themselves and in blends for their attraction to diabroticite beetles in varying field settings. They also tested synthetic analogues of these compounds by varying functional group type, position, etc. Metcalf and Lampman (1991) speculated that ancestral diabroticite rootworms coevolved with primitive Cucurbitaceae species and that antennal sensory receptors evolved from those tuned to such compounds as cinnamaldehyde into receptors more specifically attuned to compounds that are currently being produced from cucurbit plants. Metcalf and Metcalf (1992, and references therein) discussed a series of experiments which resulted in the development of a three-component blend of 1,2,4-trimethoxybenzene, indole and cinnamaldehyde (the TIC mixture), which was attractive to WCR in the field. The blend had a syn- ergistic effect in that the combination showed more than a twofold increase in rootworm response than would have been expected from an additive effect of the individual compounds. Metcalf and Metcalf (1992) referred to this blend as a highly simplified Cucurbita blossom volatile aroma. Although a behavioural bioassay is the only reliable means of evaluating an entire complement of compounds in a semiochemical blend, electroantennogram (EAG) recording is usually helpful in detecting the most important components in the blend. This technique is a simple method for electrophysiological detection of the responses of insect antennae to volatile semiochemicals (Roelofs, 1984). Hibbard et al. (1997b) used an EAG-driven isolation and identification schemes to identify (E,E)-3,5-octadien-2-one, (E,Z)-2,6-nonadienal, (E)-2-nonenal, 2- phenethanol, benzyl alcohol, and 6,10-dimethyl-5,9-undecadien-2-one chap04.qxd 24/11/04 1:38 PM Page 82

82 D.W. Tallamy et al.

from extracts of buffalo gourd (C. foetidissima Humboldt, Bonpland and Kunth) root powder, the behaviourally active component of semiochemical baits for adult corn rootworm control programmes. Cossé and Baker (1999) isolated several of these same compounds from buffalo gourd root powder and demonstrated that (E,E)-3,5-octadien-2-one was attractive to D. barberi adults.

Volatile Semiochemicals from Maize

Prystupa et al. (1988) demonstrated that WCR are attracted to maize silk. Abou-Fakhr et al. (1996) demonstrated that the senescing portions (the brown portion that protrudes from the tip of the ear) of maize silks elicited strong EAG responses from WCR adults, while the green portions (under the husk or very young, green silk beyond the husk) did not elicit a signiﬁcant EAG response. Hibbard et al. (1997a) went on to isolate and identify the primary EAG-active components from brown maize silk as tridecan-2-one, (E,E)-3,5-octadien-2-one, (E,Z)-2,6-nonadienal and (E)-2- nonenal. Hammack (1996) demonstrated that (E)-6,10-dimethyl-5,9- undecadien-2-one (geranylacetone) was highly attractive to D. barberi and also attractive to WCR. In a re-evaluation of the most EAG-active fractions from maize silk, Hibbard et al. (1997b) found 6,10-dimethyl-5,9-undecadien-2-one as one of the smaller peaks present. The phenyl propanoids 2- phenethanol and benzyl alchohol were also present, but were not found in the fractions with the most EAG activity.

Volatile Semiochemicals from Both Maize and Cucurbita: a Role Connecting the Poaceae, the Cucurbitaceae and the virgifera Group?

Both maize silk extracts and buffalo gourd root extracts are composed predominantly of free fatty acids and hydrocarbons; yet in both extracts neither the fatty acid portion of the extract nor the hydrocarbon fraction was EAG-active (Hibbard et al., 1997a,b). When EAG-driven isolations and identiﬁcations were used with WCR, the primary compounds with EAG activity in both extracts included (E,E)-3,5-octadien-2-one and 6,10- dimethyl-5,9-undecadien-2-one (Hibbard et al., 1997a,b), both of which have been shown to be attractants for D. barberi (Hammack, 1996; Cossé and Baker, 1999). The phenyl propanoids 2-phenethanol and benzyl alchohol were also present from both plant species, which again are attractive to D. barberi (Metcalf and Metcalf, 1992). The same general chemistry appears to play a role in both the Poaceae and the Cucurbitaceae in terms of biological activity for WCR and D. barberi adults. As noted above, despite specialization on the Poaceae, adult WCR and D. barberi feed compulsively on bitter cucurbitacins when presented with the opportunity, and they are attracted to volatiles from Cucurbita blossoms (Metcalf and Metcalf, 1992, and references therein). According chap04.qxd 24/11/04 1:38 PM Page 83

Western Corn Rootworm, Cucurbits and Curcubitacins 83

to Metcalf and Metcalf (1992) the Cucurbitaceae were the ancestral host of all Diabroticites, but the virgifera group, which includes the WCR, currently specializes on the Poaceae. If the ancestral host hypothesis is correct, volatile compounds which are produced from both the Cucurbitaceae and the Poaceae may have been partially responsible for the host switch from the Cucurbitaceae to the Poaceae for the virgifera group of Diabrotica. The results of Gillespie et al. (2003), though, call into question the ancestral host hypothesis as an explanation for the afﬁnity of some Diabroticites to the Cucurbitaceae and give additional credence to the loose receptor hypothesis (Tallamy et al., 1999) described above. Ancestral members of the virgifera group that specialized on the Poaceae may have initially found Cucurbitaceae plants through volatile attractants which are released by both groups. These compounds may still play a role today, although compounds not yet identiﬁed from maize appear to play a more important role.

Parakairomones

Given the drastically different responses of different Diabrotica species to small changes in chemical structure, Metcalf and Metcalf (1992, and references therein) evaluated the attractiveness of a series of compounds mod- iﬁed slightly from Cucurbita blossom volatiles. Adding a methoxy group to natural compounds dramatically increased its effectiveness in attracting adult beetles. Although 4-methoxycinnamaldehyde has not yet been documented to be produced by Cucurbita blossoms, it is 2750-fold more attractive than cinnamaldehyde, which was isolated and identified from Cucurbita blossoms as being attractive. Similarly, 4-methoxyphenethanol was 6200-fold more attractive to D. barberi than phenethanol (Metcalf and Lampman, 1991). Metcalf and Metcalf (1992) used this information to speculate on the make-up of the receptors responsible for detecting these chemicals. It is these more attractive methoxy analogues of natural compounds which are generally used as lures today.

Use of Volatiles and Cucurbitacin-based Baits in Rootworm Management

Semiochemicals have been used in a number of ways to assist in corn rootworm management. Shaw et al. (1984) developed a ‘vial trap’ made from 60 ml amber-coloured plastic snap caps in which holes were drilled to allow rootworm beetles to enter. Levine and Gray (1994) then used this design to develop thresholds for predicting economic damage in areas with the rotation-resistant biotype (see also Spencer et al., Chapter 6, this volume). Acetate transparency ﬁlm was sprayed with a mixture of water and carbaryl and sprinkled with squash powder before placing it in the vial. Cucurbitacins from the squash served as a feeding stimulant and chap04.qxd 24/11/04 1:38 PM Page 84

84 D.W. Tallamy et al.

arrestant, but did not attract beetles. Beetles typically fed on the carbaryl and died. Rondon and Gray (2003) went beyond the maize–soybean rotation and looked at beetle numbers in oat (Avena sativa L.) stubble and lucerne (Medicago sativa L.) with both the vial trap and the Pherocon® AM yellow sticky trap. Unfortunately, the vial trap is not currently commercially available. Whitworth et al. (2002) evaluated a series of traps with and without volatile attractants. They found that traps baited with volatile attractant lures captured more beetles than unbaited traps. Lure constituents affected the species of beetle attracted to the trap. Traps baited with 4- methoxycinnamaldehyde attracted more WCR, traps baited with eugenol were more attractive to D. barberi and traps baited with trans-cinnamaldehyde were most attractive to southern corn rootworm larvae. They also found that a new trap developed by Trécé (Salinas, California) was as effective as or more effective than traditional monitoring techniques and, because it lacks sticky material, may be accepted more by consultants and growers, who are often averse to using sticky traps. The trap uses volatile attractants and a ‘stun pill’ containing buffalo gourd root powder and 3.9% carbaryl. Although not reported in Whitworth et al. (2002), Trécé’s website (http://www.trece.com/) reports a threshold of 200 beetles/trap/week to reduce egg laying below economic levels the subsequent year. This number is reduced by half when areawide management (see Gerber et al., Chapter 11, this volume) is in place. The trap is commercially available (Trécé, Salinas, California). Pruess et al. (1974) demonstrated that adult control prior to egg laying in continuous maize could be used to prevent economic damage from larval feeding the following season. Adult control in conjunction with field scouting has been used or recommended by a number of professional crop consultants in Nebraska and several other states (Meinke, 1995). Metcalf et al. (1987) documented that a dry bait containing cucurbitacins impregnated with a reduced rate of insecticide resulted in substantial reduction in adult corn rootworm populations. Since that time, a large amount of effort has gone into optimizing the formulations. Current ver- sions of the bait contain cucurbitacins, a toxicant (which one depends on the product) and a non-toxic edible carrier. The bait uses a 95–98% less toxic active ingredient than traditional beetle management insecticides. Commercial products have been made by several companies (Slam and Adios, Microflow Co., Memphis, Tennessee; Invite, Florida Foods Products, Inc., Eustis, Florida; and CideTrak, Trécé, Inc., Salinas, California). Use of a semiochemical bait for controlling adult corn rootworm beetles has been applied on an areawide basis with some success (Chandler, 2003; Gerber et al., Chaper 11, this volume), but whether this programme will be taken over by growers after it is over remains to be seen. Metcalf et al. (1987) suggested that the effectiveness of cucurbitacin- based baits could be increased should bait efficiency increase, but, according to Lance and Sutter (1991), the addition of volatile attractants did not increase the efficiency of adult rootworm control and commercial chap04.qxd 24/11/04 1:38 PM Page 85

Western Corn Rootworm, Cucurbits and Curcubitacins 85

companies have not added volatile attractants to their formulation. Hammack (2003) suggested that attractants could be used to concentrate beetles in an area of the ﬁeld that is subsequently controlled with cucurbitacin-based baits, but this approach has not been attempted. While a complete understanding of WCR–cucurbit relations has not been achieved, considerable data have been generated. The sum of these data indicate that the most parsimonious explanation for the relationship is the loose receptor hypothesis ﬁrst proposed by Tallamy et al. (1999). Regardless of the true explanation for the relationship, it provides unique opportunities for the management of WCR. Utilizing cucurbit-based attractants and feeding stimulants in WCR management may prove even more fruitful down the road than they have been to date.

References

Abe, M., Matsuda, K. and Tamaki, Y. (2000) Differences in feeding response among three cucurbitaceous feeding leaf beetles to cucurbitacins. Applied Entomology and Zoology 35, 137–142. Abou-Fakhr, E.M., Hibbard, B.E., Jewett, D.K. and Bjostad, L.B. (1996) Electroantennogram responses of western corn rootworm adults (Coleoptera: Chrysomelidae) in relation to maize silk morphology and phenology. Environmental Entomology 25, 430–435. Andersen, J.F. (1987) Composition of the floral odor of Cucurbita maxima Duchesne (Cucurbitaceae). Journal of Agricultural and Food Chemistry 35, 60–62. Andersen, J.F. and Metcalf, R.L. (1987) Factors influencing distribution of Diabrotica spp. in blossoms of cultivated Cucurbita spp. Journal of Chemical Ecology 13, 681–699. Andersen, J.F., Plattner, R.D. and Weisleder, D. (1988) Metabolic transformations of cucurbitacins by Diabrotica virgifera virgifera LeConte and D. undecimpunctata howardi Barber. Insect Biochemistry 19, 71–78. Bach, C.E. (1977) Distribution of Acalymma vittata and Diabrotica virgifera (Coleoptera: Chrysomelidae) on Cucurbita. Great Lakes Entomologist 10, 123–125. Barber, M. (1971) Chemistry and biochemistry of pollens. Progress in Phytochemistry 2, 1–34. Bar-Nun, N. and Mayer, A.M. (1989) Cucurbitacins – repressor of induction of laccase formation. Phytochemistry 28, 1369–1371. Boppré, M. (1978) Chemical communication, plant relationships, and mimicry in the evolution of danaid butterflies. Entomologia Experimentalis et Applicata 24, 64–77. Boppré, M. (1984) Redefining ‘pharmacophagy’. Journal of Chemical Ecology 10, 1151–1154. Boppré, M. (1990) Lepidoptera and pyrrolizidine alkaloids: exemplification of complexity in chemical ecology. Journal of Chemical Ecology 16, 165–180. Boppré, M., Wickler, W. and Seibt, W. (1984) Pharmacophagy in grasshoppers. Entomologia Experimentalis et Applicata 35, 115–117. Branson, T.F. and Krysan, J.L. (1981) Feeding and oviposition behavior and life cycle strategies of Diabrotica: an evolutionary view with implications for pest management. Environmental Entomology 10, 826–831. Branson, T.F. and Ortman, E.E. (1967) Host range of larvae of the western corn rootworm: further studies. Journal of Economic Entomology 60, 201–203. Branson, T.F. and Ortman, E.E. (1970) The host range of larvae of the western corn rootworm: further studies. Journal of Economic Entomology 63, 800–803. chap04.qxd 24/11/04 1:38 PM Page 86

86 D.W. Tallamy et al.

Branson, T.F., Reyes, R.J. and Valdés, H.M. (1982) Field biology of Mexican corn rootworm, Diabrotica virgifera zeae (Coleoptera: Chrysomelidae), in Mexico. Environmental Entomology 11, 1078–1083. Brower, L.P., McEvoy, P.B., Williamson, K.L. and Flannery, M.A. (1972) Variation in cardiac glycoside content of Monarch butterflies from natural populations in Eastern North America. Science 177, 426–429. Cardellina, J.H., Gustafson, K.R., Beutler, J.A., McKee, C., Hallock, Y.F., Fuller, R.W. and Boyd, M.R. (1990) Human medicinal agents from plants. In: Kinghorn, A.D. and Balandrin, M.F. (eds) National Cancer Institute Intramural Research on Human Immunodeficiency Virus Inhibitory and Antitumor Plant Natural Products. American Chemical Society, Washington, DC, pp. 218–227. Chambliss, O.L. and Jones, C.M. (1966) Cucurbitacins: specific insect attractants in Cucurbitaceae. Science 153, 1392–1393. Chandler, L.D. (2003) Corn rootworm areawide management program: United States Department of Agriculture–Agricultural Research Service. Pest Management Science 59, 605–608. Chiang, H.C. (1973) Bionomics of the northern and western corn rootworms. Annual Review of Entomology 18, 47–72. Chyb, S., Eichenseer, H., Hollister, B., Mullin, C.A. and Frazier, J.L. (1995) Identification of sensilla involved in taste mediation in adult western corn rootworm (Diabrotica virgifera virgifera LeConte). Journal of Chemical Ecology 21, 313–329. Clark, T.L. and Hibbard, B.E. (2004) A comparison of non-maize hosts to support western corn rootworm (Coleoptera: Chrysomelidae) larval biology. Environmental Entomology 33, 681–689. Clark, T.L., Meinke, L.J. and Foster, J.E. (2001) Molecular phylogeny of Diabrotica beetles (Coleoptera: Chrysomelidae) inferred from analysis of combined mitochondrial and nuclear DNA sequences. Insect Molecular Biology 10, 303–314. Contardi, H.G. (1939) Estudios geneticos en Cucurbita. Physis L8, 331–347. Cossé, A.A. and Baker, T.C. (1999) Electrophysiologically and behaviorally active volatiles of buffalo gourd root powder for corn rootworm beetles. Journal of Chemical Ecology 25, 51–66. Curtis, P.S. and Meade, P.M. (1971) Cucurbitacins from Crusiferae. Phytochemistry 10, 3081–3083. David, A. and Vallance, D.K. (1955) Bitter principles of Cucurbitaceae. Journal of Pharmacy and Pharmacology 7, 295–296. DeHeer, C.J. and Tallamy, D.W. (1991) Cucumber beetle larval affinity to cucurbitacins. Environmental Entomology 20, 775–788. Dethier, V.G. (1980) Evolution of receptor sensitivity to secondary plant substances with special reference to deterrents. American Naturalist 115, 45–66. Dinan, L., Whiting, P., Girvalt, J., Lapont, R., Dhadialla, T.S., Cress, D.E., Mugat, B., Antoniewski, C. and Lepesant, J. (1997a) Cucurbitacins are insect steroid hormone antagonists acting at the ecdysteroid receptor. Biochemical Journal 327, 643–650. Dinan, L., Whiting, P., Sarker, S.D., Kasai, R. and Tamasaki, K. (1997b) Cucurbitane-type compounds from Hemsleya carnosiflora antagonize ecdysteroid action in the Drosophila melanogaster B11 cell line. Cellular and Molecular Life Sciences 53, 271–274. Dryer, D.L. and Trousdale, E.K. (1978) Cucurbitacins in Purshia tridentate. Phytochemistry 17, 325–326. Dussourd, D.E. (1986) Adaptations of insect herbivores to plant defenses. PhD dissertation, Cornell University, Ithaca, New York. Dussourd, D.E. (1993) Foraging with finesse: caterpillar adaptations for circumventing chap04.qxd 24/11/04 1:38 PM Page 87

Western Corn Rootworm, Cucurbits and Curcubitacins 87

plant defenses. In: Stamp, N.E. and Casey, T. (eds) Caterpillars: Ecological and Evolutionary Constraints on Foraging. Chapman & Hall, New York, pp. 92–131 Dussourd, D.E., Harvis, C.A., Meinwald, J. and Eisner, T. (1991) Pheromonal advertisement of a nuptial gift by a male moth (Utethesia ornatrix). Proceedings of the National Academy of Sciences, USA, 88, 9224–9227. Eben, A.M., Barbercheck, B. and Aluja, S.M. (1997) Mexican diabroticite beetles: II. Test for preference of cucurbit hosts by Acalymma and Diabrotica spp. Entomologia Experimentalis et Applicata 82, 63–72. Edgar, J.A. (1982) Pyrrolizidine alkaloids sequestered by Solomon Island Danaine butterflies. The feeding preferences of the Danainae and Ithomiinae. Journal of Zoology, London 196, 385–399. Edgar, J.A., Culvenor, C.C.J. and Pliske, T.E. (1974) Coevolution of danaid butterflies and their host-plants. Nature – London 250, 646–648. Eisner, T., Smedley, S.R., Young, D.K., Eisner, M., Roach, B. and Meinwald, J. (1996) Chemical basis of courtship in a beetle (Neopyrachroa flabellate): Cantharidin as ‘nuptial gift’. Proceedings of the National Academy of Sciences, USA 93, 6499–6503. Erhardt, A. and Baker, I. (1990) Pollen amino acids – an additional diet for a nectar feeding butterfly? Plant Systematics and Evolution 169, 111–121. Feeny, P.P., Paauwe, K.L. and Demong, N.J. (1970) Flea beetles and mustard oils: host plant specificity of Phyllotreta cruciferae and P. striolata adults (Coleoptera: mildewed bug). Annals of the Entomological Society of America 64, 1173–1175. Feldlaufer, M.F., Buchmann, S.L., Lusby, W.R., Weinrich, G.F. and Svoboda, J.A. (1993) Neutral sterols and ecdysteroids of the solitary cactus bee Diadosia rinconis Cokerell (Hymenoptera: Anthophnidae). Archives of Insect Biochemistry and Physiology 23, 91–98. Ferguson, J.E. and Metcalf, R.L. (1985) Cucurbitacins: plant derived defense compounds for Diabroticina (Coleoptera: Chrysomelidae). Journal of Chemical Ecology 11, 311–318. Ferguson, J.E., Metcalf, E.R., Metcalf, R.L. and Rhodes, A.M. (1983) Influence of cucurbitacin content in cotyledons of Cucurbitaceae cultivars upon feeding behavior of Diabroticina beetles (Coleoptera: Chrysomelidae). Journal of Economic Entomology 76, 47–57. Fielding, D.J. and Ruesink, W.G. (1985) Varying amounts of bait influences numbers of western and northern corn rootworms (Coleoptera: Chrysomelidae) caught in cucurbitacins traps. Journal of Econonomic Entomology 78, 1138–1144. Fisher, J.R., Branson, T.F. and Sutter, G.R. (1984) Use of common squash cultivars, Cucurbita spp., for mass collection of corn rootworm beetles, Diabrotica spp. (Coleoptera: Chrysomelidae). Journal of the Kansas Entomological Society 57, 409–412. Fraenkel, G. (1959) The raison d’être of secondary plant substances. Science 129, 1466–1470. Frazier, J.L. (1986) The perception of plant allelochemicals that inhibit feeding. In: Brattsten, L.B. and Ahmad, S. (eds) Molecular Aspects of Insect–Plant Associations. Plenum Press, New York, pp. 1–42. Frazier, J.L. (1992) How animals perceive secondary plant compounds, second edition. In: Rosenthal, G.A. and Berenbaum, M.R. (eds). Herbivores: Their Interaction with Secondary Plant Metabolites; Evolutionary and Ecological Processes, Vol. 2. Academic Press, San Diego, California, pp. 89–134. Frenzel, M. and Dettner, K. (1994) Quantitation of cantharidin in canthariphilous cerato- pogonids (Diptera: Ceratopogonidae), anthomyids (Diptera: Anthomyidae) and cantharidin-producing oedemerids (Coleoptera: Oedemeridae). Journal of Chemical Ecology 20, 1795–1812. chap04.qxd 24/11/04 1:38 PM Page 88

88 D.W. Tallamy et al.

Frenzel, M., Dettner, K., Wirth, D., Waibel, J. and Boland, W. (1992) Cantharidin analogues and their attractancy for ceratopogonid flies (Diptera: Ceratopogonidae). Experientia 48, 106–111. Fronk, W.D. and Slater, J.H. (1956) Insect fauna of cucurbit flowers. Journal of the Kansas Entomological Society 29, 141–145. Fuller, R.W., Cardellina, J.H., II, Cragg, G.M. and Body, M.R. (1994) Cucurbitacin: differential cytotoxicity, dereplication and first isolation from Gonystylus keithii. Journal of Natural Products 57, 1442–1445. Galinat, W.C. (1977) The origin of corn. In: Sprague, G.F. (ed.) Corn and Corn Improvement. American Society of Agronomy, Madison, Wisconsin, pp. 1–47. Gibbs, R.D. (1974) Chemotaxonomy of Flowering Plants 2. McGill-Queen’s University Press. Montreal and London, pp. 829–830, 843, 1255–1259. Gillespie, J.J., Kjer, K.M., Duckett, C.N. and Tallamy, D.W. (2003) Convergent evolution of cucurbitacin-feeding and pharmacophagy in spatially isolated rootworm taxa (Coleoptera: Chrysomelidae; Galerucinae, Luperini). Molecular Phylogenetics and Evolution 29, 161–175. Gillespie, J.J., Kjer, K.M., Riley, E.G. and Tallamy, D.W. (2004) The evolution of cucurbitacin pharmacophagy in rootworms: insight from Luperini paraphyly. In: Jolivet, P.H., Santiago-Blay, J.A. and Schmitt, M. (eds) New Developments in the Biology of Chrysomelidae. Kluwer Academic Publishers, Boston, Massachusetts, 37–57. Guha, J. and Sen, S.P. (1975) The cucurbitacins – a review. Journal of Plant Biochemistry 2, 12–28. Halaweish, F.T. (1987) Cucurbitacins in tissue cultures of Brionia dioica Jacq. PhD thesis, University of Wales, Cardiff, UK. Halaweish, F.T., Tallamy, D.W. and Santana, E. (1999) Cucurbitacins: a role in cucumber beetle nutrition? Journal of Chemical Ecology 25, 2373–2383. Hammack, L. (1996) Corn volatiles as attractants for northern and western corn rootworm beetles (Coleoptera: Chrysomelidae: Diabrotica spp.). Journal of Chemical Ecology 22, 1237–1253. Hammack, L. (2003) Volatile semiochemical impact on trapping and distribution in maize of northern and western corn rootworm beetles (Coleoptera: Chrysomelidae). Agricultural and Forest Entomology 5, 113–122. Hibbard, B.E. and Bjostad, L.B. (1988) Behavioral responses of western corn rootworm larvae to volatile semiochemicals from corn seedlings. Journal of Chemical Ecology 14, 1523–1539. Hibbard, B.E., Randolph, T.L., Bernklau, E.J., Abou-Fakhr, E.M. and Bjostad, L.B. (1997a) Electroantennogram-active components of maize silk for adults of the western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 26, 285–295. Hibbard, B.E., Randolph, T.L., Bernklau, E.J. and Bjostad, L.B. (1997b) Electro-antenno- gram-active components of buffalo gourd root powder for western corn rootworm adults (Coleoptera: Chrysomelidae). Environmental Entomology 26, 1136–1142. Hill, R.A., Cooper, A., Roberts, A.D. and MacDonald, F.M. (1991) Dictionary of Steroids: Chemical Data, Structures, and Bibliographies. Chapman & Hall, New York. Hill, R.E. and Mayo, Z. (1980) Distribution and abundance of corn rootworm species as influenced by topography and crop rotation eastern Nebraska. Environmental Entomology 9, 122–127. Hollister, B. and Mullin, C.A. (1998) Behavioral and electrophysiological dose–response relationshiops in adult western corn rootworm (Diabrotica virgifera virgifera LeConte) for host plant amino acids. Journal of Insect Physiology 44, 463–470. Howe, W.L. and Rhodes, A.M. (1976) Phytophagous insect association with Cucurbita in Illinois. Environmental Entomology 5, 747–751. chap04.qxd 24/11/04 1:38 PM Page 89

Western Corn Rootworm, Cucurbits and Curcubitacins 89

Jeffrey, C. (1980) A review of the Cucurbitaceae. Botanical Journal of the Linnean Society 81, 233–247. Jewett, D.K. and Bjostad, L.B. (1996) Dichloromethane attracts diabroticite larvae in a laboratory behavioral bioassay. Journal of Chemical Ecology 22, 1331–1344. Jolivet, P. and Hawkeswood, T. (1995) Host-plants of Chrysomelidae of the World: an Essay about Relationships between the Leaf Beetles and Their Food Plants. Backhuys Publishers, Leiden, The Netherlands. Kim, J.H. and Mullin, C.A. (1998) Structure–phagostimulatory relationships for amino acids in adult western corn rootworm, Diabrotica virgifera virgifera LeConte. Journal of Chemical Ecology 24, 1499–1511. Krasnoff, S.B. and Dussourd, D.E. (1989) Dihydropyrrolizine attractants for arctiid moths that visit plants containing pyrrolizidine alkaloids. Journal of Chemical Ecology 15, 47–56. Krysan, J.L. (1982) Diapause in the nearctic species of the virgifera group of Diabrotica: evidence for tropical origin and temperate adaptations. Annals of the Entomological Society of America 75, 136–142. Krysan, J.L. and Branson, T.F. (1977) Inheritance of diapause intensity in D. virgifera. Journal of Heredity 68, 415–417. Krysan, J.L. and Smith, R.F. (1987) Systematics of the virgifera species group of Diabrotica (Coleoptera: Chrysomelidae: Galerucinae). Entomography 5, 375–484. Krysan, J.L., Branson, T.F. and Diaz Castro, G. (1977) Diapause in Diabrotica virgifera (Coleoptera: Chrysomelidae): a comparison of eggs from temperate and subtropical cli- mates. Entomologia Experimentalis et Applicata 22, 81–89. Krysan, J.L., Smith, R.F., Branson, T.F. and Guss, P.L. (1980) A new subspecies of Diabrotica virgifera (Coleoptera: Chrysomelidae): description, distribution, and sexual compatibility. Annals of the Entomological Society of America 73, 123–130. Krysan, J.L., McDonald, I.C. and Tumlinson, J.H. (1989) Phenogram based on allozymes and its relationship to classical biosystematics and pheromone structure among eleven Diabroticites (Coleoptera: Chrysomelidae). Annals of the Entomological Society of America 82, 574–581. LaMunyon, C.W. and Eisner, T. (1993) Postcopulatory sexual selection in an arctiid moth (Utethesia ornatrix). Proceedings of the National Academy of Sciences USA 90, 4689–4692. Lance, D.R. and Sutter, G.R. (1991) Semiochemical-based toxic baits for Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae): effects of particle size, location, and attractant content. Journal of Economic Entomology 84, 1861–1868. Lavie, D. and Glotter, E. (1971) The cucurbitacins, a group of tetracyclic triterpenes. Fortschritte der Chemie Organischer Naturstoffe 29, 306–362. LeConte, J.L. (1868) New Coleoptera collected on the survey for the extension of the Union Pacific Railway, E.D. from Kansas to Fort Craig, New Mexico. Transcripts of the American Entomological Society 2, 49–59. Levine, E. and Gray, M.E. (1994) Use of cucurbitacin vial traps to predict corn rootworm (Coleoptera: Chrysomelidae) larval injury in a subsequent crop of corn. Journal of Entomological Sciences 29, 590–600. Lewis, P.A. and Metcalf, R.A. (1996) Behavior and ecology of Old World Luperini beetles of the genus Aulacophora (Coleoptera: Chrysomelidae). Chemoecology 7, 150–155. Lin, S. and Mullin, C.A. (1999) Lipid, polyamide and flavonol phagostimulants for adult western corn rootworm from sunflower (Helianthus annus L.) pollen. Journal of Agricultural and Food Chemistry 47, 1223–1229. Mafra-Neto, A. and Jolivet, P. (1994) Entomophagy in Chrysomelidae: adult Aristobrotica angulicollis (Erichson) feeding on adult meloids (Coleoptera). In: Jolivet, P.H., Cox, chap04.qxd 24/11/04 1:38 PM Page 90

90 D.W. Tallamy et al.

M.L. and Petitpierre, E. (eds) Novel Aspects of the Biology of the Chrysomelidae. Kluwer Academic Publishers, Boston, Massachusetts, pp. 171–178. Mangelsdorf, P.C. (1974) Corn: Its Origin, Evolution, and Improvement. Belkamp Press of Harvard University Press, Cambridge, Massachusetts. Marquardt, V. and Adam, G. (1991) Recent advances in brassinosteroid research. In: Adam, G. (ed.) Herbicide Resistance – Brassinosteroids, Gibberellins, Plant Growth Regulations. Springer-Verlag, Berlin, pp. 103–139. Matsuda, K. (1988) Feeding stimulants of leaf beetles. In: Jolivet, P., Petitpierre, E. and Hsiao, T.H. (eds) Biology of the Chrysomelidae, Dr W. Junk, Dordrecht, The Netherlands, pp. 41–56. Mehta, P.K. and Sandhu, G.S. (1992) Inﬂuence of cucurbitacins on the feeding activity of red pumpkin beetle, Aulacophora foeveicollis (Lucas). Journal of Insect Science 5, 187–189. Meinke, L.J. (1995) Adult Corn Rootworm Management. Miscellaneous Publications 63, Agricultural Research Division, University of Nebraska. Lincoln, Nebraska. Meinwald, J., Meinwald, Y.C. and Mazzocchi, P.H. (1969) Sex pheromone of queen but- terﬂy: chemistry. Science 164, 1174–1175. Metcalf, R.L. (1979) Plants, chemicals, and insects: some aspects of coevolution. Bulletin of the Entomological Society of America 25, 30–35. Metcalf, R.L. (1985) Plant kairomones and insect pest control. Bulletin III Natural History Survey 35, 175. Metcalf, R.L. (1994) Chemical ecology of Diabroticites. In: Jolivet, P.H., Cox, M.L. and Petitpierre, E. (eds) Novel Aspects of the Biology of the Chrysomelidae. Kluwer Academic Publishers, Boston, Massachusetts, pp. 153–169. Metcalf, R.L. and Lampman, R.L. (1991) Evolution of diabroticite rootworm beetle (Chrysomelidae) receptors for Cucurbita blossom volatiles. Proceedings of the National Academy of Sciences USA 88, 1869–1872. Metcalf, R.L. and Metcalf, E.R. (1992) Plant Kairomones in Insect Ecology and Control. Routledge, Chapman & Hall, New York. Metcalf, R.L., Metcalf, R.A. and Rhodes, A.M. (1980) Cucurbitacins as kairomones for diabroticite beetles. Proceedings of the National Academy of Sciences, USA 17, 3769–3772. Metcalf, R.L., Ferguson, J.E., Lampman, R. and Andersen, J.F. (1987) Dry cucurbitacin-containing baits for controlling diabroticite beetles (Coleoptera: Chrysomelidae). Journal of Economic Entomology 80, 870–875. Miro, M. (1995) Cucurbitacins and their pharmacological effects. Phytotherapy Research 9, 159–168. Monroe, D.D. and Smith, R.F. (1980) A revision of the systematics of Acalymma sensu stricto Barber (Coleoptera: Chrysomelidae) from North America including Mexico. Memoirs of the Entomological Society of Canada 112, 1–92. Mullin, C.A., Mason, C.H., Chou, J.C. and Linderman, J.R. (1992) Phytochemical an- tagonism of aminobutyric acid based resistances in Diabrotica. In: Mullin, C.A. and Scotts, G. (eds) Molecular Mechanisms of Insecticide Resistance Diversity among Insects. Symposium Series 505, American Chemical Society, Washington, DC, pp. 288–308. Mullin, C.A., Hollister, B., Chyb, S., Eichenseer, H. and Frazier, J.L. (1993) Chemical basis for pollen – western corn rootworm (Diabrotica virgifera virgifera LeConte) interactions. In: Schultz, J.C. and Raskin, J. (eds) Plant Signals in Interactions with Other Organisms. Current Topics in Plant Physiology, Vol. 2. American Society of Plant Physiologists, Rockville, Maryland, pp. 226–229. Mullin, C.A., Chyb, S., Eichenseer, H., Hollister, B. and Frazier, J.L. (1994) Neuroreceptor chap04.qxd 24/11/04 1:38 PM Page 91

Western Corn Rootworm, Cucurbits and Curcubitacins 91

mechanisms in insect gestation: a pharmacological approach. Journal of Insect Physiology 40, 913–931. Musza, L.L., Speight, P., McElhiney, S., Brown, C.T., Gillum, A.M., Cooper, R. and Killar, L.M. (1994) Cucurbitacins: cell adhesion inhibitors from Conobea scoparioides. Journal of National Products 57, 1498–1502. Nielsen, J.K., Larsen, M. and Sorenson, H.J. (1977) Cucurbitacins E and I in Iberis amara feeding inhibitors for Phyllotreta nemorum. Phytochemistry 16, 1519–1522. Nishida, R. and Fukami, H. (1990) Sequestration of distasteful compounds by some pharmacophagous insects. Journal of Chemical Ecology 16, 151–164. Nishida, R., Yokoyama, M. and Fukami, H. (1992) Sequestration of cucurbitacin analogs by New and Old World chrysomelids leaf beetles in the tribe Luperini. Chemoecology 3, 19–24. Oyediran, I.O., Hibbard, B.E. and Clark, T.L. (2004) Prairie grasses as alternate hosts of western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 33, 740–747. Pliske, T.E. (1975) Attraction of Lepidoptera to plants containing pyrrolizidine alkaloids. Environmental Entomology 4, 455–473. Pohlman, J. (1975) Die Cucurbitacine in Bryonia alba und Bryonia dioica. Phytochemistry 14, 1587–1589. Pruess, K.P., Witkowski, J.F. and Raun, E.S. (1974) Population suppression of western corn rootworms by adult control with ULV Malathion. Journal of Economic Entomology 67, 651–655. Prystupa, B., Ellis, C.R. and Teal, P.E.A. (1988) Attraction of adult Diabrotica (Coleoptera: Chrysomelidae) to corn silks and analysis of the host-ﬁnding response. Journal of Chemical Ecology 14, 635–651. Purdy, R.H., Morrow, A.L., Blinn, J.R. and Paul, S.M. (1990) Synthesis, metabolism, and pharmacological activity of 3+-hydroxy steroids which potentiate GABA-receptor- mediated chloride uptake in rat cerebral cortical symaptoneurosomes. Journal of Medicinal Chemistry 33, 1572–1582. Reddi, C.S. and Aluri, R.J.S. (1993) Chemical ecology of insect pollination – an overview, In: Ananthakrishnan, T.N. and Raman, A. (eds) Chemical Ecology of Phytophagous Insects. International Science Publishers, New York, pp. 211–225. Roelofs, W.L. (1984) Electroantennogram assays: rapid and convenient screening procedures for pheromones. In: Hummel, H.E. and Miller, T.A. (eds) Techniques in Pheromone Research. Springer, New York, pp. 131–159. Rondon, S.I. and Gray, M.E. (2003) Captures of western corn rootworm (Coloeptera: Chrysomelidae) adults with Pherocon AM and vial traps in four crops in East Central Illinois. Journal of Economic Entomology 96, 737–747. Rosenthal, G.A. and Janzen, D.H. (eds) 1979. Herbivores: Their Interaction with Secondary Plant Metabolites. Academic Press, New York. Samuelson, G.A. (1994) Pollen consumption and digestion by leaf beetles. In: Jolivet, P.H., Cox, M.L. and Petitpierre, E. (eds) Novel Aspects of the Biology of the Chrysomelidae. Kluwer Academic Publishers, Boston, Massachusetts, pp. 179–183. Schneider, D., Boppré, M., Zweig, J., Horsley, S.B., Bell, T.W., Meinwald, J., Hansen, K. and Diehl, E.W. (1982) Scent organ development in a Creatonotos moth: regulation by pyrrolizidin alkaloids. Science 215, 1264–1265. Schoonhoven, L.M., Blaney, W.M. and Simmonds, M.S.J. (1992) Sensory coding of feeding deterrents in phytophagous insects, In: Bernays, E. (ed.) Insect–Plant Interactions, Vol. 4. CRC Press, Boca Raton, Florida, pp. 59–79. Seeno, T.N. and Wilcox, J.A. (1982) Leaf beetle genera (Coleoptera: Chrysomelidae). Entomography 1, 1–221. chap04.qxd 24/11/04 1:38 PM Page 92

92 D.W. Tallamy et al.

Shaw, J.T., Ruesink, W.G., Briggs, S.P. and Luckman, W.H. (1984) Monitoring populations of corn rootworm beetles (Coleoptera: Chrysomelidae) with a trap baited with cucurbitacins. Journal of Economic Entomology 77, 1495–1499. Simmonds, M.S.J., Blaney, W.M. and Fellows, L.E. (1990) Behavioral and electrophysiological study of antifeedant mechanisms associated with polyhydroxy alkaloids. Journal of Chemical Ecology 16, 3167–3196. Sinha, A.K. and Krishna, S.S. (1970) Further studies on the feeding behavior of Aulacophora foveicollis on cucurbitacin. Journal of Economic Entomology 63, 333–334. Smith, R.F. (1966) The distribution of Diabroticites in Western North America. Bulletin of the Entomological Society of America 12, 108–110. Smith, R.F. and Lawrence, J.F. (1967) Clariﬁcation of the type specimens of Diabroticites (Coleoptera, Chrysomelidae, Galerucinae). University of California Publications in Entomology 45, 1–174. Städler, E. (1991) Behavioral responses of insects to plant secondary compounds. In: Rosenthal, G.A. and Berenbaum, M.R. (eds) Herbivores: Their Interaction with Secondary Plant Metabolites, Evolutionary and Ecological Processes, 2nd edn, vol. 2. Academic Press, San Diego, California, pp. 44–88. Stanley, R.G. and Linskens, H.F. (1974) Pollen: Biology, Biochemistry, Management. Springer-Verlag, New York. Strnad, S.P., Bergman, M.K. and Fulton, W.C. (1986) First-instar western corn rootworm (Coleoptera: Chrysomelidae) response to carbon dioxide. Environmental Entomology 15, 839–842. Tallamy, D.W. and Krischik, V.A. (1989) Variation and function of cucurbitacins in Cucurbita: an examination of current hypotheses. American Naturalist 133, 766–786. Tallamy, D.W., Stull, J., Erhesman, N. and Mason, C.E. (1997a) Cucurbitacins as feeding and oviposition deterrents in nonadapted insects. Environmental Entomology 26, 678–688. Tallamy, D.W., Gorski, P.M. and Pesek, J. (1997b) Intra- and interspeciﬁc genetic variation in the gustatory perception of cucurbitacins by diabroticite rootworms (Coleoptera: Chrysomelidae). Environmental Entomology 26, 1364–1372. Tallamy, D.W., Whittington, D.P., Defurio, F., Fontaine, D.A., Gorski, P.M. and Gothro, P. (1998) The effect of sequestered cucurbitacins on the pathogenicity of Metarhizium anisopliae (Moniliales: Moniliaceae) on spotted cucumber beetle eggs and larvae (Coleoptera: Chrysomelidae). Environmental Entomology 27, 366–372. Tallamy, D.W., Mullin, C.A. and Frazier, J.L. (1999) An alternative route to insect pharmacophagy: the loose receptor hypothesis. Journal of Chemical Ecology 25, 1987–1997. Tallamy, D.W., Gorski, P.M. and Burzon, J.K. (2000) The fate of male-derived cucurbitacins in spotted cucumber beetle females. Journal of Chemical Ecology 26, 413–427. Tallamy, D.W., Powell, B.E. and McClafferty, J.A. (2002) Male traits under cryptic female choice in the spotted cucumber beetle (Coleoptera: Chrysomelidae). Behavioral Ecology 13, 511–518. Thorne, R.F. (1981) Phytochemistry and angiosperm phylogeny, a summary statement. In: Young, D.A. and Seigler, D.F. (eds) Phytochemistry and Angiosperm Phylogeny. Praeger, New York, pp. 233–295. Trigo, J.R. and Molta, P.C. (1990) Evolutionary implications of pyrrolizidine alkaloid assimilation by danaine and ithomine larvae (Lepidoptera: Nymphalidae). Experientia 46, 332. Watt, J.M. and Breyer-Brandwijk, M.G. (1962) The Medicinal and Poisonous Plants of Southern and Eastern Africa, 2nd edn. E. & S. Livingstone, Edinburgh, UK. Webster, F.M. (1895) On the probable origin, development and diffusion of North chap04.qxd 24/11/04 1:38 PM Page 93

Western Corn Rootworm, Cucurbits and Curcubitacins 93

American species of the genus Diabrotica. Journal of the New York Entomological Society 3, 158–166. Whitworth, R.J., Wilde, G.E., Shufran, R.A. and Milliken, G.A. (2002) Comparison of adult corn rootworm (Coleoptera: Chrysomelidae) sampling methods. Journal of Economic Entomology 95, 96–105. Wilcox, J.A. (1965) A Synopsis of the North American Galerucinae (Coleoptera: Chrysomelidae). Bulletin No. 400, New York State Museum and Science Service, Albany, New Jersey, 226 pp. Wilcox, J.A. (1972a) Coleopterorum Catalogus Supplementa (Chrysomelidae: Galerucinae: Luperini: Aulacophorina: Diabroticina), Pars 78, Fasc. 2. 2nd edn. Dr W. Junk, Dordrecht, The Netherlands. Wilcox, J.A. (1972b) Coleopterorum Catalogus Supplementa. (Chrysomelidae: Galerucinae: Luperini: Luperina), Pars 78, Fasc. 3, 2nd edn. Dr W. Junk, Dordrecht, The Netherlands. Young, D.K. (1984) Field studies of cantharidin orientation by Neopyrochroa ﬂabellate (Coleoptera: Pyrochroidae). Great Lakes Entomology 17, 23–30. chap04.qxd 24/11/04 1:38 PM Page 94 chap05.qxd 24/11/04 1:40 PM Page 95

5 Natural Mortality Factors Acting on Western Corn Rootworm Populations: a Comparison between the United States and Central Europe

STEFAN TOEPFER AND ULRICH KUHLMANN CABI Bioscience Switzerland Centre, Delémont, Switzerland

Introduction

The maize plant, Zea mays L. (Poaceae), and the western corn rootworm, Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae), are non-native species in the USA and in Europe. It is believed that both D. v. virgifera and its host plant evolved together in the subtropics of Mexico and Central America (Branson and Krysan, 1981; Krysan, 1982). The beetles’ major host plant, Z. mays, was domesticated as a crop more than 7000 years ago in Central America (Steffey et al., 1999). Several hundred years ago, maize became a successful agricultural crop in Europe, and by the early 1800s maize cultivation was widespread throughout the USA (Steffey et al., 1999). Thereafter, D. v. virgifera started to move northwards and was first recognized as a pest in the southern USA in 1912 (Steffey et al., 1999). From the 1950s onwards, the species invaded new areas at a rate of 64 to 80 km/year, and the practice of continuous maize planting without crop rotation allowed D. v. virgifera to become a major pest (Steffey et al., 1999). Approximately 50 years later, D. v. virgifera was accidentally introduced into Europe and was first observed near Belgrade, Yugoslavia (now Serbia), in 1992 (Baca, 1994). Within 10 years, D. v. virgifera successfully invaded the majority of maize producing areas in Central Europe (Kiss et al., 2001), and threatened to become a serious pest throughout Europe (Baufeld and Enzian, 2002). In order to successfully combat D. v. virgifera, we need to thoroughly understand the population dynamics of this invasive pest species in the two invaded regions, Europe and the USA (Dent and Walton, 1998; Dent, 2000). Regarding this objective, clarification is needed to determine: (i) if

96 S. Toepfer and U. Kuhlmann

natural mortality factors potentially regulating the population dynamics of D. v. virgifera differ between the USA and Europe; (ii) if indigenous natural enemies from Mexico and Central America followed the northward spread of D. v. virgifera and contributed towards an increased pest mortality in the USA; and (iii) if present, whether such natural enemies are significant mortality factors affecting population growth rates in the USA compared to Central Europe, where no effective and specific natural enemies have been detected (see Toepfer and Kuhlmann, 2004; Kuhlmann et al., Chapter 13, this volume). This chapter provides details on natural mortality factors acting on the life stages of D. v. virgifera, and summarizes statistically reliable information compiled and published over the last 40 years. Such information is not available for the area of origin of D. v. virgifera in Mexico and Central America; however, a large number of research studies exist in the USA, where natural mortality factors have been investigated or information has been obtained as a secondary result. Reported information on mortality factors acting on the egg, larval, pupal and adult stage of D. v. virgifera in the USA have been compiled into comparable units; this information was compared directly with results obtained from similar life-table studies on D. v. virgifera in Hungary, representing mortality factors in Central Europe. This knowledge was then used to: (i) elucidate differences or similarities in published or observed natural mortality factors acting on D. v. virgifera populations in the two invaded regions; (ii) rank the mortality factors regarding their intensity in reducing D. v. virgifera populations; and finally (iii) discover key mortality factors acting on D. v. virgifera life stages which thereby influence population growth.

Natural Mortality Factors Acting on the D. v. virgifera Egg Stage

Mortality factors can act on different age intervals in the D. v. virgifera egg stage, such as: (i) on fertilized or unfertilized eggs in the adult gonads; (ii) after oviposition on pre-diapausing eggs; (iii) on overwintering or diapausing eggs; and (iv) on post-diapaused eggs and ﬁrst-instar larvae hatching from these eggs.

United States The literature review revealed no information regarding the proportion of unfertilized eggs within D. v. virgifera populations. During the pre-diapause age interval, ﬁrst-instar larvae hatched from eggs in low numbers approximately 8 weeks after oviposition during late summer when cli- matological conditions were favourable; however, these larvae died when winter appeared in Minnesota (Chiang, 1973). According to the literature reviewed, no values were available on mortality factors affecting the overwintering egg stage. In the post-diapause egg stage, ﬁrst-instar larvae that failed to hatch from otherwise viable overwintered eggs accounted chap05.qxd 24/11/04 1:40 PM Page 97

Natural Mortality Factors Acting on Western Corn Rootworm Populations 97

for 38–42% mortality in the soil at 170 mm depth in South Dakota (Fisher, 1989). However, when eggs were incubated at 25°C after overwintering for 6 months at 8.5°C in a climate chamber, approximately 32–33% of viable South Dakota first-instar larvae failed to hatch from eggs (Fisher, 1989). After overwintering for 6 months at 7–8°C, 2–5% of viable Nebraska eggs (Branson, 1987) and 15% of South Dakota eggs (Weiss et al., 1985) failed to hatch. Total egg mortality, which is the sum of the pre- diapausing mortality, the overwintering mortality and the failure of first- instar larvae to hatch from post-diapaused eggs, varied widely between 35% and 70%, depending on the soil depth location of overwintering eggs and on the yearly climate fluctuations during winter (Table 5.1; Godfrey et al., 1995). In the laboratory, total egg mortality was 22–46% at 4°C and 80–100% relative humidity (r.h.) in a similar time period (Godfrey et al., 1995).

Central Europe As in the USA, there is no information available on the proportion of unfertilized eggs within D. v. virgifera in Europe. In the pre-diapause egg stage, mortality accounted for 15–22% (2-year mean = 17.3% ± 3.9 standard deviation (SD)), and a low additional morality of < 0.1% resulted from early hatching of first-instar larvae. Mortality in overwintering eggs reached 38–63% (2-year mean = 46.3% ± 14.5 SD). There was no significant difference found between the mortality of overwintering eggs in the field and those which were incubated in darkness at 4–8°C and 80–100% r.h. (Kolmogorov–Smirnov test, chi squared = 2.9, P = 0.46). After successful overwintering, D. v. virgifera eggs were incubated in darkness at 25°C and 80–100% r.h.; during this post-diapause egg stage, the proportion of first instars that failed to hatch from viable overwintered eggs reached 17.5–19.6% (mean = 18.4% ± 1 SD). In Central Europe, total egg mortality accounted for a mean of 63.2% ± 11.8 SD, with the largest proportion of mortality occurring in the overwintering egg stage, and due to the failure of first-instar larvae to hatch from post-diapaused eggs (mean = 50.9% ± 16.8 SD).

Factors influencing egg mortality The mortality data from the USA and Central Europe are comparable; however, differences are hard to determine due to high variations between years, regions and experimental designs. In the USA and Central Europe, pre-diapausing eggs appear to be reduced to a smaller number due to early eclosion of first-instar larvae during prolonged warm weather conditions in late summer and autumn (Chiang, 1973). In the laboratory, George and Ortman (1965) demonstrated that up to 60% of first-instar larvae had the potential to hatch when eggs were not exposed to cold climatic conditions. Total egg mortality measured in the field was consistently higher than chap05.qxd 24/11/04 1:40 PM Page 98

98 S. Toepfer and U. Kuhlmann . D. v. virgifera ). No hatching of viable ). No hatching of viable ., 1995 ., 1995 ., 1995 ., 1995 ., 1995 SD SD et al et al et al et al et al ) overwintering mortality and 17.5–19.6% (mean = 18.4% ± 1 ) overwintering mortality and 17.5–19.6% (mean = 18.4% ± 1 SD SD – 4 Nebraska, USA Lab Godfrey 70–120 Natural Hungary, EU Field S. Toepfer and U. Kuhlmann, unpublished Egg mortality from overwintering eggs until hatching of ﬁrst-instar larvae post-diapaused a b 35–70 Several depths Natural44 and 57 – Nebraska, USA Field 4 and 6–8 Godfrey Hungary, EU Lab S. Toepfer and U. Kuhlmann, unpublished 827057.741–70 75 150 300 Natural Natural Natural Nebraska, USA Nebraska, USA Nebraska, USA Field Field Field Godfrey Godfrey Godfrey 22–46 Divided into 38–63% (mean = 46.3% ± 14.5 Divided into 41–50% (mean = 45.2% ± 6.6 post-diapaused eggs. post-diapaused eggs. Table 5.1. Mortality (%) Soil depth (mm) Temperature (°C) Locationa Lab/ﬁeldb Source chap05.qxd 24/11/04 1:40 PM Page 99

Natural Mortality Factors Acting on Western Corn Rootworm Populations 99

that measured in the laboratory in the US studies (Godfrey et al., 1995), whereas no differences between mortality in the field and in the laboratory were observed in Central Europe. This could be an effect of the unfavourable winter climate or of soil-inhabiting natural enemies. Indeed, Stoewen and Ellis (1991) mentioned that predatory mesostigmatic Acari attack eggs of D. v. virgifera, thereby occasionally influencing the pest population dynamics in the USA; conversely, no effect of any natural enemies on egg populations has been found to date in Central Europe (Toepfer and Kuhlmann, 2004). However, egg mortality in the USA, and most probably also in Europe, is mainly influenced by environmental factors rather than by natural enemies. A literature review was carried out which revealed that the following factors influence egg mortality in the USA: (i) lack of rainfall or snow cover; (ii) minimum soil temperature of less than –7°C and chilling periods of less than –10°C; (iii) the sum of negative degree-days below –7°C; and (iv) length of the overwintering period. The impact of factors (i), (ii) and (ii) on egg mortality is influenced by characteristics of the overwintering habitat, including plant coverage, the soil depth at which overwintering occurs and soil texture (Lawson, 1986; Gray and Tollefson, 1988; Brust and House, 1990; Godfrey et al., 1995). With regard to point (i), a lack of rain- or snowfall increased egg mortality near soil surfaces in South Dakota, especially when soil temperatures were below the freezing-point (Calkins and Kirk, 1969). The desiccation occurring near soil surfaces had a greater influence on egg mortality than soil temperatures below –8°C (Calkins and Kirk, 1969). In addition, multiple regressions of climatic factors revealed the positive influence of snow cover and snowfall on egg survival in Nebraska (Godfrey et al., 1995). Interestingly, D. v. virgifera eggs from South Dakota were more susceptible to desiccation than Mexican eggs (Krysan et al., 1977). Referring to points (ii) and (iii), simple regression analyses showed a significant relationship between egg mortality and minimum soil temperature as well as between egg mortality and negative degree-days in Nebraska (Godfrey et al., 1995). The latter correlation was found only in years when soil temperature was less than –7°C for at least 80 negative degree-days (Godfrey et al., 1995). When temperature had an influence, the minimum soil temperature had a more significant effect than the number of negative degree-days (Godfrey et al., 1995). In the laboratory, temperatures of 4 or 5°C were found to be more suitable for diapause than 0°C or below 0°C. Overwintered eggs chilled at –10°C for 1-, 2- and 4-week intervals resulted in 43–50%, 23% and 0% larval hatch, respectively (Chiang, 1973), demonstrating that prolonged periods of chilling result in an increase in egg mortality. Concerning point (iv), overwintering egg mortality significantly increased with the length of the overwintering period in the field (Godfrey et al., 1995). In contrast, in a climate chamber at 4°C, egg overwintering mortality did not depend on time (Godfrey et al., 1995). In the laboratory, eggs from South Dakota remained in definite diapause for a shorter period of time compared to Mexican eggs (Krysan et al., 1977). In the USA, the failure of first instars to hatch from eggs in spring is influ- chap05.qxd 24/11/04 1:40 PM Page 100

100 S. Toepfer and U. Kuhlmann

enced by the sum of negative degree-days, as well as by the minimum and average soil temperatures during egg overwintering (Godfrey et al., 1995). The post-diapause development threshold is reported to be about 11°C in Ontario, Canada (Schaafsma et al., 1993) and in South Dakota (Fisher, 1989), 11.1°C in Minnesota and about 12.8°C in Kansas (Chiang, 1973). Krysan et al. (1977) reported that a high relative humidity in the soil increased the success of ﬁrst-instar larvae hatching from eggs.

Natural Mortality Factors Acting on the D. v. virgifera Larval and Pupal Stage

Mortality factors can act on different age intervals in the D. v. virgifera larval and pupal stage such as on the ﬁrst, second and third instars as well as on pupae.

United States Interestingly, separated field data on the impact of different mortality factors acting on the first, second and third instars as well as on pupae of D. v. virgifera are not available. Weiss et al. (1985) reported that the total mortality for the three larval instars and the pupae, measured under greenhouse conditions, reached 60–81% in D. v. virgifera males and about 77–85% in D. v. virgifera females. During 1 week of rearing in the laboratory, a mortality of 59–100% (mean = 78.7% ± 19.5 SD) in the first instars was determined after exposure to different starvation periods ranging from 0 to 24 h (Oloumi and Levine, 1989). Branson (1989) reported a mortality of 48–95% in first instars during different starvation periods of 0 to 72 h.

Central Europe Total mortality for the three larval instars and the pupae reached about 98%, indicating that only 2% of the emerged first-instar larvae survived and finally reached the adult stage. This total mortality was divided into 94.2% ± 1 SD mortality among emerged first instars, 46.5% ± 31.9 SD mortality among the remaining second instars, and 36% ± 32.2 SD mortality in the remaining third instars and pupae (2-year means given).

Factors influencing larval and pupal mortality Due to the lack of sufficient mortality data for each larval instar, it is difficult to compare mortality factors acting on individual developmental stages of D. v. virgifera in the USA and Central Europe. However, it is suggested that US studies assessing factors influencing larval mortality are also valuable for Europe, as the total mortality from the larval to the adult stage is comparable between the USA and Central Europe (Table 5.2). Moreover, it is evident that in both continents, D. v. virgifera populations chap05.qxd 24/11/04 1:40 PM Page 101

Natural Mortality Factors Acting on Western Corn Rootworm Populations 101 ., 1985 et al unpublished , summarizing the mortality from newly emerged first-instar larvae until adult D. v. virgifera /plant Starving (h) Temperature (°C) Location Lab/field Source Diabrotica Total larval and pupal mortality of – [78.7] 10 0–24 27 Illinois, USA Lab Olouni and Levine, 1989 48 (0)71.6 (2)81.5 (5)95.3 (24) 74 7297.9 [94.2] 71 73 10–80 0 24 Unknown 48 72 Natural 25 25 25 25 Hungary, EU South Dakota, USA South Dakota, USA South Dakota, USA Lab Field South Dakota, USA Lab Lab Lab Branson, 1989 S. Toepfer and U. Kuhlmann, Branson, 1989 Branson, 1989 Branson, 1989 68.7–81.7 24–73 Unknown 25 Nebraska, USA Greenhouse Weiss ( ), % mortality of first-instar larvae before reaching the roots; [ ], only in trials over 1 week. emergence (studied on maize plants). Table 5.2. Mortality (%) chap05.qxd 24/11/04 1:40 PM Page 102

102 S. Toepfer and U. Kuhlmann

are significantly reduced due to a high mortality in the first-instar larval stage. In general, this is probably a result of larvae starving during their search for maize roots. Strnad and Bergman (1987) and Gustin and Schumacher (1989) reported that approximately 70% of newly emerged larvae died from starvation and other causes prior to root penetration. Starvation of first-instar larvae for 24 h would cause an additional mortality of at least 40–45% (Strnad and Bergman, 1987; Branson, 1989), and starvation lasting for 1 or more days rapidly increased first-instar mortality and reduced the ability of larvae to bore into maize roots (Branson, 1989; Oloumi and Levine, 1989). In the laboratory, larvae survived a sufficient time to reach maize roots for most soil types when the soil was moist (Strnad and Bergman, 1987; Macdonald and Ellis, 1990). However, MacDonald and Ellis (1990) stated that mortality in first-instar larvae increased and movement decreased in extremely wet or flooded soil or very dry soil. In addition, compact soil with increased soil bulk density, lower air-filled porosity and lower air permeability, can occasionally increase mortality in D. v. virgifera larvae and reduce plant injury under field conditions (Ellsbury et al., 1994). In many studies, the soil bulk density did not affect larval density in the field because larval movement is probably reduced only at bulk densities of more than 1.0 mg/cm3 and in very wet or dry soil conditions (Strnad and Bergman, 1987; Gustin and Schumacher, 1989). Moreover, Gustin and Schumacher (1989) noted that soil density and humidity are not the only physical factors in the soil that influence first-instar movement but soil pore distribution also influences movement. Considering the soil texture, D. v. virgifera larvae move further through silty clay or loam than through loamy sand (Macdonald and Ellis, 1990) and larval mortality is higher in sandy loam soils in comparison to silty clay soils (Turpin and Peters, 1971). In addition, soil drought and/or high temperature can increase mortality in D. v. virgifera larvae, primarily in the first-instar stage. For example, an increase in air temperature from 21°C to 27°C resulted in increased mortality; however, this was probably a combined effect of increased air temperature and lower humidity (Oloumi and Levine, 1989). MacDonald and Ellis (1990) demonstrated that increasing soil temperature from 15°C to 25°C had the same effect on first-instar larval mortality as decreasing relative soil humidity from 100% to 80% or less. In general, higher temperatures can be tolerated under more humid conditions. Optimum larval development and survival occurred within the range of 18–30°C constant temperature in the laboratory (Fisher, 1986; Jackson and Elliott, 1988). According to Oloumi and Levine (1989) larvae emerging during the first half of the egg hatching period suffered less mortality than later-hatching larvae. On the other hand, the time period of emergence of larvae did not influence the ability of those larvae to establish on maize roots (Oloumi and Levine, 1989). In greenhouse tests conducted by Weiss et al. (1985) the mortality in the second and third instars was slightly increased as dry root weight decreased. In the chap05.qxd 24/11/04 1:40 PM Page 103

Natural Mortality Factors Acting on Western Corn Rootworm Populations 103

field, larval mortality appeared to be reduced and more adults were reported to emerge when: (i) maize roots were enlarged as a result of fertilization with nitrogen or irrigation; or (ii) the maize plant density was increased in the fields (Spike and Tollefson, 1988; Brust and House, 1990; for details of nutritional effects on larval survival, see Chapter 3, this volume). In the field, mortality of larvae is rarely density-dependent; however, at very high larval densities, larval mortality can increase due to stress from lack of food, and/or longevity of adults can be shortened (Branson et al., 1980; Branson and Sutter, 1985).

Natural Mortality Factors Acting on the D. v. virgifera Adult Stage

Adult longevity

United States In the field, D. v. virgifera beetles were reported to live on average 52 days (max. = 86), whereas, in the laboratory, mean adult longevity ranged from between 50 and 62 days (Elliott et al., 1990b) or 78 days (Hill, 1975) to 95 days (Branson and Johnson, 1973) depending on laboratory conditions (Table 5.3). In the field, male populations disappeared more rapidly than females and it has been suggested that the mortality in males is greater (Ball, 1957). However, laboratory studies suggest a comparable longevity of males and females. In the laboratory, after 42 days mortality in D. v. virgifera beetles rapidly increased independent of the type of diet (Elliott et al., 1990b), with approximately 50% of the adults dying after 50 days (Branson and Sutter, 1985; Elliott et al., 1990b). Hein and Tollefson (1987) used a mathematical model simulating adult mortality, which suggested a mortality of 1.25% on the first day of adult life and thereafter increasing by 1.25% each succeeding day.

Central Europe Under ﬁeld conditions, mean longevity of adult females varied between 1 month and 1.5 months (in 2000: 28 days ± 7 SD, max. = 45 days; in 2001: 36 days ± 9 SD, max. = 58; in 2002: 48 days ± 14 SD, max. = 68). The number of females decreased linearly over time (linear regression, R2 = 0.94, females = –12.3 × (days) + 140.6) and about 21% ± 3.8 SD of females did not reach maturity. Fifty per cent of females died after about 43 days. In the laboratory, the mean longevity of ﬁeld-collected females was approximately 3 months, with a maximum of 5 months (mean = 87 days ± 38 SD, max. = 159, light:dark (L:D) 14:10; L:D 25°C:15°C; 40–60% r.h.). The number of females decreased linearly over time (linear regression, R2 = 0.98, females = –0.799 × (days) + 122.8) and, after 80 to 85 days, 50% of the females had died. chap05.qxd 24/11/04 1:40 PM Page 104

104 S. Toepfer and U. Kuhlmann ., 1970 eedling leaves, h, maize after et al ., 1990b ., 1990b ., 1990b et al et al et al unpublished unpublished females (field cage and laboratory experiments with field-collected beetles). D. v. virgifera 47 0 h 24 – South Dakota, USA Lab Elliott 50 0 f,g 24 – South Dakota, USA Lab Elliott 52 86 b Natural Natural Nebraska, USA62 Field 0 i Ball, 1957 24 – South Dakota, USA Lab Elliott 28.7 5574 42 146 68 87 a 159 c Natural Natural 25:15 Hungary, EU 14:10 Hungary, EU Field S. Toepfer and U. Kuhlmann, Lab S. Toepfer and U. Kuhlmann, Oviposition period and longevity of Pre- Mean Mean Max. Mean Max. Diet (°C) L :D (h) Location Lab/field Source period: period Longevity Temperature Table 5.3. pollen, honey; e, maize as in field, later squash; f, prior to silking and pollination; g, pollination maize; pollination and silking; i, pollen, green maize silks, leaves all life long. 20–23Diet: a, maize as in field + kernels of soft stage; b, silk and tassels; c, artificial diet seedling leaves; d, s – – 18:6 Nebraska, USA Lab Short and Hill, 1972 12.223.3 76.4 100 78 132 e 28: 25/18 16:8 Nebraska, USA – Lab – Hill, 1975 12:12 Illinois, USA Lab Kuhlmann oviposition Reproductive chap05.qxd 24/11/04 1:40 PM Page 105

Natural Mortality Factors Acting on Western Corn Rootworm Populations 105

Factors influencing adult longevity According to the above-mentioned longevity estimates from Europe and the USA, adults can potentially survive until environmental conditions are no longer favourable, i.e. lack of food sources, or until first frost (Hein and Tollefson, 1987; Elliott et al., 1990b). Neither climatic factors, i.e. above-freezing temperatures or wind, nor the occurrence of natural enemies could be found to be statistically related to D. v. virgifera population changes in the USA (Hein et al., 1988; Elliott and Hein, 1991). Field populations of D. v. virgifera are rapidly reduced when maize is ripening and food is limited, indicating an increased adult mortality and adult emigration activity. The latter two can be even great enough to cause a population decline while emergence is still occurring (Hein and Tollefson, 1987). In the laboratory, it was clearly shown that adult mortality is primarily dependent on the nutritional value of the plants upon which they feed (Elliott et al., 1990b). Mortality increased as plants aged (Elliott et al., 1990b) and adult longevity was at its longest on a diet containing green maize silks, pollen and leaves throughout D. v. virgifera’s adult life. In contrast, longevity was shortest on a diet with pre- or post- flowering plants (Elliott et al., 1990b). Lance et al. (1989) and Naranjo (1991) reported that, before and after maize plants have flowered, there is a net movement of beetles out of maize fields. The simulation model of Elliott and Hein (1991) suggested that death due to ageing is not important in determining population change when compared to death and emigration after declining food quality.

Fecundity and oviposition

United States In comparison to laboratory populations, US field populations were observed to reach up to one-third of their potential size and to have a realized fecundity of one-fifth of their potential fecundity (Hein and Tollefson, 1987). The mean lifetime fecundity of D. v. virgifera in the field is approximately 300–400 eggs/female (Ball, 1957; Gustin, 1979; Table 5.4); however, the fecundity can be less in dry years (Gustin, 1979). In the laboratory, the mean oviposition period is reported to be 76–82 days (Branson and Johnson, 1973; Hill, 1975), however, no field data are available. Lifetime fecundity in the laboratory reached 200–500 eggs/female, depending on diet and laboratory test conditions (Elliott et al., 1990b; Fisher et al., 1991; Table 5.4). Ball (1957) reported that the maximum potential fecundity of D. v. virgifera is about 1000–1100 eggs/female. Considerable variation exists with regard to egg distribution in the soil profile in different locations and in different growing seasons (Hein and Tollefson, 1985; Gray et al., 1992). D. v. virgifera females are capable of ovipositing to a depth of up to 300 mm (Gray et al., 1992). Based on the two studies carried out by Gray and Tollefson (1988) and Gray et al. chap05.qxd 24/11/04 1:40 PM Page 106

106 S. Toepfer and U. Kuhlmann ., 1991 ., 1990b ., 1990b ., 1990b ., 1990b et al et al et al et al et al (field cage and laboratory experiments with field-collected beetles). D. v. virgifera Lifetime fecundity of 234.6178.9 g h 24 24 South Dakota, USA South Dakota, USA Lab Lab Elliott Elliott 322.2142.2 1045353476.7 626 b 956 a441.4 c Natural d Natural L:D 25:15 Nebraska, USA i 25 Hungary, EU Hungary, EU Field 24 South Dakota, USA Field Lab Ball, 1957 Lab S. Toepfer and U. Kuhlmann, unpublished South Dakota, USA S. Toepfer and U. Kuhlmann, unpublished Fisher Lab Elliott 125.3 f 24 South Dakota, USA Lab Elliott Mean Max. Diet Temperature (°C) Location Lab/field Source Eggs per female Table 5.4. Diet: see Table 5.3. chap05.qxd 24/11/04 1:40 PM Page 107

Natural Mortality Factors Acting on Western Corn Rootworm Populations 107

(1992), about 14% or 21% of the eggs are laid in the soil up to 100 mm depth, 26% or 44.5% between 100 and 200 mm depth and about 34% or 60% between 200 and 300 mm depth.

Central Europe

The oviposition period lasted on average 29 days ± 8 SD under field conditions in the 2 years of investigation in Hungary (max. = 55 days). In the laboratory, a lifetime fecundity of 353 eggs ± 237 SD was found, which was slightly correlated with the length of the oviposition period (R2 = 0.28, fecundity = 3.2 (days) + 130.2, analysis of variance (ANOVA), degrees of freedom (d.f.) = 1; 33, F = 12.49, P < 0.005). The maximum number of eggs laid per female in the laboratory was 956, and can be considered as the maximum potential fecundity of D. v. virgifera. However, only 19–20% of this value was realized under field conditions in Hungary over a 2-year period. The mean realized lifetime fecundity of D. v. virgifera females reached 142 eggs ± 63 SD under field conditions, with a mean maximum of 451 eggs, demonstrating that fecundity can strongly vary between years. For example, in 2001, the realized lifetime fecundity reached 187 eggs ± 143 SD (max. = 626), being significantly more than the 97 eggs ± 82 SD reached in 2002 (max. = 276; M. Whitney U test, P < 0.001).

Factors influencing fecundity and oviposition US studies have shown that oviposition and fecundity are at least influenced by soil moisture, temperature (Chapman, 1971), female age (Hein and Tollefson, 1987; Elliott et al., 1990a), quality of adult food (Elliott et al., 1990b) and the level of competition between larvae for food (see larval stage section) (Branson and Sutter, 1985) and these factors are probably also valid for Europe. No differences were found in the potential fecundity of D. v. virgifera in the USA and Central Europe. The lower fecundity realized in some years in Central Europe compared to the USA was most probably the result of dry weather conditions, high temperatures and low food quality after fast ripening of maize. While research indicates that more eggs are laid in soil with higher moisture content than in soil with lower moisture content, Gustin (1979) found that soil moisture had no influence on the depth of oviposition. In contrast, Weiss et al. (1983) reported that spatial distribution of eggs can depend upon soil moisture and that eggs were laid deeper in the soil under dry land conditions than in irrigated fields. However, oviposition by D. v. virgifera species was higher in irrigated fields (Brust and House, 1990). Fecundity was found to be higher when green maize silks, pollen and leaves were available throughout the D. v. virgifera adult stage; this effect was also observed when plant diets prior to silking and pollen shed or following pollination and silking were either provided in the laboratory (Elliott et al., 1990b). When D. v. virgifera beetles were fed with immature tassels and leaves chap05.qxd 24/11/04 1:40 PM Page 108

108 S. Toepfer and U. Kuhlmann

only, the reproductive period was delayed by up to 3 weeks and fecundity was strongly reduced (Elliott et al., 1990b).

Constructing a Life Table for D. v. virgifera in Central Europe

Life table studies have been conducted in Hungary (Table 5.5) with the aim of assessing the intensity of mortality factors reducing D. v. virgifera populations and discovering the key mortality factors acting on different D. v. virgifera life stages that may inﬂuence population growth.

Total mortality during a D. v. virgifera generation

Central Europe Total mortality during a single generation of D. v. virgifera from pre- diapausing eggs until adult emergence was found to be very high, often reaching more than 99%, in conventional monoculture maize ﬁelds in Hungary. The total generational mortality Ktotal, i.e. the negative loga- rithm of total survival, is a mortality value independent from population size, and was found to be 2.48 ± 0.23 SD in a life table study in the same ﬁelds (Tables 5.6 and 5.7). The mortality from post-diapaused eggs until adult emergence ranged from 98.5% to 99.1%.

United States The total mortality from pre-diapausing eggs until adult emergence was reported to be 98.7–99.4% under different tillage systems in South Dakota ﬁeld sites (Gustin, 1981) and accounted for a generational mortality Ktotal of 1.9–2.2. The total mortality from post-diapaused eggs until adult emergence ranged from 94% to 98% under different nitrogen fertilization regimes in Iowa (Spike and Tollefson, 1988), from 77.2% to 89.6% at different planting dates in South Dakota (Fisher et al., 1991), from 98.5% to 99.8% at different infestation densities in South Dakota (Branson and Sutter, 1985) and from 96.1% to 96% in a simulation model for South Dakota (Elliott and Hein, 1991) (Table 5.7).

Survivorship curve

In Central Europe, populations of D. v. virgifera studied followed a sigmoid log10 survivorship curve (Fig. 5.1). Highest survival of the population was shown in the egg stage prior to overwintering and from the third instar onwards to the adult stage. Thus, prior to reaching the ﬁrst instar stage, the log10 survivorship curve represented the convex type, exhibiting low initial losses that increase markedly with age (Fig. 5.1A Type I after Deevey (1947) or Type II after Slobodkin (1962)). From chap05.qxd 24/11/04 1:40 PM Page 109

Natural Mortality Factors Acting on Western Corn Rootworm Populations 109 ) /K x k b ) (100 x k factor mortality 2.48 k a )% ( x r 8.10 0.088 3.53 38.25 0.270 10.88 (100 36 77 n 174 174 ...00 ...00 SD 1.040.00 701.04 22 8.16 –0.06 –0.0007 0.088 –0.03 3.56 1.00 67 34.18 1.2390.55 49.94 188 0.39 0.336 13.54 1 0 1 1 0 )% ± x q mortality mortality ) (100 x d xx )( x l No. living at No. dying Intensity % of averaged for three populations studied in southern Hungary between 2000 and 2002. ) 187 143 x m D. v. virgifera ) 0.617 99.67 0 R ) 617.40 x m )( * f x x l d No hatching from eggsHatched from 2nd overwinteringSubtotal (hatching loss) –0.6 81.6 –0.14 81.0 18.40 18.26 Pre-diapause mortalityPre-diapause hatchingSubtotal (1st egg loss)Overwintering mortality 173.3 0.5 173.8 17.33 382.5 17.38 0.05 3.87 46.30 71 0.08 17.33 174 14.54 Unrealized progeny 0.05 0.083Realized fecundity ( 17.38 0.0002 3.33 0.083 0.01 3.34 2539 80.44 80.44 0.709 22.21 Unknown deathUnknown deathUnknown deathUnknown death Sex ratioPotential fecundity 341.8 94.23 9.73 956 4.0 46.53 31.92 36.00 67 32.16 0.97 67 3.86 0.40 53.87 0.272 0.194 10.96 237 7.81 Age-speciﬁc life table for )( Total mortality. Total generational mortality. x Diapaused eggs 443.7 Table 5.5. Eggs laidPre-diapause eggs 1000.0 826.2 Potential progenyRealized net reproductive rate ( a b 3156 1st-instar larvae2nd-instar larvae3rd-instar larvaePupae Adults emergedAdult females 362.7 20.9 11.2 7.2 3.30 Realized progeny ( ( Ageinterval factors Mortality beginning of during Apparent Real mortality generational chap05.qxd 24/11/04 1:40 PM Page 110

110 S. Toepfer and U. Kuhlmann ., 1991 ., 1991 et al et al unpublished unpublished until adult emergence in maize ﬁelds with conventional tillage until adult emergence in the following year maize ﬁelds without fertilizer Planting Infestation D. v. virgifera D. v. virgifera Fertilizer Planting Infestation 63,000 to 48,000 to 48,000 to Total mortality on pre-diapause eggs of Total mortality acting on post-diapaused eggs of 88.2 1,000 32,000 0 158 158 South Dakota, USA Fisher 96.395.495.8 300–60095.7–95.6 300–60097.1 63,000 300–600 300–60097.3 63,00084.1 87,000 63,000 0 186 200–600 336 200–600 1,000 32,000 0 137–152 137–152 32,000 137–152 137–152 137–152 137–152 0 32,000 137–152 Iowa, USA Iowa, USA 137–152 0 Iowa, USA 131–146 Iowa, USA 0 131–146 131–146 131–146 Spike and Tollefson, 1988 Spike and Tollefson, 1988 138 South Dakota, USA Spike and Tollefson, 1988 Lab strain, USA Fisher, 1992 Spike and Tollefson, 1988 148 Fisher, 1992 South Dakota, USA Fisher 97.2–97.8 470–50096.298.3 52,000 470–500 0 80–140 52,000 88,000 – 0 0 – – 105–120 148–152 South Dakota, USA Hungary, EU – Elliott and Hein, 1991 S. Toepfer and U. Kuhlmann, Model, USA Elliott and Hein, 1991 Mortality (%) Eggs/plant Plants/ha Tillage day day Location Source Mortality (%) Eggs/plant Plants/ha kg N/ha day day Location Source Table 5.6. 98.7 to 99.699.4 1,50099.2 1,500 53,400 80–140 Ridge 53,400 88,000 Conventional Conventional 113 113 105–120 September September Hungary, EU South Dakota, USA September Gustin and Schumacher, 1989 South Dakota, USA Gustin and Schumacher, 1989 S. Toepfer and U. Kuhlmann, (egg exposure in September, all mortality data are 2-year means). Table 5.7. (eggs exposed in spring). chap05.qxd 24/11/04 1:40 PM Page 111

Natural Mortality Factors Acting on Western Corn Rootworm Populations 111

1000 0.0 367.8 B ) –0.5 x l

10 –1.0 A –1.5 21.7 –2.0 6.7

Survivorship (log –2.5

–3.0 Sex ratio Diapause First instar Pre-diapause Second instar Adults emerged Diapaused eggs Third instar/pupae Age-speciﬁc intervals (x)

Fig. 5.1. Survivorship in a generation of D. v. virgifera representing a sigmoid curve from Curve A, exhibiting low initial losses increasing with age, until Curve B, exhibiting high initial losses followed by a period of much lower mortality. Numbers in the ﬁgure indicate the number of surviving individuals at the beginning of an age interval. (Survivorship curve as

log10 (lx) plotted against age; lx = individuals living at the beginning of each age interval; data from two sites in southern Hungary in 2001 and 2002; A = Type I after Deevey, 1947 or Type II after Slobodkin, 1962. B = Type III after Deevey, 1947 or Type IV after Slobodkin, 1962.

first–instar larvae onwards, a concave curve was found, exhibiting high initial losses followed by a period of much lower and relatively constant mortality until adult emergence (Fig. 5.1B Type III after Deevey (1947) or Type IV after Slobodkin (1962)). For example, a single emerged D. v. virgifera female in Hungary in 2001 originated from 49 emerged first instars, from 61 successfully overwintered eggs and from 210 laid eggs. In 2002, a single female originated from 159 emerged first instars, from 193 successfully overwintered eggs and from 365 laid eggs.

Ranked intensity of mortality factors

Table 5.5 presents an averaged life table from two ﬁelds in Hungary investigated over 2 years, and includes the mortality measured in each age interval of D. v. virgifera. Figure 5.2 then presents the ranking of mortality factors reducing D. v. virgifera populations. Ranks represent the proportion of the mortality intensity in each age interval (kx-) compared to the total generational mortality (Ktotal, here 2.48). Mortality intensities are population size-independent values of the measured mortality in each chap05.qxd 24/11/04 1:40 PM Page 112

112 S. Toepfer and U. Kuhlmann

Population reduction Marginal death rate factor % total generational mortality 0 10 20 30 40 No.

Mortality in L1 to L2 48.6% –94.2 ± 1.0

Unrealized fecundity 25.7% –80.4

Sex ratio of adults 13.1% –53.9 ± 0.5

Mortality in L2 to L3 11.9% –46.5 ± 32

Overwintering of eggs 11.3% –46.3 ± 14.5

Mortality in L3 to adults –8.5% –36.0 ± 32.2

No hatching from eggs in spring –3.4% –18.4 ± 1.0

Pre-diapause mortality in eggs –3,3% –17.3 ± 3.9

Pre-diapause hatching from eggs –0.01% –0.05 ± 0.08

Hatching from eggs after 2nd winter –0.02% –0.14 ± 0.0

Fig. 5.2. Population reduction factors and their marginal death rates per age interval as absolute observed mortality values (No.) and as a percentage of the total generational mor-

tality, Ktotal = 2.48 (mean of three life tables in southern Hungary, 2000/01 to 2001/02).

age interval (Van Driesche and Bellows, 1996). The highest intensity of mortality was found in the first-instar larvae, averaging a kx of 1.24 ± 0.07 SD, which equals already 49.9% of the total generational mortality. The intensity of the unrealized fecundity reached a value of 0.7, representing 26% of the total generational mortality. The reduction in the sex ratio had an intensity of 0.34, which equals 13.5% of the total generational mortality. An intensity of 0.2–0.3 was reached by the mortality in overwintering eggs and in the second-instar larval stage, demonstrating a mortality of 10–11% in each stage. The intensity of the mortality from third instars to adults reached 0.19, representing 7.8%. Mortality factors acting on the pre-diapausing eggs, during emergence of first instars from post-diapaused eggs and during emergence of first instars from pre-diapaused eggs reduced population size only to a minor extent (Fig. 5.2).

Population growth rates

In Central Europe, the averaged difference between the maximum potential fecundity and the realized fecundity at each site and year resulted in chap05.qxd 24/11/04 1:40 PM Page 113

Natural Mortality Factors Acting on Western Corn Rootworm Populations 113

a value referred to as unrealized fecundity, of about 80.4% (Table 5.5, Fig. 5.2). Based on the realized fecundity, the realized net reproductive rate (R0) ranged from 0.21 to 0.89 in the D. v. virgifera populations studied in Hungary, indicating a decline in population growth or an almost stable D. v. virgifera population. Life table results demonstrated that a high fecundity can compensate for a high generational mortality, and lead towards an increase in the population growth. The maximum potential for D. v. virgifera population growth is estimated to increase by a factor of three from year to year (RP = 3.16) based on the fact that each female has a maximum potential fecundity of 956 eggs. The realization of the potential fecundity per D. v. virgifera female strongly varied between years and locations in Central Europe. This natural variation was a main factor in influencing the outcome of the life tables, indicating population growth or decline (correlation of kunrealized fecundity and generational mortality 2 Ktotal had R = 0.99, P < 0.05). The failure of the population to realize their potential fecundity was also described as the main reason (Banerjee, 1979; Bellows et al., 1992) for populations of D. v. virgifera adults being often far below the potential in the USA (Hein and Tollefson, 1978). In Central Europe, high natural variation in mortality factors was also found between years and sites for overwintering eggs as well as for second- and third-instar larvae (see SD in Table 5.5). Mortality factors acting on these three stages have the potential to change the generational mortality and thus influence population growth (Southwood, 2000). In a Hungarian D. v. virgifera population studied in 2000 to 2001, a combination of comparatively high mortality in the overwintering egg stage as well as a low realized fecundity of females resulted in a lower net reproductive rate (R0). In the same area in 2002, a combination of comparatively high mortality in second- and third-instar larval stages as well as a low realized fecundity resulted again in a low net reproductive rate (R0). It can be concluded that the high mortality observed in the first-instar larval stage was a key factor in reducing the population size, but had only a limited potential to cause changes in the total generational mortality, since its value deviated only slightly between studies. In the USA, Elliott and Hein (1991) developed a simulation model for D. v. virgifera’s population dynamics wherein many population- influencing functions were combined, such as egg or larval development and mortality, as well as ageing, development, oviposition, mortality, immigration and dispersal of adults (renewed model of Hein and Tollefson (1987)). Major factors that potentially influence those functions include egg density and root availability for larvae; physiological age, food quality and dispersal of adults; oviposition dependence on physiological age of adults, temperature, larval density and food quality; and finally maize plant development. While the importance of the depth of oviposition was emphasized in terms of egg survival, this factor was not included in the model. According to the model, a change in population growth could be mainly caused by: (i) reduced rate or delay of maize plant development; (ii) decreased larval development rate; (iii) changes in chap05.qxd 24/11/04 1:40 PM Page 114

114 S. Toepfer and U. Kuhlmann

the reproductive development of adults; and finally (iv) changes in larval mortality. In contrast, changes in adult mortality due to physiological age had a minor influence on population growth, as they were overlaid by mortality or emigration due to declining food quality. In conclusion, a total mortality of over 95% during the life cycle of D. v. virgifera appears to be typical for this maize pest in Central Europe and the USA. Despite this, the pest species’ populations are still able to grow. Populations of D. v. virgifera are mainly reduced by mortality factors acting on the first-instar larval stage, the unrealization of fecundity and a relatively high mortality in the second-instar larval and overwintering egg stage. None the less, large variations in the realization of fecundity, in the overwintering mortality and in mortality of late larval instars seemed to have the highest impact on changing population growth rates of D. v. virgifera.

Assessing Differences between Mortality Factors Acting on D. v. virgifera populations in the USA and Central Europe

Interestingly, mortality factors methodically comparable between the USA and Central Europe, as well as the values for the total mortality, showed a high degree of coincidence (Table 5.2, 5.6 and 5.7). It can be concluded that the population ecology of D. v. virgifera and the impact of environmental conditions on the beetle’s mortality appear to be similar in the USA and Central Europe. In the egg stage, the mortality measured in the field was consistently higher than that measured in the laboratory in US studies, whereas no differences between mortality in the field and in the laboratory were observed in Central Europe, indicating a possible impact of soil-living natural enemies, such as mesostigmatic Acari (Stoewen and Ellis, 1991). However, soil structure and winter conditions can differ significantly between field sites and years, so that even overwintering mortality would be comparable between the USA and Central Europe. In general, the impact of natural enemies on D. v. virgifera’s egg, larval or adult stage appears to be relatively low in the two invaded regions (Kuhlmann and Burgt, 1998; Toepfer and Kuhlmann, 2004a). The following conclusions and assumptions are made: (i) specific natural enemies from the area of origin in Mexico and/or Central America did not follow the northward spread of D. v. virgifera into the USA; (ii) natural enemy populations were continuously reduced due to the current chemical-based pest management strategy in the USA; and (iii) natural enemies were left behind when D. v. virgifera was introduced into Central Europe (Toepfer and Kuhlmann, 2004a). Data on mortality factors are missing for many age intervals of D. v. virgifera in the USA, and therefore an elucidation of the impact of specific mortality factors on D. v. virgifera population growth remains difficult even when similarities have been suggested for the USA and Central Europe. Finally, the importance of life table studies on D. v. virgifera and chap05.qxd 24/11/04 1:40 PM Page 115

Natural Mortality Factors Acting on Western Corn Rootworm Populations 115

maize in the area of origin (Mexico and/or Central America) should be emphasized in order to help understand the importance of natural enemies, host plant performance and cultural practices on the population dynamics of D. v. virgifera. Life tables presented here could serve as a reference for further studies on natural D. v. virgifera populations and their natural enemy associations in the USA and Central America.

Consequences and Implications for the Integrated Pest Management (IPM) of D. v. virgifera

Using the above-described knowledge of mortality factors acting on D. v. virgifera populations, it appears that there are no reasons to apply different pest management strategies in Central Europe and in the USA. Life table studies in Central Europe showed that D. virgifera has a high potential for population growth, although the sum of the mortality factors acting during its life cycle is very high. It is difficult to make theoretical generalizations about which stage of D. v. virgifera might be the most suitable as a target for biological or chemical control actions because the knowledge about interactions between mortality factors is rather limited (Aeschlimann, 1979). There are two potential scenarios for the action of control measures which could be applied. First, the control measures could target a certain life stage of D. v. virgifera and act together with other natural mortality factors on the same age interval (Iwao, 1970; Buonaccorsi and Elkinton, 1990). Secondly, the control measures could kill D. v. virgifera and cause mortalities independently and in addition to natural mortality factors in each age interval. The second case is more desirable, and control measures would have the highest impact in reducing population growth when this additional mortality is acting in combination with high natural mortality, like during the first-instar larval stage. However, since it is not likely that all mortality factors operate independently and additionally, the first scenario is more realistic. Therefore it is assumed that control measures could be applied to any of the D. v. virgifera age intervals and have a similar chance of causing a significant change in the survival and in reducing population growth. In other words, each age interval of D. v. virgifera would be suitable for control by a biological control agent or chemical control measures. Thus, D. v. virgifera, like many herbivorous insects in natural or cultivated habitats, seems to be a good candidate for different sources of top-down control, regardless of the age interval targeted (Hawkins et al., 1999).

Acknowledgements

This work was possible due to the kind hospitality of the Plant Health Service in Hodmezovasarhely in Hungary, offered by Ibolya Zseller, Kataline Buzas, Erzebet Dormannsne, Piroska Varga, Andras Varga and chap05.qxd 24/11/04 1:40 PM Page 116

116 S. Toepfer and U. Kuhlmann

others. The work of the following summer students Lars Reimer, Edit Kiss, Szabolcs Meszaros, Orsolya Fulop, Martha Szabo, Monika Meszes, Peter Varga, Benko Edit and Marianna Szucs in Hungary as well as Manfred Grossrieder and Emma Hunt in Switzerland is very much appreciated. We would like to thank the other European Union project partners for optimizing joint efforts and creating an effective team in Hungary, such as Dr Ludger Wennemann, Dr Joachim Moeser, Mr Franck Muller and Dr Jozsef Kiss. We appreciated very much the help from Tara Gariepy for reviewing the English text. Funding was provided from the Bundesamt für Bildung und Wissenschaft (BBW), Switzerland, within the EU Project QLK5-CT-1999-01110.

References

Aeschlimann, J.P. (1979) Sampling methods and construction of life tables for Sitona humeralis populations (Col., Curculionidae) in Mediterranean climatic areas. Journal of Applied Entomology 16, 405–415. Baca, F. (1994) New member of the harmful entomofauna of Yugoslavia D. virgifera virgifera LeConte (Coleoptera, Chrysomelidae). Zastita Bilja 45, 125–131. Ball, H.J. (1957) On the biology and egg-laying habits of the western corn rootworm. Journal of Economic Entomology 50, 126–128. Banerjee, B. (1979) A key-factor analysis of population fluctuation in Andraca bipunctata Walker (Lepidoptera: Bombycidae). Bulletin of Entomological Research 69, 195–201. Baufeld, P. and Enzian, S. (2002) Spreading scenarios of western corn rootworm (D. virgifera virgifera) and possible consequences for selected EU member states. In: Proceedings of the International Symposium on Ecology and Management of Western Corn Rootworm. Göttingen University: Göttingen, Germany, p. 26. Bellows, T.S., Van Driesche, R.G. and Elkinton, J.S. (1992) Life-table construction and analysis in the evaluation of natural enemies. Annual Review of Entomology 37, 587–614. Branson, T.F. (1987) The contribution of prehatch and posthatch development to protandry in the chrysomelid, D. virgifera virgifera. Entomologia Experimentalis et Applicata 43, 205–208. Branson, T.F. (1989) Survival of starved neonate larvae of D. virgifera virgifera LeConte (Coleoptera: Chrysomelidae). Journal of the Kansas Entomological Society 62, 521–523. Branson, T.F. and Johnson, R.D. (1973) Adult western corn rootworms: oviposition, fecundity, and longevity in the laboratory. Journal of Economic Entomology 66, 417–418. Branson, T.F. and Krysan, J.L. (1981) Feeding and oviposition behavior and life cycle strategies of D. v. virgifera: an evolutionary view with implications for pest management. Environmental Entomology 10, 826–831. Branson, T.F. and Sutter, G.R. (1985) Influence of population density of immatures on size, longevity, and fecundity of adult D. virgifera virgifera (Coleoptera: Chrysomelidae). Environmental Entomology 14, 687–690. Branson, T.F., Sutter, G.R. and Fisher, J.R. (1980) Plant response to stress induced by artificial infestations of western corn rootworm. Environmental Entomology 9, 253–257. Brust, G.E. and House, G.H. (1990) Effects of soil moisture, no-tillage and predators on southern corn rootworm (D. undecimpunctata howardi) survival in corn agro ecosystems. Agriculture Ecosystems and Environment 31, 199–216. chap05.qxd 24/11/04 1:40 PM Page 117

Natural Mortality Factors Acting on Western Corn Rootworm Populations 117

Buonaccorsi, J. and Elkinton, J.S. (1990) Estimation of contemporaneous mortality factors. Research in Population Ecology 32, 151–171. Calkins, C.O. and Kirk, V.M. (1969) Effect of winter precipitation and temperature on overwintering eggs of northern and western corn rootworm. Journal of Economic Entomology 62, 541–543. Chapman, R.F. (1971) The Insects: Structure and Function. Elsevier, New York, 819 pp. Chiang, H.C. (1973) Bionomics of the northern and western corn rootworms. Annual Review of Entomology 18, 47–72. Deevey, E.S. (1947) Life tables for natural populations of animals. Quarterly Review of Biology 22, 283–314. Dent, D.R. (2000) Insect Pest Management. CAB International, Wallingford, UK, 410 pp. Dent, D.R. and Walton, M.P. (1998) Methods in Ecological and Agricultural Entomology. CAB International, Wallingford, UK, 389 pp. Elliott, N.C. and Hein, G.L. (1991) Population dynamics of the western corn rootworm: formulation, validation, and analysis of a simulation model. Ecological Modelling 59, 93–122. Elliott, N.C., Lance, D.R. and Hanson, S.L. (1990a) Quantitative description of the influence of fluctuating temperatures on the reproductive biology and survival of the western corn rootworm, D. virgifera virgifera LeConte (Coleoptera: Chrysomelidae). Canadian Entomologist 122, 59–68. Elliott, N.C., Gustin, R.D. and Hanson, S.L. (1990b) Influence of adult diet on the reproductive biology and survival of the western corn rootworm, D. virgifera virgifera. Entomologia Experimentalis et Applicata 56, 15–22. Ellsbury, M.M., Schumacher, T.E., Gustin, R.D. and Woodson, W.D. (1994) Soil com- paction effect on corn rootworm populations in maize artificially infested with eggs of western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 23, 943–948. Fisher, J.R. (1986) Development and survival of pupae of D. virgifera virgifera and D. undecimpunctata howardi (Coleoptera: Chrysomelidae) at constant temperatures and humidities. Environmental Entomology 15, 626–630. Fisher, J.R. (1989) Hatch of D. virgifera virgifera (Coleoptera: Chrysomelidae) eggs exposed to two different overwintering and hatch regimes. Journal of the Kansas Entomological Society 62, 607–610. Fisher, J.R., Sutter, G.R. and Branson, T.F. (1991) Influence of corn planting date on the survival and on some reproductive parameters of D. virgifera virgifera (Coleoptera: Chrysomelidae). Environmental Entomology 20, 185–189. Fisher, J.R., Sutter, G.R. and Jackson, J.J. (1992) Damage by and survival to adulthood of progeny of Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae) from field and laboratory origins. Journal of Kansas Entomological Society 65, 435–332. George, B.W. and Ortman, E.E. (1965) Rearing the western corn rootworm in the laboratory. Journal of Economic Entomology 58, 375–377. Godfrey, L.D., Meinke, L.J., Wright, R.J. and Hein, G.L. (1995) Environmental and edaphic effects on western corn rootworm (Coleoptera: Chrysomelidae) overwintering egg survival. Journal of Economic Entomology 88, 1445–1454. Gray, M.E. and Tollefson, J.J. (1988) Influence of tillage systems on egg populations of western and northern corn rootworms (Coleoptera: Chrysomelidae). Journal of the Kansas Entomological Society 61, 186–194. Gray, M.E., Hein, G.L., Boetel, M.A. and Walgenbach, D.D. (1992) Western and northern corn rootworm (Coleoptera: Chrysomelidae) egg densities at three soil depths: implications for future ecological studies. Journal of the Kansas Entomological Society 65, 354–356. chap05.qxd 24/11/04 1:40 PM Page 118

118 S. Toepfer and U. Kuhlmann

Gustin, R.D. (1979) Effect of two moisture and population levels on oviposition of the western corn rootworm. Environmental Entomology 8, 406–407. Gustin, R.D. (1981) Soil temperature environment of overwintering western corn rootworm (D. virgifera virgifera) eggs. Environmental Entomology 10, 483–487. Gustin, R.D. and Schumacher, T.E. (1989) Relationship of some soil pore parameters to movement of first-instar western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 18, 343–346. Hawkins, B.A., Mills, N.J., Jervis, M.A. and Price, P.W. (1999) Is the biological control of insects a natural phenomenon? Oikos 86, 493–506. Hein, G.L. and Tollefson, J.J. (1985) Seasonal oviposition of northern corn rootworms (D. barberi) and western corn rootworms (D. virgifera virgifera) (Coleoptera: Chryso- melidae) in continuous cornfields. Journal of Economic Entomology 78, 1238–1241. Hein, G.L. and Tollefson, J.J. (1987) Model of the biotic potential of western corn rootworm (Coleoptera: Chrysomelidae) adult populations, and its use in studying population dynamics. Environmental Entomology 16, 446–452. Hein, G.L., Tollefson, J.J. and Foster, R.E. (1988) Adult northern and western corn rootworm (Coleoptera: Chrysomelidae) population dynamics and oviposition. Journal of the Kansas Entomological Society 61, 214–223. Hill, R.E. (1975) Mating, oviposition patterns, fecundity, and longevity of the western corn rootworm. Journal of Economic Entomology 68, 311–315. Iwao, S. (1970) Analysis of contagiousness in the action of mortality factors on the western tent caterpillar population by using the m–m relationship. Research of Population Ecology 12, 100–110. Jackson, J.J. and Elliott, N.C. (1988) Temperature-dependent development of immature stages of the western corn rootworm, D. virgifera virgifera (Coleoptera: Chrysomelidae). Environmental Entomology 17, 166–171. Kiss, J., Edwards, C.R., Allara, M., Sivcev, I., Igrc-Barc˘ic´, J., Festic´, H., Ivanova, I., Princzinger, G., Sivicek, P. and Rosca, I. (2001) A 2001 update on the western corn rootworm (D. virgifera virgifera) in Europe. In: Proceedings of the XXI IOBC IWGO Conference. Veneto Agricoltura, Padua-Venice, Italy, p. 83. Krysan, J.L. (1982) Diapause in the Nearctic species of the virgifera group: evidence for tropical origin and temperate adaptations. Annals of the Entomological Society of America 75, 136–142. Krysan, J.L., Branson, T.F. and Diaz Castro, G. (1977) Diapause in D. v. virgifera (Coleoptera: Chrysomelidae): a comparison of eggs from temperate and subtropical cli- mates. Entomologia Experimentalis et Applicata 22, 81–89. Kuhlman, D.E., Loewe, W.L. and Luckmann, W.H. (1970) Development of immature stages of the western corn rootworm at varied temperatures. In: Proceedings of the NC Branch of the Entomological Society of America. pp. 93–95. Kuhlmann, U. and Burgt, W.A.C.M. (1998) Possibilities for biological control of the western corn rootworm, D. virgifera virgifera LeConte, in Central Europe. Biocontrol News and Information 19, 59–68. Lance, D.R., Elliott, N.C. and Hein, G.L. (1989) Flight activity of Diabrotica spp. at the borders of corn fields and its relation to ovarian stage in D. barberi. Entomologia Experimentalis et Applicata 50, 61–67. Lawson, E.C. (1986) Influence of Tillage and Depth in the Soil on Soil Temperature and Survival of Overwintering Western Corn Rootworm Eggs. Iowa State University, Ames, 63 pp. Levine, E., Oloumi, S.H. and Ellis, C.R. (1992) Thermal requirements, hatching patterns, and prolonged diapause in western corn rootworm (Coleoptera: Chrysomelidae) eggs. Journal of Economic Entomology 85, 2425–2432. chap05.qxd 24/11/04 1:40 PM Page 119

Natural Mortality Factors Acting on Western Corn Rootworm Populations 119

MacDonald, P.J. and Ellis, C.R. (1990) Survival time of unfed, first-instar western corn rootworm (Coleoptera: Chrysomelidae) and the effects of soil type, moisture, and com- paction on their mobility in soil. Environmental Entomology 19, 666–671. Naranjo, S.E. (1991) Movement of corn rootworm beetles, Diabrotica spp. (Coleoptera: Chrysomelidae), at corn field boundaries in relation to sex, reproductive status, and crop phenology. Environmental Entomology 20, 230–240. Oloumi, S.H. and Levine, E. (1989) Effect of starvation and time of egg hatch on larval survival of the western corn rootworm, D. virgifera virgifera (Coleoptera: Chrysomelidae), in the laboratory. Journal of the Kansas Entomological Society 62, 108–116. Schaafsma, A.W., Fuentes, J.D., Gillespie, T.J., Whitfield, G.H. and Ellis, C.R. (1993) Performance of a model for egg hatching of the western corn rootworm, D. virgifera virgifera LeConte, using measured and modelled soil temperatures as input. International Journal of Biometeorology 37, 11–18. Short, D.E. and Hill, R.E. (1972) Adult emergence, ovarian development and oviposition sequence of the western corn rootworm in Nebraska. Journal of Economic Entomology 65, 685–689. Slobodkin, L.B. (1962) Growth and Regulation of Animal Populations. Holt, Rinehart and Winston, New York, 184 pp. Southwood, T.R.E. (2000) Ecological Methods. Blackwell Science, Oxford, UK, 575 pp. Spike, B.P. and Tollefson, J.J. (1988) Western corn rootworm (Coleoptera: Chrysomelidae) larval survival and damage potential to corn subjected to nitrogen and plant density treatments. Journal of Economic Entomology 81, 1450–1455. Steffey, K.L., Rice, M.E., All, J., Andow, D.A., Gray, M.E. and Van Duyn, J.W. (1999) Handbook of Corn Insects. Entomological Society of America, Lanham, Maryland, 164 pp. Stoewen, J.F. and Ellis, C.R. (1991) Evaluation of a technique for monitoring predation of western corn rootworm eggs, D. virgifera virgifera (Coleoptera: Chrysomelidae). Proceedings of the Entomological Society of Ontario 122, 27–33. Strnad, S.P. and Bergman, M.K. (1987) Movement of first-instar western corn rootworms (Coleoptera: Chrysomelidae) in soil. Environmental Entomology 16, 975–978. Toepfer, S. and Kuhlmann, U. (2004) Survey for natural enemies of the alien invasive chrysomelid, D. virgifera virgifera, in Central Europe, Biocontrol 49, 385–395. Turpin, F.T. and Peters, D.C. (1971) Survival of corn rootworm larvae in relation to soil texture. Journal of Economical Entomology 64, 1448–1451. Van Driesche, R.G. and Bellows, T.S. (1996) Biological Control. Chapman & Hall, New York, 424 pp. Weiss, M.J., Mayo, Z.B. and Newton, J.P. (1983) Influence of irrigation practices on the spatial distribution of corn rootworm (Coleoptera: Chrysomelidae) eggs in the soil. Environmental Entomology 12, 1293–1295. Weiss, M.J., Seevers, K.P. and Mayo, Z.B. (1985) Influence of western corn rootworm (D. virgifera virgifera) larval densities and damage on corn rootworm survival, developmental time, size and sex ratio (Coleoptera: Chrysomelidae). Journal of the Kansas Entomological Society 58, 397–402. chap05.qxd 24/11/04 1:40 PM Page 120 chap06.qxd 24/11/04 1:42 PM Page 121

6 Movement, Dispersal and Behaviour of Western Corn Rootworm Adults in Rotated Maize and Soybean Fields

JOSEPH L. SPENCER,1 ELI LEVINE,1 SCOTT A. ISARD2 AND TIMOTHY R. MABRY3 1Center for Economic Entomology, Illinois Natural History Survey, Champaign, Illinois, USA; 2Department of Geography, University of Illinois, Urbana, Illinois, USA; 3Holden Foundation Seeds, Williamsburg, Iowa, USA

Introduction

Natural selection plays the central role in shaping the biological world; yet its glacial pace makes the evolutionary change occurring around us appear all but invisible. Occasionally, exceptional circumstances permit us to witness the process of natural selection. Such circumstances exist today in the eastern Corn Belt of the USA; the behaviour of the western corn rootworm (WCR), Diabrotica virgifera virgifera, an important pest of maize, is changing under intensive selection by crop rotation, allowing this pest to circumvent our most cost-effective and environmentally benign management tool. The WCR, D. v. virgifera LeConte, is the most serious insect pest of maize across the US Corn Belt. Metcalf (1986) estimated that this insect and the related northern corn rootworm (NCR), Diabrotica barberi Smith and Lawrence, cost US maize producers over US$1 billion in yield losses and management expenditures. As large as that ﬁgure is, it is probably an underestimate of today’s annual cost, owing to ongoing insecticide resistance problems, increased chemical costs, technology fees associated with transgenic maize varieties and the rise of rotation-resistant WCR populations. The recent introduction and spread of the WCR into Europe has increased the potential impact of rootworm injury on maize production signiﬁcantly. No longer just a North American problem, the WCR may now truly be called an international pariah. The WCR is a human-made pest (Krysan, 1993). Throughout its pro- gression from a sparsely distributed herbivore on native grass species to

122 J.L. Spencer et al.

its current position as a multiply resistant, globe-trotting pest of commercial maize, human intervention has succeeded in making rootworm problems worse. Failure to appreciate the resilient and adaptable nature of the WCR is a common thread running through the history of our interactions with this pest. As European pest managers face an expanding WCR problem, it is worthwhile to consider the key role of movement in this pest’s colonization of the US Corn Belt.

WCR Biology and History

The univoltine adults of the WCR are present in maize fields from July until the first frost and feed on maize foliage, pollen, silks and developing kernels. Oviposition has traditionally taken place almost exclusively in maize fields from late July through to mid-September (Shaw et al., 1978; Levine and Oloumi-Sadeghi, 1991). The diapausing eggs spend the autumn and winter in the soil until late May and early June, when they hatch and the larvae begin to feed on maize roots (Levine and Oloumi- Sadeghi, 1991). Larval feeding reduces the amount of water and nutrients available to developing plants, disrupting root system function, which, in turn can reduce grain yield. Larval feeding also may facilitate infection by root- and stalk-rot fungi, resulting in further damage. Extensive root injury makes plants more susceptible to lodging and direct yield losses may result from the difficulty in harvesting this maize. Finally, high densities of adults (usually > 5 WCR/plant) may interfere with maize pollination due to silk feeding (Levine and Oloumi-Sadeghi, 1991). The WCR was first collected in 1865 and described in the report of LeConte (1868), who observed it on a wild gourd (probably Cucurbita foetidissima) near Fort Wallace, Kansas, USA. Smith (1966) suggested that the native hosts of WCR populations in the Colorado–New Mexico–Arizona region of the USA were probably Tripsacum (any of approximately 15 species of perennial grass closely related to maize, Zea mays). Despite a close host association with plants in the Graminaceae, WCR adults will feed compulsively on species in the Cucurbitaceae containing cucurbitacins B and E (Metcalf, 1979). This compulsion is taken as evidence of diabroticite coevolution with Cucurbitaceae prior to a host shift onto graminaceous plants, after which the cucurbitacin compulsion was retained (Metcalf, 1983). Gillette (1912) was first to observe WCR as a pest of cultivated maize (sweetcorn) in 1909 and 1910 near Fort Collins and Loveland, Colorado, respectively. In subsequent years, the distribution of WCR expanded slowly eastward across the western maize-growing region through the 1920s–1940s (reviewed by Metcalf, 1983). The WCR was first noticed and caused some damage in south-west Nebraska in 1929 and 1930 and caused heavy damage in the early 1940s (Tate and Bare, 1946). Large-scale application of soil insecticides against the WCR began in Nebraska in 1949; treatment with aldrin, chlordane and heptachlor followed into the mid-1950s. chap06.qxd 24/11/04 1:42 PM Page 123

Movement, Dispersal and Behaviour of Adults in Maize 123

By 1954, WCR were present along the Missouri River from South Dakota to Missouri. The first WCR control failures were reported in 1959; by 1961 WCR in central Nebraska were 100-fold more resistant to cyclodiene insecticides than susceptible populations in the eastern portion of the state. Concurrent with the rise of widespread cyclodiene resistance was an accel- eration in the rate of WCR population expansion (Metcalf, 1982, 1983). Between 1909 and 1948, WCR moved 756 km from Colorado to the Missouri River, averaging just 19 km/year. In Kansas during 1953, the WCR range expanded 56 km (Burckhardt and Bryson, 1955). From 1961 to 1964 cyclodiene-resistant WCR moved 579 km from Nebraska to Eau Claire, Wisconsin (c. 193 km/year), and by 1968 had advanced 805 km to north-western Indiana (c. 113 km/year). By 1979, WCR were found throughout most of the Corn Belt; even at the leading edge of the expanding front some still retained 1000- to 2500-fold resistance to cyclodienes (reviewed by Metcalf, 1983). In the 1980s, the eastward WCR expansion finally reached US East Coast maize-producing areas. Metcalf (1983) hypothesized that the increased movement rate was related to the increased fitness among cyclodiene-resistant beetles and to a behavioural change associated with the genes for resistance. A very mobile WCR may be a legacy of the chemical era. Though insecticides are a favourite WCR management tool, cultural controls have been recommended for nearly as long as the WCR has been a recognized pest. In a paper focusing on the NCR, Gillette (1912) noted that the obligate host relationship between corn rootworm beetles and maize suggested an obvious method of pest management, crop rotation, a cultural control that has been recommended for control of both the NCR and the WCR ever since. A maize and soybean rotation (the most common rotational pattern in the Midwest USA, accounting for much of the 80% of US rotated maize (Power and Follett, 1987)) disrupts the rootworm life cycle because corn rootworm larvae cannot survive on soybean roots. Crop rotation proved so effective that farmers using it could safely discount the possibility of significant corn rootworm larval injury to the roots of first- year maize. Despite a history of insecticide resistance and encouragement to use crop rotation, the availability of new insecticides and inexpensive nitrogen fertilizers worked against widespread adoption of cultural control in favour of more profitable continuous maize production. The WCR first entered north-west Illinois in 1965 and moved across the northern counties to reach Indiana by 1970. In Illinois, the use of soil- applied insecticide in maize increased through the late 1960s when > 60% of maize hectares were treated for corn rootworms (primarily for the NCR; problems with the NCR’s prolonged diapause predate the arrival of the WCR (Levine et al., 1992)). Insecticide usage remained at ≥ 50% of hectares until the late 1970s when, thanks to educational efforts that emphasized rotation as an alternative to chemicals, insecticide use began to decline steadily into the 1990s, by which time approximately only 13% of maize was treated (Pike and Gray, 1992). The shift from broad-scale prophylactic chemical application to cultural control constituted a great chap06.qxd 24/11/04 1:42 PM Page 124

124 J.L. Spencer et al.

victory for integrated pest management. Ironically, even this level of insecticide use was probably excessive. Steffey et al. (1992) conducted surveys of root injury to maize after soybean (first-year maize) around Illinois from 1986 to 1989. Only 1.7% of rotated maize fields had injury levels that exceeded the theoretical economic injury level of 4.0 on the Iowa State University root-injury rating scale (equal to one full node of roots removed (Hills and Peters, 1971)). Steffey et al. (1992) concluded that Illinois maize producers rarely needed to apply soil insecticides to prevent rootworm injury in maize rotated with soybean. At the time, most of the economic injury reported by Steffey et al. (1992) was attributed to NCR’s prolonged diapause. East-central and north-eastern Illinois were the areas identified as being most at risk; these districts encompass the Ford County epicentre of rotation resistance and the areas where the WCR problem is currently most serious. Unfortunately, it now appears that the enthusiastic adoption of crop rotation over a broad area (98% adoption in parts of east-central Illinois) combined with the great effectiveness of the technique created conditions that favoured an existing, but rare, WCR variant with reduced egg-laying fidelity to maize fields (Onstad et al., 2001). The presence of some WCR expressing what we would now call ‘variant behaviour’ may be evident in late-1970s to late-1980s records of root injury for first-year maize (Shaw et al., 1978; Steffey et al., 1992). This unexpected, but not alarming, injury was frequently attributed to the presence of volunteer maize or grassy weeds in rotated soybean fields that were known to be attractive to ovipositing WCR and NCR (Shaw et al., 1978). Even if this injury had been viewed as a manifestation of a natural variation in WCR host fidelity, it seems unlikely that scientists at the time could have imagined that a rapid development of resistance was already under way. An appreciation for widespread crop rotation as a selective force was lacking. As late as 1993, it was suggested that ‘it is highly unlikely that the WCR could become adapted to crop rotation by oviposition in the alternate crop’ (Krysan, 1993). There are large areas in Illinois (e.g. Monmouth, located c. 220 km north-west of Champaign-Urbana, in western Illinois) where WCR can be very abundant in maize, but only low numbers of WCR adults are ever found outside of maize fields and there is no evidence of rotation resistance. The distance between these areas and the rotation-resistance epicentre (see below) has protected them from the problem; however, even these areas will eventually experience rotation resistance as the problem continues to spread north and west in Illinois (Onstad et al., 1999, 2000; Fig. 6.1). Where crop rotation is applied nearly universally, the practice imposes a strong (though unintended) selection that favours individuals expressing reduced ovipositional fidelity to maize fields. Where maize is rotated, females with reduced fidelity to maize fields realize a reproductive advantage over females with perfect ovipositional fidelity to maize fields since they lay some of their eggs in non-host fields (e.g. soybean) which are rotated to maize with high probability. Eggs deposited in maize chap06.qxd 24/11/04 1:42 PM Page 125

Movement, Dispersal and Behaviour of Adults in Maize 125

1997 1998 1999

2000 2001 2002

2003 WCR Abundance in Illinois Soybean Fields: 1997–2003 (WCR/100 sweeps)

0 11–50

≤ 10 > 50

Not sampled

Fig. 6.1. WCR abundance (mean number of WCR beetles per 100 sweeps per county with a 38 cm diameter sweep net) in soybean ﬁelds for Illinois (USA). Each mean represents four to eight samples. Collections were made during the last week of July and ﬁrst week of August 1997–2003. Sampled counties have a heavy border outline, unsampled counties have thin outlines with a stippled interior.

fields rotated to soybean are lost. We hypothesize this selection generated a WCR population in which a high proportion of females have some propensity to exit maize fields and oviposit in soybean fields (or other locations in addition to maize fields). Extensive adult monitoring and chap06.qxd 24/11/04 1:42 PM Page 126

126 J.L. Spencer et al.

analysis of soybean field soil samples confirm that WCR resistance to crop rotation occurs because females lay eggs in crops rotated with maize (eggs are still also laid in maize fields) (Levine et al., 2002; Pierce, 2003). The first hints that the crop rotation success story was failing came in 1987 when unexpected WCR root injury to rotated seed maize was reported near Piper City in Ford County, Illinois. Local use of pyrethroid insecticide in seed production maize fields was initially proposed as a mechanism that could have repelled gravid WCR from treated maize fields and into surrounding untreated seed soybean fields where females oviposited (Levine and Oloumi-Sadeghi, 1996). The pyrethroid hypothesis was found to be false as the primary cause of the problem in 1993 when circumvention of crop rotation occurred in commercial maize fields where pyrethroids were not used. However, the possibility that pyrethroids provided the ‘push’ needed to concentrate a critical mass of WCR with variant behaviour outside of maize fields remains an intriguing hypothesis for how the more widespread problem of behavioural resistance in WCR was initiated. By 1993 and 1994, WCR injury to rotated maize was reported in several east-central Illinois counties and, in 1995, failure of crop rotation was devastating in nine Illinois counties and 15 adjacent Indiana counties (Levine et al., 2002). As a consequence of rotation resistance, growers in east-central Illinois and western Indiana have experienced economic losses due to WCR larval injury to first-year maize since 1995. For these growers and those in parts of Michigan and Ohio, where this problem has now spread, the only reliable option to manage this pest is a planting-time application of soil insecticide (Levine et al., 2002). Modelling studies of the spread of the problem also strongly suggest that it originated very close to Piper City (Onstad et al., 1999; Levine et al., 2002).

Response to Rotation Resistance

Unable to rely on crop rotation as a management strategy for corn rootworm, producers across a large portion of the eastern Corn Belt in Illinois, Indiana, Ohio and Michigan affected by rotation-resistant WCR have returned to soil insecticide application as their primary management tool. This response has led to a costly reversal in what was once a declining trend in soil insecticide use throughout the affected area. Since the rise of the rotation-resistant variant, the use of planting-time soil insecticides (approximate cost = US$40.00/ha) has escalated on rotated maize hectares. Based on surveys indicating high WCR abundance in soybean ﬁelds, producers in 31 Illinois counties will face a moderate to high risk of WCR larval injury to maize hectares during the 2004 growing season (nine more counties than in 2002). These counties are located primarily in east-central and north-eastern regions of Illinois with 1.93 million maize hectares at risk. This expansion of rotation resistance costs Illinois producers in these affected counties US$76.2 million annually in insecti- chap06.qxd 24/11/04 1:42 PM Page 127

Movement, Dispersal and Behaviour of Adults in Maize 127

cidal treatments. If all of the threatened maize hectares were affected by rotation resistance, added costs to growers could eventually reach more than US$120 million annually in just the state of Illinois. The return to broad-scale prophylactic use of insecticides for WCR management comes at a time when the US regulatory climate (under the 1996 Food Quality Protection Act) has cooled to the prospect of escalat- ing pesticide inputs. Though economic injury is likely each year in continuous and first-year maize fields throughout east-central Illinois and northern Indiana, along the edges and locally within the affected region there are areas where WCR egg-laying in soybean is too low to cause economic injury to rotated maize. Development of a threshold for adult WCR abundance in rotated soybean fields (O’Neal et al., 2001) provided growers with a needed tool to assess their risk of economic injury and guide decision-making about the judicious use of soil insecticides in rotated maize. O’Neal et al. (2001) describe a method where 12 Pherocon® AM (Trece, Inc., West Adair, OK) traps are deployed in soybean fields from the last week of July until the third week of August. Growers recover the traps each week to count the WCR and determine the average number of WCR captured per trap per day. Counts of ≥ 5 WCR/trap/day are likely to result in economic injury to first-year maize planted in that field the following year. Proper adherence to WCR monitoring recommendations can help growers avoid unnecessary soil insecticide use.

WCR Movement

Rotation resistance is a problem of movement. The landscape of crop rotation probably created the rotation-resistant WCR by selecting for a once uncommon phenotype with a greater tendency to move out of maize fields and lay eggs. Where crop rotation was broadly adopted, a reduced fidelity to maize fields endowed those beetles with a reproductive advantage over WCR with strong fidelity to maize fields. If we are to succeed in restoring crop rotation as a WCR management tool, avoid rapid insect resistance development to transgenic maize varieties and contain the WCR in Europe, a greater appreciation for the influence of the local landscape/environment on WCR biology (and movement in particular) is needed. The following sections describe characteristics and patterns of WCR movement across multiple spatial and temporal scales and discuss the role(s) played by plants in proximate mechanisms hypothesized to facilitate rotation resistance and interfield movement in general.

Movement and the WCR Life Cycle

Female WCR are mated by protandrous males, who often intercept teneral virgins shortly after they emerge from the soil and begin releasing sex pheromone (Ball, 1957; Hill, 1975; Branson et al., 1977; Sherwood and chap06.qxd 24/11/04 1:42 PM Page 128

128 J.L. Spencer et al.

Levine, 1993; Hammack, 1995). Mating pairs are commonly observed in copulo near the base of maize plants. Once mated, WCR females feed on the maize tissues available in the field where they emerged. After a period of several days, between 15% (Coats et al., 1986) and 24% (Naranjo, 1990) of the newly mated females engage in sustained, migratory flights of > 30 min duration (as measured with tethered flight mills) that would facilitate long-distance dispersal from their natal field. In the field, many young WCR females may be observed ascending during the hours just before dusk from mid-July to early August. Those that engage in migratory flight are believed to fly for a period at an elevation significantly higher than the plant canopy before they descend into another maize field, where they resume feeding and mature their first clutch of eggs. In laboratory tests, only WCR between 2 and 10 days old engaged in sustained flights (Coats et al., 1987). Long-distance, sustained flight is also associated with moderate juvenile hormone (JH) titres (Coats et al., 1987). Application of a JH mimic or inhibitor had significant effects on WCR sustained (and trivial) flight tendencies and capabilities. Concurrent collection of adult WCR from within the canopies of maize and soybean fields, and those flying at 1 and 10 m above soybean canopy level reveal that populations vary significantly between elevation and collection site. The highest-flying (10 m) population is composed primarily of young, newly mated females (containing a recognizable spermatophore, and sperm in their spermathecae, but lacking any mature eggs) with residual maize tissue in their gut contents (Table 6.1). Adults collected from within the maize canopy are similar to those collected from high elevation; however, there is a higher proportion of unmated females in the maize canopy (7%) than at high elevation (< 2%). The similarities suggest WCR at 10 m recently originated from maize fields. Insects collected within the soybean canopy have soybean tissues in their gut contents, are all mated and carry oocytes that are significantly more mature than those of females from in or above maize fields. The presence of spermatophores of any size was significantly less common among these insects than in those collected elsewhere. Females flying within 1 m of the soybean canopy are very similar to those within the soybean field canopy. The proportion of female WCR among insects collected in the soybean canopy and those flying just above the canopy is significantly less than that among the high-flying population, though it is still strongly biased toward females. Overall, the array of characteristics suggests that the WCR in and around soybean fields are more mature than those collected at high elevation. We hypothesize that females moving in the soybean canopy have dispersed from their natal field and begun periodic movement between maize and soybean fields. A several day’s to a week delay in the movement of WCR adults from their natal maize fields into soybean (or other non-host fields) is typical of WCR abundance patterns previously characteristic of continuous vs. rotated maize fields (Godfrey and Turpin, 1983). Prior to rotation resistance, few WCR were present in rotated crops and few WCR emerged in chap06.qxd 24/11/04 1:42 PM Page 129

Movement, Dispersal and Behaviour of Adults in Maize 129

Table 6.1. The characteristics of dissected female WCR captured in maize and soybean ﬁelds in 2000 at Urbana, Illinois, USA. Values in the same row bearing the same letter are not statistically different at α = 0.05. Female weight was compared using Fisher’s protected LSD test following signiﬁcant analysis of variance (ANOVA) at α = 0.05. Per cent data were compared as proportions using the methods of Zar (1996) for comparing multiple proportions.

WCR collection location and method Maize ﬁelds Soybean ﬁelds Live Aerial net Aerial net Sweep net collection at 10 ma at 1 m in canopyb Measured quantity n = 502 n = 502 n = 113 n = 120

Female weight (g/103) 12.3 ab 11.7 b 99.1 a 12.9 a % mated 93.0 b 99.1 a 44.5 ab 0.0 a % with maize tissue in gutc 79.5 a 44.5 b 12.0 c 8.3 c % with soybean tissue in gut 1.3 d 12.0 c 84.0 b 85.8 a % with spermatophoresd 59.6 b 84.0 a 43.7 c 49.2 bc % with large spermatophores 47.4 a 43.7 a 0.6 c 25.8 b % with mature eggse 1.3 ab 0.6 b 2.7 ab 7.5 a

a10 m aerial collections were made from the top of scaffolding towers. bCollection made with a 37 cm diam. net. cPlant tissues were visually identified in gut contents at 100× magnification. dIn newly mated females, the spermatophore fills the bursa copulatrix and the milky lobe of the spermatophore is as large as or larger than the pink lobe. The milky lobe shrinks and is gone within 3 days, evidence of the spermatophore is gone by 7 days post-mating (Lew and Ball, 1980). eMature eggs had full sculpturing on the chorion. LSD, least significant difference.

first-year maize. WCR abundance in first-year maize remained low until females mated and dispersed from continuous maize fields. Because females also predominate among the dispersing adults, sex ratios outside continuous maize were significantly biased toward females. Today, due to rotation resistance, there is now a lagging relationship between WCR abundance in first-year maize and soybean (as well as other crops in rotation with maize). This pattern is characteristic of rotation-resistant WCR populations (O’Neal et al., 1999; Levine et al., 2002; Rondon and Gray, 2003; Fig. 6.2). The effect of rotation resistance on WCR abundance and movement between maize and soybean fields is illustrated with a comparison of seasonal WCR abundance patterns between areas with and without rotation resistance. Beetle abundance patterns are dramatically different between Monmouth (Warren County), Illinois, and Urbana (Champaign County), Illinois (Fig. 6.3). Abundance patterns in these locations (separated by 200 km) typify rotation-susceptible and rotation-resistant populations. At Monmouth WCR are uncommon visitors to soybean fields (as measured chap06.qxd 24/11/04 1:42 PM Page 130

130 J.L. Spencer et al.

Warren County Champaign County Rotation-susceptible population Rotation-resistant population 1.3 WCR/100 sweeps 234 WCR/100 sweeps

) 35 120 SEM 30 Maize 100 25 Soybean 80 20 60 15 10 40 5 20

0 0 Mean no. WCR/trap/day (± 187 194 201 208 215 222 229 236 243 250 187 194 201 208 215 222 229 236 243 250 Julian date Julian date

) 20 Maize Soybean 50 Maize Soybean

SEM 18 16 40 14 12 30 10 8 20 6 4 10 2

Mean no. WCR/trap/day (± 0 0 C5 C4 C3 C2 C1 S1 S2 S3 S4 S5 C5 C4 C3 C2 C1 S1 S2 S3 S4 S5 Transect position Transect position

Fig. 6.2. 2001 WCR abundance in Illinois soybean fields and WCR abundance data for maize and soybean fields located at Monmouth (Warren County; rotation-susceptible population) and Urbana (Champaign County; rotation-resistant population), Illinois (USA). Bar charts present seasonal average capture rates for equally spaced positions along a linear transect of traps intersecting the interface between a maize and soybean field. Line charts present weekly WCR abundance for maize (n = 5) or soybean (n = 5) traps along a linear transect over the 2001 growing season. Bar charts depict mean daily capture rates of WCR adults in insecticide + cucurbitacin-baited vial traps positioned at ear height in maize, or the same traps at the top of the soybean plant canopy.

with vial traps and sweep samples) while they are very common in Urbana soybean fields. In recent years, rotation-resistant WCR populations have made significant advances into northern and north-west Illinois. At Shabbona (De Kalb County), Illinois, WCR abundance in soybean fields has changed from the Monmouth-like pattern to one like Urbana since 1998 as rotation-resistant populations have become firmly established in north-eastern Illinois (Levine et al., 2002). Although WCR are capable of long, sustained flight, the bulk of WCR movement is probably quite local (trival adult flight duration aver- chap06.qxd 24/11/04 1:42 PM Page 131

Movement, Dispersal and Behaviour of Adults in Maize 131

Maize 50 Soybean

40 ) SEM 30 (Mean ± 20 1999 WCR/vial trap/day

174 181 188 195 202 209 216 223 230 237 244 251 258 Julian date of weekly sampling period midpoint

Fig. 6.3. Pattern of mean weekly WCR abundance (± SEM) in adjacent maize and soybean ﬁelds in an area where rotation-resistant WCR beetles are present. WCR per trap were measured with insecticide + cucurbitacin-baited vial traps positioned at ear height in maize and at the top of the plant canopy in soybean. Data are from Urbana, Illinois (USA), in 1999.

ages 3.1 min/flight and covers an average of just 68 min/flight (Coats et al., 1986)). While sustained flight-mill flights were much longer (averaging 71.8 min/flight) and could result in calculated daily displace- ments of up to 36 km/day, Illinois weather conditions probably limit the intervals when sustained flight would be possible. Based on 181.7 h of meteorological observations concurrent with flying WCR collections at our Urbana location (18 July to 5 September 1997), Onstad et al. (1999) estimated that WCR adults would have only 1.45 h suitable for sustained flight and a typical beetle could travel only c. 4.4 km/year during that period without the aid of wind. The historical record of WCR population spread and the recent expansion of rotation-resistant WCR populations greatly exceed 4.4 km/year and indicate that there must be another contributing mechanism that facilitates long-distance dispersal. Before rotation resistance was evident, Grant and Seevers (1989) documented long-distance adult WCR transport associated with passage of summertime convective storms from 1984 to 1986. They report 16 occasions when large numbers of WCR beetles washed up on the southern shores of Lake Michigan after the passage of weather fronts in July and August. Onstad et al. (1999) used assumptions related to the WCR sustained-flight capacity and other weather-related indicators to show that storm- and wind-supported movement could explain much of the expan- chap06.qxd 24/11/04 1:42 PM Page 132

132 J.L. Spencer et al.

sion of rotation resistance between 1987 and 1997. With the aid of wind, it was calculated that an adult WCR could move 10–30 km/year, depending on storms and prevailing winds (Onstad et al., 1999). Onstad et al. (2003) revisited this problem and, using additional weather and landscape data, illustrated how various models for the spread of rotation resistance were more or less sensitive to other factors (e.g. local landscape diversity) that were not considered in the original model. Allowing for combinations of dispersal-promoting conditions, a more accurate fit between observed root injury and various measures of WCR abundance was possible. Annual statewide sampling of WCR adults in soybean fields and root-injury evaluations for maize fields in 30 or more select Illinois counties have provided data that have proved useful for evaluating the yearly spread of the rotation-resistance problem area (Shroeder and Ratcliffe, 2003). The Onstad et al. (1999, 2003) models suggest that an important factor in the evolution and spread of rotation resistance is the local proportion of land that is not in rotation with maize and does not provide host plants for WCR larvae (i.e. ‘extra vegetation’). These areas provide a counteract- ing selection against oviposition outside of maize and become a sink for eggs. Onstad et al. (2001) suggest that, where less than c. 80% of the plant landscape is rotated, the presence of ≥ 20% of the land cover as ‘extra vegetation’ (i.e. non-maize and non-rotated soybean vegetation) does not provide adequate selective advantage to maintain rotation resistance when populations move into these areas. The east-central Illinois epicentre of the rotation-resistance problem has a very low percentage of extra vegetation in the landscape; six of the nine counties have ≤ 10% extra vegetation and the remainder have ≤ 20% extra vegetation (Onstad et al., 2003). In western and southern Illinois, most of the counties have ≥ 20%, with many having ≥ 30–50% extra vegetation (Onstad et al., 2003). In a 3- year sticky trap survey of Michigan counties, O’Neal et al. (2003) report that very few soybean fields supported WCR populations indicative of a rotation-resistance threat. In contrast to Illinois, no Michigan counties have percentages of extra vegetation at or below the 20% hypothesized by Onstad et al. (2001) as necessary to sustain rotation resistance. O’Neal et al. (2003) report high adult WCR abundance in some soybean fields, but suggest that normal WCR responses to crop maturity (Darnell et al., 2000) and farming practices (e.g. early season harvest of silage maize) can adequately explain beetle movement out of maize fields. They also suggest that low levels of adult WCR emergence in Michigan fields of first-year maize after soybean is consistent with the innate tendencies of WCR beetles to oviposit outside maize at some low level rather than expression of rotation resistance. Slower than predicted expansion of rotation- resistant WCR populations into Michigan may be a consequence of the greater vegetational diversity in this state, as hypothesized by Onstad et al. (2003). chap06.qxd 24/11/04 1:42 PM Page 133

Movement, Dispersal and Behaviour of Adults in Maize 133

Factors Inﬂuencing Movement

Once a mated female disperses from her natal field to a new maize field, daily interfield movements are subject to several limiting influences. WCR movement between maize fields and locations outside maize (principally adjacent soybean fields) exhibits a diel periodicity (Isard et al., 1999, 2000). Extensive sweep sampling and capture of flying WCR using malaise traps positioned at maize and soybean field interfaces reveals that WCR abundance in soybean fields is significantly greater in the morning and evening, with significantly lower populations in the middle of the day. These interfield movement patterns are confirmed by gut-content dissection studies showing that the proportion of WCR beetles collected in soybean fields with both maize and soybean tissues in their gut contents (indicative of recent movement into soybean from maize fields) rises and falls as WCR move in and out of soybean fields (Spencer et al., 1999b). Movement between maize and soybean fields (and movement in general) is gated by atmospheric conditions; WCR flight is not initiated during darkness, nor do beetles fly when temperatures exceed 33°C or when winds are greater than c. 3.3 m/s. Under conditions when flight is possible, no single factor dominates the flight probability function of the WCR (Isard et al., 1999). Instead, the likelihood that WCR will move between fields is closely related to atmospheric stability, measured as the gradient Richardson number (Rosenberg et al., 1983). WCR flight is most likely under permissive conditions when the atmosphere is unstable (negative Richardson numbers; temperatures tend to decrease with elevation leading to buoyant conditions that favour the upward movement of air parcels) or neutral (Richardson numbers around 0). A stable atmosphere (temperatures rise with elevation (Richardson numbers ≤ 0)) is unfavourable for interfield flight. Apparently, as long as the atmosphere is not working against WCR buoyancy, beetles are apt to fly. Recent short-interval analyses of flight and atmospheric conditions have further clarified flight-favouring conditions and it has been hypothesized that WCR initiate flight during periods when the atmosphere is in transition from stable to neutral or from unstable to neutral (Isard et al., 2004). WCR are apparently cueing into changes in atmospheric stability. A second important finding arising from the same observations strongly suggests that flying WCR collected at 10 m elevation are in the initial phases of dispersal from their natal fields, rather than part of an ongoing stream of migrants that have travelled a long distance. This conclusion is based on the observation that WCR flight halted when local atmospheric conditions were unfavourable for flight initiation; if WCR at 10 m were long-distance migrants, the stream of oncoming beetles would not stop when conditions just above the plant canopy are unfavourable for local take-offs. Since local flight rose and fell with local conditions, we hypothesize that most WCR caught at 10 m have recently initiated flight and are in the process of ascending to an elevation for longer-distance movement. chap06.qxd 24/11/04 1:42 PM Page 134

134 J.L. Spencer et al.

The elevation where migrating WCR may be found is not known. However, various popular press reports of WCR beetles accumulated high (49 and 50 storeys) on the outsides of tall downtown Chicago, Illinois buildings (Gray, 2001; Nevala, 2001) suggest they may be moving at c. 130 m above the earth’s surface when not transported in convective storms. These observations are consistent with patterns for other migrating insect species that become concentrated at the top of nocturnal inversion layers in warm jet streams at altitudes of several hundred metres (Isard and Gage, 2001).

Measuring Movement

Interest in the topic of local intra- and interfield adult WCR movement grew in advance of the registration and commercialization of the first transgenic maize variety for corn rootworm in 2003: YieldGard® Rootworm. This variety is active against root-feeding diabroticite beetle larvae, including the WCR (EPA Office of Pesticide Programs, 2003). The potential for development of resistance to a particular Bacillus thuringiensis (Bt)-plant-incorporated protectant and Bt technology in general war- rants action to reduce the risk of Bt resistance and to preserve both the current and the future utility of Bt (EPA Office of Pesticide Programs, 2001). To this end, the US Environmental Protection Agency (EPA) has mandated insect resistance management (IRM) plans for all transgenic crop varieties with Bt-plant-incorporated protectants (EPA Office of Pesticide Programs, 2001). A key element of IRM plans for Bt crops is the planting of non-Bt refuges that provide a large reservoir of Bt-susceptible insects that will disperse and randomly mate with potentially Bt-resistant adults emerging from nearby Bt crops. The IRM plan for YieldGard® Rootworm includes a 20% structured non-Bt refuge (EPA Office of Pesticide Programs, 2003). How well a particular refuge design promotes adequate mating between rootworms from the Bt crop and the non-Bt refuge will depend on how well the relative size and placement of crop and refuge fields are matched to the movement capabilities of WCR adults. The defining role that adult WCR movement plays in the design and success of refuge configurations has sparked a renewed interest in quantifying individual beetle behaviour and movement patterns in the field. Detailed insect activity data are often obtained by directly observing insects or applying mark–release–recapture methods to quantify movement parameters. Many techniques have been used to mark and release individual insects so that they may be identified if later recaptured at another location (reviewed by Hagler and Jackson, 2001). Mass marking techniques have been used in WCR field and laboratory dispersal studies (Lance and Elliott, 1990; Naranjo, 1990; Oloumi-Sadeghi and Levine, 1990; Spencer et al., 1999a). In 1999 and 2000, extensive observations of WCR movement were made in an east-central Illinois maize field using mark–release–recapture methods. During 1999 and 2000, 4726 and 13,090 WCR adults, chap06.qxd 24/11/04 1:42 PM Page 135

Movement, Dispersal and Behaviour of Adults in Maize 135

respectively, were captured in maize fields and marked with fluorescent powder before being released in the centre of a concentric circular array of vial traps positioned in a 4-acre maize field. Marked insect recovery rates were disappointingly low. In 1999, 21 marked males and three marked females were recaptured (0.51% recovery); in 2000, four marked males and four marked females were recaptured (0.06% recovery). Based on distance from release point and interval between release and recovery, we calculated movement rates for males and females. Male movement rates were 4.6 m/day in 1999 and 16.1 m/day in 2000. Female movement rates were 2.6 m/day in 1999 and 21.8 m/day in 2000. While these rates fall broadly within the Coats et al. (1987) estimate of < 30 m/day for movement by tethered females engaging in trivial flight, the low recovery rates were troublesome. These results suggested that monitoring beetle dispersal using mark–release–recapture methods was very challenging, if not impossible, given the extremely high WCR abundance in east-central Illinois. The logistics of collecting, handling and marking enough WCR to ensure that a reasonable number of marked insects were recovered led us to consider a novel alternative to standard mark–release–recapture methods (Spencer et al., 2003). For several years, observation of maize and soybean tissues in the gut contents of dissected WCR that were collected in soybean and maize fields, respectively, had been used as a ‘marker’ to indicate recent interfield movement. Knowledge of average gut passage time for various ingested plant tissues allowed scientists to estimate rates of interfield movement. Though useful in a general sense, because both tissues were common in the landscape, it was not possible to know with any certainty where a particular insect originated; furthermore, the process of dissection and identification was slow. An ideal marker for monitoring WCR movement would be one the insects acquired during the course of normal behaviour and it would have an isolated, identifiable source and be unambiguously detectable. During WCR studies in transgenic maize plots, it was discovered that the presence of tissue from YieldGard® Rootworm maize (expressing the Cry3Bb1 protein) could be detected in the bodies of adult WCR that had recently fed on transgenic plant material (there are no acute toxic effects of Cry3Bb1 protein on adult WCR) (Spencer et al., 2003). Ingested transgenic tissue is detected using lateral flow test strips (Bt-Cry3Bb1 ImmunoStrips, AgDia, Elkhart, Indiana, USA) that react specifically to the presence of the Cry3Bb1 protein expressed in YieldGard® Rootworm corn. The strips indicate the presence (two lines appear) or absence (one line appears) of Cry3Bb1 protein when inserted into a buffer solution containing a pulverized WCR beetle (Fig. 6.4). In calibration tests, CryBb1 protein was detectable inside WCR beetles up to 24 h after they last consumed YieldGard® Rootworm tissue (50% were Cry3Bb1-positive after 12–16 h). Owing to the diel movement patterns of WCR adults, the 24 h detection interval for Cry3Bb1 protein using the AgDia ImmunoStrip is well suited to monitoring daily movement by WCR. chap06.qxd 24/11/04 1:42 PM Page 136

136 J.L. Spencer et al.

Transgenic tissue detection (TTD) has several advantages over traditional mark–release–recapture methods of monitoring insect movement. There is no handling of the insects to apply a marker since WCR acquire the marker during feeding. Because the YieldGard® Rootworm maize is continually available to insects in the ﬁeld, new insects are constantly ‘marked’ as they feed. The short detection interval ensures that any observed displacement must have occurred within 24 h. Lastly, the probability of detecting marked insects is much higher using TTD because the pool of marked insects is higher (and constantly being renewed). By growing isolated blocks of transgenic maize that are contiguous with non- transgenic maize varieties or other crops in rotation with maize, it is possible to measure rates of WCR movement out of transgenic maize. Since the distance from a source of Cry3Bb1-expressing maize is known and the detection interval is also known, we can calculate rates of movement for captured WCR between the transgenic maize and other locations. Over 3 years, we have found good coincidence between movement rate estimates derived from TTD and those generated from mark–recapture methods and have enjoyed a tremendous increase in the proportion of marked insects we recover. Nearly 15% (109/737) of WCR captured on 1 day in 2001

Pestles with Cry3Bb1-speciﬁc abrasive immunostrips Completed Cry3Bb1 detection tests

WCR Adult with transgenic maize tissues (expressing Cry3Bb1 Cry3Bb1 Cry3Bb1) in gut contents (+) (–)

Tubes with buffer solution

Fig. 6.4. Immunostrip test for detection of Cry3Bb1 protein (expressed in YieldGard® Rootworm maize; Monsanto Company, St Louis, Missouri, USA) contained in the gut contents of individual WCR beetles. Cry3Bb1-speciﬁc Immunostrips are available from AgDia, Inc., Elkart, Indiana (USA). The presence of two lines on an Immunostrip after development (10 min) in a processed WCR sample indicates that Cry3Bb1 protein was expressed in maize tissue ingested by the sample insect. The post-feeding detection interval for Cry3Bb1 protein in WCR gut contents is 24 h. chap06.qxd 24/11/04 1:42 PM Page 137

Movement, Dispersal and Behaviour of Adults in Maize 137

were Cry3Bb1-positive in plots surrounding YieldGard® Rootworm maize, compared to the 0.106% (9/8497) of ﬂuorescent powder-marked WCR recovered within our trapping array in 1999. Within maize plots and between maize and soybean plots, c. 85–90% of male and female WCR moved ≤ 4.6–9.1 m/day (Spencer et al., 2003).

Diet and Movement

Significant effects of diet on WCR movement/activity and reproductive behaviour have recently been measured (Mabry, 2002; Mabry and Spencer, 2003; Mabry et al., 2004). Understanding how herbivory might influence beetle behaviour has been of interest since the earliest reports of WCR herbivory on soybean foliage. Initial hypotheses suggested that rotation-resistant WCR beetles might be attracted to soybean fields to feed and/or lay eggs (Sammons et al., 1997). Subsequent studies falsified these hypotheses and demonstrated that there is no particular WCR attraction to soybean fields or any unique tendency to feed on soybean tissue among rotation-resistant beetles (Spencer et al., 1999a; Levine et al., 2002; O’Neal et al., 2002; Mabry and Spencer, 2003). In fact, soybean herbivory is documented for WCR populations regardless of their rotation-resistance status (O’Neal et al., 2002; Mabry and Spencer, 2003). Dissection data reveal that the proportion of WCR adults that feed on soybean tissues while in soybean fields (c. 55%) is the same, independent of the rotation- resistance status of the collection area (Levine et al., 2002). While the likelihood of feeding on soybean is the same among insects that leave maize fields, the proportion of maize field individuals that exit maize fields and feed on soybean foliage is higher among rotation-resistant populations. The abundance of rotation-resistant WCR moving between fields and feeding on soybean has led to speculation that WCR may become significant vectors of soybean diseases, such as bean pod mottle virus (BPMV) (Mabry et al., 2003). Good coincidence between the proportion of WCR positive for BPMV-infected tissue in gut contents and plant infection levels in fields suggests that the activities of abundant WCR in soybean fields may make them an efficient disease sampling ‘tool’. Despite its prevalence among insects collected from soybean fields, soybean herbivory offers very little nutritional advantage to WCR beetles (Mabry, 2002; Mabry and Spencer, 2003; Mabry et al., 2004). In extensive feeding test–mortality assays comparing rotation-resistant and susceptible populations and individuals with and without a tendency to feed on soybean tissue, there was no differential advantage to soybean herbivory. An exclusive diet of soybean tissues kills a majority of WCR females within 3–6 days regardless of their origin or tendency towards soybean herbivory. In the context of their assays, Mabry and Spencer (2003) found that the consequences of soybean herbivory were very similar to those of starvation (insects had continuous access to water, but no food). While an exclusive soybean diet is soon fatal to WCR, beetles that chap06.qxd 24/11/04 1:42 PM Page 138

138 J.L. Spencer et al.

feed on maize tissues after soybean herbivory are as vigorous as those that feed exclusively on maize tissues. Under laboratory conditions, adult WCR can recover from up to 3 days of starvation or soybean herbivory if subsequently given access to maize tissues. This finding also illustrates how rapidly rotation-resistant WCR that temporarily abandon maize fields to feed and lay eggs in soybean fields can be restored to a state of vigour like that of beetles that always feed on maize. The ability to rapidly mitigate the consequences of soybean herbivory allows WCR that lay eggs in soybean fields to gain a fitness advantage over beetles laying eggs only in maize fields without incurring a nutritional penalty for leaving maize and feeding on soybean tissue (Mabry and Spencer, 2003). The fecundity of WCR maintained on diets that alternated between maize and soybean tissues was equivalent to the fecundity of insects that consumed only maize tissues (Mabry et al., 2004). Since even variation in maize diet quality had been shown to reduce WCR egg production (Elliott et al., 1990), it was expected that beetles given diets that alternated between maize and soybean tissues would produce fewer eggs than beetles with continuous access to maize. In a model for the evolution of WCR behavioural resistance to crop rotation, Onstad et al. (2001) assumed that the fecundity of females moving between maize and soybean fields and feeding in both locations was reduced in proportion to the area of non- maize habitat in the landscape. However, since the consequences of a key behavioural tendency associated with rotation resistance – leaving a maize field to lay eggs in the crop rotated with maize – result in little or no fecundity cost to a female, it is not surprising that rotation resistance was quickly established. Independent of other costs associated with interfield movement (e.g. increased risk of predation (not considered here)), the fitness advantage gained by insects that lay eggs outside of maize is already tremendous in east-central Illinois (Onstad et al., 2001). The demonstration that there was no fecundity penalty among beetles that eat mixed diets suggests that selection for a rotation-resistant population could proceed much more quickly than predicted (Onstad et al., 2001).

Diet and Mechanisms of Rotation Resistance

Effects of soybean herbivory on movement are evident on a day-to-day scale; soybean herbivory increases beetle activity compared to beetles feeding on maize tissues (Mabry, 2002; Mabry et al., 2004). Earlier studies indicated that WCR feeding on soybean tissues (or on water-only diets) initiated flight significantly faster than WCR feeding on maize diets and were also more agitated prior to flight initiation (J.L. Spencer, unpublished laboratory data). Mabry (2002) confirmed that soybean feeding led to increased beetle activity, using an assay that measured the speed of WCR movement upward through a vertical glass tube. Individual female movement rates were significantly greater on days when they had access to only soybean tissue than on days with access to maize tissues. WCR chap06.qxd 24/11/04 1:42 PM Page 139

Movement, Dispersal and Behaviour of Adults in Maize 139

from rotation-susceptible and rotation-resistant populations both became significantly more active when tested on days when they had access to soybean tissues; however, the magnitude difference between activity rates on days with maize vs. soybean access days was greater among rotation- susceptible beetles. This population difference led Mabry (2002) to pos- tulate that rotation-susceptible WCR may be more sensitive to short- interval (day-to-day) dietary heterogeneity than beetles from rotation- resistant populations. Greater activity in response to non-hosts may prevent the rotation-susceptible beetles from becoming arrested for long periods and thus reducing their sustained exposure to non-hosts outside maize fields. Exposure to soybean or soybean herbivory (or starvation) led to an unexpected effect on reproduction; WCR females were significantly more likely to deposit eggs on days when they had access to soybean tissues (Mabry, 2002). This was true regardless of the interval between access to maize and soybean diets, which ranged between 1 and 3 days. Increased oviposition in response to soybean exposure and soybean herbivory may be a key proximate mechanism underlying rotation resistance. Mabry and Spencer (2003) also measured the reproductive characteristics of WCR females flying out of maize and towards a soybean field. Only 1.5% of 203 females leaving maize in late July were gravid (determined by the presence of mature, chorionated eggs in their ovaries); furthermore, only 20% were capable of eventually producing eggs without additional feeding on maize tissues. O’Neal et al. (1999) previously noted that a similar percentage of gravid WCR females was collected above the soybean field canopy (average = 1.3%) and hypothesized that gravid females may not be as prone to capture on Pherocon® AM sticky traps. The low representation of gravid females among the population moving out of maize and into soybean fields suggests that the proximate mechanism for exiting maize is not strongly dependent on possession of mature eggs and imminent oviposition. Rather, these findings support hypotheses that the initial steps related to rotation resistance may be tied to increased activity or a general increase in the propensity of WCR adults to leave maize fields. The combination of an increase in activity/flight propensity and oviposition in response to soybean exposure suggests a plausible scenario for explaining how soybean herbivory (or perhaps any non-host feeding) may facilitate rotation resistance. It is important to note that neither oviposition nor feeding appears to be the origin of the primary stimulus causing entry into soybean fields. We hypothesize that the presence of WCR in soybean fields, and other locations outside maize fields, is probably a function of a change in general WCR activity, which may be symp- tomatic of a relaxed affinity to maize fields that was selected for by crop rotation. Once outside the maize field, soybean fields (and other non-host plant fields) provide an environment where herbivory (or starvation) results in stress, which promotes egg-laying and greater activity. This response helps sustain the phenomenon of rotation resistance because it chap06.qxd 24/11/04 1:42 PM Page 140

140 J.L. Spencer et al.

leads to oviposition in fields likely to be rotated to maize and facilitates activities that increase the likelihood that a beetle makes a timely return to a host plant field. While recent feeding studies suggest that the soybean herbivory scenario provides a possible mechanistic explanation for oviposition and activity outside maize fields, debate still rages over the mechanism responsible for movement of females out of maize fields and into non-host fields. The first step in rotation resistance is periodic female movement out of maize fields; dispersal from the natal maize field (and into another maize field to feed and lay eggs) is already part of normal female WCR behaviour (Godfrey and Turpin, 1983). Clearly, WCR entry into a soybean field is not due to the nutritional benefit available in soybean, nor is it driven by imminent oviposition. O’Neal et al. (2002) have demonstrated an influence of maize phenology on the likelihood for WCR to feed on maize and soybean tissue in the laboratory. The presence of mature maize (foliage and silks) in choice tests with soybean tissues led to greater consumption of soybean tissues. The phenology hypothesis of O’Neal et al. (2002) suggests that movement out of maize fields occurs because WCR are abandoning late vegetative stage/early reproductive stage maize in search of less mature maize tissues. While it is very evident that adult WCR distribution can be affected by maize phenology (Darnell et al., 2000), it does not account for season-long movement of WCR between maize and soybean fields, which commences shortly after first adult emergence and continues even while maize is pollinating (Levine et al., 2002). In a test of the phenology hypothesis, Pierce (2003) exaggerated phenological differences between adjacent maize and soybean fields in an area without rotation resistance (Monmouth, Illinois) in an attempt to move WCR adults into soybean, but failed to see significant movement. The selection-based model does not depend on crop phenology differences but on reduced female fidelity to maize fields (due to crop rotation imposed selection for initially rare females with reduced fidelity) to account for increased interfield movement. At the core of this hypothesis is the assumption that there are genetic differences between individuals from rotation-susceptible and -resistant populations. Studies focusing on differential gene expression between WCR populations are under way.

Summary

1. The WCR is a human-made pest. Human activities have repeatedly contributed to increasing the potential for the WCR to inﬂict economic damage on commercial maize. WCR populations that became resistant to insecticide in the 1950s were more mobile than previous populations; increased mobility probably helped the WCR to colonize the eastern US Corn Belt by the early 1980s. 2. Near universal adoption of crop rotation in portions of Illinois imposed selection favouring WCR beetles with reduced ovipositional ﬁdelity to chap06.qxd 24/11/04 1:42 PM Page 141

Movement, Dispersal and Behaviour of Adults in Maize 141

maize fields. WCR became behaviourally resistant to crop rotation by laying eggs in crops like soybean that were growing in annual rotation with maize. An initially isolated problem has now grown to encompass portions of at least four US states (Illinois, Indiana, Ohio and Michigan). 3. WCR mobility has contributed to its pest status. Rotation resistance is a problem of movement. Understanding WCR movement patterns and capabilities is crucial to dealing with current management challenges and anticipating future ones. 4. Significant episodes of WCR long-distance movement and local dispersal occur shortly after adult emergence and mating. Some WCR adults may move 10s of kilometres over several days, others move a few kilometres in their entire lifetime. WCR movement between maize and soybean fields has a strong diel periodicity but is subject to limitations imposed by local weather conditions. Beetles in soybean fields consume soybean foliage with a high probability. 5. Soybean herbivory has significant consequences for WCR survival, movement and reproduction. Negative effects of soybean herbivory may be mitigated by subsequent feeding on maize tissues. Soybean herbivory increases WCR activity and stimulates oviposition. Behavioural effects of soybean herbivory may be the proximate mechanism for rotation resistance. 6. Local movement rates of 85–90% of male and female WCR beetles in maize and soybean fields are between 4.6 and 9.1 m/day.

References

Ball, H.J. (1957) On the biology and egg-laying habits of the western corn rootworm. Journal of Economic Entomology 50, 126–128. Branson, T.F., Guss, P.L. and Jackson, J.J. (1977) Mating frequency of the western corn rootworm. Annals of the Entomological Society of America 70, 506–508. Burckhardt, C.C. and Bryson, H.R. (1955) Notes on the distribution of the western corn rootworm Diabrotica virgifera Lec. in Kansas. Journal of the Kansas Entomological Society 28, 1–3. Coats, S.A., Tollefson, J.J. and Mutchmor, J.A. (1986) Study of migratory flight in the western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 15, 620–625. Coats, S.A., Mutchmor, J.A. and Tollefson, J.J. (1987) Regulation of migratory flight by juvenile hormone mimic and inhibitor in the western corn rootworm (Coleoptera: Chrysomelidae). Annals of the Entomological Society of America 80, 697–708. Darnell, S.J., Meinke, L.J. and Young, L.J. (2000) Influence of corn phenology on adult western corn rootworm (Coleoptera: Chrysomelidae) distribution. Environmental Entomology 29, 587–595. Elliott, N.C., Gustin, R.D. and Hanson, S.L. (1990) Influence of adult diet on the reproductive biology and survival of the western corn rootworm, Diabrotica virgifera virgifera. Entomologia Experimentalis et Applicata 56, 15–21. EPA Office of Pesticide Programs (2001) Biopesticides registration action document: Bacillus thuringiensis plant-incorporated protectants. http:www.epa.gov/pesticides/ biopesticdes/ingredients/reds/brad-b-pip2.htm EPA Office of Pesticide Programs (2003) Biopesticides registration action document: event chap06.qxd 24/11/04 1:42 PM Page 142

142 J.L. Spencer et al.

MON 863. Bacillus thuringiensis Cry3Bb1 corn. http:www.epa.gov/pesticides/ biopesticides/ingredients/factsheets Gillette, C.P. (1912) Diabrotica virgifera LeC. as a corn root-worm. Journal of Economic Entomology 5, 364–366. Godfrey, L.D. and Turpin, F.T. (1983) Comparison of western corn rootworm (Coleoptera: Chrysomelidae) adult populations and economic thresholds in first-year and continuous cornfields. Journal of Economic Entomology 76, 1028–1032. Grant, R.H. and Seevers, K.P. (1989) Local and long-range movement of adult western corn rootworm (Coleoptera: Chrysomelidae) as evidenced by washup along southern Lake Michigan shores. Environmental Entomology 18, 266–272. Gray, M. (2001) Western corn rootworm densities reach alarming levels in rural and urban areas. Pest Management and Crop Development Bulletin 20 (August), 10. Hagler, J.R. and Jackson, C.G. (2001) Methods for marking insects: current techniques and future prospects. Annual Review of Entomology 46, 511–543. Hammack, L. (1995) Calling behavior in female western corn rootworm beetles (Coleop- tera: Chrysomelidae). Annals of the Entomological Society of America 88, 562–569. Hill, R.E. (1975) Mating, oviposition patterns, fecundity and longevity of the western corn rootworm. Journal of Economic Entomology 68, 311–315. Hills, T.M. and Peters, D.C. (1971) A method of evaluating post-planting insecticide treatments for control of western corn rootworm larvae. Journal of Economic Entomology 64, 764–765. Isard, S.A. and Gage, S.H. (2001) Flow of Life in the Atmosphere: an Airscape Approach to Understanding Invasive Organisms. Michigan State University Press, East Lansing, Michigan, 240 pp. Isard, S.A., Nasser, M.A., Spencer, J.L. and Levine, E. (1999) The influence of weather on western corn rootworm flight activity at the borders of a soybean field in east central Illinois. Aerobiologia 15, 95–104. Isard, S.A., Spencer, J.L., Nasser, M.A. and Levine, E. (2000) Aerial movement of western corn rootworm, Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae): diel periodicity of flight activity in soybean fields. Environmental Entomology 29, 226–234. Isard, S.A., Spencer, J.L., Mabry, T.R. and Levine, E. (2004) The influence of atmospheric conditions on high elevation flight of western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 33, 350–356. Krysan, J.L. (1993) Adaptation of Diabrotica to habitat manipulations. In: Kim, K.C. and McPheron, B.A. (eds) Evolution of Insect Pests/Patterns of Variation. John Wiley & Sons, New York, pp. 361–373. Lance, D.R. and Elliott, N.C. (1990) Marking western corn rootworm beetles (Coleoptera: Chrysomelidae): effects on survival and a blind evaluation for estimating bias in mark–recapture data. Journal of the Kansas Entomological Society 63, 1–8. LeConte, J.L. (1868) New Coleoptera collected on the survey for the extension of the Union Pacific Railroad, E.D. from Kansas to Fort Craig, New Mexico. Transactions of the American Entomological Society 2, 49–59. Levine, E. and Oloumi-Sadeghi, H. (1991) Management of diabroticite rootworms in corn. Annual Review of Entomology 36, 229–255. Levine, E. and Oloumi-Sadeghi, H. (1996) Western corn rootworm (Coleoptera: Chrysomelidae) larval injury to corn grown for seed production following soybeans grown for seed production. Journal of Economic Entomology 89, 1010–1016. Levine, E., Oloumi-Sadeghi, H. and Fisher, J.R. (1992) Discovery of multiyear diapause in Illinois and South Dakota northern corn rootworm (Coleoptera: Chrysomelidae) eggs and incidence of the prolonged diapause trait in Illinois. Journal of Economic Entomology 85, 262–267. chap06.qxd 24/11/04 1:42 PM Page 143

Movement, Dispersal and Behaviour of Adults in Maize 143

Levine, E., Spencer, J.L., Isard, S.A., Onstad, D.W. and Gray, M.E. (2002) Adaptation of the western corn rootworm, Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae) to crop rotation: evolution of a new strain in response to a cultural management practice. American Entomologist 48, 94–107. Lew, A.C. and Ball, H.J. (1980) Effect of copulation time on spermatozoa transfer of Diabrotica virgifera (Coleoptera: Chrysomelidae). Annals of the Entomological Society of America 73, 360–361. Mabry, T.R. (2002) The effects of soybean herbivory on the behavior and ecology of the western corn rootworm (Diabrotica virgifera virgifera LeConte) variant. MSc thesis, University of Illinois, Urbana, Illinois. Mabry, T.R. and Spencer, J.L. (2003) Survival and oviposition of a western corn rootworm variant feeding on soybean. Entomologia Experimentalis et Applicata 109, 113–121. Mabry T.R., Hobbs, H.A., Steinlage, T.A., Johnson, B.B., Pedersen, W.L., Spencer, J.L., Levine, E., Isard, S.A., Domier, L.L. and Hartman, G.L. (2003) Distribution of leaf feeding beetles and bean pod mottle virus (BPMV) in Illinois and insect transmission of BPMV in soybean. Plant Disease 87, 1221–1225. Mabry, T.R., Spencer, J.L., Levine, E. and Isard, S.A. (2004) Western corn rootworm (Coleoptera: Chrysomelidae) behavior is affected by alternating diets of corn and soybean. Environmental Entomology 33, 860–871. Metcalf, R.L. (1979) Plants, chemicals, and insects: some aspects of coevolution. Bulletin of the Entomological Society of America 25, 30–35. Metcalf, R.L. (1982) Insecticides in pest management. In: Metcalf, R.L. and Luckmann, W.H. (eds) Introduction to Insect Pest Management, 2nd edn. Wiley-Interscience, New York, pp. 217–277. Metcalf, R.L. (1983) Implications and prognosis of resistance to insecticides. In: Georghiou, G.P. and Saito, T. (eds) Pest Resistance to Pesticides. Plenum Press, New York, pp. 703–733. Metcalf, R.L. (1986) Foreword. In: Krysan, J.L. and Miller, T.A. (eds) Methods of the Study of Pest Diabrotica. Springer-Verlag, New York, pp. vii–xv. Naranjo, S.E. (1990) Comparative flight behavior of Diabrotica virgifera virgifera and Diabrotica barberi in the laboratory. Entomologia Experimentalis et Applicata 55, 79–90. Nevala, A.E. (2001) Insects taking city by swarm: weather’s quirks comfy for critters. Chicago Tribune, 19 August, p. 1. http://chicagotribune.com/news/local/Chicago/ chi-010890305aug19.story Oloumi-Sadeghi, H. and Levine, E. (1990) A simple, effective, and low-cost method for mass marking adult western corn rootworms (Coleoptera: Chrysomelidae). Journal of Entomological Science 25, 170–175. O’Neal, M.E., Gray, M.E. and Smyth, C. (1999) Population characteristics of a western corn rootworm (Coleoptera: Chrysomelidae) strain in east-central Illinois corn and soybean fields. Journal of Economic Entomology 92, 1301–1310. O’Neal, M.E., Gray, M.E., Ratcliffe, S. and Steffey, K.L. (2001) Predicting western corn rootworm (Coleoptera: Chrysomelidae) larval injury to rotated corn with Pherocon AM traps in soybeans. Journal of Economic Entomology 94, 98–105. O’Neal, M.E., DiFonzo, C.D. and Landis, D.A. (2002) Western corn rootworm (Coleoptera: Chrysomelidae) feeding on corn and soybean leaves affected by corn phenology. Environmental Entomology 31, 285–292. O’Neal, M.E., DiFonzo, C.D., Landis, D.A. and Meek, D. (2003) Monitoring Diabrotica virgifera virgifera (LeConte) in Michigan soybean fields and subsequent adult emergence in rotated and continuous cornfields. Great Lakes Entomologist 35, 173–181. Onstad, D.W., Joselyn, M.G., Isard, S.A., Levine, E., Spencer, J.L., Bledsoe, L.W., Edwards, C.R., DiFonzo, C.D. and Willson, H. (1999) Modeling the spread of western corn root- chap06.qxd 24/11/04 1:42 PM Page 144

144 J.L. Spencer et al.

worm (Coleoptera: Chrysomelidae) populations adapting to soybean–corn rotation. Environmental Entomology 28, 188–194. Onstad, D.W., Spencer, J.L., Guse, C.A., Levine, E. and Isard, S.A. (2001) Modeling evolution of behavioral resistance by an insect to crop rotation. Entomologia Experimentalis et Applicata 100, 195–201. Onstad, D.W., Crowder, D.W., Isard, S.A., Levine, E., Spencer, J.L., O’Neal, M.E., Ratcliffe, S.T., Gray, M.E., Bledsoe, L.W., DiFonzo, C.D., Eisley, J.B. and Edwards, C.R. (2003) Does landscape diversity slow the spread of rotation-resistant western corn rootworm (Coleoptera: Chrysomelidae)? Environmental Entomology 35, 992–1001. Pierce, C.F. (2003) Case study of a variant of western corn rootworm, Diabrotica virgifera virgifera LeConte, in east central Illinois. PhD dissertation, University of Illinois, Champaign, Illinois, 216 pp. Pike, D.R. and Gray, M.E. (1992) A history of pesticide use in Illinois. In: Proceedings of Eighteenth Annual Illinois Crop Protection Workshop, 3–5 March, 1992. University of Illinois, Champaign-Urbana, Illinois, pp. 43–52. Power, J.F. and Follett, R.F. (1987) Monoculture. Scientific American 256, 78–86. Rondon, S.I. and Gray, M.E. (2003) Captures of western corn rootworm (Coleoptera: Chrysomelidae) adults with Pherocon AM and vial traps in four crops in east central Illinois. Journal of Economic Entomology 96, 737–747. Rosenberg, N.J., Blad, B.L. and Verma, S.B. (1983) Microclimate: the Biological Environment. Wiley, New York. Sammons, A.E., Edwards, C.R., Bledsoe, L.W., Boeve, P.J. and Stuart, J.J. (1997) Behavioral and feeding assays reveal a western corn rootworm (Coleoptera: Chrysomelidae) variant that is attracted to soybean. Environmental Entomology 26, 1336–1342. Shaw, J.T., Paullus, J.H. and Luckmann, W.H. (1978) Corn rootworm oviposition in soybeans. Journal of Economic Entomology 71, 189–191. Sherwood, D.R. and Levine, E. (1993) Copulation and its duration affects female weight, oviposition, hatching patterns, and ovarian development in the western corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 86, 1664–1671. Shroeder, J.B. and Ratcliffe, S.T. (2003) 2003 Variant western corn rootworm on-farm survey. University of Illinois Extension, Champaign, Illinois. http://ipm.uiuc.edu/ wcrsurvey/index.html Smith, R.F. (1966) Distributional patterns of selected western North American insects. The distribution of diabroticites in western North America. Bulletin of the Entomological Society of America 12, 108–110. Spencer, J.L., Isard, S. and Levine, E. (1999a) Free flight of western corn rootworm (Diabrotica virgifera virgifera LeConte) to corn and soybean plants in a walk-in wind tunnel. Journal of Economic Entomology 92, 146–155. Spencer, J.L., Isard, S. and Levine, E. (1999b) Western corn rootworm injury in first-year corn: what’s new? In: 1999 Proceedings of the Illinois Crop Protection Technology Conference. Cooperative Extension Service, University of Illinois, Urbana-Champaign, Illinois, pp. 13–27. Spencer, J.L., Mabry, T.R. and Vaughn, T.T. (2003) Use of transgenic plants to measure insect herbivore movement. Journal of Economic Entomology 96, 1738–1749. Steffey, K.L., Gray, M.E. and Kuhlman, D.E. (1992) Extent of corn rootworm (Coleoptera: Chrysomelidae) larval damage in corn after soybeans: search for the expression of the prolonged diapause trait in Illinois. Journal of Economic Entomology 85, 268–275. Tate, H.D. and Bare, O.S. (1946) Corn Rootworms. Agricultural Experiment Station Bulletin 381, University of Nebraska, 12 pp. Zar, J.H. (1996) Biostatistical Analysis, 3rd edn. Prentice Hall, Upper Saddle River, New Jersey, 662 pp. chap07.qxd 12/11/04 1:07 PM Page 145

7 Within-ﬁeld Spatial Variation of Northern Corn Rootworm Distributions

Michael M. Ellsbury,1 Sharon A. Clay,2 David E. Clay2 and Douglas D. Malo2 1Northern Grain Insects Research Laboratory, USDA-ARS, Brookings, South Dakota, USA; 2Plant Science Department, South Dakota State University, Brookings, South Dakota, USA

Introduction

Corn rootworms, Diabrotica spp. (Coleoptera: Chrysomelidae), are major pests of maize in North America. The geographical range of the western corn rootworm, Diabrotica virgifera virgifera LeConte, grew rapidly following its initial discovery at the western edge of the Corn Belt and now extends to the eastern seaboard of the USA. Western corn rootworms have also appeared during the past decade in Eastern Europe and their distribution in Europe is currently expanding. A second species, the northern corn rootworm, Diabrotica barberi Smith and Lawrence, is indigenous to the North American prairie grasslands and does not occur outside that region. Early management practices for corn rootworms centred on larval control by planting-time application of a soil insecticide or on reduction of ovipositing adult populations by application of insecticide during the later stages of maize growth. Insecticides have often been applied routinely regardless of actual rootworm populations in relation to economic thresholds. Rapid development of resistance to pesticides in some populations of corn rootworms, human safety issues and ecological concerns have motivated the development and adoption of environmentally rational management practices. One of these, areawide management by means of a semiochemical-based adulticidal bait, has shown promise for rootworm management by using a targeted, low-dose, toxic bait formulation. Success of this approach depends greatly on universal adoption by growers in a relatively large area and requires coordinated sampling and monitoring of rootworm density to determine whether populations have reached an economic action threshold.

146 M.M. Ellsbury et al.

A 2-year maize/soybean crop rotation is generally accepted as a non- pesticidal tactic for corn rootworm management, but in some cases rotation fails to provide adequate long-term control. Northern corn rootworms, D. barberi Smith and Lawrence, have adapted to crop rotation by producing extended-diapause eggs that are capable of surviving in the soil over more than one winter, thus bridging the gap between maize phases in rotated maize and soybean cropping systems. The incidence of economically significant northern corn rootworm infestations following soybean is increasingly frequent in the northern reaches of the Corn Belt, such as eastern South Dakota and south-western Minnesota, where a 2- year maize/soybean rotation is commonly used. The central Corn Belt has also seen failures of rotation as a management strategy because of the appearance of an adapted western corn rootworm strain that lays eggs in soybean fields and thus produces populations that infest the maize phase of the rotation during the following cropping season. Regardless of the management approach, distributions of rootworm populations sampled for management decisions are usually considered as a field mean expressed as a density of insects per unit area or per sample unit for a given field. In reality, actual distributions of pests vary spatially at any given time, i.e. they are not constant over the entire expanse of a field. Spatial variability in adult northern corn rootworm populations was documented by Rossi et al. (1993) on a regional scale in north-western Iowa, USA, and by Steffey and Tollefson (1982) on a production-field scale. Ellsbury et al. (1998) described field-scale spatial variability in western and northern corn rootworm adult emergence. Existence of spatial variability suggests that a site-specific approach to the management of corn rootworms is feasible. This concept is attractive because of the environmental benefits from possible reduction in pesticide usage, lower input costs and increased safety for the grower/applicator. However, spatial patterns of pest dispersion may change through time as the population develops and food resources change. Spatiotemporal variability in pest dispersion would make site- specific pest management difficult, if not impossible, to undertake, particularly if targeted to a highly mobile stage such as the adult of the corn rootworm. Larvae of corn rootworms are dependent on availability of suitable host plant roots for survival and are not capable of moving from oviposition sites more than about 1 m through the soil (Short and Luedtke, 1970; Gustin and Schumacher, 1989). Because of the relative immobility of their soil-dwelling stages, corn rootworms are potential candidates for a site-specific approach to management. Fundamental to development of site-specific pest management is a determination of the nature and predictability of spatial variability present in pest populations, particularly for the target stages. Thus the initial objective of our studies was the characterization of spatial variability in distributions of eggs and larvae and of adult emergence for corn rootworms, undertaken as part of multidisciplinary precision agriculture research in eastern South Dakota. chap07.qxd 12/11/04 1:07 PM Page 147

Within-ﬁeld Spatial Variation of Northern Corn Rootworm Distributions 147

Materials and Methods

Study sites

Long-term study sites were established in two production fields of 64.8 ha (160 acres) located about 5 km (3 miles) apart in Brookings and Moody Counties, South Dakota, USA. Both fields were farmed as a maize/soybean rotation on 76 cm (30 in.) row centres. Only one field was planted to maize in any given year. Both fields were zone-tilled in the autumn during application of anhydrous ammonia. The Moody site averaged about 523 m above sea level, with undulating topography and 16.5 m of topographic relief from the highest to the lowest point. Toeslope and footslope areas of the Moody field were typically wet because of water seepage and poor drainage, even though the lowest areas were tiled for drainage. Average elevation of the Brookings site was about 509 m above sea level, with little undulation, and sloping with about 14.4 m of topographic relief from the highest (north-east) to the lowest (south-west) corner. The soils of the Brookings County field were generally wet except at the north-east corner and were well drained in the south-west quad- rant, where a tiled drainage system had been installed.

Spatial variability

Soil properties in both ﬁelds were characterized from georeferenced data on soil electromagnetic induction (EMI) (Fritz et al. 1999), taken over grid transects at 15 m spacing using an EM38 meter (Geonics Limited, Mississauga, Ontario, Canada). Corn rootworms were also grid-sampled at each study site at grid nodes located 65–75 m apart in transects running from south to north. Sample sites within transects were also 65–75 m apart and were offset about 33 m from sites in adjacent transects. Additional sites were established within selected transects to provide data at lag distances of less than 65 m between sites. Rootworm egg populations were estimated from four 1 l soil core samples taken at each study site during mid-spring or early autumn, bulked together and sub- sampled to 1 l. Eggs were washed from the 1 l subsamples using the methods of Shaw et al. (1976), counted and identiﬁed to verify that they were northern corn rootworm eggs. Rootworm larval populations were estimated by extraction of larvae from roots of four whole-plant samples taken from grid nodes at each site when the maize was in the V4–V5 leaf stage. Larval rootworm injury was assessed from three whole-root samples that were collected later in the growing season (late July) at each grid node, and were rated for injury severity on a 1 to 9 scale (Welch, 1977), where 1 was assigned for no damage and 9 indicated three root nodes destroyed. chap07.qxd 12/11/04 1:07 PM Page 148

148 M.M. Ellsbury et al.

Data analysis

Spatial variability of corn rootworm life stages was characterized as semivariograms computed as described in Ellsbury et al. (1998). Details of the method applied to insect populations can be found in Rossi et al. (1992), Liebhold et al. (1993) and Roberts et al. (1993). Data were transformed to ln(z + 1) and semivariances were divided by the population variance to permit comparison on a common scale (Rossi et al. 1992). Contour maps of rootworm life stage data were produced from grid files created in Surfer® (version 8; Golden Software, Golden, Colorado). Data for rootworm adult populations, root ratings and EM-38 values were assigned to landscape positions (n = 3) defined by arbitrary elevation limits using Arcview (ESRI, Redlands, California). Arbitrary zones of landscape position, each encompassing about one-third of the total topographic relief, were established. For the Moody County field, these were: footslope/toeslope, < 525 m (below 1723 ft); backslope, 525–530 m (1723–1740 ft); and shoulder/summit, > 530 m (above 1740 ft). In the Brookings County field the landscape zones were: footslope/toeslope, < 509 m (below 1670 ft); backslope, 509–513 m (1670–1683 ft); and shoulder/summit, > 513 m (above 1683 ft). Rootworm population parameters, EM-38 readings and root injury ratings for each landscape position zone were subjected to analysis of variance, semivariance analysis, which was used to produce contour maps of response variables in relation to elevation.

Results and Discussion

Semivariance analysis

Representative semivariograms are shown in Fig. 7.1 for egg, larval and adult stages of northern corn rootworms. Semivariance was plotted as a percentage of sample variance so that the figures could be compared on a common scale (Rossi et al., 1992). Distributions of northern corn rootworm eggs were best described by linear semivariogram models that were horizontal or nearly so for both the Moody County and Brookings County field sites. A typical semivariogram is shown in Fig. 7.1A for rootworm eggs sampled from the Brookings County field. The horizontal form of this semivariogram indicates little spatial dependence in distribution of northern corn rootworm eggs at the time of sampling. This interpretation is supported by mean egg densities that varied little from lowest (footslope and toeslope) to highest (summit and shoulder) landscape positions (Tables 7.1 and 7.2). Oviposition by corn rootworms thus may occur more or less randomly on a whole-field scale but may also be variable at a given location within a field. It should also be noted that the pure nugget effect seen in the semi- chap07.qxd 12/11/04 1:07 PM Page 149

Within-ﬁeld Spatial Variation of Northern Corn Rootworm Distributions 149

1.2 Eggs, Brookings Field, 1996 A 1.0 0.8 Linear model: 0.6 Co = 1.019; Co + C = 1.019; 0.4 Ao = 514.17; r2 = 0.069; 0.2 RSS = 0.0494 0.0 0 100 200 300 400 500 600

B 1.2 Larvae, Moody Field, 1999 1.0 0.8 Linear model: 0.6 Co = 0.590; Co + C = 1.020; 0.4 Ao = 360.73; r2 = 0.976;

Semivariance 0.2 RSS = 0.593 0.0 0 100 200 300 400 500 600

1.2 Adults, Moody Field, 1999 C 1.0 0.8 Exponential model: 0.6 Co = 0.546; Co + C = 1.092; 2 0.4 Ao = 178.00; r = 0.943; 0.2 RSS = 0.0227 0.0 0 100 200 300 400 500 600

Separation distance (m)

Fig. 7.1. Representative semivariograms for eggs (A), larvae (B) and adults (C) of the northern corn rootworm. Semivariance is expressed as a percentage of the sample variance to allow comparison on a common scale. Parameters: Co, nugget; C, silk; Ao, range; RSS, residual sum of squares.

variograms for egg samples indicates that it may not be possible to obtain enough samples to adequately describe the ﬁeld-scale spatial variability of rootworm eggs (Krajewski and Gibbs, 2001). Data from larval sampling produced semivariograms that were also best described by a linear model (Fig. 7.1B) but with positive slopes and a deﬁnite nugget effect. The nugget effect suggested relatively high variability between samples at short distances (Krajewski and Gibbs, 2001) chap07.qxd 12/11/04 1:07 PM Page 150

150 M.M. Ellsbury et al.

Table 7.1. Northern corn rootworm egg and adult population densities, soil electrical conductivity (EM-38) and larval injury to maize roots in relation to landscape position for the Moody County, South Dakota study site. Adult emergence was the mean total for the season at each site and eggs were sampled in the autumn. Numbers in parentheses are standard errors; means followed by the same letter were not signiﬁcantly different, Fisher’s protected LSD (P < 0.05).

Landscape Elevation EM-38 Eggs per Adults Root position interval (m) (mS/m) l soil per 0.5 m2 injury

Footslope/toeslope < 525 41.08 a 5.1 a 66.8 a 3.9 a (0.18) (1.8) (9.4) (0.2) Backslope 525–530 35.90 b 5.0 a 107.0 b 4.0 a (0.17) (1.2) (13.2) (0.2) Shoulder/summit > 530 31.02 b 4.8 a 115.5 b 4.5 a (0.07) (1.4) (18.1) (0.3)

LSD, least signiﬁcant difference.

and the positive slope indicated more spatial dependence among larval sample locations than was found for egg data. For the semivariogram shown in Fig. 7.1B the intercept with the vertical axis indicates that about 50% of the variation in larval numbers cannot be attributed to spatial correlation. We hypothesize that edaphic variability associated with the effects of mortality factors such as temperature or moisture acting on egg populations may be responsible for the differences in degree of spatial variation between egg and larval rootworm populations. Semivariograms for adult emergence densities from both study sites showed stronger spatial dependence than that seen for eggs or larvae, with a nugget effect in most cases. Semivariograms for adult emergence

Table 7.2. Northern corn rootworm larval densities, soil electrical conductivity (EM38), and larval injury to maize roots in relation to landscape position for east and west halves of study area in Brookings County, South Dakota. Adult emergence was mean total for the season at each site, eggs were sampled in the autumn. Numbers in parentheses are standard errors; means followed by the same letter were not signiﬁcantly different, Fisher’s protected LSD (P < 0.05).

Landscape Elevation EM-38 Eggs per Adults Root position interval (m) (mS/m) l soil per 0.5 m2 injury

Footslope/toeslope < 509 33.85 a 2.5 a 41.6 a 3.3 a (0.54) (0.9) (5.0) (0.3) Backslope 509–513 31.14 b 1.6 a 45.3 a 3.7 a (0.25) (0.2) (3.9) (0.2) Shoulder/summit > 513 30.67 b 1.4 a 30.6 a 4.1 a (0.34) (0.4) (20.2) (0.6) chap07.qxd 12/11/04 1:07 PM Page 151

Within-ﬁeld Spatial Variation of Northern Corn Rootworm Distributions 151

were best described by exponential or spherical models with distinct sills, suggesting a higher degree of spatially dependent variation for adult emergence than was evident in distributions of the egg or larval stages. Figure 7.1C shows a semivariogram of exponential form for northern corn rootworm adult emergence density from the Brookings County ﬁeld. The low slope is indicative of a relatively gradual change in variability with increasing distance between samples (Krajewski and Gibbs, 2001).

Distribution Maps

Distributions of northern corn rootworm eggs are shown as contour maps in Plate 1. Mean egg densities varied from 1.4 to 2.6 eggs/l in the Brookings County field and 4.8 to 5.1 eggs/l of soil in the Moody County field. Differences in the appearance of the contour maps reflect the fact that numbers of eggs in soil samples were higher in the Moody County field than in the Brookings field, even though spatial dependence was not strong at either site. Larval sampling produced generally lower numbers than did egg sampling at both sample sites (Plate 2). As with the egg samples, consistently more larvae were recovered from plant samples taken in the Moody County field than in the Brookings County field. Adult emergence densities for northern corn rootworm are shown as contour maps in Plate 3. Adult densities in the Brookings field were highest in an area apparent as a band of higher values running diagonally from north-west to south-east through an area that coincided with the location of a subsurface drainage system. This is most evident in Plate 3B. Similarly higher densities of adult rootworms occurred in better-drained areas of the Moody field at shoulder and backslope areas. The mean number of northern corn rootworm adults emerged per cage (0.5 m2) was 44.6 ± 3.5 in the Brookings County field (Plate 3A–C) and 91.5 ± 7.5 in the Moody County field (Plate 3D–F). As with the egg and larval samples, populations of adult northern corn rootworms were consistently higher at the Moody site than at the Brookings site. It should be noted that comparison among variograms for egg, larval, and adult samples may be ques- tionable because of the differences in support (Krajewski and Gibbs, 2001) among the data sets in terms of methodology by which data for the different life stages were obtained.

Landscape effects

Landscape position signiﬁcantly affected EMI readings at the Moody County site (F = 811.73, degrees of freedom (d.f.) = 2, 5277, P < 0.0001) and at the Brookings County site (F = 18.74251, d.f. = 2, 622, P < 0.001). In both ﬁelds, the highest EMI readings were associated with moisture- laden soil at the footslope/toeslope landscape positions (Tables 7.1 and chap07.qxd 12/11/04 1:07 PM Page 152

152 M.M. Ellsbury et al.

7.2). In the Moody County field (Table 7.1), adult emergence appeared to be negatively correlated with the higher EMI readings, at the footslope/toeslope positions. In contrast, adult emergence in the Brookings County field was highest at the footslope/toeslope positions, where EMI readings were also high, and yet root injury was low at these positions. In the examples from our Moody and Brookings County study sites, adult rootworm populations appeared to behave quite differently in the two fields with respect to field topography, root injury and soil EMI properties. Comparison of the two study sites suggests that it will be difficult to develop generalized guidelines for site-specific management of corn rootworms based on any single sampling protocol or indicator variable.

Conclusions

Distribution of adult emergence for northern corn rootworms is affected by soil-mediated changes in the distribution of the soil-dwelling immature stages from the time eggs are deposited until the damaging larval populations occur. Because of the influence of the soil environment on survival of the immature stages of corn rootworm, we hypothesize that measurable soil properties, such as soil electrical conductivity, may be used as ancillary variables to predict where corn rootworms are most likely to survive and cause economic loss. Intensive grid-sampling, georeferred by the use of a global positioning system (GPS), provides potentially valuable knowledge in geographical information system (GIS)-managed data layers incorporating rootworm distribution in relation to soil properties, fertility, weeds, landscape and yields in maize fields. However, it should be noted that the usefulness of the information presently is limited by the complexity of data layer interpretation and by the cost of sampling for corn rootworms. The necessary GIS/GPS capabilities are available but have not yet been effectively combined into systems incorporating map-driven application technology with economical scouting methods or real-time monitoring and mapping of corn rootworm variability. The complexity and field-to-field variability of insect pest/soil interactions is such that a multivariate approach rather than use of a single predictor such as soil electrical conductivity may be necessary for site- specific management of corn rootworms in the soil environment. The focus of continuing research will be to identify relations among soil characteristics and yield-limiting factors in order to define boundaries of zones for site-specific management.

References

Ellsbury, M.M., Woodson, W.D., Clay, S.A., Malo, D., Schumacher, J., Clay, D.E. and Carlson, C.G. (1998) Geostatistical characterization of the spatial distribution of adult chap07.qxd 12/11/04 1:07 PM Page 153

Within-ﬁeld Spatial Variation of Northern Corn Rootworm Distributions 153

corn rootworm (Coleoptera: Chrysomelidae) emergence. Environmental Entomology 27, 910–917. Fritz, R.M., Malo, D.D., Schumacher, T.E., Clay, D.E., Carlson, C.G., Ellsbury, M.M. and Dalsted, K.J. (1999) Field comparison of two soil electrical conductivity measurement systems. In: Proceedings of the 4th International Conference on Precision Agriculture. American Society of Agronomy, St Paul, Minnesota, pp. 1211–1217. Gustin, R.D. and Schumacher, T.E. (1989) Relationship of some soil pore parameters to movement of first-instar western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 18, 343–346. Krajewski, S.A. and Gibbs, B.L. (2001) Variograms explained. In: Understanding Contouring: a Practical Guide to Spatial Estimation Using a Computer and Variogram Interpretation. Gibbs Associates, Boulder, Colorado, 46 pp. Liebhold, A.M., Rossi, R.E. and Kemp, W.P. (1993) Geostatistics and geographic information systems in applied insect ecology. Annual Review of Entomology 38, 303–327. Roberts, E.A., Ravlin, F.W. and Fleischer, S.J. (1993) Spatial data representation for integrated pest management programs. American Entomologist 39, 92–107. Rossi, R.R., Mulla, D.J., Journel, A.G. and Franz, E.H. (1992) Geostatistical tools for modeling and interpreting ecological spatial dependence. Ecological Monographs 62, 277–314. Rossi, R.R., Borth, P.W. and Tollefson, J.J. (1993) Stochastic simulation for characterizing ecological spatial patterns and appraising risk. Ecological Applications 3, 719–735. Shaw, J.T., Ellis, R.O. and Luckmann, W.H. (1976) Apparatus and procedure for extracting corn rootworm eggs from soil. Illinois Natural History Survey Biological Notes No. 96. Short, D.E. and Luedtke, R.J. (1970) Larval migration of the western corn rootworm. Journal of Economic Entomology 63, 325–326. Steffey, K.L. and Tollefson, J.J. (1982) Spatial dispersion patterns of northern and western corn rootworm adults in Iowa cornﬁelds. Environmental Entomology 11, 283–286. Turpin, F.T., Dumenil, L.C. and Peters, D.C. (1972) Edaphic and agronomic characters that affect potential for rootworm damage to corn in Iowa. Journal of Economic Entomology 65, 1615–1619. Welch, V.A. (1977) Breeding corn for rootworm resistance or tolerance. In: Oden, H.D. and Wilkinson, D. (eds) Proceedings of the 32nd Annual Corn and Sorghum Research Conference. American Seed Trade Association, Washington, DC, pp. 131–142. chap07.qxd 12/11/04 1:07 PM Page 154 chap08.qxd 15/11/04 2:51 PM Page 155

8 Heterogeneous Landscapes and Variable Behaviour: Modelling Rootworm Evolution and Geographical Spread

David W. Onstad, Charles A. Guse and Dave W. Crowder Department of Natural Resources and Environmental Sciences, University of Illinois, Urbana, Illinois, USA

The western corn rootworm (WCR), Diabrotica virgifera virgifera LeConte, is the most serious insect pest of maize grown after maize (Zea mays (L.)) in the north-central USA (Levine and Oloumi-Sadeghi, 1991). WCR adults are present in maize fields from July until the first frost. From late July to September, egg laying occurs primarily in maize fields; few eggs are normally laid in other crops. The eggs of this univoltine insect overwinter in the soil and hatch from late May and early June. Larvae survive only on the roots of maize and a limited number of grasses (Levine and Oloumi-Sadeghi, 1991), making crop rotation an effective pest management strategy. For many years, growers controlled the larvae of this beetle without using soil insecticide treatments by rotating maize with soybean (Glycine max (L.)) or another non-host crop. However, since 1993 some growers in Illinois and Indiana who had successfully used crop rotation for control have suffered serious crop losses. This insect has a long history of evolv- ing resistance to insecticides (Ball and Weekman, 1962, 1963; Metcalf, 1983; Meinke et al., 1998; Miota et al., 1998; Scharf et al., 1999). We now suspect that the intense use of maize–soybean rotation has selected for an insect strain that circumvents crop rotation by laying eggs outside maize fields (Onstad et al., 1999, 2003a). In this chapter, we describe a variety of models that help explain the current distribution of the rotation-resistant phenotype over geographical space (Onstad et al., 1999, 2003a) and the evolution of this insect to crop rotation (Onstad et al., 2001b), and examine strategies for delaying resistance to transgenic maize. In all cases the heterogeneity of the landscape and the resulting selection pressure have serious consequences for root-

156 D.W. Onstad et al.

worm management as well as for environmental protection. In all of these models, we assume that these are autosomal, single-locus, two-allele genetic systems. As we describe the models, we shall demonstrate how adult behaviours are a key component of evolution and geographical distribution in an ecological system.

Models of the Spread of the Rotation-resistant Western Corn Rootworm

In areas of intensive maize–soybean rotation, rootworm larvae from eggs that are oviposited into and overwinter in soybean emerge in maize the following spring. Alternatively, eggs laid in maize fields hatch the following year in a non-host field (i.e. soybean). Given that WCR larvae do not feed on soybean and eggs do not exhibit an extended diapause (Levine and Oloumi-Sadeghi, 1996; Levine et al., 2002), there is selection pressure in a landscape predominantly rotated between maize and soybean that favours WCR that lay eggs in soybeans. Both issues are important to managing this pest. We created a set of simple meteorological and behavioural models to predict the spread of the beetle infesting soybeans throughout the north- central USA. Data collected from 1987 to 2001 in Illinois, Indiana, Michigan and Ohio identified geographical areas where WCR in soybean fields exceeded a detection threshold of 20 beetles per 100 sweeps and two beetles per yellow sticky trap per day. Counts above a detection threshold represent populations that lack fidelity to maize and are adapted to circumvent maize–soybean rotation. Maps of these observations were used for evaluation of the model. Figure 8.1 presents the observed infestations of soybean by WCR over time in Illinois, Indiana, Michigan and Ohio. The counties are shaded according to the year during which WCR adults were first observed in excess of the detection threshold in soybean (20 beetles per 100 sweeps or two beetles per trap per day). Since 1986, the rootworm has expanded its range in Illinois to the west, north and south. The southern, eastern and northern fronts in Indiana, Ohio and Michigan, respectively, did not change much after 1997. Analysis of the counties exceeding the higher detection threshold indicated that the rate of spread from 1986 to 1997 was approximately 27 km/year to the east and 8.5 km/year to the west. From 1998 to 2001, the rate of spread slowed to approximately 16 km/year to the east and 7.75 km/year to the west. This indicates that some factor has limited the spread of the rootworm in these directions. The models are based on wind speed and direction, the direction of storms with rainfall over 2.54 cm, knowledge of beetle flight speeds and the probability of a beetle flying long distances (Onstad et al., 1999, 2003a). We assumed that beetles can be carried a maximum of 33 km by storm fronts (Onstad et al., 1999, 2003a). Onstad et al. (1999) described the first dozen years of the geographical spread of rotation-resistant WCR. chap08.qxd 15/11/04 2:51 PM Page 157

Heterogeneous Landscapes and Variable Behaviour 157

1986 1996–1997

1992–1993 1998–1999

1994–1995 2000–2001

Fig. 8.1. Counties in the north-central USA with soybeans infested by western corn rootworm at 20 beetles per 100 sweeps or 2.0 beetles per trap per day with initial year of observation indicated by colour and shading.

Model results supported the hypothesis that the population of WCR infesting soybean originated in Ford County, Illinois. The predictions of the simple model ﬁtted an independent set of observations well on three of four fronts or directions up to 1997. Some of the newer models invoked a landscape-diversity function that included the proportion of non-maize, non-rotated soybean vegetation on farmland in each county (Onstad et al., 2003a). An example of how the values of extra vegetation were used in the model is shown in Fig. 8.2. East-central Illinois and western Indiana have the lowest levels in the region. The proportions increase in Ohio, Michigan and north-eastern and southern Indiana. We assume that landscape diversity increases as the proportion of this extra vegetation increases. The best model for the period 1997 to 2001 reduces the distance that beetles spread each year by the proportion of extra vegetation in a county. This version is superior to the model published in 1999 and to two new chap08.qxd 15/11/04 2:51 PM Page 158

158 D.W. Onstad et al.

< 30 50–70

30–50 > 70

Fig. 8.2. Percentage of extra vegetation on farmland in each county in the north-central USA.

models that do not consider landscape diversity but reduced the spread each year by a constant factor (0.85 or 0.80). Most of the models predicted spread at too high a rate between 1997 and 2001 compared to observations, but a few new models with rates of spread reduced by a landscape- diversity function matched the observations relatively well. Figure 8.3 compares the results of two models and the model published in 1999 with the observations. The dark contour lines represent the 12th (inner line for 1997) and 16th (outer line for 2001) years of the model simulations. A model that reduced the distance that beetles spread each year by the mean extra vegetation (MEV) per county performed relatively well by 2001 but failed to predict the earlier infestations in Ohio (Fig. 8.3a). The use of a threshold level of extra vegetation (30%), where there is no reduction in the distance a beetle can travel from counties with levels of extra vegetation below 30%, improved the 1997 predictions, especially in Ohio, but by 2001 the predicted wave front extended too far south-east, south-west and north into Wisconsin (Fig. 8.3b). All three of chap08.qxd 15/11/04 2:51 PM Page 159

Heterogeneous Landscapes and Variable Behaviour 159

1986–1997

1998–2001

Fig. 8.3. Comparison of model results to observations (20 beetles per 100 sweeps or 2.0 beetles per trap per day) with the dark contours representing the 12th (1997) and 16th (2001) years of the simulations. Multipliers used in the model are (a) 1-MEV, (b) 1-MEV for EV > 0.3, (c) no multiplier (Onstad et al., 1999).

these models outperformed the previously published 1999 model, which overpredicted the spread on the northern, western and eastern fronts by 2001 (Fig. 8.3c). The results of these models suggest that landscape diversity affects the spread of rotation-resistant individuals primarily to the north and east, while wind is a limiting factor to dispersal on the western and southern fronts. Results suggest that the conclusions based on a linear model using proportion of extra vegetation as the key parameter are likely to be robust. Thus, we hypothesize that, as the landscape diversity represented by the proportion of non-maize and non-rotated soybean vegetation in a geographical region increases, the rate of regional spread of the rotation- resistant WCR decreases over several years.

Models of Rootworm Evolution to Crop Rotation

In addition to understanding the geographical spread of rotation-resistant rootworms, we were also interested in understanding how landscape chap08.qxd 15/11/04 2:51 PM Page 160

160 D.W. Onstad et al.

management had contributed to the evolution of this phenotype. We created a simple model of adult behaviour, population genetics and genetic expression of behaviour in a landscape with different levels of maize–soybean rotation (Onstad et al., 2001b). The landscape consists of four plant patches: maize grown every year (continuous maize fields), a maize patch that follows a non-maize patch in a rotation (first-year maize fields), and two non-maize patches, one preceding maize in a rotation (soybean) and the other never being rotated to maize (extra vegetation). Normal monophagous individuals emerge from the natal maize patch and distribute themselves (and their eggs) across the two maize patches according to the relative area of each maize patch. Polyphagous (rotation- resistant) individuals distribute themselves across all patches according to the proportional area of each patch in the whole region. When expression of polyphagy is additive, heterozygotes move to all crop types but are biased towards maize. The original model describing the evolution of behavioural resistance (Onstad et al., 2001b) used a fitness cost for feeding in soybean. Recent data of Mabry (2002) demonstrate that there is no fitness cost. In an improved model (Onstad et al., 2003b), the simple fitness parameter was replaced with population dynamics from the model of rootworm adaptation to transgenic maize (Onstad et al. 2001a) described later in this chapter. Ignoring gender, fecundity was set to 220 eggs per female and overwintering mortality at the egg stage was set to 50% (Godfrey et al., 1995). We assume density-dependent mortality occurs as competition for feeding sites among neonate larvae and is applied following overwintering mortality and any mortality due to toxins. The rotation level, R, is the sum of the proportional areas of rotated maize and soybean, which are always equal in the model. The baseline simulation involves a 2-year rotation with the landscape defined as R = 0.85, extra vegetation is 0.05 and the proportion of continuous maize is 0.10. The initial frequency of the allele for polyphagy and resistance is 0.0001. Figure 8.4 shows how the resistance allele frequency changes over 20 years. The allele for rotation resistance reached 3% in years 13, 3 and 3 and 50% in years 14, 6 and 5 when the resistance allele is recessive, additive or dominant, respectively. A dominant allele for resistance permits the frequency of the resistance allele to increase the fastest, but, after several years, the resistance allele frequency is actually greatest when expression of rotation resistance is additive or recessive. Figure 8.5 shows how the rotation level influences the resistance allele frequency after 15 years. As R increases and the proportion of extra vegetation decreases, the allele frequency for rotation resistance increases as expected. The allele frequency for rotation resistance declines as the proportion of continuous maize or extra vegetation increases. The model suggests that evolution of polyphagy and behavioural resistance may have resulted from selection on a single gene that could only develop where grower practices involved high levels of rotation (R > 80% of landscape). In less diverse landscapes, crop rotation selects for the expansion chap08.qxd 15/11/04 2:51 PM Page 161

Heterogeneous Landscapes and Variable Behaviour 161

1.0

0.9

0.8 x = y

0.7

0.6 Y > x 0.5

0.4 X > y

Y allele frequency 0.3

0.2

0.1

0.0 02468101214161820 Years

Fig. 8.4. Resistance allele frequencies over 20 years with a 2-year rotation (R = 0.85) and an initial resistance allele frequency of 0.0001 when the allele for rotation resistance is recessive (X > y), additive (x = y), or dominant (Y > x).

of host preferences and reduction in fidelity to maize by adults. Diverse landscapes may delay resistance to crop rotation depending on the fitness costs and the nature of the genetic system. Onstad et al. (2003b) used the improved model to indicate that only three management scenarios prevent or delay the development of WCR resistance to crop rotation compared to a 2-year rotation scenario (Fig. 8.6). (We did not consider soil insecticide use in first-year maize.) One scenario maintains the 2-year rotation but replaces the current soybean varieties with cultivars that repel 90% or more of the WCR adults before egg laying occurs. The second is a 3-year rotation with regular soybean and wheat, which is unattractive for oviposition. In this scenario, the proportion of land planted to soybeans, wheat and rotated maize is 0.30 while the proportion of land planted to continuous maize is 0.10. Wheat that repels or never attracts 90% of WCR adults is always planted before maize. The third scenario involves transgenic maize planted in a 2-year rotation with soybean. Transgenic rotated maize kills 90% of WCR larvae hatching in rotated maize. We did not model the use of transgenic maize in continuous maize fields. When the allele for rotation resistance is recessive, these strategies kept the resistance allele frequency fixed at 0.0001 (Fig. 8.6). The use of repellent soybeans was able to slow the evolution of resistance significantly when the expression of rotation resistance was additive, but not as well when the allele for resistance was dominant (Fig. 8.6). The use of chap08.qxd 15/11/04 2:51 PM Page 162

162 D.W. Onstad et al.

1.0

0.9

0.8 x = y

0.7

0.6 Y > x 0.5

0.4 X > y Y allele frequency 0.3

0.2

0.1

0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Rotation level

Fig. 8.5. Resistance allele frequency in year 15 as a function of the level of rotation, keeping the proportion of land planted to continuous maize ﬁxed at 0.05, when the allele for rotation resistance is recessive (X > y), additive (x = y) or dominant (Y > x).

transgenic rotated maize or a 3-year rotation with unattractive wheat were the most effective management strategies for slowing the evolution of resistance. Each of these strategies kept the resistance allele frequency at low values (≤ 0.02) over 15 years with each type of gene expression (Fig. 8.6). Only the repellent soybean and transgenic maize were economically valuable as well and, of these two technologies, only transgenic maize is close to commercialization. This model did not consider resistance to transgenic maize during the 15-year time horizon.

Models of Rootworm and Transgenic Maize

Onstad et al. (2001a) created a deterministic simulation model of the population dynamics and genetics of the WCR, for a landscape of maize, soybean and other crops. The region, 100 ha of cropland, consists of one or two fields containing only maize or as many as four crops with a maximum of six fields, four of which can be maize. The four crops are continuous maize, rotated maize, soybean and an extra non-maize crop (e.g. oats or lucerne, not Cucurbitaceae). All maize fields are phenologically the same. Transgenic maize in the landscape can be configured in blocks (six fields) or as row strips (four fields). Block configurations assume the transgenic maize is planted to the same field in successive years. The model was validated by comparing model output for the distribution of females in first-year maize fields and the distribution of ovipo- chap08.qxd 15/11/04 2:51 PM Page 163

Heterogeneous Landscapes and Variable Behaviour 163

2-year Rsoy Trans 3-year UE

1.0

0.9

0.8

0.7

0.6 0.5

0.4

Y allele frequency 0.3

0.2

0.1

0.0 X > y Y > x x = y

Fig. 8.6. Resistance allele frequency in year 15 for four management strategies: 2-year rotation (2-year), repellent soybeans (Rsoy), transgenic-rotated maize (Trans) and a 3-year rotation with unattractive wheat (3-year UE), with the allele for rotation resistance as recessive (X > y), additive (x = y) or dominant (Y > x).

sition in non-transgenic maize fields to published literature. Dispersal rates produced adult female WCR populations in first-year maize fields comparable to observations of Godfrey and Turpin (1983). The period of oviposition in the model matched observations by Hein and Tollefson (1985). Expression of the allele for resistance to the transgenic toxin and toxin dose in the maize plant were the two most important factors affecting resistance development. A dominant resistance allele allowed rapid evolution of resistance to transgenic maize, whereas a recessive resistance allele delayed resistance more than 99 years. With high dosages of toxin and additive expression of resistance, the time required to reach 3% resistance allele frequency ranged from 13 to more than 99 years. Table 8.1 shows how refuge size, configuration and dose of toxin affect the length of time it takes for the resistant allele frequency to reach 3% when gene expression is additive. Without a refuge in continuous transgenic maize, the length of time to reach 3% resistant allele frequency increases as toxicity decreases. In the block configuration, with toxin doses that allowed 0.001 or fewer susceptibles to survive, the resistant allele frequency did not exceed 3% in 99 years. The number of years necessary to reach 3% resistant allele frequency ranged from 5 at 5% refuge with 5% of susceptibles surviving, to 9 for 30% refuge and 20% of susceptibles surviving in transgenic maize (Table 8.1). chap08.qxd 15/11/04 2:51 PM Page 164

164 D.W. Onstad et al.

Table 8.1. Year in which the allele frequency for resistance to the transgenic toxin reached 0.03 in a region of continuous maize (no rotated maize), additive expression of resistance and an initial allele frequency of 0.0001. Survival of susceptibles is a function of the dose of toxin.

Survival of susceptibles Proportion in refuge 0.0 0.001 0.05 0.2

No refuge 0.0 1336 Refuge as adjacent block 0.3 > 99 > 99 7 9 0.2 > 99 > 99 6 8 0.05 > 99 > 99 5 6 Refuge as row strips 0.3 > 99 34 4 6 0.2 > 99 29 3 5 0.05 52 13 2 4

A row-strip configuration resulted in much shorter times to reach 3% resistant allele frequency compared with the block configuration (Table 8.1). If the concentration of toxin in the crop corresponds to a survival rate for susceptible larvae of 0.001, the allele frequency for resistance to transgenic maize never reached 3% resistant allele frequency with the block refuge configuration. However, with a row-strip configuration, the time to reach 3% resistant allele frequency ranged from 13 years at 5% refuge to 34 years at 30% refuge. Submodels of the probability of females finding a mate (Kuno, 1978) were incorporated to address the possibility of genetic isolation in block refuge configurations. WCR are protandrous, with males emerging about 1 week prior to the first females. The sex ratio of beetles observed in first- year maize and soybeans suggested that the male dispersal rate was one- quarter the dispersal rate of adult females. Given the additional week, even the lowest levels of male dispersal were enough to mate all teneral females in all maize fields. Consequently, the dispersal rate for adult beetles did not significantly affect the time to 3% allele frequency with a block configuration. When sublethal effects of transgenic maize cause susceptible phenotypes to emerge later than normal in a block configuration, resistance can develop much faster when expression of resistance is recessive. When resistance to transgenic maize is recessive and only homozygous susceptibles are delayed 6 days, the time to 3% allele frequency was shortened in the two lowest toxin doses. For 10% survival of susceptibles, the time to 3% allele frequency ranged from 51 to 95 years, respectively, for 5 to 30% refuge. For 20% survival of susceptibles, the time to 3% allele frequency ranged from 22 to 46 years, respectively. These time periods should be compared to the standard 99 plus years. With recessive expres- chap08.qxd 15/11/04 2:51 PM Page 165

Heterogeneous Landscapes and Variable Behaviour 165

sion where both susceptible phenotypes are delayed 6 days, the allele frequency did not change in the 99 years of using transgenic maize. When the homozygous susceptibles were delayed by 3 to 9 days with additive allele expression, resistance developed more quickly relative to the standard, but the number of years required to reach 3% allele frequency did not change by more than 1 year (and then only for the lowest dose). Storer (2003) created a stochastic, spatially explicit computer model that simulates the adaptation by WCR to transgenic maize. The model reflects the ecology of the rootworm in much of the Corn Belt of the USA. It includes functions for crop development, egg and larval mortality, adult emergence, mating, egg laying, mortality and dispersal to simulate the population dynamics of WCR and compares alternative methods of rootworm control. The allele for resistance to transgenic maize varies from incompletely recessive to incompletely dominant, depending on the efficacy of the toxin in the crop. Validation was achieved by comparing populations from the model with field data on population dynamics, and with field data documenting WCR adaptation to cyclodienes and organophosphates. The model was used to compare the rate at which the resistance allele spread through the population under different refuge deployment scenarios, and with crops of different efficacy. For a given refuge size, the model indicated that placing the refuge in a block within a transgenic maize field would be likely to delay WCR resistance longer than planting the refuge in separate fields in varying locations. If a portion of the refuge were planted in the same fields or the same in-field blocks each year, WCR adaptation would be substantially delayed. Storer (2003) conducted a brief analysis of the need for insecticide use in refuges because results suggested that resistance to transgenic maize would be unaffected by soil insecticide treatment in the refuge. In this analysis, refuge insecticide treatments were warranted for the first few years of transgenic deployment until the regionwide population was reduced. The smaller the proportion of fields planted to non-transgenic maize, the smaller the proportion of them that require treatment.

Discussion

WCR invaded east-central Illinois between 1968 and 1970 (Metcalf, 1983), about 16 years before the ﬁrst observation of resistance to crop rotation in Ford County, Illinois (Onstad et al., 1999, 2003a). Since the late 1960s, continuous maize has varied from 2 to 9% in the region of Ford County and the adjacent Champaign County (Onstad et al., 2001a). According to our models, the high levels of crop rotation imposed intense selection pressure, which promoted the evolution of behavioural changes and resistance to crop rotation. Our models also suggest that the evolution of resistance is delayed in areas where crop rotation is not as extensively employed for control of WCR. chap08.qxd 15/11/04 2:51 PM Page 166

166 D.W. Onstad et al.

In areas where the allele frequency for resistance to crop rotation is already high, either from local evolution or from invasion, it is difﬁcult, if not practically impossible, to reverse or halt further evolution of resistance. However, we hypothesize that the rotation-resistant WCR cannot persist over the long term in small areas with high landscape diversity. Uncertainty about the timing of invasion by the rotation-resistant variant and initial gene frequency makes it difﬁcult to choose among management strategies. The costs of resistance are not limited to reductions in farmer returns. If the use of soil insecticides increases as a result of rotation resistance, there may be broader social costs. From this perspective, it may be desirable to subsidize farmers in the present so that they have an incentive to change practices now and delay the development of resistance in the future. The history of the WCR provides several examples of its ability to evolve around our attempts to manage population densities or damage. Future management strategies should be used cautiously. Transgenic maize, active against corn rootworms, may provide the only economical and environmentally sound control for rotation-resistant rootworms. Given the adaptability of this pest with regard to crop rotation and previous insecticides, further research must be done into alternatives or addi- tions to the refuge strategy to assess the risk of evolution of resistance to transgenic maize. Future management strategies should include biological control, host plant resistance and other feasible tactics and offset these selection pressures with regional landscape heterogeneity. Heterogeneous landscapes are the stage on which these stories of evolution are played out. Natural variation in adult behaviour is acted upon by natural selection caused by society’s choices of land use and crop selection. Any pest management strategy that is very effective over the short term will cause strong selection pressure on a pest. Thus, even crop rotation and host plant resistance can be misused over evolutionary time. Individual farmers and society must recognize that simple solutions that ignore landscapes, insect behaviour, ecology and evolution will always lead to complications.

References

Ball, H.J. and Weekman, G.T. (1962) Insecticide resistance in the adult western corn rootworm in Nebraska. Journal of Economic Entomology 55, 439–441. Ball, H.J. and Weekman, G.T. (1963) Differential resistance of corn rootworms to insecticides in Nebraska and adjoining states. Journal of Economic Entomology 56, 553– 555. Godfrey, L.D. and Turpin, F.T. (1983) Comparison of western corn rootworm (Coleoptera: Chrysomelidae) adult populations and economic thresholds in ﬁrst-year and continuous corn ﬁelds. Journal of Economic Entomology 76, 1028–1032. Godfrey, L.D., Meinke, L.J., Wright, R.J. and Hein, G.L. (1995) Environmental and edaphic effects on western corn rootworm (Coleoptera: Chrysomelidae) overwintering egg survival. Journal of Economic Entomology 88, 1445–1454. chap08.qxd 15/11/04 2:51 PM Page 167

Heterogeneous Landscapes and Variable Behaviour 167

Hein, G.L. and Tollefson, J.J. (1985) Seasonal oviposition of northern and western corn rootworms (Coleoptera: Chrysomelidae) in continuous cornfields. Journal of Economic Entomology 78, 1238–1241. Kuno, E. (1978) Simple mathematical models to describe the role of mating in insect populations. Researches on Population Ecology 20, 50–60. Levine, E. and Oloumi-Sadeghi, H. (1991) Management of diabroticite rootworms in corn. Annual Review of Entomology 36, 229–255. Levine, E. and Oloumi-Sadeghi, H. (1996) Western corn rootworm (Coleoptera: Chrysomelidae) larval injury to corn grown for seed production following soybeans grown for seed production. Journal of Economic Entomology 89, 1010–1016. Levine, E., Spencer, J.L., Isard, S.A., Onstad, D.W. and Gray, M.E. (2002) Adaptation of the western corn rootworm to crop rotation: evolution of a new strain in response to a management practice. American Entomologist 48, 94–107. Mabry, T.R. (2002) The effects of soybean herbivory on the behavior and ecology of the western corn rootworm (Diabrotica virgifera virgifera LeConte) variant. MS thesis, University of Illinois, Urbana, Illinois. Meinke, L.J., Siegfried, B.D., Wright, R.J. and Chandler, L.D. (1998) Adult susceptibility of Nebraska western corn rootworm (Coleoptera: Chrysomelidae) populations to selected insecticides. Journal of Economic Entomology 91, 594–600. Metcalf, R.L. (1983) Implications and prognosis of resistance to insecticides. In: Georghiou, G.P. and Saito, T. (eds) Pest Resistance to Pesticides. Plenum, New York, pp. 703–733. Miota, F., Scharf, M.E., Ono, M., Marcon, P., Meinke, L.J., Wright, R.J., Chandler, L.D. and Siegfried, B.D. (1998) Mechanisms of methyl and ethyl parathion resistance in the western corn rootworm (Coleoptera: Chrysomelidae). Pesticide Biochemistry and Physiology 61, 39–52. Onstad, D.W., Joselyn, M.G., Isard, S.A., Levine, E., Spencer, J.L., Bledsoe, L.W., Edwards, C.R., DiFonzo, C.D. and Willson, H. (1999) Modeling the spread of western corn rootworm (Coleoptera: Chrysomelidae) populations adapting to soybean–corn rotation. Environmental Entomology 28, 188–194. Onstad, D.W., Guse, C.A., Spencer, J.L., Levine, E. and Gray, M.E. (2001a) Modeling the dynamics of adaptation to transgenic corn by western corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 94, 529–540. Onstad, D.W., Spencer, J.L., Guse, C.A., Isard, S.A. and Levine, E. (2001b) Modeling evolution of behavioral resistance by an insect to crop rotation. Entomologia Experimentalis et Applicata 100, 195–201. Onstad, D.W., Crowder, D.W., Isard, S.A., Levine, E., Spencer, J.L., O’Neal, M., Ratcliffe, S., Gray, M.E., Bledsoe, L.W., DiFonzo, C.D., Eisley, B. and Edwards, C.R. (2003a) Does landscape diversity slow the spread of rotation-resistant western corn rootworm (Coleoptera: Chrysomelidae)? Environmental Entomology 32, 992–1001. Onstad, D.W., Crowder, D.W., Mitchell, P.D., Guse, C.A., Spencer, J.L., Levine, E. and Gray, M.E. (2003b) Economics versus alleles: balancing IPM and IRM for rotation- resistant western corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 96, 1872–1885. Scharf, M.E., Meinke, L.J., Siegfried, B.D., Wright, R.J. and Chandler, L.D. (1999) Carbaryl susceptibility, diagnostic concentration determination, and synergism for US populations of western corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 92, 33–39. Storer, N.P. (2003) A spatially explicit model simulating western corn rootworm (Coleoptera: Chrysomelidae) adaptation to insect-resistant maize. Journal of Economic Entomology 96, 1530–1547. chap08.qxd 15/11/04 2:51 PM Page 168 chap09.qxd 12/11/04 1:14 PM Page 169

9 Sampling Devices and Decision Rule Development for Western Corn Rootworm (Diabrotica virgifera virgifera LeConte) Adults in Soybean to Predict Subsequent Damage to Maize in Indiana

Corey K. Gerber,1 C. Richard Edwards,1 Larry W. Bledsoe,1 John L. Obermeyer,1 Gyorgy Barna2 and Ricky E. Foster1 1Department of Entomology, Purdue University, W. Lafayette, Indiana, USA; 2Syngenta Seeds Kft., Alkotas, Hungary

Introduction

In the Midwest, western corn rootworm (WCR), Diabrotica virgifera virgifera LeConte, adult emergence begins in late June to early July, with adult peak activity occurring in the first 3 weeks of August (Godfrey and Turpin, 1983; Levine and Gray, 1996), with oviposition primarily occurring in August and September (Hill, 1975). WCR eggs overwinter in the soil and hatch in May and early June (Levine and Oloumi-Sadeghi, 1991). Corn rootworm larvae are primarily subterranean and feed almost exclusively on maize, Zea mays L., root systems (Branson and Ortman, 1970). Feeding damage by WCR larvae can result in significant yield losses (Sutter et al., 1990; Gibb and Higgins, 1991; Spike and Tollefson, 1991). In the mid-1980s, corn rootworm control costs and crop losses in the USA were estimated at US$1 billion/year (Metcalf, 1986). Therefore, the WCR is the most destructive and costly maize pest in North America. Before the late 1980s, WCR larval injury was found infrequently in first-year maize (primarily maize planted after soybean) (Levine and Oloumi-Sadeghi, 1991; Steffey et al., 1992) because WCR females oviposited a majority of eggs in maize fields. During the late 1980s and throughout the 1990s, the frequency and severity of WCR damage in first- year maize increased in north-western Indiana and east-central Illinois.

170 C.K. Gerber et al.

Researchers suggested that a variant of WCR beetle, that prefers to oviposit in areas other than maize, had been selected through the widespread adoption of crop rotation (Edwards, 1996; Levine and Gray, 1996; Sammons et al., 1997). To avoid economic losses, producers applied soil insecticides to first-year maize fields as well as to continuous maize fields (maize planted after maize). Consequently, the amount of maize treated with soil insecticides increased from 10% to 65% in northern Indiana and from 13% to 75% in east-central Illinois between 1994 and 1996 (Edwards, 2000). Because this increase was not the result of sampling decisions, it can be assumed that some fields may have been treated unnecessarily. To reduce unnecessary insecticide applications, economic thresholds have been developed and implemented to trigger the control of pest species before economic injury levels are reached (Stern, 1973). The economic threshold, also known as the action threshold, is the number of pests at which treatment actions should occur to prevent the pest density from reaching the economic injury level. An economic injury level is the lowest number of pests that cause economic damage (the value of loss that equals the cost of avoiding the loss). Economic injury levels and/or economic thresholds have been established for corn rootworm eggs (Davis, 1994), larvae (Reed et al., 1991) and adults (Hein and Tollefson, 1985; Kuhar and Youngman, 1998) infesting continuous maize. To determine if economic injury levels or economic thresholds are met, sampling the targeted pest population is required. Tollefson (1975) demonstrated that sampling adult rootworms with sticky traps and counting adults on maize plants in continuous maize could predict rootworm larval damage the following growing season. Over the past three decades, a number of adult corn rootworm traps and techniques have been compared and tested as population density estimators in continuous maize, and as prediction tools for subsequent rootworm larval damage (Tollefson, 1975; Hein and Tollefson, 1984; Shaw et al., 1984; Levine and Gray, 1994; Kuhar and Youngman, 1998). Hein and Tollefson (1985) suggested that, in continuous maize fields, unbaited Pherocon® AM yellow sticky traps (Trécé Inc., Salinas, CA 93912) and the whole-plant count method were equally effective in predicting larval damage. Beginning in the mid-1990s, monitoring tools and techniques for corn rootworms were tested and adapted for soybean production systems (Barna et al., 1998; Edwards et al., 1998; Spencer et al., 1998; O’Neal et al., 1999). In Indiana, economic injury levels and economic thresholds have not been determined for WCR adults infesting soybean preceding maize. The objectives of this study were to develop a sampling programme for adult WCR in the maize/soybean rotational system and to establish an economic injury level and an economic threshold based on WCR adult density estimates in soybean for the potential damage created by WCR larvae in subsequent maize fields. chap09.qxd 12/11/04 1:14 PM Page 171

Sampling Devices and Decision Rule Development 171

Materials and Methods

WCR sampling devices

Study sites Six soybean fields in Benton County, Indiana, were used during 1997 and 1999, and six adjacent soybean fields were used in 1998 and 2000. These fields were selected based on the relatively high abundance of the variant WCR beetle (C.R. Edwards, L.W. Bledsoe, J.L. Obermeyer and R.L. Blackwell, unpublished data). Field sizes ranged from 12 to 42 ha. Tillage, planting date and other crop inputs were not controlled in this experiment.

Sampling methods In 1997 and 1998, four sampling devices were used to sample WCR adults: the Pherocon® AM unbaited yellow sticky trap, the corn rootworm (CRW) non-lure trap (Trécé Inc., Salinas, CA 93912), a cucurbitacin vial trap (Shaw et al., 1984) and a standard 38.1 cm diameter muslin sweep net. In each field, three transects were established running the length of the field, each approximately one-quarter of the field width apart. Three trap lines were established on these transects and each line consisted of one of the three trap types. Eight traps were placed equidistant in each line. Based on the length of each field, traps in each trapping line were placed approximately 50 to 100 m apart. It was assumed that traps were far enough apart for trap interference not to occur. At each Pherocon® AM trap and vial trap deployment site, 1.5 m wood lath stakes were driven approximately 0.3 m into the soil. Each trap was positioned at the top of the canopy, with adjustments each week to maintain height relative to the top of the soybean canopy. At each CRW trap deployment site, 1.8 m × 1.27 cm polyvinyl chloride (PVC) pipes were driven approximately 0.2 m into the soil. Traps were placed directly on the PVC piping located above the canopy. Trap adjustment relative to the soybean canopy was not possible. Eight samples of 30 sweeps each were obtained in one of the trap lines in each of the soybean fields. Sweeps were made with a pendulum motion in the upper quarter to third of the soybean canopy. In 1999 and 2000, sampling devices included only Pherocon® AM and CRW traps. Two transects were established running the length of each field, each approximately one-third of the field width apart. Each transect contained either eight Pherocon® AM traps or eight CRW traps. Trap placement on stakes and piping, as well as the distance between traps, was identical to the protocol in 1997 and 1998.

Trapping periods Traps were monitored over a 6–7-week period. Adult WCR peak activity in the eastern Midwest (Illinois, Indiana, Michigan, Ohio and Wisconsin) chap09.qxd 12/11/04 1:14 PM Page 172

172 C.K. Gerber et al.

occurs during the ﬁrst 3 weeks of August (Levine and Gray, 1996). In 1997 and 1998, trap deployment began on 22 and 14 July, respectively. Trap deployment in 1999 began on 20 July and in 2000 deployment began on 12 July. All traps were replaced every 7–8 days during the trapping period. The number of WCR beetles collected per week was recorded. The 6–7-week trapping period ended by 10 September in each of the years WCR beetles were sampled.

Data analysis For the data set of each sampling device for each year, a linear regression of log s2 on log x– was computed. Minimum sample sizes required for 20% and 25% levels of precision were determined for each sampling device using Taylor’s power law (Ruesink, 1980): n = ax– b – 2 c2 where n is the number of sampling devices, x– is the mean, a and b are the values of the y intercept and the slope of the line, respectively, based on the linear regression equations of log s2 on log x– for each sampling device, and c is the precision expressed as a fraction of the mean.

WCR economic injury levels and threshold

Study sites Sixteen pairs of adjacent first-year maize and soybean fields, in an annual maize/soybean rotation, in six north-west Indiana counties – Benton (seven pairs), Clinton (two pairs), Fountain (one pair), Jasper (two pairs), Newton (three pairs) and White (one pair) – were used in this study during 1998–2000. In 1998, WCR adult populations were sampled in each soybean field. Maize was planted into these fields in 1999, and WCR larval damage was determined on maize roots from each field. In that same year, sampling WCR adult populations was conducted in the paired soybean fields. In 2000, maize was planted into these soybean fields, and again WCR larval damage was determined on maize roots from each field. Field sizes ranged from 12 to 47 ha. Fields were selected based on the relatively high abundance of the variant WCR beetle (C.R. Edwards, L.W. Bledsoe, J.L. Obermeyer and R.L. Blackwell, unpublished data). The 16 pairs of maize and soybean fields were considered replicates. Tillage, planting date and other crop inputs were not controlled in this experiment.

Population sampling Adult WCR were sampled using the Pherocon® AM unbaited sticky traps in 32 soybean ﬁelds. Twelve of the 32 soybean ﬁelds were used in the pre- chap09.qxd 12/11/04 1:14 PM Page 173

Sampling Devices and Decision Rule Development 173

viously described WCR sampling study. In the remaining 20 soybean fields, a single transect was established running the length of each field, approximately one-half of the field width apart. A trap line consisted of six equidistantly placed Pherocon® AM traps. Based on the length of each field, traps in each trapping line were placed approximately 50 to 150 m apart. Establishment of traps at each deployment site was identical to the protocol in the WCR sampling study. Even though eight traps were used in 12 fields, and six traps in 20 fields, we believe the data were comparable and therefore they were pooled.

Trapping protocol Trap deployment began on 13 and 14 July, 1998, and 19 and 20 July, 1999. Traps were replaced every 7–8 days and the number of WCR beetles collected per week was recorded. The 6-week trapping period, during the 1998 and 1999 growing seasons, ended by 2 September.

Subsequent WCR maize root damage assessment To examine rootworm larval damage, roots were extracted from rows in which no soil insecticides were placed, approximately along the same transects where trap deployment occurred the preceding year in all fields. Maize roots were dug up between 9 July and 18 August in 1999 and 6 July and 28 July in 2000. Fifty roots were dug up equally spaced throughout each untreated row in the 20 fields where a single trap line had been established in soybean in the preceding year. Twenty-five roots were dug up in each untreated row in the 12 fields that contained the two trap lines in the preceding year. Each root was labelled, washed and rated using the Hills and Peters (1971) 1 to 6 root damage rating scale.

Data analysis Linear regression was used to determine the coefﬁcient of determination between mean number of beetles collected per trap, during the sampling dates in 1998 and 1999, and the mean root rating values obtained in 1999 and 2000 (Supernova, 1989).

Validation study of the proposed economic threshold

A follow-up study was conducted from 2000 to 2002 to validate the proposed economic threshold. Twelve and 11 soybean fields, located in north-west Indiana and east- central Illinois, were monitored in 2000 and 2001, respectively. Deployment of the Pherocon® AM traps in each field was similar to the protocols in the WCR sampling device and the WCR economic injury level and threshold studies. However, in 12 of the 23 soybean fields, the chap09.qxd 12/11/04 1:14 PM Page 174

174 C.K. Gerber et al.

trapping scheme was established as two rows of four traps, uniformly distributed throughout each field. Even though two trapping schemes were used, we believe the data were comparable and therefore they were pooled. In 2001 and 2002, WCR larval damage was assessed on maize roots extracted from rows located approximately along the same transects where trap deployment occurred the preceding year in all fields. Fifty roots were dug up equally spaced throughout each untreated row in the 11 fields where a single trap line had been established. Twenty roots were dug up in each untreated row in the 12 fields that contained the two trap lines. Damage on maize roots by WCR larvae was evaluated using the Hills and Peters (1971) root damage rating scale. Data obtained from this study (23 fields) and from the WCR economic injury level and threshold study (32 fields) were pooled and a contingency table was developed to determine the accuracy of predicting WCR larval damage in first-year maize, based on the economic threshold of WCR adults in soybean.

Results

WCR sampling devices

The minimum sample sizes required for the Pherocon® AM trap, the CRW trap, the cucurbitacin vial trap and the sweep net sampling device, at two levels of precision, were calculated in 1997 and 1998 for densities ranging from one to 25 WCR beetles per trap per day. These densities produced ranges of numbers of traps per field for each trap type tested (Tables 9.1, 9.2, 9.3 and 9.4). However, based on the economic threshold presented later in this chapter (five WCR beetles per trap per day), minimum sample sizes are only discussed for the mean WCR beetle density of five beetles per trap per day. During the 1997 monitoring period, at the 20% and 25% levels of precision, the minimum sample sizes for the Pherocon® AM (y = 1.555x – 0.283; r2 = 0.635; P = 0.0001) and CRW traps (y = 1.581x – 0.065; r2 = 0.720; P = 0.0001) were determined as seven and five traps and 11 and eight traps per field, respectively. At these same levels of precision, the sweep net sampling method (y = 0.966x + 0.387; r2 = 0.376; P = 0.0001) resulted in minimum sample sizes of 12 and eight sweep sets per field. The cucurbitacin vial traps (y = 1.500x + 0.288; r2 = 0.629; P = 0.0001) required a higher number of traps to adequately sample the adult WCR population. In 1998, the Pherocon® AM (y = 1.768x – 0.606; r2 = 0.702; P = 0.0001) and CRW traps (y = 1.507x – 0.240; r2 = 0.658; P = 0.0001) required the least number of traps per field at each precision level at the mean WCR beetle density of five beetles per trap per day. The sample size of Pherocon® AM traps at the 20% precision level was five traps per field chap09.qxd 12/11/04 1:14 PM Page 175

Sampling Devices and Decision Rule Development 175

Table 9.1. Number of Pherocon® AM unbaited yellow sticky traps necessary to estimate densities of WCR beetles in soybean ﬁelds at various mean densities (x–) (WCR/trap/day) and two levels of precision (c).

No. of samples 1997 1998 (x–) c = 0.20 c = 0.25 c = 0.20 c = 0.25

1149 74 57553 10 5 3 4 3 25 4 2 3 2

1999 2000 c = 0.20 c = 0.25 c = 0.20 c = 0.25

1211474 59643 10 6 4 4 2 25 4 3 3 2

and at the 25% precision level, the sample size was three traps per field. The minimum sample sizes for the CRW traps, at these same levels of precision, were determined as seven and five traps per field. The number of sweep samples (y = 1.209x + 0.074; r2 = 0.891; P = 0.0001) required per field, at the 20% and 25% levels of precision, were calculated to be nine

Table 9.2. Number of corn rootworm (CRW) non-lure traps necessary to estimate densities of WCR beetles in soybean ﬁelds at various mean densities (x–) (WCR/trap/day) and two levels of precision (c).

No. of samples 1997 1998 (x–) c = 0.20 c = 0.25 c = 0.20 c = 0.25

122141510 5118 75 10 9 6 5 3 25 6 4 3 2

1999 2000 c = 0.20 c = 0.25 c = 0.20 c = 0.25

14932117 5149 74 10 8 6 5 4 25 4 3 4 3 chap09.qxd 12/11/04 1:14 PM Page 176

176 C.K. Gerber et al.

Table 9.3. Number of cucurbitacin vial traps necessary to estimate densities of WCR beetles in soybean ﬁelds at various mean densities (x–) (WCR/trap/day) and two levels of precision (c).

No. of samples 1997 1998 (x–) c = 0.20 c = 0.25 c = 0.20 c = 0.25

149322919 522141611 10 16 10 13 8 25 10 7 9 6

and six, respectively. As noted in 1997, cucurbitacin vial traps (y = 1.624x + 0.063; r2 = 0.880; P = 0.0001) required a higher number of traps to adequately sample WCR adults. Based on the smallest sample size required to determine adult WCR populations, only the Pherocon® AM and CRW traps were further tested in 1999 and 2000 (Tables 9.1 and 9.2). During the 1999 sampling period, the minimum sample size for the Pherocon® AM trap (y = 1.431x – 0.085; r2 = 0.572; P = 0.0001) at the 20% precision level was nine traps per field. Six traps per field were calculated at the 25% precision level. The sample sizes for the CRW trap (y = 1.207x + 0.289; r2 = 0.494; P = 0.0001) at the 20% and 25% levels of precision were 14 and nine traps per field, respectively. In 2000, the minimum sample sizes for the Pherocon® AM (y = 1.704x – 0.611; r2 = 0.830; P = 0.0001) and CRW traps (y = 1.689x – 0.394; r2 = 0.827; P = 0.0001) recorded the lowest values in the 4-year study. The minimum sample sizes for the Pherocon® AM and CRW traps at the 20% and 25% levels of precision were calculated to be four, three, seven and four traps per field, respectively. Due to the variability of the sample sizes by year, for each of the mean beetle densities used, the Pherocon® AM trap data were pooled

Table 9.4. Number of sweep net samples necessary to estimate densities of WCR beetles in soybean ﬁelds at various mean densities (x–) (WCR/trap/day) and two levels of precision (c).

No. of samples 1997 1998 (x–) c = 0.20 c = 0.25 c = 0.20 c = 0.25

161403020 5128 96 10 6 4 5 4 25 3 2 3 2 chap09.qxd 12/11/04 1:14 PM Page 177

Sampling Devices and Decision Rule Development 177

A 5 y = 1.624x – 0.423 r 2 = 0.805 P = 0.0001 4

3 2 s Log 2

0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Log 0

5 y = 1.679x – 0.403 B r 2 = 0.809 P = 0.0001 4

3 2 s Log 2

0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Log 0

Fig. 9.1. Regression of log s2 on log x– for WCR beetles collected by Pherocon® AM traps (A) and CRW traps/week (B) in 1997, 1998, 1999 and 2000.

(y = 1.624x – 0.423; r2 = 0.805; P = 0.0001) for the 4 sampling years (Fig. 9.1), and the minimum sample sizes were calculated using Taylor’s power law (Table 9.5). The CRW trap data were also pooled (y = 1.679x – 0.403; r2 = 0.809; P = 0.0001) for the 4 sampling years (Fig. 9.1) and the minimum sample sizes also calculated for the 4 sampling years (Table 9.5). The minimum sample sizes for the Pherocon® AM trapping method chap09.qxd 12/11/04 1:14 PM Page 178

178 C.K. Gerber et al.

Table 9.5. Number of Pherocon® AM unbaited yellow sticky traps and corn rootworm (CRW) non-lure traps necessary to estimate densities of WCR beetles in soybean ﬁelds at various mean densities (x–) (WCR/trap/day) and two levels of precision (c) (1997 to 2000).

No. of samples Pherocon® AM traps CRW traps (x–) c = 0.20 c = 0.25 c = 0.20 c = 0.25

1107107 56464 10 4 3 5 4 25 3 2 4 3

at the 20% and 25% levels of precision were determined to be six and four traps per ﬁeld. The CRW trapping method also revealed minimum sample sizes of six and four traps per ﬁeld at the 20% and 25% levels of precision.

WCR economic injury levels and threshold

The regression analyses showed that there were no correlations between CRW trap catches in 1998 and 1999 and root damage ratings in 1999 and 2000 (y = 0.006x + 2.315; r2 = 0.067; P = 0.6207 (1998 to 1999) and y = 0.012x + 2.809; r2 = 0.008; P = 0.8658 (1999 to 2000) (Fig. 9.2)). Similar regressions showed that Pherocon® AM trap catches were positively correlated with root damage ratings during the same time period (y = 0.026x + 1.457; r2 = 0.579; P = 0.0006 (1998 to 1999) and y = 0.048x + 1.562; r2 = 0.360; P = 0.0139 (1999 to 2000) (Fig. 9.3), respectively). Pooling the Pherocon® AM trap data for 2 years revealed an r2 value of 0.46 (y = 0.045x + 1.307; P = 0.0001 (Fig. 9.4)), and, by solving the linear regression equation for a root damage rating of 3.50, an economic injury level of 6.96 beetles per trap per day was calculated.

Validation study of the proposed economic threshold

Five WCR beetles per Pherocon® AM trap per day was hypothesized as the economic threshold once the economic injury level was established. The contingency table (Table 9.6) revealed that, of the 55 fields used in this study, no fields below the economic threshold reached an overall root damage rating of 3.50. A Type 2 error was not detected. Fifteen fields at or above this threshold level reached a root damage rating of 3.50. On the other hand, ten fields were below the beetle threshold and did not exceed the 3.50 rating. The final 30 fields resulted in beetle numbers over the eco- chap09.qxd 12/11/04 1:14 PM Page 179

Sampling Devices and Decision Rule Development 179

6 y = 0.006x + 2.315 A r 2 = 0.067 P = 0.621 5

2 Root ratings (untreated) 1999

0 0 20 40 60 80 100 WCR beetles/trap/week 1998 6 y = 0.012x + 2.809 B r 2 = 0.008 P = 0.866 5

2 Root ratings (untreated) 2000

0 0 20 40 60 80 100 WCR beetles/trap/week 1999 Fig. 9.2. Regression analysis of mean western corn rootworm population estimates/trap/ week from six north-west Indiana soybean ﬁelds using CRW traps in 1998 versus root ratings taken in these same ﬁelds during 1999 (A), and using CRW traps in 1999 versus root ratings taken in 2000 (B).

nomic threshold, but the root damage ratings were below 3.50. Therefore, Type 1 errors were detected. Similar results were also noted when four and six WCR beetles per trap per day were tested (Table 9.6). However, when six WCR beetles per trap per day were tested, a Type 2 error was detected. Therefore, the economic threshold for WCR beetles in soybean is determined as ﬁve WCR beetles per Pherocon® AM trap per day. chap09.qxd 12/11/04 1:14 PM Page 180

180 C.K. Gerber et al.

Table 9.6. Contingency tables comparing a threshold of 4, 5 and 6 WCR beetles/trap/day in soybean with subsequent maize root damage at a rating level of 3.50.

Beetles/trap/day Maize root damage rating < 4.00 > 4.00

< 3.50 9 31 ≥ 3.50 0 15

< 5.00 > 5.00

< 3.50 10 30 ≥ 3 50 0 15

< 6.00 > 6.00

< 3.50 12 28 ≥ 3 50 1 14

Discussion

WCR sampling devices

Cucurbitacin vial traps have been documented as a reliable sampling tool for adult WCR populations in maize, by determining larval damage in maize the following year (Levine and Gray, 1994). Of the four sampling devices evaluated, the cucurbitacin vial trapping method required the highest number of traps to adequately sample WCR beetle populations in soybean. Based on these numbers, it can be assumed that producers will not select and use a trapping device that demands trap numbers of this magnitude in each field. The sweep net sampling method has been utilized as a monitoring tool to determine pest populations in field crops. Examples include potato leafhoppers, Empoasca fabae (Harris), in lucerne (Fleischer and Allen, 1982), tarnished plant bugs, Lygus lineolaris (Palisot de Beauvois), in cotton (Snodgrass, 1998), western spotted cucumber beetles, Diabrotica undecimpunctata undecimpunctata Mannerheim, in snap beans (Weinzierl et al., 1987), and three-cornered alfalfa hoppers, Spissistilus festinus (Say), corn earworms, Helicoverpa zea (Boddie), soybean loopers, Pseudoplusia includens (Walker), grasshoppers, Melanoplus femurrubrum (De Geer) and Melanoplus differentialis (Thomas), and WCR in soybean (Johnson and Mueller, 1989; Studebaker et al., 1991; Browde et al., 1992; O’Neal et al., 1999). Utilization of a sweep net as a sampling device in Indiana soybean fields does have drawbacks. Firstly, a majority of soybean fields in Indiana are planted in narrow rows. Walking through narrow row soybean fields in midsummer chap09.qxd 12/11/04 1:14 PM Page 181

Sampling Devices and Decision Rule Development 181

6 A

2 y = 0.026x + 1.457 Root ratings (untreated) 1999 r 2 = 0.079 1 P = 0.0006

0 0 20 40 60 80 100 WCR beetles/trap/week 1998

6 B

2 y = 0.048x + 1.562

Root ratings (untreated) 2000 r 2 = 0.36 1 P = 0.0139

0 0 20 40 60 80 100 WCR beetles/trap/week 1999

Fig. 9.3. Regression analysis of mean western corn rootworm population estimates/trap/ week from 16 north-west Indiana soybean ﬁelds using Pherocon® AM traps in 1998 versus root ratings taken in these same ﬁelds during 1999 (A), and using Pherocon® AM traps in 1999 versus root ratings taken in 2000 (B).

can be a physical challenge and, with limited space for walking, walking on soybean plants, resulting in plant damage, is inevitable. Secondly, error is high when different individuals are involved in taking sweep samples. Thirdly, using sampling devices that measure pest abundance at chap09.qxd 12/11/04 1:14 PM Page 182

182 C.K. Gerber et al.

6 y = 0.045x + 1.307 r 2 = 0.046 5 P = 0.0001

Root ratings (untreated) 1999 and 2000 1

0 0 20406080100 WCR beetles/trap/week 1998 and 1999

Fig. 9.4. Regression analysis of mean western corn rootworm population estimates/ trap/week from 32 north-west Indiana soybean ﬁelds using Pherocon® AM traps from 26 and 27 July to 31 August 1998 and 1999 versus root ratings taken in these same ﬁelds during 1999 and 2000.

single points in time may be less reliable than those sampling devices that integrate captures throughout a time period. Although the sweep net method results revealed that fewer samples were required than when using the vial trap method, specifically at the mean density level of five WCR beetles per trap per day, the overall drawbacks with using sweep nets in soybean fields will dissuade producers and crop consultants from relying on this sampling method. A more recent sampling device, the CRW trap, was tested and introduced to producers and crop consultants in the late 1990s (P.D. Lingren, Trécé Inc., personal communication). The CRW trap has been designed for monitoring corn rootworms in maize fields. CRW traps are user- friendly and the cost of a CRW trap is approximately US$4.00. CRW traps are reusable and can be utilized for three to four growing seasons. Although user-friendly and quite durable in soybean fields, one drawback was noted. Due to strong thunderstorms and high winds, CRW traps were easily dislodged from the PVC piping. Since PVC piping can be utilized over multiple growing seasons, the additional cost of PVC piping was not considered an indirect drawback of the CRW trapping method. Based on the number of CRW traps required to adequately sample WCR populations in soybean, along with the added benefits, producers and crop consultants may be more inclined to use the CRW trap in the future to monitor WCR beetles. chap09.qxd 12/11/04 1:14 PM Page 183

Sampling Devices and Decision Rule Development 183

During the past two decades, the Pherocon® AM trap has been tested as a monitoring device for adult WCR in maize (Hein and Tollefson, 1984, 1985; Karr and Tollefson, 1987; Hesler and Sutter, 1993; Midgarden et al., 1993; Gray and Steffey, 1995; Youngman et al., 1996). As noted with the CRW trap, the Pherocon® AM trap is user-friendly, with the exception of the adhesive coating located on one side of the trap. Unlike the instabil- ity of CRW traps mounted on the end of PVC pipes in soybean, Pherocon® AM traps are very stable in the field when properly attached to wood lath stakes. The cost of an individual Pherocon® AM trap is approximately US$1.00. Although Pherocon® AM traps are less expensive than the CRW traps per unit, since each Pherocon® AM trap is recommended to be replaced weekly, the cost of monitoring WCR beetles with six Pherocon® AM traps (based on the 20% level of precision and at the mean density of five WCR beetles per trap per day) during an entire growing season is approximately US$36.00 a year, compared to US$24.00, the cost of six CRW traps. However, if at any time during the sampling period the WCR beetle economic threshold is reached, monitoring can cease, which reduces the overall cost of the Pherocon® AM trapping method. Since Pherocon® AM traps are attached to wood lath stakes, the cost of these stakes is minor and not considered a drawback of the Pherocon® AM trapping method. Even though a price difference can exist, the benefits and number of Pherocon® AM traps required per field might encourage producers and crop consultants to utilize these traps for monitoring WCR beetles in soybean fields.

WCR economic injury levels and threshold

Although corn rootworm injury can be accurately rated, the point on the rating scale where economic injury for corn rootworm larval damage occurs is a controversial issue. According to Turpin et al. (1972), the economic injury level was estimated as 2.50 on the Hills and Peters (1971) 1 to 6 root rating scale. Data suggested that damage ratings above 2.50 were related linearly to a decrease in yields where a damage rating increase of 1.00 resulted in a 627 kg/ha yield reduction (Turpin et al., 1972). Using an artificial rootworm infestation, Branson et al. (1980) determined that a root rating of 2.00 produced a yield reduction of approximately 4%, which is equivalent to the control cost if a yield measures 9411 kg/ha. Mayo (1986) determined that a majority of investigators are convinced that the potential for economic damage exists if average root damage ratings exceed 2.75 to 3.00 on a 1 to 6 scale. Additional field experiments have suggested other economic injury levels. For example, Sutter et al. (1990) suggested that ratings might reach 4.00 to 5.00 before yield loss occurs. Davis (1994), working with an artificial rootworm infestation, estimated an economic injury level of 3.5–4.6 depending on yield potential and crop value. Even though root ratings can determine a level of super- chap09.qxd 12/11/04 1:14 PM Page 184

184 C.K. Gerber et al.

ficial injury by corn rootworms, since yield is influenced by a number of agronomic, biological and environmental factors, determining a specific economic injury level for rootworm larval feeding damage on maize roots has been a difficult task for researchers. With the number of factors that can have an adverse impact on yield, prevention of economic root damage by WCR larvae is one producers can control. Currently, soil insecticides and crop rotation are implemented in pest management programmes to protect maize roots from WCR larvae. As a direct result of the discovery of the crop rotation-adapted WCR variant (Edwards, 1996; Levine and Oloumi-Sadeghi, 1996), soil insecticide use for managing WCR has dramatically increased in the Midwest. To minimize the number of unnecessary soil insecticide applications, economic thresholds need to be implemented in pest management programmes to alert producers when soil insecticides are required to protect maize roots from WCR larvae, before economic injury occurs. Research conducted by Hein and Tollefson (1985) determined an economic threshold, within Iowa maize fields, of six beetles per trap per day (r2 = 0.26) at a root damage rating equivalent to 3.00, based on the Hills and Peters (1971) root damage rating scale. O’Neal et al. (2001) determined an economic threshold, in Illinois soybean fields, of five beetles per trap per day (r2 = 0.27) at a root damage rating of 3.00. Our study revealed that in Indiana, if economic root damage ratings of 3.00 and 3.50 are used, 5.37 and 6.96 beetles per Pherocon® AM trap per day are the economic injury levels (r2 = 0.46). Although economic injury levels can be well defined mathematically, the selection of a value for the economic threshold is less precise. By definition, the economic threshold should be established at a level that will allow pest management actions to take place in order to prevent pest densities from reaching the economic injury level. In the corn rootworm system, the stage of the pest that is sampled is not the primary stage that causes damage or the primary target of control measures. Control measures will not be required for 6 to 8 months once sampling has been completed. Therefore, there is no reason to factor a lag time between sampling and the application of a pest management action into the establishment of the economic threshold. As a result, the primary reason to establish the economic threshold for corn rootworms below the economic injury level is to allow for sampling error. Our proposed sampling plan provides 25% precision, which means that the true population mean should lie within 25% of the sample mean approximately 95% of the time. If our economic injury level root damage rating of 3.50 is accurate, then setting the economic threshold at 5.0 beetles per trap per day should result in fewer than 5% Type 2 errors. In our study, a Type 2 error occurs when sampling indicates economic beetle thresholds are not reached, but economic damage occurs, and a Type 1 error occurs when sampling indicates that economic beetle thresholds are reached, but economic damage does not occur. Maize producers are more inclined to accept a Type 1 error, rather than a Type 2 error. Because of uncertainty in the accuracy of the 3.5 root chap09.qxd 12/11/04 1:14 PM Page 185

Sampling Devices and Decision Rule Development 185

damage rating, we established our economic threshold at ﬁve beetles per trap per day.

References

Barna, G., Kiss, J., Edwards, C.R., Gerber, C. and Bledsoe, L.W. (1998) Comparisons of Hungarian pheromone and Pherocon AM® traps under economic western corn rootworm populations in Indiana, USA. Pflanzenschutzberichte 57, 52–56. Branson, T.F. and Ortman, E.E. (1970) The host range of larvae of the western corn rootworm: further studies. Journal of Economic Entomology 63, 800–803. Branson, T.F., Sutter, G.R. and Fisher, J.R. (1980) Plant response to stress induced by artificial infestations of western corn rootworm. Environmental Entomology 9, 253–257. Browde, J.A., Pedigo, L.P., Degooyer, T.A., Higley, L.G., Wintersteen, W.K. and Zeiss, M.R. (1992) Comparison of sampling techniques for grasshoppers (Orthoptera: Acrididae) in soybean. Journal of Economic Entomology 85, 2270–2274. Davis, P.M. (1994) Comparison of economic injury levels for western corn rootworm (Coleoptera: Chrysomelidae) infesting silage and grain corn. Journal of Economic Entomology 87, 1086–1090. Edwards, C.R. (1996) The dramatic shift of the western corn rootworm to first-year corn. In: 1996 Proceedings of the Illinois Agricultural Pesticides Conference. Cooperative Extension Service, University of Illinois at Urbana-Champaign, Illinois, pp. 14–15. Edwards, C.R. (2000) Managing the western corn rootworm variant and controlling European corn borer with Bt corn. In: Proceedings of the 2000 Crop Management Workshops. Purdue Pest Management Program, Purdue University, West Lafayette, Indiana, pp. 17–23. Edwards, C.R., Bledsoe, L.W. and Obermeyer, J.L. (1998) Western Corn Rootworm Sweep Net Survey in Soybean. Pest Management and Crop Production Newsletter 28, Purdue Cooperative Extension Service, Purdue University, West Lafayette, Indiana. Fleischer, S.J. and Allen, W.A. (1982) Field counting efficiency of sweep-net samples of adult potato leafhoppers (Homoptera: Cicadellidae) in alfalfa. Journal of Economic Entomology 75, 837–840. Gibb, T.J. and Higgins, R.A. (1991) Aboveground dry weight and yield response of irrigated field corn to defoliation and root pruning stresses. Journal of Economic Entomology 84, 1562–1576. Godfrey, L.D. and Turpin, F.T. (1983) Comparison of western corn rootworm (Coleoptera: Chrysomelidae) adult populations and economic thresholds in first-year and continuous corn fields. Journal of Economic Entomology 76, 1028–1032. Gray, M.E. and Steffey, K.L. (1995) On-farm validation of Pherocon AM trap utility for corn rootworms (Coleoptera: Chrysomelidae): outlook for producer acceptance in Illinois. Journal of Sustainable Agriculture 81, 1398–1403. Hein, G.L. and Tollefson, J.J. (1984) Comparison of adult corn rootworm (Coleoptera: Chrysomelidae) trapping techniques as population estimators. Environmental Entomology 13, 266–271. Hein, G.L. and Tollefson, J.J. (1985) Use of the Pherocon AM trap as a scouting tool for predicting damage by corn rootworm (Coleoptera: Chrysomelidae) larvae. Journal of Economic Entomology 78, 200–203. Hesler, L.S. and Sutter, G.R. (1993) Effect of trap color, volatile attractants, and type of toxic bait dispenser on captures of adult corn rootworm beetles (Coleoptera: Chrysomelidae). Environmental Entomology 22, 743–750. chap09.qxd 12/11/04 1:14 PM Page 186

186 C.K. Gerber et al.

Hill, R.E. (1975) Mating, oviposition patterns, fecundity, and longevity of the western corn rootworm. Journal of Economic Entomology 68, 311–315. Hills, T.M. and Peters, D.C. (1971) A method of evaluating postplanting insecticide treatments for control of western corn rootworm larvae. Journal of Economic Entomology 64, 764–765. Johnson, M.P. and Mueller, A.J. (1989) Flight activity of the three cornered alfalfa hopper (Homoptera: Membracidae) in soybean. Journal of Economic Entomology 82, 1101–1105. Karr, L.L. and Tollefson, J.J. (1987) Durability of the Pherocon AM trap for adult western corn rootworm (Coleoptera: Chrysomelidae) sampling. Journal of Economic Entomology 80, 891–896. Kuhar, T.P. and Youngman, R.R. (1998) Olson yellow sticky trap: decision-making tool for sampling western corn rootworm (Coleoptera: Chrysomelidae) adults in ﬁeld corn. Journal of Economic Entomology 91, 957–963. Levine, E. and Gray, M.E. (1994) Use of cucurbitacin vial traps to predict corn rootworm (Coleoptera: Chrysomelidae) larval injury in a subsequent crop of corn. Journal of Entomological Science 29, 590–600. Levine, E. and Gray, M.E. (1996) First-year corn rootworm injury: east-central Illinois research progress to date and recommendations for 1996. In: 1996 Proceedings of the Illinois Agricultural Pesticides Conference. Cooperative Extension Service, University of Illinois at Urbana-Champaign, Illinois, pp. 3–13. Levine, E. and Oloumi-Sadeghi, H. (1991) Management of diabroticite rootworms in corn. Annual Review of Entomology 36, 229–255. Levine, E. and Oloumi-Sadeghi, H. (1996) Western corn rootworm (Coleoptera: Chrysomelidae) larval injury to corn grown for seed production following soybeans grown for seed production. Journal of Economic Entomology 89, 1010–1016. Mayo, Z.B., Jr (1986) Field evaluation of insecticides for control of larvae of corn rootworms. In: Krysan, J.L. and Miller, T.A. (eds) Methods for the Study of Pest Diabrotica. Springer-Verlag, New York, pp. 183–203. Metcalf, R.L. (1986) Foreword. In: Krysan, J.L. and Miller, T.A. (eds) Methods for the Study of Pest Diabrotica. Springer-Verlag, New York, pp. vii–xv. Midgarden, D.G., Youngman, R.R. and Fleischer, S.J. (1993) Spatial analysis of counts of western corn rootworm (Coleoptera: Chrysomelidae) adults on yellow sticky traps in corn: geostatistics and dispersion indices. Environmental Entomology 22, 1124–1133. O’Neal, M.E., Gray, M.E. and Smyth, C.A. (1999) Population characteristics of a western corn rootworm (Coleoptera: Chrysomelidae) strain in east-central Illinois corn and soybean ﬁelds. Journal of Economic Entomology 92, 1301–1310. O’Neal, M.E., Gray, M.E., Ratcliffe, S. and Steffey, K.L. (2001) Predicting western corn rootworm (Coleoptera: Chrysomelidae) larval injury to rotated corn with Pherocon AM traps in soybeans. Journal of Economic Entomology 94, 98–105. Reed, J.P., Hall, F.R., Taylor, R.A.J. and Wilson, H.R. (1991) Development of a corn rootworm (Diabrotica, Coleoptera: Chrysomelidae) larval threshold for Ohio. Journal of the Kansas Entomological Society 64, 60–68. Ruesink, W.G. (1980) Introduction to sampling theory. In: Kogan, M. and Herzog, D.C. (eds) Sampling Methods in Soybean Entomology. Springer-Verlag, New York, pp. 61–78. Sammons, A.E., Edwards, C.R., Bledsoe, L.W., Boeve, P.J. and Stuart, J.J. (1997) Behavioral and feeding assays reveal a western corn rootworm (Coleoptera: Chrysomelidae) variant that is attracted to soybean. Environmental Entomology 26, 1336–1342. Shaw, J.T., Ruesink, W.G., Briggs, S.P. and Luckman, W.H. (1984) Monitoring populations of corn rootworm beetles (Coleoptera: Chrysomelidae) with trap baited with cucurbitacins (Diabrotica barberi). Journal of Economic Entomology 77, 1495–1499. chap09.qxd 12/11/04 1:14 PM Page 187

Sampling Devices and Decision Rule Development 187

Snodgrass, G.L. (1998) Distribution of the tarnished plant bug (Heteroptera: Miridae) within cotton plants. Environmental Entomology 27, 1089–1093. Spencer, J.L., Isard, S.A. and Levine, E. (1998) Western corn rootworms on the move: monitoring beetles in corn and soybeans. In: 1998 Proceedings of the Illinois Agricultural Pesticides Conference. Cooperative Extension Service, University of Illinois at Urbana-Champaign, Illinois, pp. 10–23. Spike, B.P. and Tollefson, J.J. (1991) Yield response of corn subjected to western corn rootworm (Coleoptera: Chrysomelidae) infestation and lodging. Journal of Economic Entomology 84, 1585–1590. Steel, R.G.D. and Torrie, J.H. (1980) Principles and Procedures of Statistics, a Biometrical Approach, 2nd edn. McGraw-Hill, New York, 633 pp. Steffey, K.L., Tollefson, J.J. and Hinz, P.N. (1992) Sampling plan for population estimation of northern and western corn rootworm adults in Iowa cornﬁelds. Journal of Economic Entomology 11, 287–291. Stern, V.M. (1973) Economic thresholds. Annual Review of Entomology 18, 259–280. Studebaker, G.E., Spurgeon, D.W. and Mueller, A.J. (1991) Calibration of ground cloth and sweep net sampling methods for larvae of corn earworm and soybean looper (Lepidoptera: Noctuidae) in soybean. Journal of Economic Entomology 84, 1625–1629. Supernova (1989) Supernova: Accessible General Linear Modeling. Abacus Concepts, Berkeley, California, 316 pp. Sutter, G.R., Fisher, J.R., Elliott, N.C. and Branson, T.F. (1990) Effect of insecticide treatments on root lodging and yields of maize in controlled infestations of western corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 83, 2414–2420. Tollefson, J.J. (1975) Corn rootworm adult and egg sampling techniques as predictors of larval damage. PhD dissertation, Iowa State University, Ames. Turpin, F.T., Dumenil, L.C. and Peters, D.C. (1972) Edaphic and agronomic characteristics that affect potential for rootworm damage to corn in Iowa. Journal of Economic Entomology 65, 1615–1619. Weinzierl, R.A., Berry, R.E. and Fisher, G.C. (1987) Sweep-net sampling for western spotted cucumber beetle (Coleoptera: Chrysomelidae) in snap beans: spatial distribution, economic injury level, and sequential sampling plans. Journal of Economic Entomology 80, 1278–1283. Youngman, R.R., Kuhar, T.P. and Midgarden, D.G. (1996) Effect of trap size on efﬁciency of yellow sticky traps for sampling western corn rootworm (Coleoptera: Chrysomelidae) adults in corn. Journal of Entomological Science 31, 277–285. chap09.qxd 12/11/04 1:14 PM Page 188 chap10.qxd 24/11/04 1:44 PM Page 189

10 Western Corn Rootworm (Diabrotica virgifera virgifera LeConte) and the Crop Rotation Systems in Europe

József Kiss,1 Judit Komáromi,1 Khosbayar Bayar,1 C. Richard Edwards2 and Ibolya Hatala-Zsellér3 1Department of Plant Protection, Szent István University, Gödöllö, Hungary; 2Department of Entomology, Purdue University, W. Lafayette, Indiana, USA; 3Csongrád County Plant and Soil Protection Service, Hódmezövásárhely, Hungary

Introduction

The western corn rootworm (WCR), Diabrotica virgifera virgifera LeConte, has spread over a large area from its first detection point, near Belgrade, Serbia and Montenegro, in Europe. The total area infested by WCR in Europe exceeded 300,000 km2 in 2003 and economic damage to maize was present on more than 20,000 ha. WCR was also detected a considerable distance from its actual spread line (for details on spread see Kiss et al., Chapter 2, this volume). The total area of maize harvested in Europe in 2002 was 12.4 million ha (http://apps.fao.org/default.htm), which provides the opportunity for a large establishment area for WCR. Until found in Europe, the WCR was an economic pest of maize only in the USA. The WCR is a univoltine insect and overwinters in the egg stage. WCR beetles feed on the leaves, silks, pollen and young kernels of maize, although pollen is the preferred food. After the silking and pollination period, they tend to remain in maize stands and feed on foliage (Ludwig and Hill, 1975). In some cases, with the drying of the silks a small number of adult beetles move to other crops, such as lucerne (Ball, 1957) and soybean (Shaw et al., 1978). The larvae of WCR primarily feed on the root system of maize, but can feed to some degree on some species of Gramineae (Branson and Ortman, 1967, 1970), and adults were produced from five of nine tested grassy weed species (Clark and Hibbard, 2003; Moeser and Hibbard, Chapter 3, this volume). Since WCR beetles tend to lay their eggs in fields where they feed (Branson and Krysan,

190 J. Kiss et al.

1981), females deposit most of their eggs in maize fields (Shaw et al., 1978). This ovipositional habit of WCR females is critical for larval survival, as larvae can develop on the roots of maize, but high mortality occurs on the roots of grassy-type weeds (Branson and Ortman, 1970). Therefore, crop rotation has been recommended for many years as the primary non-chemical control method in the USA. The rotation of maize to soybean in the Corn Belt of the USA has been a successful and widely used management strategy for the control of WCR (Gray et al., 1998). However, in the state of Illinois (USA) in 1987 in inbred seed maize fields that had been planted after seed-production soybean, which was free of weeds and volunteer maize, severe WCR larval injury was observed (Levine and Oloumi-Sadeghi, 1996). In the neighbouring state of Indiana, WCR larval injury in first-year maize following soybean was also reported (Edwards et al., 1996). Crop rotation of maize is also the primary integrated pest management (IPM) control option in Europe, either from a management or from an eradication viewpoint. The rotation of maize complies with the requirements of integrated production as described in the International Organization for Biological Control (IOBC) Western Palaearctic Regional Section (WPRS) guidelines for arable Crops (Boller et al., 1997). In areas of Europe with well-established WCR populations, several national regulations also prescribe obligatory rotation of maize once larval injury occurs in the maize field (for example, Decree No. 7/2001 by the Minister of Agriculture in Hungary). For new detections of WCR in the European Union (detection of adults in areas that were previously known to be free of WCR), the relevant Commission Decision also prescribes an obligatory crop rotation for 2 years (in the focus zone) or 1 year (in the safety zone) (2003/766 EC). For farmers and decision makers the question is the same. Is crop rotation in Europe an effective tool for eradication or for managing WCR populations or will the rotation-tolerant variant become a problem? Several hypotheses were developed as to the possible reasons for this WCR behavioural change (details given later). It is not at present known whether the WCR population that is present in Europe originates from the crop rotation-tolerant variant or if it has the ability to change its behaviour to adapt itself to the crop rotation systems in Europe, or if any signs of the adaptation process can be identified in the WCR population. As written before, the key element of present non-chemical control tools against WCR is interfering with its normal biological process through rotating maize to non-maize crop(s). Thus, emerging larvae from eggs laid in the maize stand the previous year will not find their primary food source, the maize roots. Therefore, in our study we approached the behaviour of WCR under European conditions and asked questions as follows:

Do WCR larvae feed, survive and develop into the adult stage on the root system of non-maize crops (if maize is rotated to non-maize crops or eggs are laid in ﬁelds with non-maize crops in the subsequent year)? chap10.qxd 24/11/04 1:44 PM Page 191

Western Corn Rootworm and Crop Rotation Systems in Europe 191

Do WCR adults leave maize stands for egg laying in non-maize crop stands that may be rotated to maize in the subsequent year and can emerging larvae find their necessary food source? To what extent does the fragmented landscape in Europe with crop stands (including their weed cover), with non-crop habitats (orchards, pastures, field margins and other ecological compensation areas, set- aside fields) favour or not favour the dispersal, immigration and population build-up of WCR?

In any case, the question is how the WCR population will behave in the various ecosystems of Europe, or, in other words, to what extent the existing typical crop rotation systems will favour or not favour the build-up of the WCR population. Therefore, which type of crop rotation, if any, contributes to a decrease in the WCR population over time and space?

Hypotheses for the Development of the New Rotation-tolerant Variant in the USA

In order to assess the impact of crop rotation systems in Europe on WCR population levels, one must review the history of and present knowledge of the rotation-tolerant WCR variant in the USA. The crop rotation- tolerant variant is probably a mixed population of WCR beetles, a portion of the population preferring soybean, while the remainder prefer maize (Sammons et al., 1997). The WCR has been forced to adapt to crop rotation in this part of the USA, as approximately 74% of acreage in Indiana is in the annual crop rotation of maize and soybean (Sammons et al., 1997). Decreasing the host plant area forced the WCR to ﬁnd alternative plants to feed on and to oviposit in so as to increase the chance of larval survival, as maize may not be present in the same location that the adults emerge in (Coats et al., 1986). Thus, the double strategy of oviposition can be the key to survival. Since the northern corn rootworm (NCR), Diabrotica barberi Smith & Lawrence, adapted itself to crop rotation with a prolonged diapause (Krysan et al., 1984; Levine et al., 1992), there was speculation as to whether WCR could do the same. Prolonged diapause, however, cannot be responsible for such larval injury in maize after soybean (Levine and Oloumi-Sadeghi, 1996). Soil samples taken in July in Illinois showed no sign of WCR eggs, thus suggesting that prolonged diapause was not the cause of the larval injury in ﬁrst-year maize (Levine et al., 2002). Widely used applications of permethrin (synthetic pyrethroid insecticide) in seed maize could also have forced WCR adults to move from maize fields to soybean fields, as permethrin has repellent and antifeedant effects on several beetle types (Dobrin and Hammond, 1985). It is unlikely that this effect caused the resistance and the severe damage caused by WCR larvae (Levine and Oloumi-Sadeghi, 1996). Intermating between the NCR and WCR is frequently seen within chap10.qxd 24/11/04 1:44 PM Page 192

192 J. Kiss et al.

fields (Ball, 1957). The offspring of these species morphologically resem- ble the WCR (Hintz and George, 1979). However, this hybrid may have inherited traits from both species. NCR tend to feed on pollen from different crop types after pollination of maize, whereas WCR generally stay in maize fields and feed on the foliage of maize (Hill and Mayo, 1980; Branson and Krysan, 1981). All of the above results indicate the behavioural change of WCR population that can take place and result in the laying of their eggs outside maize fields. The question is: which factors, if any, influence behaviour especially under European conditions? One of the factors may be the attraction of different crops to WCR adults and their feeding on non- maize crops. This possibility is of great importance under European conditions where crop plant heterogeneity over the landscape is much higher compared to that in the Corn Belt in the USA.

Attraction of Different Crops to WCR Adults and Their Feeding on Non-maize Crops in the USA

In the Corn Belt, mostly maize and soybean are available for WCR adults on large acreages. The fact that WCR adults feed on other crops in addition to maize is known in high-percentage crop-rotation regions. In some regions smaller numbers of WCR were found after the silking period of maize feeding in lucerne (Medicago sativa L.) and kochia (Kochia scoparia L.) fields (Ball, 1957). Although the longevity of WCR females is significantly reduced when maintained on host diets other than maize, this species is capable of feeding on a number of alternative host plants, including sunflower (Helianthus annuus L.), squash (Cucurbita maxima Duch.) and golden rod (Solidago canadensis L.) (Siegfried and Mullin, 1990). Besides the above plants, WCR adult feeding was observed on red clover (Trifolium pratense L.), velvetleaf (Abutilon theophrasti Medic.), giant ragweed (Ambrosia trifida L.) and Jerusalem artichoke (Helianthus tuberosus L.) (Levine et al., 2002). In the US state of Nebraska, WCR was abundant in lucerne fields and also in pasture (Hixson et al., 1947). WCR populations were also found in soybean fields, where volunteer maize was abundant (Shaw et al., 1978). No economic population was observed at that time, either on non-maize crops where WCR was feeding or in first-year maize. Since 1987, when the problem in first-year maize after soybean in the eastern Corn Belt was observed, experiments have been established to study the behaviour of the variant. Due to volatiles given off by maize residue in soybean fields (Edwards et al., 1996) or because reduced affinity for maize fields has appeared (Spencer et al., 1999) or the soybean itself served as an attractant for WCR adults (Sammons et al. 1997), changes occurred. WCR may feed and lay its eggs in non-maize plots. Host preference assays showed in Indiana that WCR beetles significantly preferred soybean without maize residues to all chap10.qxd 24/11/04 1:44 PM Page 193

Western Corn Rootworm and Crop Rotation Systems in Europe 193

the other hosts, including pollinating and pollinated maize (Sammons et al., 1997). Contrary to the previous results, adults collected from the same region where behavioural tolerance to crop rotation developed did not show significant association to soybean, but a significantly greater proportion of WCR was recorded on maize plants (Spencer et al., 1999; Levine et al., 2002). The difference between the two experiments was that Sammons et al. allowed the beetles to choose their host(s) for 36 h while Spencer et al. allowed only 0.5 h. Samples of variant WCR adults on problem areas from problem soybean and maize fields in Illinois and analysis of their gut content revealed that soybean herbivory is not exclusive to rotation-adapted WCR populations. The tendency to feed on soybean has not changed much in the WCR population, but the percentage of adults with this tendency is likely to increase in problem areas (Levine et al., 2002).

Development of WCR Adults in Maize Following Different Crops in the USA

In 1996, corn rootworm larval injury to maize following oats, wheat and soybean was observed (Levine et al., 2002). In Illinois, injury to maize roots following soybean showed root damage ratings of 3.5 and 4.2 on the Iowa 1–6 scale (Levine et al., 2002). In Illinois, emergence cages were placed in maize rows in ﬁrst-year maize ﬁelds. All 94 beetles captured under the cages were WCR beetles (Levine et al., 2002). In 1968 in Nebraska, an average of 7.3, 14.0 and 0.5 WCR adults emerged in maize after soybean, milo (sorghum) and oat stubble, respectively. Cages covered single maize plants (Hill and Mayo, 1980). Thus, even as early as 1968, emergence of WCR adult was observed in maize after non-maize crops.

Population Level of WCR Adults in Different Crop Stands in the USA

In US states where it is likely that the crop rotation-tolerant variant is present, several experiments were carried out with different trap types and sampling methods to obtain data on WCR presence and abundance in different crop stands. Historical data from Urbana County, Illinois, showed six to 16 WCR adults in 100 sweeps with a sweep net between 1979 and 1982 in mid-August (the peak period for oviposition). In problem areas (rotation-tolerant variant), WCR counts in soybean ﬁelds ranged between 23 and 100 beetles per 100 sweeps. In 1996, the greatest densities were 59–102 beetles per 100 sweeps, and the majority of beetles caught were females (Levine et al., 2002). Thus an increase in WCR adult numbers is expected in non-maize crops in areas where the new strain develops. Comparing the captures in Pherocon® AM traps (Trécé Inc., Salinas, chap10.qxd 24/11/04 1:44 PM Page 194

194 J. Kiss et al.

CA 93912) in maize and soybean fields, there was only 1 week in 7 when significantly more WCR beetles were collected in soybean fields (O’Neal et al., 1999). The percentage of females was significantly higher in 4 weeks of 7 in soybean fields than in maize fields in Illinois. In the percentage of females, there was no significant difference considering the Pherocon® AM traps’ location in the interior or exterior part of the soybean field. Similarly to this, the location of vial traps in the field did not affect the number of beetles captured or the percentage of females in either maize or soybean fields (O’Neal et al., 1999). In field sampling in 1996 and 1997 in Indiana, WCR adults were present in both maize and soybean fields in the same time period. Although the build-up patterns were similar in both years, significantly higher numbers of WCR adults were found in maize fields (Barna et al., 1998). In this state, however, as many as 1081 WCR adults per 100 sweeps with a sweep net were caught in soybean in 1997 (Blackwell, 1997). Captures of female beetles were significantly higher in soybean fields compared to maize fields, except one year (1998) when weather conditions were not favourable for larval development and adult emergence (Barna et al., 1999). In Champaign County, Illinois, vial traps caught significantly more WCR beetles in soybean than in maize, oat and lucerne fields (Levine et al., 2002). The above data demonstrate the complexity of WCR adult movement patterns and behaviour among different crop stands in space and time. Most probably, one important factor for adult WCR flight into crop stands is the presence of food outside maize fields. However, WCR adult movement from maize to soybean happened even while fresh to brown maize silks were available (Levine et al., 2002).

Materials and Methods

The 3-year crop rotation experiment was established by the Plant Protection Department, Szent István University, Hungary, in 2000 in Szeged in southern Hungary. The crops included in the study were similar to those normally grown in rotation with or next to maize in many areas of Europe, especially typical of maize-growing regions of Central Europe. The ﬁrst WCR adults were detected in the region of the crop rotation experiment in 1995. Therefore, the population has been exposed for up to an 8-year period where behavioural changes could be taking place.

Preconditions for the experiment

The rotation study was established in fields dedicated to this research activity for a period of 3 years. The study area was noted as being one of the first WCR-infested areas in Hungary with significant WCR population densities (Kiss et al., 2001a). Maize was the pre-crop (1999) for the study. chap10.qxd 24/11/04 1:44 PM Page 195

Western Corn Rootworm and Crop Rotation Systems in Europe 195

Soil conditions (silt and clay content 66.1%, organic matter 1.8%) favoured egg laying and larval development. Soils high in clay content have a greater deﬁned secondary structure with a higher number of con- nected pores than sandy soils (Edwards, 2000). Crops and soil in the rotation were not treated with insecticides during the 3-year period so as to avoid WCR disturbance. For herbicides, application of pre-emergence products with less persistence and potential carry-over were used to minimize any effect on eggs in the soil or larval development. Regular production practices, as applied in the region for the targeted crops, were used so as to mimic those normally carried out by farming communities.

Experimental layout

The layout was designed to allow for plot sizes large enough for reliable data to be generated within each plot. Six replications of each treatment were used for the study. The test site was about 3 ha. The study area was divided into two sections. One half was grown to maize (continuous maize) and the other half was planted to strips of the various crops that were in rotation with maize (maize, soybean, sunflower and winter wheat). In both sections continuous maize was present in each year. Therefore, we often refer to two data sets for continuous maize. One data set refers to strips of continuous maize among non-maize crops, while the other refers to the continuous maize strips among rotated maize strips. Due to starting the study in the spring, oats were planted in the first year as the cereal crop instead of winter wheat. Winter wheat was then planted in the early autumn for the subsequent growing cycles. The two sections of the test field were annually rotated for the duration of the study. Each plot where the various crops were planted was marked with a permanent marker. The field was laid out in the shape of a rectangle. The size of each of the two sections that made up the study was 47 m × 217 m; the distance between the two sections was 9 m (a roadway to allow for movement and manoeuvring of machinery, etc.). The uncropped area on the outside perimeter (outer boundary) of sections was also 9 m. Each section was par- titioned into 24 plots (six replications × four crops) of equal size. Crops within replications were randomized, but the same crop from each of two replications could not border each other. Also, to reduce the chance of moving WCR eggs from one area of a field to another or from plot to plot, equipment was kept as clean as possible by hand removing of soil. Maize (‘Borbála’ SC hybrid, Cereal Research Non-Profit Company, Szeged, Hungary) was planted between 16 and 25 April depending on the year. The planting depth was 6 cm. The seeding rate was 60,000 seeds/ha. The row width was 70 cm. Fertilizers and pre-emergence herbicide treatment (active ingredient: metholachlor) applications were uniform in maize, sunflower and soybean stands as usually used by farmers in this region in southern Hungary. For soil preparation, disking and ploughing were used. chap10.qxd 24/11/04 1:44 PM Page 196

196 J. Kiss et al.

The experimental layout was designed to establish an experiment that gave reliable data and was feasible to carry out with available inputs. Therefore, the number of crops was limited to four. Plot size was maxi- mized with the condition that treatments (crops) must be replicated six times.

Plot layout

In order to properly place three traps in each plot, plots were arranged in an elongated, rectangular shape. Each plot was at least 8.6 m × 47 m (404.20 m2), thus giving approximately 11 rows of maize, soybean and sunﬂower and 63 rows of cereal. There was approximately 1 m distance between each plot.

Sampling for WCR population

In the crop rotation experiment, various types of sampling were conducted to survey all WCR developmental stages (eggs, larvae, pupae and adults) to estimate WCR population density where applicable and to follow its biology in the various crops in rotation over the 3 years.

Sampling for WCR adult population

Since the crop rotation-tolerant variant has been identified, the monitoring of WCR adult populations in non-maize crops has become important. Several traps are used in the USA for sampling WCR adults, include the Pherocon® AM No Bait trap, Pherocon® corn rootworm (CRW) kairomone trap, vial trap and Olson yellow sticky trap. The Pherocon® AM trap was found to be the most efficient for estimating WCR adult population levels in Indiana. This trap showed the dynamics of the WCR population (adults per trap, sex ratio) in accordance with that of the natural population (Barna et al., 1998; Barna, 2001). Comparisons of the Hungarian Csalomon® pheromone (PAL) trap with the Pherocon® AM trap under high WCR population levels demonstrated that the Pherocon® AM trap better mimics the natural field population (Barna, 2001). In our experiment, when selecting a trap for WCR adult sampling, we had to take into consideration several points. The limited size of plots and the various crop plant heights (or even stubble only on harvested winter wheat plots) forced us to select a visual trap that has a limited attraction circle if placed deep into the crop stand. We also wanted to avoid or decrease the trapping out of the WCR adult population, which would highly influence egg laying and the following year’s WCR population. The WCR adult population was monitored from mid-June to mid- September during the growing seasons in 2000–2002. Emergence cages chap10.qxd 24/11/04 1:44 PM Page 197

Western Corn Rootworm and Crop Rotation Systems in Europe 197

and Pherocon® AM unbaited yellow sticky traps were used as population estimating tools.

Adult activity near the soil surface

Three Pherocon® AM yellow sticky traps were placed equidistant within each plot in the middle row with the first being 7 m from the entrance of the plot and the last 7 m from the back of the plot. Pherocon® AM traps were attached to the stems of maize and sunflower at about 40 cm height and on a stake in the soybean and wheat plots 5 cm above the soil surface. This method was used to capture WCR adults showing activity above the soil surface (laying eggs in the soil in any crop in the rotation or emerging from the soil expected in maize only). Admittedly, adult WCR movement into crop stands and their potential feeding there could be better followed by traps near the crop canopy level; however, in our experiment, the placement of the traps close to ground surface in soybean and cereals, especially after harvesting the latter in mid-July, was designed to decrease the visual attraction of adults from some distance away. Traps were checked and changed weekly and removed to the laboratory for identification and counting. Data recorded for WCR adult presence in maize and in other crops in rotation (on Pherocon® AM traps) were as follows: Number of WCR adults trapped. Sex ratio of adults (to follow and compare sex ratio by crop over the period of adult activity). Number of eggs being carried by WCR females over the trapping period. The appearance of adults in a given crop stand (%; A) was calculated as follows: NC A = catch × 100% NCtotal

where NCcatch = the number of adults captured in different crop stands, and NCtotal = total number of adults captured in all four crop stands. The frequency of adult catches (%; F) was calculated using the formula:

Ncatch F = × 100% Ntotal

where Ncatch = number of traps with adult captures, and Ntotal = total number of traps (= 18). Finally, the relative adult captures (%; R) were calculated using the formula: T R = rotated × 100% Tcontinuous chap10.qxd 24/11/04 1:44 PM Page 198

198 J. Kiss et al.

where Trotated = total number of captured adults in rotated maize, and Tcontinuous = total number of captured adults in continuous maize.

Adult emergence

Three emergence cages were placed equidistant within each plot in the middle row. Each year the exact location of each plot and cage was marked to allow for the proper future placements. For emergence cages, the inner dimension of the frame was 1.25 m length × 0.4 m width (0.5 m2). Maize plants were at the growth stages of V6 to V11 each year when the cages were installed. Each cage was centred over three plants within a row in maize and sunflower. One plant was left to develop, and two plants were trimmed back under each cage so as to avoid screen damage, but to allow root development to continue in case larval feeding was taking place. In cereal and soybean plots, the emergence cages were simply placed over as many plants as would fit under the frame in the case of the cereal crop and as many soybean plants within the row that would fit under the frame in soybean plots. In the case of the cereals, cages were removed from the plot just before harvest (mid-July) and immediately replaced so as not to miss any WCR adult emergence that might occur. The WCR emergence investigation was established to prove the presence of eggs in the soil from the previous year and development of larvae to the adult stage in the crop in the rotation in the present year. The number of needed emergence cages for one plot was calculated by the formula of Southwood (Southwood, 1978). Traps and cages in contiguous maize mirrored those in the rotated plots. As noted before, the exact location of the plots was marked so as to allow for the proper placement of future plots. The total number of Pherocon® AM traps in each section was 72 (144 total for two sections). These traps were changed weekly and the number of WCR adults in the traps was counted and recorded. Also, all the females trapped in Pherocon® AM traps were dissected and the number of eggs recorded. The emergence cages were equal in number (72 + 72 = 144) to the Pherocon® AM traps. The presence of WCR adults was assessed and quantified by using the Pherocon® AM traps and emergence cages from mid-June to mid-September. Emerged adults were collected weekly from cages by using a half Pherocon® AM trap. The number and sex of adults in each collection were recorded. For adult emergence under emergence cages in different crop stands, the same (F, frequency, and R, relative emergence) formulae were used as for adult appearance.

Sampling for WCR egg population

Samplings for WCR eggs were carried out on 30 August 2000 and 24 September 2001 in maize and in non-maize crops. By this time a major- chap10.qxd 24/11/04 1:44 PM Page 199

Western Corn Rootworm and Crop Rotation Systems in Europe 199

ity of WCR eggs should have been laid by females. Soil samples were taken to a depth of 16 cm using a spade. Each soil sample (0.5 l container) consisted of four subsamples within 2 m2 grid within plant rows. Sampling was taken randomly in two (2000) and four (2001) replications in each plot. In contiguous maize in 2000 a total of eight samples, while in 2001 a total of four samples, were taken regardless of the pre-crop. The samples were labelled and transported to the laboratory and stored in a deep freezer until processed. Samples were washed and eggs extracted by using Schuiling cyst nematode washing equipment. All extracted eggs were identiﬁed to species by chorion sculpturing (Atyeo et al., 1964) and counted. Data obtained from the egg samplings were analysed using the binomial population density estimation based upon frequency of occurrence (Gerrard and Chiang, 1970; Gerrard and Cook, 1972; Southwood, 1978). The k value (aggregation index of the population at negative binomial) was calculated as: x2 k = s2 – x where s = standard deviation and x = mean number of eggs per sampling unit. The estimated egg density per sample (λ) was calculated using the

formula:

 1

 1  k 

λ  = k  – 1 1 – θ  where θ = probability of egg occurrence in a sampling unit.

Estimating WCR larval population

Root rating for larval activity For larval activity, root systems of maize plants were evaluated using the Iowa 1–6 root rating scale (Hills and Peters, 1971) for damage. The roots of ten maize plants from each plot were rated between 11 July and 19 July each year when maize plants were at the growth stages of R1 and R2. This period was determined on the basis of our experiences related to peak larval development and activity in south Hungary. At this time, root damage was expected to be the highest, with possible root regeneration the lowest.

Larval sampling Since larval damage to even the continuously cropped maize was moderate in the ﬁrst year (2000), it was decided not only to check for larval activity (root damage to maize), but also to search for larvae the following year. Sampling for WCR larvae was done once on 14 June in 2001 only in continuous maize and, to double-check their presence, twice (15 June and chap10.qxd 24/11/04 1:44 PM Page 200

200 J. Kiss et al.

14 July) in 2002 in continuous maize, in maize after non-maize crops and in winter wheat. Maize plants were at growth stages of V10–V11 (14–15 June) and R2 (14 July). The sampling unit was a soil cube measuring approximately 16 cm on all sides with a centrally located maize root system. Each plot was sampled ten times (ten maize plants). One sample (root and soil) was about 0.004 m3 in volume. Immediately after collection, the labelled samples were taken to the laboratory and visually searched for WCR larvae. After identiﬁcation (Mendoza and Peters, 1964), all larvae were separated as to their developmental stages, counted and recorded. Data, including numbers of larvae, were analysed using the method as described in the data analysis paragraph. Mean numbers of larvae sampled on each maize plant were correlated to larval root injury indices and to cumulative adult numbers in emergence cages in the same year.

Additional data recorded

Crop phenologies over the sampling period were recorded weekly for cereals, soybean, sunﬂower and maize (Special Report No. 48, Iowa State University, USA). Weed species, relevant meteorological and agronomic data and other crops around the experiment were also recorded.

Data analysis

For the statistical analyses EXCEL and MINITAB statistics programs were used. Beetle counts on Pherocon® AM traps and in emergence cages were normalized by using the formula (x + 0.5)1/2. This formula was also used when the number of larvae and root injury were compared on maize samples after different crop stands. Transformed data were used only for statistical analyses, Student t test and Welsh test (Ló´kös Tóth, 2002). In the case of adult captures by Pherocon® AM traps in continuous maize and non-maize crops, an estimation was made for the subsequent adult populations in emergence cages. Data obtained from adult counts in emergence cages were analysed as an emergence curve over time, which was calculated from cumulative adult numbers emerged (total of three cages in six replications) in continuous maize.

Results

The impact of crop rotation on WCR population levels and the development of WCR populations in a rotational system can be better understood and interpreted if the population build-up data in continuous maize in the same area and experiment are available. Therefore, relevant data on this are included. (For more details on WCR population build-up in continuous maize in the rotation experiment, see Bayar et al., 2003.) chap10.qxd 24/11/04 1:44 PM Page 201

Western Corn Rootworm and Crop Rotation Systems in Europe 201

WCR population build-up in continuous maize versus rotated crops

Adult WCR population on Pherocon® AM traps In 2000, a total of 1466 WCR adults were collected on 144 Pherocon® AM traps in the crop rotation experiment. A majority of adults (98.98%) were caught in maize stands. Few WCR adults were trapped in other crop stands (0.55% for sunﬂower, 0.27% for oats and 0.20% for soybean). WCR adult numbers averaged 273 and 294.5 beetles per six plots on three traps each in continuous maize (all maize plots were continuous due to maize pre-crop in 1999). Maximum densities were between one and two adults per trap per week at the peak, which was still far from the level of an economic population. First-year results demonstrated a very strong adult preference for maize ﬁelds at the WCR population densities measured. In 2001 and 2002, the total number of WCR adults trapped in Pherocon® AM traps in one crop stand ranged from 273 (maize in rotation) to 1228 (continuous maize) (Table 10.1). The WCR adult population consisted of 17–48% females, which created the overwintering egg population (Table 10.1). This population level was already close to an economic population for the following year’s larval damage.

Table 10.1. Total number of WCR adults on Pherocon® AM traps (n = 18) in rotated and continuous maize over the growing period (June–September) with percentage of females, 2001 and 2002, Szeged, Hungary.

Pre-crop Cereal Soybean Sunﬂower Maize

Crop Maize Maize Maize Maize 2001 273 (44%) 353 (39%) 339 (40%) 1228 (48%) 2002 314 (45%) 346 (39%) 276 (37%) 1102 (17%)

Emerging WCR adult population in maize after different pre-crops In 2001, a total of five, four and 98 WCR adults emerged in maize with cereal, soybean and maize (continuous maize) pre-crops, respectively. The mean numbers of WCR adults that emerged in maize fields after oat, soybean and maize (continuous maize) were 0.83 (± 1.6), 0.66 (± 0.51) and 16.33 (± 11.36), respectively. No adults emerged in maize fields after sunflower (Table 10.2). In 2002, a higher number of adults were caught in all cases in emergence cages in maize after different pre-crops compared to the same rotations in 2001. A total of 29, 41 and 19 WCR adults emerged in rotated maize (cereal, soybean and sunflower pre-crops), respectively. From continuous maize, a total of 263 adults were recorded in emergence cages. The mean numbers of adults that emerged from maize after winter wheat, soybean, sunflower and maize were 4.83 (± 1.72), 6.83 (± 3.31), chap10.qxd 24/11/04 1:44 PM Page 202

202 J. Kiss et al.

Table 10.2. Within emergence cages the total number of WCR adults emerging from soil in rotated and continuous maize in 2001 and 2002 (% females in parentheses), Szeged, Hungary. Numbers followed by different letters are signiﬁcantly different within the same year (Student t and/or Welsh test, P = 0.01).

Pre-crop Cereal Soybean Sunﬂower Maize

Crop Maize Maize Maize Maize 2001 5 (60%) a 4 (100%) a 0 98 (58%) b 2002 29 (100%) ab 41 (88%) a 19 (95%) b 263 (60%) c

3.16 (± 2.78), and 48.83 (± 12.35), respectively. The emerging adult population was still low (Table 10.2).

WCR egg population in the soil in maize

In 2000, egg sampling yielded a total of zero in continuous maize among non-maize crops and two eggs in contiguous maize. In 2001, the number of samples was increased in continuous maize (n = 24). From contiguous maize only four random samples (500 cm3 each) were taken. A total of seven eggs were found in continuous maize plots in contiguous maize, while three eggs were found in continuous maize among the non-maize crop strips. Based on frequency of eggs sampled, the estimated egg population densities in 2000 and 2001 ranged from zero to 0.20 eggs/soil sample in continuous maize among non-maize crops, and from 0.31 to 0.11 eggs/soil sample in contiguous maize. As estimated, the quantities of zero and 250,000 eggs/ha in continuous maize among non-maize crops, and 137,500 to 387,500 eggs/ha in contiguous maize were present in the soil to a depth of 16 cm in continuous maize (Table 10.3).

Table 10.3. Density-related parameters of the egg population of Diabrotica virgifera virgifera in continuous maize (sampling dates: 30 August 2000; 24 September 2001), Szeged, Hungary.

Estimated Estimated Total egg density egg k Probability number per sample quantity Year Field value (θ) of eggs λ per ha

2000 Maize strips – 0.00 (n = 12) 0 0.00 – Contiguous maize 1.56 0.25 (n = 8) 2 0.31 387,500 2001 Maize strips 0.16 0.13 (n = 24) 7 0.20 250,000 Contiguous maize 0.21 0.08 (n = 24) 3 0.11 137,500 chap10.qxd 24/11/04 1:44 PM Page 203

Western Corn Rootworm and Crop Rotation Systems in Europe 203

WCR larval population

Larval activity In 2001, low levels of WCR larval injury were measured on the roots of rotated maize where pre-crops were cereal, soybean and sunflower. Using the Iowa 1–6 scale, the larval activity on maize roots resulted in damage ratings of 1.3 (± 0.30) after oats, 1.08 (± 0.10) after soybean, 1.37 (± 0.28) after sunflower and 2.68 (± 0.49) in continuous maize (Table 10.4). Significantly higher (Student t test, (x + 0.5)1/2 transformed data, P = 0.01) damage was noted on roots of continuous maize compared to maize after non-maize crops. In continuous maize, the root damage rating yielded a lower larval feeding injury score than the established economic threshold level of 2.5–3.5 (Iowa 1–6 scale) for the US Corn Belt. In 2002, the larval activity on the maize root system was the same as that observed in 2001. Root damage ratings were made on maize roots taken for the second larval sampling. Larval injury determinations, calculated as the average of ten samples in one crop plot, were 1.53 (± 0.24) in maize fields after winter wheat, 1.49 (± 0.26) after soybean, 1.50 (± 0.13) after sunflower and 2.96 (± 0.28) in continuous maize (see bottom line in Table 10.4). Damage on roots in continuous maize fields differed significantly from that after non-maize crops (P = 0.01), but there was no significant difference in damage on maize roots after soybean, sunflower and winter wheat. In the case of continuous maize, the larval injury to the roots approached the economic threshold (Table 10.4). No plant lodging was observed in the crop rotation field experiment from 2000 to 2002.

Larval population density In 2001, soil samples were taken in continuous maize ﬁelds to determine the WCR larval density in the soil. In 2002, however, all maize plots were sampled (root system of maize plants with soil) twice for WCR larvae regardless of the pre-crops.

Table 10.4. WCR larval activities in rotated and continuous maize based on the root injury index (1–6 scale) (± STDEV, average of ten maize roots, six replications), Szeged, Hungary. Means followed by different letters are signiﬁcantly different within the same year (Student t and/or Welsh test, P = 0.05).

Pre-crop Cereal Soybean Sunﬂower Maize

Crop Maize Maize Maize Maize 2001 1.30 ± 0.30 a 1.08 ± 0.10 b 1.37 ± 0.28 a 2.68 ± 0.49 c 2002 1.53 ± 0.24 a 1.49 ± 0.26 a 1.50 ± 0.13 a 2.96 ± 0.28 b

SEM, standard error of the mean. chap10.qxd 24/11/04 1:44 PM Page 204

204 J. Kiss et al.

Table 10.5. Density-related parameters of the larval population of Diabrotica virgifera virgifera in continuous maize (n = 60; sampling dates: 14 June 2001; 15 June 2002), Szeged, Hungary.

Highest Estimated Estimated larval number of larval density quantity per k Probability larvae per per maize ha at 60,000 Year Field value (θ) maize plant plant λ seeding rate

2001 Maize strips 0.21 0.30 11 0.90 54,000 Contiguous maize 0.72 0.10 2 0.11 6,600 2002 Maize strips 0.90 0.60 5 1.56 93,600 Contiguous maize 1.04 0.72 12 2.49 149,600

In 2001, larval sampling yielded a total of six and 36 larvae, while the highest larval numbers were two and 11 per maize plant in contiguous maize and in maize in strips with non-maize crops, respectively. Based on frequency of larval occurrence, the estimated larval population densities were 0.11 and 0.90 larvae per maize plant in contiguous maize and in maize in strips with non-maize crops. Consequently, the estimated quantities of 6600 and 54,000 larvae/ha occurred to the soil depth of 16 cm over a field at the 60,000 seeding rate (Table 10.5). In continuous maize, a total of 51 and 147 larvae were found in maize strips with non-maize crops and in contiguous maize, respectively. The highest larval numbers per maize plant were five and 12, respectively. Based on frequency of occurrence, the estimated larval population densities were 1.56 and 2.49 larvae/maize plant. Consequently, the estimated quantities of 93,600 and 149,600 larvae/ha occurred to the soil depth of 16 cm on the field at the 60,000 seeding rate (Table 10.5). Most immature stages had completed their development to the adult stage by mid-July in 2002 since only one pupa was found in 120 soil samples on 14 July in contiguous maize. The number of larvae found in continuous maize was significantly higher ((x + 0.5)1/2 transformed data, P = 0.01) than in maize with non- maize pre-crops. Larval numbers found in maize after winter wheat, soybean and sunflower did not differ significantly from each other.

WCR population in non-maize crops

WCR population, especially adult activity, near the soil surface in crops other than maize was one of the focuses of the crop rotation experiment. The development of the rotation-tolerant variant in the USA and the unknown origin of the European WCR population raised the question as to how the European WCR adults will behave under the local crop rotation system. chap10.qxd 24/11/04 1:44 PM Page 205

Western Corn Rootworm and Crop Rotation Systems in Europe 205

Table 10.6. Mean seasonal trap capture of WCR adults in three Pherocon® AM traps in different crops and in maize (± STDEV, six replications), Szeged, Hungary. Means followed by different letters are signiﬁcantly different within the same year (Student t and/or Welsh test, P = 0.05).

Year Cereal Soybean Sunﬂower Maize

2000 0.66 ± 1.63 a 0.5 ± 0.83 a 1.33 ± 0.51 a 45.5 ± 20.71 b 2001 40.16 ± 7.13 a 3.66 ± 2.06 b 13.33 ± 13.1 c 551.0 ± 177.7 d 2002 5.66 ± 3.44 a 0.16 ± 0.4 b 2.5 ± 2.58 c 74.83 ± 28.44 d

Adult activity on Pherocon® AM traps in non-maize crops During 2000 the mean number of WCR captures over the growing season in four different crop stands (oats, soybean, sunflower and maize strips) was 0.66 (± 1.63) in oats, 0.5 (± 0.83) in soybean, 1.33 (± 0.51) in sunflower and 45.5 (± 20.71) in maize, respectively (Table 10.6). The population change and sex ratio could not be demonstrated in non-maize crops due to the low WCR population density (Table 10.7). The population peak of WCR adults on 19 July was in maize plots, at the blister stage (R2) of maize. In July on four sampling dates, 151 males and 60 females (2.51:1) were caught in 18 Pherocon® AM traps, while in August on five sampling dates four males and 29 females (0.13:1) were caught in the same number of traps. WCR females removed from Pherocon® AM traps and dissected showed the presence of gravid females from 6 July to 24 August in maize. In soybean, one gravid WCR with 59 eggs was found on 27 July. In 2001, the WCR population increased and a total of 3306 WCR beetles were found in 72 Pherocon® AM traps over the growing period, established in four different crops. A total of 9.4% of the WCR adult population (343 adults) was trapped in non-maize crop strips. The mean number of WCR beetles in three Pherocon® AM traps (one plot) in different crops stands for the whole growing season was 40.16 (± 7.13) in winter wheat, 3.66 (± 2.06) in soybean, 13.33 (± 13.1) in sunflower and 551 (± 177.7) in maize (Table 10.6). The number of WCR

Table 10.7. Summary of total seasonal trap captures of WCR adults in Pherocon® AM traps (n = 18) in different crops and in maize in 2000, 2001 and 2002 (% of females in parentheses), Szeged, Hungary.

Year Cereal Soybean Sunﬂower Maize

2000 4 (75%) 3 (33%) 8 (62%) 273 (33%) 2001 241 (54%) 22 (55%) 80 (51%) 3306 (52%) 2002 34 (71%) 1 (100%) 15 (40%) 449 (21%) chap10.qxd 24/11/04 1:44 PM Page 206

206 J. Kiss et al.

beetles caught by Pherocon® AM traps in non-maize crop stands was significantly less compared to maize, but also differed from each other (P = 0.01). The sampling period in 2001 started at the beginning of the maize silking stage (R1) (6 July). The population peak of WCR adults in maize fields was on 27 July, at the milk stage (R3) of maize. In winter wheat, the population peak was somewhat later, 10 August, while in sunflower it was on 3 August. In soybean, no defined population peak was observed. The ratio of males to females fluctuated over the growing season, but the percentage of females was higher until the end of the growing season from 17 August in soybean, sunflower and maize, while in winter wheat the percentage of females started to increase on 24 August. In 2001, more beetles were caught in all crop stands; therefore these data were more representative of the expected sex ratio. In winter wheat, male WCR beetles were 2.38 times more abundant than females (2.38:1) in July, while in August the ratio of females was higher (0.74:1). In soybean in both months (July and August), the ratio of females to males was higher (1:0.75 and 1:0.87), although the ratio of males was increasing in August. In sunflower, the ratio of females was higher in July (1:0.3) and decreased for August (1:1.16). In maize, the situation was similar to that in sunflower; thus the ratio of females to males was higher in July (1:0.83) than in August (1:1.07). Based on the trap captures of Pherocon® AM traps (female dissection), gravid WCR were present from 20 July to 31 August in maize. A total of three gravid WCR with 35–86 eggs were caught on 17 and 24 August in the cereal crop. In 2002, the total number of WCR adults caught by Pherocon® AM traps in all crop stands declined six-fold compared to 2001, as a total of 499 beetles were caught in maize among the non-maize crop strips. The mean WCR abundance in different crop stands in winter wheat, soybean, sunflower and maize was 5.66 (± 3.44), 0.16 (± 0.4), 2.50 (± 2.58) and 74.83 (± 28.44), respectively (Table 10.6). WCR abundance was significantly different in all crop stands, and significantly more beetles were caught in maize than in other crops. Only males were detected on the first sampling date on 21 June. The silking period (R1) of maize started on 28 June, in the third week of the sampling period. The population peak of WCR was on 12 July in maize and sunflower. In winter wheat, the population peak occurred 2 weeks later. Only one female was caught in 18 Pherocon® AM traps over the growing period in soybean on 26 July. In winter wheat, more male WCR beetles were present in July, but in August more females were detected. The ratio of males to females was 1.5:1 and 0.06:1, respectively. In sunflower, the ratio was 1.8:1 in July, and in August no male beetles were caught. In maize, the ratio was 4.98:1, while in August the ratio of females was higher (0.61:1). chap10.qxd 24/11/04 1:44 PM Page 207

Western Corn Rootworm and Crop Rotation Systems in Europe 207

Emerging WCR adult population in non-maize crops Over the 3-year period, only one WCR beetle was found in the emergence cage in winter wheat (17 August 2001). This adult may have developed on the root system of winter wheat or may have crawled under the emergence cage from the outside through soil breaks, etc.

WCR egg population in non-maize crops

In 2001, no WCR eggs were found in oats, soybean and sunﬂower. In 2002, no egg samples were taken, because the experiment in 2001 showed the low effectiveness of this sampling method at the population levels that occurred in the project ﬁeld.

Discussion

WCR population in continuous maize

Adult activity The adult WCR population density as measured by Pherocon® AM traps showed a 4.5–12-fold increase from the year 2000 to 2001 and some decrease by year 2002 in continuous maize plots. The adult WCR population was significantly less in maize rotated with cereal, soybean and sunflower than in continuous maize in all years of the experiment. The percentage of WCR females, which lay the eggs that hatch in the subsequent year’s maize, was highly variable. According to our data, much lower numbers of adults were observed in rotated maize when compared to continuous maize. Therefore it is assumed that the largest portion of the WCR adult population has a preference for continuous maize stands in Hungary. Consequently, a crop rotation system including cereals, soybean and sunflower is effective at this time in minimizing WCR populations.

Adult emergence WCR adult emergence showed an increase of five to 12 times from the year 2000 to 2001 and remained at almost the same level for 2002. Thus, the WCR population change (adult density) over the 3 years of the experiment was very similar, measured either by adult activity (Pherocon® AM traps) or by adult emergence (emergence cages). A significantly lower number of adults emerged from the soil in rotated maize than in continuous maize. The relative adult emergence varied from 4.0 to 15.5% in rotated maize compared to that in continuous maize (Table 10.8). Adult emergence also confirms that a crop rotation chap10.qxd 24/11/04 1:44 PM Page 208

208 J. Kiss et al.

Table 10.8. Frequency and relative adult WCR emergence in rotated and continuous maize, 2001–2002, Szeged, Hungary.

Number of cages with adult Frequency of adult Relative adult emergence emergence % (F) emergence % (R) Pre-crop Crop 2001 2002 2001 2002 2001 2002

Cereal Maize 3 14 16.6 77.7 5.0 11.0 Soybean Maize 4 16 22.2 88.8 4.0 15.5 Sunﬂower Maize 0 10 0.0 55.5 0.0 7.2 Maize Maize 16 18 88.8 100.0 100.0 100.0

system including cereals, soybean and sunflower is effective in minimizing the WCR population. Initial emergence of WCR adults in the plot area was delayed by 2 weeks, and total emergence was reduced in 2001 compared to 2002. A few weather factors, including temperature and wind, were significantly (P < 0.05) correlated with percentage population change in the USA, but none of these factors produced consistently significant correlations across consecutive sampling dates. Therefore, none could be shown to be related to population change (Hein et al., 1988). Based on the age-specific life table of WCR, starting at the egg density of 1000 laid, on average 7.2 adults emerged from the soil between 2000 and 2002 (Toepfer and Kuhlmann, Chapter 5, this volume).

WCR egg population

A mean fecundity of 226 (± 133.4 standard deviation (STDEV) and 85 (± 57.8) eggs per female was recorded over a period of 53 days in 2001 and 42 days in the laboratory in 2002, respectively. Also, results from project laboratory rearing confirmed that the field-collected gravid WCR laid two-thirds of their eggs during the first 3 weeks of oviposition (Bayar et al., 2002). In our case, extremely low egg population densities may be the result of heterogeneity of WCR eggs over the targeted field, thus making it impossible to reliably estimate autumn egg populations. Similar egg samplings in maize fields in Croatia, Serbia and Montenegro confirmed that it is very difficult to find WCR eggs in soil samples taken from maize in infested areas (Kiss and Edwards, 2002). Despite this, the subsequent adult population that emerged from the soil indicates a high egg per larval population density. In recently infested areas in Europe, the first reaction of entomologists is to survey WCR egg populations to predict the next year’s expected larval density either for field trial purposes or to assess the risk of larval damage to the next year’s maize and to decide if control is necessary. Heterogeneity of the adult WCR density over maize fields and that of their eggs in the soil result in biased egg sampling results. In chap10.qxd 24/11/04 1:44 PM Page 209

Western Corn Rootworm and Crop Rotation Systems in Europe 209

addition to this, egg washing techniques are difficult to use under normal farming conditions. Simple techniques, however, show poor efficacy. Zs. Újvari, P. Hoffmann, J. Kiss and G. Vörös (2002, unpublished results) in Hungary recovered 30% of the eggs from a liquid solution (counted number of WCR eggs added to water and filtered with a 0.25 mm screen) and recovered 3.90% of the eggs mixed with soil and then filtered.

WCR larval population The estimated larval population density per hectare ranged from 6600 to 54,000 in 2001, thus resulting in 93,600 to 149,600 larvae/ha at 60,000 seeding rate in continuous maize in southern Hungary. The number of larvae recovered was not a function of sample volume, and most larvae are aggregated in the upper 10 cm of the maize root system (Bergman et al., 1981). Variability of spatial distributions may be in part the result of behavioural and mortality factors operating at different points in the seasonal life cycle of corn rootworms (Ellsbury et al., 1998). The movement of the larvae in the soil is limited. In laboratory tests, the oligophagous larvae of WCR could not, by olfaction, distinguish the difference between the roots of host (maize) and non-host plants (soybean and sunﬂower). Tests in a soil olfactometer indicated that the larvae aggregated around an odour of source (plant roots) probably by being arrested towards the source rather than by moving randomly, thus ﬁnding the odour source by chance, and then stopping at the source by an arrestant (Branson, 1982).

WCR population and crop rotation

In order to develop a management strategy for a sustainable crop rotation system for suppressing the WCR population level over time and space, we have to approach the WCR problem in three steps:

The first step is to follow WCR development and verify the biological evidence that WCR development happens in the crop rotation system in our experiment as follows. Specifically we have to address the questions whether: (i) WCR adults do immigrate/appear in non-maize crop stands; (ii) WCR females lay (at least part of) their eggs in non-maize crop stands; (iii) WCR larvae develop on the root system of maize after non-maize crops; and finally (iv) WCR adults emerge from maize after non-maize crops. The second step is to interpret project plot experiment results to a larger scale (farm) level. The third step is to search for signs and processes of adaptation and analyse factors that may influence it. chap10.qxd 24/11/04 1:44 PM Page 210

210 J. Kiss et al.

Do WCR adults immigrate/appear in non-maize crop stands? Yes! Pherocon® AM traps did demonstrate WCR adult presence and activity near the ground surface in winter wheat, soybean and sunflower. The WCR population in the project experiment (mainly in 2001 and 2002), measured by capturing adults on Pherocon® AM traps, approached the threshold level established for economic larval damage in next year’s maize in the USA. At this population density, a smaller proportion (5–10% of the adult population) was caught outside maize. In each year, significantly higher numbers of WCR adults were trapped in maize compared to any other crop in the rotation system. WCR adults showed a strong preference for maize in all 3 years of the experiment. Pherocon® AM traps (which visually attract WCR adults) were covered by dense sunflower foliage in July and August or by soybean foliage. Therefore, adults caught by the traps in these crop stands were actively flying into these plots. In the case of winter wheat, one cannot speak about the plant stand from mid-July under our conditions. Since winter wheat was harvested around mid-July, the traps were visible from a distance, which may have contributed to increase adult numbers in traps on harvested winter wheat plots. Among non-maize crop stands, there was significant difference for adult numbers except in the year 2000. In 2001 and 2002, WCR adult density for the whole growing season was the highest in harvested winter wheat plots, followed by sunflower and soybean. Making observations of WCR adult activity in non-maize crops is justified when checking the frequency of adult WCR captures of all traps over the growing period. Frequency of capture in different crop stands over the whole season in years 2001 and 2002 was 61.1 and 5.5% in soybean, 88.8 and 44.4% in sunflower and 100 and 88.8% in winter wheat (100% means all 18 traps in the crop stand recorded WCR adults (Table 10.7). Adult density in non-maize crops over the whole season, however, was very low (0.66 to 7.57%), while captures in maize were 100% (Table 10.9). At the population density in the project experiment, there was no dis- tinguishable adult density shift from maize to non-maize crops in the growing period of maize or after silking of maize. Certainly, observations showed that WCR adult numbers increased in harvested winter wheat plots on traps and on volunteer wheat plants (feeding adults). This phenomenon raises the important question of the possibility that harvested winter wheat fields could be potential egg-laying sites for WCR females in Europe. However, the proportion of egg-laying WCR females within the population after harvesting winter wheat is still a question. Since harvest time and WCR phenology vary from one region to another in Europe, this phenomenon should be considered and checked on a regional basis in the future. chap10.qxd 24/11/04 1:44 PM Page 211

Western Corn Rootworm and Crop Rotation Systems in Europe 211 AM traps in different pre-crops and maize, 2000, 2001 2002, Szeged, ® for the whole sampling WCR adult Frequency of Relative adult Total capture in 18 traps ® AM No. of traps/crop period (WCR adult appearance in adult captures captures traps/crop with captures number) different crops % (A) % (F) % (R) Frequency of adult WCR captures in Pherocon No. Pherocon Hungary. Table 10.9. Crop 2000 OatsSoybeanSunflower 18Maize 18 18Crop 2001 Winter wheatSoybean 18 7 18Sunflower 18Maize 3 18 3Crop 2002 Winter wheat 18 18Soybean 16 18 18Sunflower 18Maize 11 8 18 1.09 3 4 18 16 18 273 8 241 80 1 22 0.60 16.66 18 2.93 3306 34 15 2.19 1.46 94.79 61.1 6.60 1.09 1 38.88 449 88.88 16.66 90.60 100 100 0.66 6.81 3.00 2.93 2.41 0.22 1.46 100 100 89.97 7.28 88.88 44.44 100 5.55 100 7.57 3.34 0.22 100 chap10.qxd 24/11/04 1:44 PM Page 212

212 J. Kiss et al.

Do WCR females lay their eggs in non-maize crops? Yes! Although we have not sampled WCR eggs in non-maize plots (and few eggs in continuous maize plots), sampling of larvae and adult emergence in maize after non-maize crops showed that egg laying did happen in non-maize crops. As already mentioned, at moderate WCR population density the heterogeneity of egg laying makes egg sampling very difficult and of low efficacy. At higher WCR population levels (Illinois, USA), 15 WCR eggs were found in 38 soil samples (500 cm3 each) in two soybean seed production fields. Thus, a total of 5.7 million eggs/ha were laid by WCR females on the two fields (Levine and Oulami-Sadeghi, 1996), while we estimated experiment plot egg population density ranging from 137,500 to 387,500 eggs/ha. In 1997, an average of 1.10 (± 0.70), 4.00 (± 1.80) and 1.70 (± 1.20) viable WCR eggs were found in 12 samples in maize, soybean and lucerne fields, respectively (Levine et al., 2002). It is expected that at higher WCR population density or with improved sampling techniques WCR eggs could be sampled from non-maize crops. This method, however, is labour- and time-intensive and therefore cannot be suggested under normal farming conditions.

Does WCR larval development happen in maize after non-maize crops? Yes! Sampling (18 maize plots after non-maize pre-crops, two times, ten samples per plot each) amounting to a total of 360 maize plant (root and soil) samples in 2002 resulted in four larvae (one each after winter wheat and sunﬂower and two after soybean). Again, as in the case of egg population, this density was rather low. One should note that the L1 stage practically cannot be found with the sampling method used due to their small size or being in the root, and the same applies for the early developmental period of L2. Therefore, the small number of larvae found was not unexpected. This larval sampling method was effective at higher population densities in continuous maize, resulting in 51 and 147 larvae in maize among non-maize crop strips and in contiguous maize, respectively. The estimated WCR larval population density in continuous maize was still low, approximately one-tenth of the economic level (eight to 12 larvae per plant, thus an average of 600,000 larvae/ha) as established in the USA. WCR larval presence and activity were also assessed by root damage ratings. As expected from adult densities observed, root damage to continuous maize was signiﬁcantly higher in 2001 and 2002 than in 2000. Larval activity (root damage) in maize after different non-maize pre- crops varied year by year. At observed population densities, no reliable evidence can be demonstrated for the effect of the pre-crop on the next year’s maize root damage by WCR larvae. In Illinois in the location of the rotation-tolerant WCR, maize planted after soybean sustained greater larval injury (2.70 (± 0.10, n = 20)) than maize planted after wheat (1.90 (± 0.10, n = 20)). However, when roots were additionally sampled 2 weeks later, injury had increased to a higher degree in the maize planted after chap10.qxd 24/11/04 1:44 PM Page 213

Western Corn Rootworm and Crop Rotation Systems in Europe 213

wheat than in maize after soybean. In maize ﬁelds after oats, larval injury to roots (1.80 (± 0.10, n = 20)) was observed (Levine et al., 2002).

Do WCR adults emerge from the soil of maize after non-maize crops? Yes! Since various pre-crops were available for egg laying for the first time in 2001, data from 2001 and 2002 can be used to evaluate this issue. In the 2001 and 2002 growing seasons, a total of 98 WCR adults emerged in maize after non-maize crops the previous year within a total of 54 emergence cages. Due to the relatively high adult population in 2001, the number of emerged adults was higher in 2002 when compared to 2001. Determining WCR adult emergence over time in maize after non- maize crops was possible through checking the frequency of adult WCR emergence in all cages over the growing period. Frequency of captures in maize after different crop stands over the whole season in years 2001 and 2002 was 0 and 55.50% in maize after sunflower, 16.60 and 77.70% after cereals and 22.20 and 88.80% after soybean. Adult density in maize after non-maize crops over the whole season, however, was very low (0 to 15.50%) when compared to captures in maize of 100%. Continuous maize always maintained significantly higher adult emergence numbers compared to maize after non-maize pre-crops. For the effect of the pre-crop on the next year’s adult population, there was no difference between winter wheat and soybean, while sunflower resulted in significantly fewer adults compared to soybean. The WCR adult population density in maize after non-maize pre- crops, even the highest one, soybean, over the whole season (5.40 adults/m2) is still far from the level experienced in emergence cages in maize fields after soybean (a maximum of 23 and 15 beetles/m2) in 1996 and 1997, respectively, in Indiana, USA (Barna et al., 1998). All of the above points to the fact that part of the WCR population immigrates to non-maize crop stands and lays eggs, and in the subsequent maize the larvae develop (thus damaging maize roots) and adults emerge. Various pre-crops may have different impacts on the various stages of WCR. As demonstrated by Hatvani and Horváth (2002), sunflower blossoms are likely to be attractive for adults. We have also observed this type of attractiveness for adults by volunteer winter wheat. Foliage, soil and microclimatic conditions within any crop stand, however, may have different impacts on WCR behaviour and development. Thus, different pre- crops may have different impacts on WCR population levels. Since the greatest damage to maize roots is caused by the L2 and L3 stages and the mortality of L2, L3 and pupal stages can be determined, it is possible to determine the potential impact of WCR adult numbers in non-maize pre- crops (using Pherocon® AM traps) and these data can be correlated to the subsequent year’s WCR adult density (using emergence cages). Based on correlation equations (Table 10.10), WCR adults strongly prefer maize plots. In all cases, there was a strong correlation between the captures in Pherocon® AM traps in pre-crops and the captures in emer- chap10.qxd 24/11/04 1:44 PM Page 214

214 J. Kiss et al.

Table 10.10. Correlations of WCR adult population densities (quantiﬁed by Pherocon® AM traps, x axis) over the whole growing season in the rotated crops to subsequent year initial adult population (emergence cage, y axis) in maize (sum of 2001 and 2002), Szeged, Hungary.

Crop Linear regression R2

Cereals y = 0.1028x + 0.7342 0.6914 Sunﬂower y = 0.1900x + 0.1897 0.6775 Soybean y = 1.5706x + 0.4780 0.7957 Contiguous maize y = 0.0517x + 14.674 0.6731

gence cages in the subsequent year’s maize. The correlation coefficients were 0.83, 0.89, 0.82 and 0.82 for cereal, soybean, sunflower and maize, respectively. In all cases, there was a significant difference (P = 0.01) when comparing the empirical (r = 0.68) and calculated correlation coefficients. There is a linear regression connection between the two data sets, as t > temp (t11= 4.73, 6.24, 4.58 and 4.53 in the case of cereal, soybean, sunflower and maize, respectively; in all cases temp = 3.106; P = 0.005). However, from the standpoint of maintaining the WCR population, cereals (oats and winter wheat) and sunflower were similar, while a few WCR adults in soybean plots resulted in the highest population in the next year’s maize. In other words, to produce the same number of WCR adults in maize, soybean plots as the pre-crop ‘needed’ the fewest number of adults (Table 10.11). These correlations were calculated based on a given population density, local environmental conditions and other biotic and abiotic factors. Therefore, these correlations must be verified over years and in other regions.

Table 10.11. Number of WCR adults in pre-crops (in Pherocon® AM traps) and in next year’s maize (under emergence cage or per plant) calculated by equations from Table 10.10, 2000–2002, Szeged, Hungary.

Cereal Soybean Sunﬂower Maize

2000/2001 on Pherocon® AM (in pre-crops) 100.00 100.00 100.00 100.00 In emergence cage 11.01 157.54 19.19 19.84 Per one maize plant 3.67 52.51 6.40 6.61

2001/2002 on Pherocon® AM (in pre-crops) 226.32 14.98 125.32 180.39 In emergence cage 24.00 24.00 24.00 24.00 Per one maize plant 8.00 8.00 8.00 8.00 chap10.qxd 24/11/04 1:44 PM Page 215

Western Corn Rootworm and Crop Rotation Systems in Europe 215

Interpretation of plot experiment results to a larger scale (farm level)

Population spread measured by the presence of WCR adults varies from 30 to 90 km/year in Europe (Kiss et al., 2001b; Kiss and Edwards, 2002). Thus, once present, there is a large expanse of area for WCR adults to find maize fields for feeding and egg laying. The project’s crop rotation experiment results originated from an established population in the Szeged area. The statement that part of the WCR adult population emigrates to other crop stands and lays eggs is valid under the above conditions and experiences with WCR. No data are available on this phenomenon in the case of the initial population. However, these findings should be valid in any region for 6–8 years after the WCR infestation is first noted. An important question is how small-scale research results relate to large- scale conditions. WCR adult density levels along maize/soybean transects in problem areas in the USA were high even in soybean fields 240 m from the nearest maize field (Levine et al., 2002). Újvari et al. (2002, unpublished results) detected numerous WCR adults in Pherocon® AM traps in a harvested winter wheat field 25, 50 and 75 m from continuous maize in Hungary in 2002. Komáromi et al. (2002) recorded a maximum adult capture of 27.60 per Pherocon® AM trap per week in a large maize field in 2001, while maximum adult captures on traps near ground level in a well-developed and dense soybean field, which was 15 m from a maize field, varied from 0.11 to 0.66 adults per Pherocon® AM trap per week. Also, emergence cages in a large (more than 100 ha) maize field after soybean caught a total of nine WCR adults over 9 weeks. Emergence cages in this maize field were 15–95 m from the field border (Komáromi et al., 2002). Therefore, the above data support the assumption that project findings under experiment conditions can be interpreted on the large field-scale level as well.

‘Signs and processes’ of adaptation and analysis of factors that may inﬂuence it

Under the observed WCR population density and experiment conditions, no special preference was observed for WCR population maintenance for any non-maize crop. Growers in the USA may have selected inadvertently for a new strain of WCR because of the routine practice of rotating maize with soybean (Gray et al., 1998). In some areas where the NCR and WCR occur together, there are striking differences in their survival under continuous maize and annual rotation regimes; rotation favours the NCR, and the planting of maize in the same field each year favours the WCR (Hill and Mayo, 1980). In contrast to the NCR, the WCR is poorly adapted to crop rotation, because it is unable to remain dormant for more than one winter (Krysan et al., 1984). Also, according to Hill and Mayo (1980), it is highly unlikely that the WCR could become adapted to crop rotation by oviposition in the alternate crop, because adults tend to remain within maize fields, i.e. the WCR for the most part move from maize to other maize fields. chap10.qxd 24/11/04 1:44 PM Page 216

216 J. Kiss et al.

Recent field experiments and results in USA have led to the assumption that the development of the rotation-tolerant variant is a consequence of the extremely high selection pressure caused by the rigid maize/soybean rotation system over a large area and extended time (Levine et al., 2002). The model of Onstad et al. (2001) suggests that the evolution of this new strain may have been the result of selection on a single gene for adult movement (leading to oviposition outside of maize fields) under a high level (> 80%) of an alternate-year rotation of crops. In some problem areas in USA, 89% of the total land area is under cultivation, 98% of that land is planted to either maize or soybeans (Kepley, 1999) and nearly all (98%) of the soybean fields are rotated to maize (Onstad et al., 2001).

Conclusions

The WCR adult population, which is near economic levels 6–8 years after the initial infestation, strongly prefers maize stands for adult feeding, egg laying and larval development. The WCR population in continuous maize showed a continuous increase, though the rate of increase can differ depending on environmental (rainfall or drought) conditions. Even under extreme dry summer and unfavourable soil conditions, the WCR population can increase the following year. Part of the WCR adult population can emigrate to other crop stands or to other fields (harvested). The reason for such emigration could be to feed on the flowers and pollen of the crop, such as sunflower, or to feed on volunteer winter wheat in harvested wheat fields. A small portion of the WCR adult population (5–15%) is active near the ground surface of these crop stands. Under the Food and Agriculture Organization (FAO) Network the same experiment was established in infested areas in Croatia (led by J. Igrc-Barc˘ic´), Serbia and Montenegro (led by I. Sivcev) and Romania (led by I. Rosca) from 2001 and 2002. Preliminary data from those experiments support our findings here (Kiss et al., 2001a; Kiss and Edwards, 2002; Kiss, 2003). Biologically, the immigration of WCR adults to other crop stands, adult feeding and/or egg laying in non-maize pre-crops and larval feeding on the roots in the next year’s maize were observed in the crop rotation experiment. Although there were differences among non-maize pre-crops regarding the robustness of correlations between the adult density of immigrat- ing individuals and the next year’s adult population density in maize, the differences were not enough to demonstrate pre-crop preferences by a critical number of adults for non-maize crops. There was no clear preference for non-maize pre-crops by WCR adults and no sign of any special adaptation of the WCR population to the crop rotation system. Rather, a random spread and movement of WCR adults chap10.qxd 24/11/04 1:44 PM Page 217

Western Corn Rootworm and Crop Rotation Systems in Europe 217

and subsequent egg laying happened under project experiment conditions. We assume that crop rotation systems in Europe will not result in the high selection pressure on the WCR at levels that have been seen in the US Corn Belt in the near future. Diverse landscape with a more diverse plant spectrum in Europe offers broader feeding sources and different egg-laying habitats compared to the landscape in the US Corn Belt, which is primarily a maize/soybean rotation system. We believe that diversity in space, as well as in time, may contribute to the lower possibility for the development of an adapted WCR population. Diversity in time refers to avoiding long-term, narrow and routine rotations of crops. This may be of special importance in the typical maize-growing areas in Europe where maize is cultivated on large areas, over long periods and in narrow rotation selections (one to two crops in rotation with maize). Existing IPM (Integrated Production and Pest Management (IPPM)) rules also support the above. While the rotation of maize with other crops is a primary control method of WCR populations and is strongly prescribed (by law or other types of regulations in some cases), especially in the population build-up phase, there are still major questions concerning the long-term management of WCR. If, theoretically, all maize plots (or a majority) will be rotated to non-maize crops, will the WCR population survive? This type of full rotation may pose a very high selection pressure. Therefore, the use of ‘refuges’ of continuous maize for 2 years within rotation areas to reduce selection pressure and preserve rotation as a means for managing the WCR population needs to be studied and policies developed if effective.

References

Atyeo, W.T., Weekman, G.T. and Lawson, D.E. (1964) The identiﬁcation of Diabrotica species by chorion sculpturing. Journal of the Kansas Entomological Society 37, 9–11. Ball, H.J. (1957) On the biology and egg-laying habits of the western corn rootworm. Journal of Economic Entomology 50, 126–128. Barna, G. (2001) Estimation of the population of Diabrotica virgifera virgifera LeConte and adaptation of decision making for treatment. (In Hungarian.) PhD thesis, Szent István University, Gödöllö, Hungary. Barna, G., Edwards, C.R., Gerber, C., Bledsoe, L.W. and Kiss, J. (1998) Management of western corn rootworm (Diabrotica virgifera virgifera) in corn based on survey information from previous soybean crop. Acta Phytopathologica et Entomologica Hungarica 33, 173–182. Barna, G., Edwards, C.R., Kiss, J., Gerber, C. and Bledsoe, L.W. (1999) Study of behavioral change of western corn rootworm beetle by crop and sex in maize and soybean ﬁelds in Northwestern Indiana, USA. Acta Phytopathologica et Entomologica Hungarica 34, 393–402. Bayar, K., Komáromi, J. and Kiss, J. (2002) Egg production of western corn rootworm (Diabrotica virgifera virgifera LeConte) in laboratory rearing. Növényvédelem (Plant Protection) 38, 543–545. chap10.qxd 24/11/04 1:44 PM Page 218

218 J. Kiss et al.

Bayar, K., Komáromi, J., Kiss, J., Edwards, C.R., Hatala-Zsellér, I. and Széll, E. (2003) Characteristics of a population of western corn rootworm (Diabrotica virgifera virgifera LeConte) in continuous corn. (In Hungarian.) Növénytermelés 52(2), 185–202. Bergman, M.K., Tollefson, J.J. and Hinz, P.N. (1981) Sampling scheme for estimating populations of corn rootworm larvae. Environmental Entomology 10, 986–990. Blackwell, R. (1997) Western corn rootworm sweep net survey in soybean. In: Pest Management and Crop Production Newsletter, 26 (19 September). Cooperative Extension Service, Purdue University, West Lafayette, Indiana, p. 3. Boller, E.F., Malavolta, C. and Jörg, E. (eds) (1997) Guidelines for Integrated Production of Arable Crops in Europe, Technical Guideline III. IOBC WPRS Bulletin 20 (5), 5–21. Branson, T.F. (1982) Olfactory response of larvae of Diabrotica virgifera virgifera to plant roots. Entomologia Experimentalis et Applicata 31, 303–307. Branson, T.F. and Krysan, J.L. (1981) Feeding and oviposition behavior and life cycle strategies of Diabrotica: an evolutionary view with implications for pest management. Environmental Entomology 10, 826–831. Branson, T.F. and Ortman, E.E. (1967) Host range of larvae of the western corn rootworm. Journal of Economic Entomology 60, 201–203. Branson, T.F. and Ortman, E.E. (1970) The host range of larvae of the western corn rootworm: further studies. Journal of Economic Entomology 63, 800–803. Clark, L.T. and Hibbard, B.E. (2003) Alternate hosts: Important factors in the western corn rootworm life cycle? In: Vidal, S. (ed.) International Symposium on Ecology and Management of Western Corn Rootworm, 19–23 January 2003, Göttingen, Germany, p. 36. Coats, S.A., Tollefson, J.J. and Mutchmor, J.A. (1986) Study of migratory flight in the western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 15, 620–625. Dobrin, G.C. and Hammond, R.B. (1985) The antifeeding activity of selected pyrethroids towards the Mexican bean beetle (Coleoptera: Coccinellidae). Journal of the Kansas Entomological Society 58, 422–427. Edwards, C.R. (2000) The interaction and impact of soil properties on corn rootworms. IWGO Newsletter 21(1–2), 5–6. Edwards, C.R., Obermeyer, J.L. and Bledsoe, L. (1996) What about controlling WCR beetles in soybean fields in the area where first year corn rootworm problems have been noted? Pest and Crop Newsletter No. 4, 2. Ellsbury, M.M., Woodson, W.D., Chandler, L.D., Malo, D., Clay, D.E., Carlson, C.G. and Clay, S.A. (1998) Spatial variability in corn rootworm distribution in relation to spatially variable soil factors and crop condition. In: Proceedings of the 4th International Conference on Precision Agriculture, St Paul, Minnesota, pp. 523–533. Gerrard, D.J. and Chiang, H.C. (1970) Density estimation of western corn rootworm egg populations based upon frequency of occurrence. Ecology 51, 237–245. Gerrard, D.J. and Cook, R.D. (1972) Inverse binomial sampling as a basis for estimating negative binomial population densities. Biometrics 28, 971–980. Gray, M.E., Levine, E. and Oloumi-Sadeghi, H. (1998) Adaptation to crop rotation: western and northern corn rootworms respond uniquely to a cultural practice. Recent Research Development in Entomology 2, 19–31. Hatvani, A. and Horváth, Z. (2002) Damage of western corn rootworm to sunflower in North Bacska (In Hungarian). Növényvédelem 38, 513–517. Hein, G.L., Tollefson, J.J. and Foster, R.E. (1988) Adult northern and western corn rootworm (Coleoptera: Chrysomelidae) population dynamics and oviposition. Journal of the Kansas Entomological Society 61, 214–223. Hill, R.E. and Mayo, Z.B. (1980) Distribution and abundance of corn rootworm species as chap10.qxd 24/11/04 1:44 PM Page 219

Western Corn Rootworm and Crop Rotation Systems in Europe 219

inﬂuenced by topography and crop rotation in eastern Nebraska. Environmental Entomology 9, 122–127. Hills, T.M. and Peters, D.C. (1971) A method of evaluating postplanting insecticide treatments for control of western corn rootworm larvae. Journal of Economic Entomology 64, 764–765. Hintz, A.M. and George, B.W. (1979) Successful laboratory hybridization of Diabrotica virgifera (western corn rootworm) and Diabrotica longicornis (northern corn rootworm) (Coleoptera: Chrysomelidae). Journal of the Kansas Entomological Society 52, 324–330. Hixson, E., Hill, R.E. and Muma, M.H. (1947) 60th Annual Report. Agricultural Experiment Station, University of Nebraska, Lincoln, 52 pp. Kepley, G. (1999) Illinois Agricultural Statistics: Annual Summary – 1999. Bulletin 99–1, Illinois Agricultural Statistics Service, Springﬁeld. Kiss, J. (2003) Final Report on Western Corn Rootworm (WCR), Diabrotica virgifera virgifera LeConte. Network Activity Report No. PR 21261, FAO. Kiss, J. and Edwards, C.R. (2002) Final Report on Western Corn Rootworm (WCR), Diabrotica virgifera virgifera LeConte. Network Activity Report No. PR 19713, FAO. Kiss, J., Bayar, K., Komáromi, J., Igrc-Barc˘ic´, J., Dobrinc˘ic´, R., Sivcev, I., Edwards, C.R. and Hatala-Zsellér, I. (2001a) Is the western corn rootworm adapting to the European crop rotation system? Results of a joint European trial. In: Proceedings Book of the XXI IWGO Conference and VIII Diabrotica Subgroup Meeting. Veneto Agricoltura, Legnaro, Italy, pp. 29–37. Kiss, J., Edwards, C.R., Allara, M., Sivcev, I., Igrc-Barc˘ic´, J., Festic´, H., Ivanova, I., Princzinger, G., Sivcev, I., Sivicek, P. and Rosca, I. (2001b) A 2001 update on the western corn rootworm, Diabrotica virgifera virgifera LeConte, in Europe. In: Proceedings Book of the XXI IWGO Conference and VIII Diabrotica Subgroup Meeting. Veneto Agricoltura, Legnaro, Italy, pp. 83–87. Komáromi, J., Bayar, K., Kiss, J., Edwards, C.R. and Széll, E. (2002) Is the development of western corn rootworm possible in corn/soybean and corn/alfalfa rotation systems? IWGO News Letter 24(1–2), 3. Krysan, J.L., Jackson, J.J. and Lew, A.C. (1984) Field termination of egg diapause in Diabrotica with new evidence of prolonged diapause in D. barberi (Coleoptera: Chrysomelidae). Environmental Entomology 13, 1237–1240. Levine, E. and Oloumi-Sadeghi, H. (1996) Western corn rootworm (Coleoptera: Chrysomelidae) larval injury to corn grown for seed production following soybeans grown for seed production. Journal of Economic Entomology 89, 1010–1016. Levine, E., Oloumi-Sadeghi, H. and Fischer, J.R. (1992) Discovery of multiyear diapause in Illinois and South Dakota northern corn rootworm (Coleoptera: Chrysomelidae) eggs and incident of the prolonged diapause trait in Illinois. Journal of Economic Entomology 85, 262–267. Levine, E., Spencer, J.L., Isard, S.A., Onstad, D.W. and Gray, M.E. (2002) Adaptation of the western corn rootworm to crop rotation: evolution of a new strain in response to a management practice. American Entomologist 48, 94–107. Lökös Tóth, K. (2002) Analysis of statistical hypothesis. (In Hungarian.) In: Szücs, I. (ed.) Applied Statistics. Agroinform Kiadó, Budapest, pp. 211–260. Ludwig, K.A. and Hill, R.E. (1975) Comparison of gut contents of adult western and northern corn rootworms in northeast Nebraska. Environmental Entomology 4, 435–438. Mendoza, C.E. and Peters, D.C. (1964) Species differentiation among mature larvae of Diabrotica undecimpunctata Howardi, D. virgifera, and D. longicornis. Journal of the Kansas Entomological Society 37, 123–125. O’Neal, M.E., Gray, M.E. and Smyth, C.A. (1999) Population characteristics of a western chap10.qxd 24/11/04 1:44 PM Page 220

220 J. Kiss et al.

corn rootworm (Coleoptera: Chrysomelidae) strain in east-central Illinois corn and soybean ﬁelds. Journal of Economic Entomology 92, 1301–1310. Onstad, D.W., Spencer, J.L., Guse, C.A., Levine, E. and Isard, S.A. (2001) Modeling evolution of behavioral resistance by an insect to crop rotation. Entomologia Experimentalis et Applicata 100, 195–201. Sammons, A.E., Edwards, C.R., Bledsoe, L.W., Boeve, P.J. and Stuart, J.J. (1997) Behavioral and feeding assays reveal a western corn rootworm (Coleoptera: Chrysomelidae) variant that is attracted to soybean. Environmental Entomology 26, 1336–1342. Shaw, J.T., Paullus, J.H. and Luckmann, W.H. (1978) Corn rootworm oviposition in soybean. Journal of Economic Entomology 71, 189–191. Siegfried, B.D. and Mullin, C.A. (1990) Effects of alternative host plants on longevity, oviposition, and emergence of western and northern corn rootworms (Coleoptera: Chrysomelidae). Environmental Entomology 19, 474–480. Southwood, T.R.E. (ed.) (1978) Ecological Methods with Particular Reference to the Study of Insect Populations, 2nd edn. Chapman & Hall, London, pp. 17–73. Spencer, J.L., Isard, S.A. and Levine, E. (1999) Free ﬂight of western corn rootworm (Coleoptera: Chrysomelidae) to corn and soybean plants in a walk-in wind tunnel. Journal of Economic Entomology 92, 146–155. chap11.qxd 12/11/04 1:20 PM Page 221

11 Application of the Areawide Concept Using Semiochemical- based Insecticide Baits for Managing the Western Corn Rootworm (Diabrotica virgifera virgifera LeConte) Variant in the Eastern Midwest

COREY K. GERBER,1 C. RICHARD EDWARDS,1 LARRY W. BLEDSOE,1 MICHAEL E. GRAY,2 KEVIN L. STEFFEY2 AND LAURENCE D. CHANDLER3 1Department of Entomology, Purdue University, W. Lafayette, Indiana, USA; 2Department of Crop Science, University of Illinois, Urbana 61801, Illinois, USA; 3RRVARC, USDA-ARS, Fargo, North Dakota, USA

Introduction

Western corn rootworms (WCR), Diabrotica virgifera virgifera LeConte, are the most destructive and costly pests of maize, Zea mays L., in the USA. Crop losses resulting directly from corn rootworm feeding damage, as well as management costs, have been estimated at US$1 billion a year (Metcalf 1986). In the eastern Midwest (Illinois, Indiana, Michigan, Ohio and Wisconsin), WCR adults emerge in late June until late August (Godfrey and Turpin, 1983; Levine and Gray, 1996). In late July, gravid females begin ovipositing in the soil, with peak oviposition occurring in early to mid- August (Hill, 1975). Female beetles typically lay clutches of approximately 80 eggs every 3 to 8 days (Hill 1975) in ﬁelds where a preferred food source, typically maize pollen and maize silks (Capinera et al., 1986), exists (Branson and Krysan, 1981). WCR eggs overwinter in the soil and larvae hatch in May and early June (Levine and Oloumi-Sadeghi, 1991). Corn rootworm larvae feed primarily on maize roots (Branson and Ortman, 1967,

222 C.K. Gerber et al.

1970), which can cause significant damage and considerable yield loss (Sutter et al., 1990; Gibb and Higgins, 1991; Spike and Tollefson, 1991). Historically, root damage caused by WCR larvae has been greatest in fields where maize is planted following maize (continuous maize). Before the early 1970s, a large percentage of maize was in continuous maize production. In the 1950s, producers primarily applied chlorinated hydrocarbon insecticides to the soil at planting to protect maize roots from corn rootworm larvae. When resistance to chlorinated hydrocarbons became apparent in the late 1950s (Ball and Weekman, 1962; Bigger, 1963), producers began managing WCR adults with aerial applications of non-chlorinated hydrocarbon insecticides (Meinke, 1995). During the 1960s, carbamate and organophosphate soil and foliar insecticides were used extensively either to manage rootworm larvae or to suppress rootworm adults. However, due to the greater level of management required for the success of foliar applications – primarily timing and placement of applications (Levine and Oloumi-Sadeghi, 1991) – and the overall management costs (scouting, application, insecticide), suppression of adults was not as economical as managing corn rootworm larvae with soil insecticides (L.W. Bledsoe, Entomology Department of Purdue University, personal communication). Therefore, with the exception of a few localized areas in the Midwest, producers relied heavily on the prophylactic use of soil insecticides. Due to the costs and environmental hazards associated with soil insecticides and comparable production costs and market values of maize and soybean, adoption of crop rotation of maize with soybean began to gain momentum in the early 1970s throughout the Midwest. Because WCR adults deposit most of their eggs in maize (Shaw et al., 1978), and because WCR larvae do not survive on soybean roots (Branson and Ortman, 1970), the annual rotation of maize and soybean crops disrupted the corn rootworm life cycle, precluding the need for soil insecticides. The reduction of soil insecticide use resulted in substantial economic savings to producers and a decrease in the deleterious effects of insecticides to the environment (Edwards, 1996). From the late 1970s to the early 1990s, the use of crop rotation was a major strategy for management of WCR. Before 1987, economic injury to first-year maize fields (maize planted following soybean and other non-maize crops, as well as fallow ground) by WCR was not observed, although low densities of WCR adults had been observed in soybean fields. In the summer of 1987, six seed maize fields grown in rotation with soybean near Piper City, Illinois, were severely damaged by WCR larvae (Levine and Oloumi-Sadeghi, 1996). In 1993, economic damage by WCR larvae in first-year maize was reported in northern Indiana (Edwards et al., 1994). Since the initial reports of WCR damage in first-year maize, the number of first-year maize fields damaged by WCR larvae has increased in east-central Illinois and northern Indiana (Sammons et al., 1997; Spencer et al., 1997). chap11.qxd 12/11/04 1:20 PM Page 223

Insecticide Baits for Managing Western Corn Rootworm 223

Researchers suggested that a variant of WCR beetle that prefers to oviposit in areas other than maize had been selected through the widespread adoption of crop rotation (Edwards, 1996; Levine and Gray, 1996; Gray et al., 1998; Levine et al., 2002). This change in behaviour eliminated crop rotation as an effective tool for managing corn rootworm in areas where the WCR variant population occurred. As a result of the discovery of the WCR variant, insecticide use for managing corn rootworms has increased dramatically in the eastern Midwest (Edwards, 2000). To avoid economic losses caused by the variant WCR, producers began applying soil insecticides in first-year maize fields. Due to this increase in soil insecticide use, pressure by the public and government agencies to protect groundwater and non-target organisms has increased. In addition, the possibilities of cancellation of some insecticides and revized regulations on insecticide use have stimulated the pursuit of environmentally sound rootworm management tactics. One of the most intensively studied, environmentally sound management tactics designed to target rootworm populations has been the use of semiochemical-based insecticide baits. Semiochemicals are chemicals that mediate the interactions between organisms (Nordlund, 1981). Semiochemical-based insecticide baits are formulated with one or more semiochemicals and insecticides (Metcalf et al., 1987). Cucurbitacins, which are rootworm adult feeding stimulants and arrestants (Metcalf et al., 1982), are the primary semiochemicals used in corn rootworm bait formulations. Because cucurbitacins evoke a compulsive feeding behaviour in adult corn rootworms, mixtures of cucurbitacins and reduced rates of insecticides provide effective control of corn rootworm adults. Metcalf et al. (1987) determined that broadcast applications of semiochemical-based insecticide baits that included approximately 0.1% cucurbitacins and 0.1% of a carbamate insecticide, effectively reduced adult corn rootworm populations in maize and cucurbit crops. The suppression of adult corn rootworms has primarily been attempted on a field-by-field basis. Due to the mobility of corn rootworm adults, a multi-hectare approach, which can include several individual fields, has been tested to manage rootworm adults. Pruess et al. (1974) attempted to manage WCR adults over a 41.5 km2 area over a 3-year period. ULV malathion was aerially applied to the management area prior to peak beetle activity. Results of this study indicated that beetle populations could be reduced the following growing season; however, the area suppression approach did not appear to be economically feasible. In the early 1990s, the concept of areawide suppression of corn rootworm populations was again tested; however, semiochemical insecticide baits were the primary rootworm management tactic used to suppress rootworm adults. By 1995, the US Department of Agriculture Agricultural Research Service (USDA-ARS) developed a management concept of suppressing corn rootworm adults before oviposition with semiochemical- based insecticide baits applied over a well-defined geographical region chap11.qxd 12/11/04 1:20 PM Page 224

224 C.K. Gerber et al.

(Chandler and Sutter, 1995). The concept of regional pest management, or areawide pest management (AWPM), described by E.F. Knipling (1978), focuses on the management of a single pest throughout a relatively large geographical area in order to reduce the abundance of that pest population. Chandler and Sutter (1995) reported that the management of corn rootworm beetles with semiochemical baits was efficacious when the baits were applied over a relatively large area. In 1996, USDA-ARS implemented five large-scale corn rootworm AWPM programmes in the USA (Chandler and Faust, 1998). The five programmes were located in Indiana and Illinois (Buhler et al., 1998), Iowa (Tollefson, 1998), Kansas (Wilde et al., 1998), South Dakota (Chandler and Hartman, 1998) and Texas (Coppedge et al., 1997). The programme sites were selected based on the uniqueness of each site, including geography, production practices and pest complexes. To evaluate the corn rootworm AWPM concept on the WCR variant located in north-western Indiana and east-central Illinois, with the use of semiochemical-based insecticide baits as the primary management strategy, four objectives for this programme were established. The first objective was to determine if a consistent decrease in the area requiring treatment with semiochemical-based baits occurred during the entirety of the programme. The second and third objectives were to determine if a consistent decrease in root damage by WCR larvae and a consistent decrease in WCR beetle emergence were detected from the beginning to the end of the programme. The fourth objective was to determine if the use of semiochemical-based insecticide baits to manage WCR adults provided a consistent yield benefit during the conduct of the programme. These objectives were established to determine if the AWPM concept was efficacious against the WCR variant.

Materials and Methods

AWPM programme time line

The Indiana/Illinois AWPM programme was initiated in 1996. Data were collected from 1997 to 2002. The programme was completed in 2003.

AWPM evaluation site

The Indiana/Illinois AWPM evaluation site (Fig. 11.1) was established in 1996 as a 41.5 km2 area located south-east of Sheldon, Illinois, and west of Raub, Indiana (Fig. 11.2). The location of the evaluation site was selected based on the presence of variant WCR populations. The programme enlisted the partnership of 52 Indiana and Illinois maize and soybean producers. Most producers in the evaluation site fol- chap11.qxd 12/11/04 1:20 PM Page 225

Insecticide Baits for Managing Western Corn Rootworm 225

Fig. 11.1. Location of the AWPM evaluation site in Indiana and Illinois.

lowed a maize/soybean rotation and practised either conservation or no tillage. The programme area consisted of approximately 4658 ha of maize and soybean. Field sizes ranged from 1.2 ha to 65.2 ha. Within the evaluation site, a 3726 ha management zone was established to test the effectiveness of semiochemical-based insecticide baits on adult WCR populations. A comparison zone, comprized of several control fields, either adjoining or within 3.22 km of the management zone, was monitored for adult rootworm populations. However, if economic thresholds were exceeded in the control fields, semiochemical-based insecticide bait treatments were not applied. To determine the overall effectiveness of the treatment applications, root damage by WCR larvae, WCR beetle emergence and yield benefits were compared between the management zone and the comparison zone.

AWPM WCR sampling protocol

Scouting firms To determine the need for and proper timing of a semiochemical-based insecticide bait treatment, maize and soybean fields were monitored weekly throughout the period of WCR adult activity by field scouts affili- ated with AFS Services (Maple Park, IL 60151 (formerly Midwest Consulting Service)) and Creek Consulting (Fowler, IN 47944). chap11.qxd 12/11/04 1:20 PM Page 226

226 C.K. Gerber et al.

Areawide Pest Management Programme

Maize ﬁelds Soybean ﬁelds

Fig. 11.2. Indiana/Illinois AWPM evaluation site.

Sampling methods The numbers of WCR adults in maize and soybean fields were estimated using two sampling methods. The sampling methods used in maize fields were different from the sampling methods used in soybean fields. WCR populations in maize fields were sampled by counting the number of beetles on two plants approximately 0.9 to 1.5 m apart in each of ten locations within a field. This sampling method is known as the whole-plant count method (Tollefson, 1975). WCR adults were monitored from 1997 to 1999 in each maize field in the management and comparison zones. The decision to treat individual maize fields for WCR adults was based on the established economic thresholds of 0.5 beetle per plant in first-year maize or 0.7 beetle per plant in continuous maize, with at least 10% of the female beetles gravid (Bergman and Edwards, 1985). In 2000, the protocol for treating fields in the management zone was chap11.qxd 12/11/04 1:20 PM Page 227

Insecticide Baits for Managing Western Corn Rootworm 227

AWPM Site Sections

10 11 12 19 2.59 km2

7 8 91718 N

4 5 61516

1 2 31314

Illinois Indiana

Fig. 11.3. The AWPM evaluation site represented as sections.

altered from individual field treatments to treating entire sections (multiple fields), with treatment decisions being made based on the scouting results from any soybean field in a section (Fig. 11.3). Because most of the fields in the management zone were in a maize/soybean rotation, the numbers of WCR adults present in soybean fields were critical for making treatment decisions to reduce the risks of WCR larval damage to maize the following year. Therefore, maize fields that had threshold levels of WCR were not used for making treatment decisions. However, at least one maize field was scouted in each section to provide information about the presence or absence of WCR adults. Approximately one-quarter of all maize fields were monitored in 2000, 2001 and 2002. The Pherocon® AM unbaited yellow sticky trap (Trécé Inc., Salinas, CA 93912) is a reliable monitoring tool used to measure WCR adult population numbers (O’Neal et al., 2001). These numbers can be used to predict WCR larval injury to rotated maize. Therefore, Pherocon® AM traps were used to monitor WCR adults in soybean fields located in the AWPM evaluation site. For making decisions to treat soybean fields to control WCR, a threshold of two WCR beetles per Pherocon® AM trap per day was selected. An economic threshold for WCR adults in soybeans, which determines the threat of economic damage caused by WCR larvae in maize the following year, was not known at the time the AWPM programme was initiated. The threshold used was determined to be low enough to prevent substantial WCR oviposition. chap11.qxd 12/11/04 1:20 PM Page 228

228 C.K. Gerber et al.

In each soybean field, two transects were established through the length of the field, each approximately one-third of the field width apart. Trap lines were established on these transects, and each line consisted of four Pherocon® AM traps. Based on field length, traps were placed approximately 50 to 200 m apart. It was assumed that traps were far enough apart for one trap not to interfere with the capture of WCR in another trap. At each trap deployment site, 1.5 m wood lath stakes were driven approximately 0.3 m into the soil. Traps were attached to the wood lath stakes and positioned at the top of the soybean canopy. Traps were replaced weekly. Beetles were counted on each trap, and the numbers of beetles per trap per day were calculated. Due to the size of the AWPM evaluation site, scouting corn and soybean fields occurred over a 4- to 5-day period. From 1997 to 1999, when the field- by-field treatment protocol was in place, approximately one-quarter of the maize and soybean fields in the evaluation site were monitored each day. However, in 2000, 2001 and 2002, three to six sections that contained various numbers of maize and soybean fields were monitored each day. In 2002, the final season of data collection, the numbers of WCR beetles collected were used only to monitor the WCR beetle population in the evaluation site, not to determine whether semiochemical-based insecticide bait applications were necessary for a given section.

Trapping periods

WCR populations were monitored with Pherocon® AM traps over a 7- week period (Table 11.1). In 1997 and 1998, trap deployment began on 21 and 13 July, respectively. Trap deployment began on 12 July in 1999 and 10 July in 2000. In 2001 and 2002, the ﬁrst Pherocon® AM traps were deployed on 9 July and 8 July, respectively. All traps were replaced every 7 to 8 days. Each trapping period ended by 31 August, with the exception of 1997, when the trapping period ended by 8 September.

Table 11.1. Time frame, by week, for monitoring WCR beetles with Pherocon® AM traps in soybean ﬁelds at the AWPM evaluation site from 1997 to 2002.

Year Week 1997 1998 1999 2000 2001 2002

1 21.7–28.7 13.7–20.7 12.7–19.7 10.7–17.7 9.7–16.7 8.7–15.7 2 28.7–4.8 20.7–27.7 19.7–26.7 17.7–24.7 16.7–23.7 15.7–22.7 3 4.8–11.8 27.7–3.8 26.7–2.8 24.7–31.7 23.7–30.7 22.7–29.7 4 11.8–18.8 3.8–10.8 2.8–9.8 31.7–7.8 30.7–6.8 29.7–5.8 5 18.8–25.8 10.8–17.8 9.8–16.8 7.8–14.8 6.8–13.8 5.8–12.8 6 25.8–1.9 17.8–24.8 16.8–23.8 14.8–21.8 13.8–20.8 12.8–19.8 7 1.9–8.9 24.8–31.8 23.8–30.8 21.8–28.8 20.8–27.8 19.8–26.8 chap11.qxd 12/11/04 1:20 PM Page 229

Insecticide Baits for Managing Western Corn Rootworm 229

Semiochemical-based insecticide baits

Two different semiochemical-based insecticide baits were used in this programme. From 1997 to 1999, the semiochemical-based insecticide bait used in the management zone was SLAM™ (Micro-Flo Company, Lakeland, FL 33807). SLAM™ consists of cucurbitacins (65%) derived from dried and ground roots of buffalo gourd, Cucurbita foetidissima H. B. K.; non-toxic edible carriers (22%); and carbaryl (13%). Application rates were 420 g and 560 g of SLAM™ per ha sprayed at 9.35 l (formulated mixture) per ha. Due to formulation inconsistencies, as well as the limited efﬁcacy of SLAM™, a new semiochemical-based bait, Invite™ EC (FFP AgroTech, Eustis, FL 32727), was used in 2000 and 2001. Invite™ EC consists of cucurbitacins (80%) derived from Hawkesbury watermelon, Citrullus vulgaris Schrad, and non-toxic edible carriers (20%). Before application, carbaryl (SEVIN® XLR PLUS, Rhône-Poulenc Ag Company, Research Triangle Park, NC 27709) was added to Invite™ EC. The application rate of the formulated mixture (Invite™ EC, 0.91 l/ha; water, 8.21 l/ha; and carbaryl, 0.23 l/ha) was 9.35 l/ha.

Application method

Semiochemical-based insecticide baits were applied with ﬁxed-wing aircraft. Due to the size of the management zone, as well as the number of applications required to suppress the adult WCR population, two aerial application companies were employed during the course of the programme – Zumwalt Aviation (Sheldon, IL 60966) and Schertz Aerial Service Corporation (Hudson, IL 61748). Zumwalt Aviation used a Grumman Agcat aircraft with a conventional aircraft spray system. Spraying Systems nozzles with D-5 tips, spinner plates removed, were used to deliver large spray droplets. Application speed was 145 kph, and the spray pressure ranged from 124 to 139 kPa. Schertz Aerial Service used Air Tractor (402 and 502) aircraft, each equipped with a conventional aircraft spray system. Both Air Tractors were equipped with a trailing edge boom system that contained Spraying Systems 0010 solid stream nozzles. Application speed was 225 kph, and the spray pressure was 276 kPa.

WCR larval damage assessment

To determine the effectiveness of SLAM™ treatments in 1997, 1998 and 1999 and Invite™ EC treatments in 2000 and 2001, maize roots were evaluated for rootworm larval damage in 1998, 1999, 2000, 2001 and 2002. Producers afﬁliated with the AWPM programme treated most of their chap11.qxd 12/11/04 1:20 PM Page 230

230 C.K. Gerber et al.

maize fields with soil insecticides at planting. The soil insecticides used in the evaluation site were Aztec® 2.1% (Bayer Corporation, Kansas City, MO 64120), Capture® 2EC (FMC Corporation, Philadelphia, PA 19103), Counter® 20CR (BASF, formally American Cyanamid, Research Triangle Park, NC 27709), Lorsban™ 15G (Dow AgroSciences, Indianapolis, IN 46268), Force® 3G (Syngenta, formally Zeneca, Greensboro, NC 27409) and Fortress® 5G (Amvac, Fortress formally owned by DuPont, Newport Beach, CA 92660). Most of the producers established check strips within fields, which consisted of one to multiple rows of maize that had not been treated with a soil insecticide. At some point during the programme, 14 producers elected either not to use a soil insecticide or to use a soil insecticide only on a selected number of maize fields. The number of fields not treated with a soil insecticide ranged from 10 to 22, depending upon growing season. In 1998, 30 roots were dug up equidistant along randomly selected soil insecticide-treated rows and untreated check rows. Thirty roots were also dug up equidistant in fields that did not receive soil insecticide application. From 1999 to 2002, 20 roots were dug up in treated and check rows in the evaluation site, and 20 roots were dug from fields in the evaluation site that were not treated with a soil insecticide. Each root was labelled. To observe rootworm larval damage, roots were washed and rated using the Hills and Peters (1971) 1 to 6 root damage rating scale. Treatments were analysed using analysis of variance (ANOVA) procedures and means were separated using Fisher’s protected least significant difference (LSD), P ≤ 0.05 (Supernova, 1989).

WCR adult emergence assessment

The adult WCR population was characterized using emergence traps in areas of maize fields that did not receive a soil insecticide application at planting. During the 1997 growing season, four fields within the management zone were monitored with emergence traps, described by Hein et al. (1985). In 1998, five fields (four fields from the management zone and a single field from the comparison zone) were monitored with the same emergence traps. In 1999 and 2000, four fields within the management zone and four fields in the comparison zone were monitored using the same emergence traps used from 1997 to 2000. In 2001, a modified emergence trap (Chaddha et al., 1993) was used. To compare the data gathered from the two different types of traps, four fields from the management zone and comparison zone were monitored using both types of traps in 2002. Placement of the different types of traps was alternated along two rows of maize in each field. Except in 1997, when only eight emergence traps were used, 12 emergence traps were placed in untreated check rows within each field at the onset of WCR beetle emergence. Traps were placed 100 m from the field chap11.qxd 12/11/04 1:20 PM Page 231

Insecticide Baits for Managing Western Corn Rootworm 231

Table 11.2. Time frame, by week, for monitoring WCR beetle emergence with emergence traps in maize ﬁelds at the AWPM evaluation site from 1997 to 2002.

Year Week 1997 1998 1999 2000 2001 2002

1 14.7–21.7 1.7–8.7 30.6–7.7 28.6–5.7 26.6–3.7 1.7–8.7 2 21.7–28.7 8.7–15.7 7.7–14.7 5.7–12.7 3.7–10.7 8.7–15.7 3 28.7–4.8 15.7–22.7 14.7–21.7 12.7–19.7 10.7–17.7 15.7–22.7 4 4.8–11.8 22.7–29.7 21.7–28.7 19.7–26.7 17.7–24.7 22.7–29.7 5 11.8–18.8 29.7–5.8 28.7–4.8 26.7–2.8 24.7–31.7 29.7–5.8 6 18.8–25.8 5.8–12.8 4.8–11.8 2.8–9.8 31.7–7.8 5.8–12.8 7 25.8–2.9 9.8–16.8 7.8–14.8

edge and 50 m apart. The traps were checked weekly (Table 11.2) until beetle emergence numbers were biologically inconsequential. Numbers of emerging WCR adults were analysed using ANOVA procedures and means were separated using Fisher’s protected least signiﬁ- cant difference (LSD), P ≤ 0.05 (Supernova, 1989).

Yield assessment

Maize yield estimates were calculated from 1999 to 2002. Yield estimates were obtained by harvesting all maize ears in 5.31 m of row for 0.76 m row spacing, and 4.19 m of row for 0.97 m row spacing, in both treated and check rows at four locations in each ﬁeld. Ear maize was weighed using a Chatillon® bulk scale (Ametek Corporation, Kew Gardens, NY 11415), and percentage grain moisture was determined with calibrated moisture testers (Farmex Electronics Corporation, Aurora, OH 44202 or Dickey-John Corporation, Auburn, IL 62615). Ear weights were converted to kg per ha of No. 2 shelled maize at 15.5% moisture. Maize yield estimates were analysed using ANOVA procedures and means were separated using Fisher’s protected LSD, P ≤ 0.05 (Supernova, 1989).

Results and Discussion

1997–2002 summary

Treatment applications, 1997 to 2001 A total of 5605 ha were treated with SLAM™ in 1997. Hectares treated in 1998 declined to 1710 ha. In this same year, heavy rainfall in early to mid- June ﬂooded a number of maize ﬁelds, which probably contributed to high WCR larval mortality, resulting in low numbers of WCR adults. In chap11.qxd 12/11/04 1:20 PM Page 232

232 C.K. Gerber et al.

1999, the area treated (5282 ha) was similar to the area treated in 1997. In 2000, a total of 8355 ha were treated with Invite™ EC. The large number of hectares treated was attributed to the change in the treatment strategy from a field-by-field approach to a multi-field approach. However, in 2001, the area treated (7684 ha) was comparable to the area treated in 2000. If the availability of the product had not been limited in 2001, the area treated in 2001 would have exceeded the area treated in 2000.

WCR larval damage assessment, 1998 to 2002 Treatments were analysed within a given year and not between years due to the high variation between years (Fig. 11.4).

EVALUATION OF THE EFFECTIVENESS OF SLAM™ – 1998, 1999, 2000. In 1998, to evaluate the impact of SLAM™ alone, root damage by WCR in the management zone, where no soil insecticides were applied, was compared to root damage in the comparison zone, where neither soil insecticides nor semiochemical baits were applied. Signiﬁcantly less WCR damage was detected in the management zone than in the comparison zone. Similar results were obtained in 1999 and 2000.

EVALUATION OF THE EFFECTIVENESS OF INVITE™ EC – 2001, 2002. In 2001 and 2002, an assessment similar to the root damage comparisons conducted in 1998 through 2000 was conducted to evaluate the impact of Invite™ EC. Signiﬁcantly less WCR damage was detected in the management zone, where only Invite EC™ was applied than in the comparison zone, where no insecticide treatments were applied.

OVERALL EVALUATION. From 1998 to 2002, based on the data collected in the management zone where only semiochemical bait treatments were applied, a consistent decrease in root damage was not detected (Fig. 11.4).

WCR adult emergence assessment, 1998 to 2002

EVALUATION OF THE EFFECTIVENESS OF SLAM™ – 1998, 1999, 2000. The numbers of WCR adults emerging from maize ﬁelds in the management zone were numer- ically lower than in the comparison zone; however, adult numbers were only signiﬁcantly lower in the management zone in 1999.

EVALUATION OF THE EFFECTIVENESS OF INVITE™ EC – 2001, 2002. In 2001, signiﬁcantly more WCR adults emerged in the comparison zone than in the management zone. However, in 2002, signiﬁcantly fewer beetles emerged from the comparison zone. This may have been the result of the limited product availability in 2001. chap11.qxd 12/11/04 1:20 PM Page 233

Insecticide Baits for Managing Western Corn Rootworm 233

4.0

3.5

3.0

2.5

2.0

1.5

Mean root damage rating 1.0 SLAM INVITE EC 0.5

0.0 1998 1999 2000 2001 2002

MZ-SI MZ-NSI CZ-SI CZ-NSI

Fig. 11.4. Root damage ratings from areas in first-year maize fields treated with a soil insecticide (SI), as well as areas not treated with a soil insecticide (NSI), in the management and comparison zones (MZ and CZ) at the AWPM evaluation site from 1998 to 2002. Columns with the same letter are not significantly different at the 0.05 level of significance, according to the Fisher LSD.

OVERALL EVALUATION. With the exception of 2002, a reduction in WCR beetle emergence in the management zone was evident (Fig. 11.5).

Yield assessment, 1999 to 2002 As with the root damage assessments, yield assessments were analysed within a given year and not between years due to the high variation between years (Fig. 11.6).

EVALUATION OF SLAM™ EFFECTIVENESS – 1999, 2000. In 1999, yield estimates were compared between the management and comparison zones to evaluate the impact of SLAM™. Results indicated no signiﬁcant differences among treatments. In 2000, the yield estimates were signiﬁcantly lower in the comparison zone than in the management zone where both soil insecticides and semiochemical baits were applied.

EVALUATION OF INVITE™ EC EFFECTIVENESS – 2001, 2002. In 2001 and 2002, yield estimates were compared to evaluate the impact of Invite™ EC. Results indicated no signiﬁcant differences among treatments.

OVERALL EVALUATION. From 1999 to 2002, based on the data collected in the chap11.qxd 12/11/04 1:20 PM Page 234

234 C.K. Gerber et al.

12 Management zone Comparison zone 10

8 2

6 WCR/m

0 1997 1998 1999 2000 2001 2002 2003 Year

Fig. 11.5. WCR adult emergence in the management and comparison zones at the AWPM evaluation site from 1998 to 2002.

management zone where only semiochemical bait treatments were applied, a consistent yield beneﬁt was not detected (Fig. 11.6).

1997–2002 conclusions

An overall decrease in the area treated with SLAM™ (1997 to 1999) or Invite™ EC (2000 and 2001) within the management zone was not detected. A consistent decrease in root damage and beetle emergence in the areas treated with semiochemical baits and a consistent increase in yield were also not detected at the completion of the programme. We speculate that three factors affected the overall results of this suppression programme:

1. During the programme, it became evident that the variant WCR population was extremely large in north-western Indiana and east-central Illinois (C.K. Gerber et al., unpublished data). In addition, previous studies have indicated that WCR adults are highly mobile (Coats et al., 1986; Grant and Seevers, 1989), readily migrating long distances, and have a very high reproductive capacity (Branson and Johnson, 1973; Hill, 1975; Krysan, 1986). Therefore, considering the dynamics of the WCR population and its propensity for movement and reproduction, it was evident that suppressing a population of this magnitude would prove very difﬁcult. chap11.qxd 12/11/04 1:20 PM Page 235

Insecticide Baits for Managing Western Corn Rootworm 235

6 kg/ha

 2 SLAM INVITE EC

0 1999 2000 2001 2002

Fig. 11.6. Yield estimates from areas in first-year maize fields treated with a soil insecticide (SI), as well as areas not treated with a soil insecticide (NSI), in the management and comparison zones (MZ and CZ) at the AWPM evaluation site from 1999 to 2002. Columns with the same letter are not significantly different at the 0.05 level of significance, according to the Fisher LSD.

2. A second factor that influenced the outcome of the programme was the poor performance of the semiochemical baits, relative to residual activity and rainfastness. The average residual activity of SLAM™ (1997 to 1999) and Invite™ EC (2000) was approximately 2 weeks (G.K. Gerber et al., unpublished data). It was evident that, when it rained, the products were easily washed off the plants. Consequently, additional treatments were needed. In 2001, the residual activity of Invite™ EC, without a rain event, was only 24 h (G.K. Gerber et al., unpublished data). We suspected that the formulation might have been altered to reduce the unexpected mortality of non-target organisms that occurred in 2000. 3. A third factor that may have contributed to the results of the AWPM programme was the field-by-field approach we initially used to make semiochemical bait treatment decisions. This approach did not account for the movement of variant WCR adults to and from maize and soybean fields within the evaluation site. An intensive large-scale treatment approach should have been implemented during the first year to drastically lower the population equilibrium level at the evaluation site. If we had taken that approach, it is possible that the number of applications of semiochemical baits from one year to the next might have been reduced. However, even that approach may not have overcome the problem of immigration of WCR from outside the evaluation site. chap11.qxd 12/11/04 1:20 PM Page 236

236 C.K. Gerber et al.

For a programme of this magnitude to be successful in north-western Indiana and east-central Illinois, we believe it is necessary for a rootworm semiochemical-based insecticide bait to have a residual activity of at least 4 weeks. Regardless of the size of the WCR population or the size of the targeted area, adequate suppression will occur only if a reliable bait product is available and used before WCR adult peak activity.

References

Ball, H.J. and Weekman, G.T. (1962) Insecticide resistance in the adult western corn rootworm in Nebraska. Journal of Economic Entomology 55, 439–441. Bergman, M.K. and Edwards, C.R. (1985) Managing Corn Rootworms. Extension Publication E–49, Purdue University Cooperative Extension Service, West Lafayette, Indiana. Bigger, J.H. (1963) Corn rootworm resistance to chlorinated hydrocarbon insecticides in Illinois. Journal of Economic Entomology 56, 118–119. Branson, T.F. and Johnson, R.D. (1973) Adult western corn rootworm: oviposition, fecundity, and longevity in the laboratory. Journal of Economic Entomology 66, 417–418. Branson, T.F. and Krysan, J.L. (1981) Feeding and oviposition behavior and life cycle strategies of Diabrotica: an evolutionary view with implications for pest management. Environmental Entomology 10, 826–831. Branson, T.F. and Ortman, E.E. (1967) Host range of larvae of the western corn rootworm. Journal of Economic Entomology 60, 201–203. Branson, T.F. and Ortman, E.E. (1970) The host range of larvae of the western corn rootworm: further studies. Journal of Economic Entomology 63, 800–803. Buhler, W.G., Edwards, C.R., Bledsoe, L.W., Gerber, C., Gray, M.E. and Steffey, K.L. (1998) Areawide pest management of western corn rootworm in Indiana and Illinois – turning the first corner. Pflanzenschutzberichte 57(2), 69–74. Capinera, J.L., Epsky, N.D. and Thompson, D.C. (1986) Effects of adult western corn rootworm (Coleoptera: Chrysomelidae) ear feeding on irrigated field corn in Colorado. Journal of Economic Entomology 79, 1609–1612. Chaddha, S., Ostlie, K.R. and Fisher, J.R. (1993) Design for an improved adult emergence trap for corn rootworm (Coleoptera: Chrysomelidae). Journal of the Kansas Entomological Society 66, 338–344. Chandler, L.D. and Faust, R.M. (1998) Overview of areawide management of insects. Journal of Agricultural Entomology 15, 319–325. Chandler, L.D. and Hartman, D. (1998) South Dakota corn rootworm areawide management study site – 1998 activities. In: 1998 Update Corn Rootworm Areawide Management Program. USDA-ARS, Northern Plains Area, Brookings, South Dakota. Chandler, L.D. and Sutter, G.R. (1995) Evaluation of corn rootworm populations within a 16 square mile area previously treated with semiochemical insecticide-baits. In: Report to NCR-46 Corn Rootworm Technical Committee (not for publication). USDA- ARS, Northern Plains Area, Brookings, South Dakota, pp. 17–18. Coats, S.A., Tollefson, J.J. and Mutchmor, J.A. (1986) Study of migratory flight in the western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 15, 1–6. Coppedge, J.R., Hoffman, W.C., Lingren, P.S., Kirk, I.W., Westbrook, J.K., O’Neil, T.M., Harp, S.J., Eyster, R.S. and Schleider, P.G. (1997) Areawide management of corn rootworm results from research program, College Station, TX – 1997. In: 1997 Update chap11.qxd 12/11/04 1:20 PM Page 237

Insecticide Baits for Managing Western Corn Rootworm 237

Corn Rootworm Areawide Management Program (not for publication). USDA-ARS, Northern Plains Area, Brookings, South Dakota, pp. 2–7. Edwards, C.R. (1996) The dramatic shift of the western corn rootworm to first-year corn. In: 1996 Proceedings of the Illinois Agricultural Pesticides Conference. Cooperative Extension Service, University of Illinois, Urbana-Champaign, Illinois, pp. 14–15. Edwards, C.R. (2000) Managing the western corn rootworm variant and controlling European corn borer with Bt corn. In: Proceedings of the 2000 Crop Management Workshops. Purdue Pest Management Program, Purdue University, West Lafayette, Indiana, pp. 17–23. Edwards, C.R., Bledsoe, L.W. and Obermeyer, J.L. (1994) Corn rootworm injury in first-year corn. In: Pest Management and Crop Production Newsletter, 21. Purdue University Cooperative Extension Service, West Lafayette, Indiana. Gibb, T.J. and Higgins, R.A. (1991) Aboveground dry weight and yield response of irrigated field corn to defoliation and root pruning stresses. Journal of Economic Entomology 84, 1562–1576. Godfrey, L.D. and Turpin, F.T. (1983) Comparison of western corn rootworm (Coleoptera: Chrysomelidae) adult populations and economic thresholds in first-year and continuous corn fields. Journal of Economic Entomology 76, 1028–1032. Grant, R.H. and Seevers, K.P. (1989) Local and long-range movement of adult western corn rootworm (Coleoptera: Chrysomelidae) as evidenced by washup along Southern Lake Michigan shores. Environmental Entomology 18, 266–272. Gray, M.E., Levine, E. and Oloumi-Sadeghi, H. (1998) Adaptation to crop rotation: western and northern corn rootworms respond uniquely to a cultural practice. In: Pandalai, S.G. (ed.) Recent Research Developments in Entomology. Research Signpost, Trivandrum, India, pp. 19–31. Hein, G.L., Bergman, M.K., Bruss, R.G. and Tollefson, J.J. (1985) Absolute sampling technique for corn rootworm (Coleoptera: Chrysomelidae) adult emergence that adjusts to fit common-row spacing. Journal of Economic Entomology 78, 1503–1506. Hill, R.E. (1975) Mating, oviposition patterns, fecundity, and longevity of the western corn rootworm. Journal of Economic Entomology 68, 311–315. Hills, T.M. and Peters, D.C. (1971) A method of evaluating postplanting insecticide treatments for control of western corn rootworm larvae. Journal of Economic Entomology 64, 764–765. Knipling, E.F. (1978) Eradication of plant pest – pro: advances in technology for insect- population eradication and suppression. Bulletin of the Entomological Society of America 24, 44–52. Krysan, J.L. (1986) Introduction: biology, distribution, and identification of pest Diabrotica. In: Krysan, J.L. and Miller, T.A. (eds) Methods for the Study of Pest Diabrotica. Springer-Verlag, New York, pp. 1–23. Levine, E. and Gray, M.E. (1996) First-year corn rootworm injury: east-central Illinois research progress to date and recommendations for 1996. In: 1996 Proceedings of the Illinois Agricultural Pesticides Conference. Cooperative Extension Service, University of Illinois, Urbana-Champaign, Illinois, pp. 3–13. Levine, E. and Oloumi-Sadeghi, H. (1991) Management of diabroticite rootworms in corn. Annual Review of Entomology 36, 229–255. Levine, E. and Oloumi-Sadeghi, H. (1996) Western corn rootworm (Coleoptera: Chrysomelidae) larval injury to corn grown for seed production following soybeans grown for seed production. Journal of Economic Entomology 89, 1010–1016. Levine, E., Spencer, J.L., Isard, S.A., Onstad, D.W. and Gray, M.E. (2002) Adaptation of the western corn rootworm to crop rotation: evolution of a new strain in response to a management practice. American Entomologist 48, 94–107. chap11.qxd 12/11/04 1:20 PM Page 238

238 C.K. Gerber et al.

Meinke, L.J. (1995) Adult Corn Rootworm Management. Miscellaneous Publication 63, Agricultural Research Division, University of Nebraska. Metcalf, R.L. (1986) Foreword. In: Krysan, J.L. and Miller, T.A. (eds) Methods for the Study of Pest Diabrotica. Springer-Verlag, New York, pp. vii–xv. Metcalf, R.L., Rhodes, A.M., Metcalf, R.A., Ferguson, J., Metcalf, E.R. and Lu, P. (1982) Cucurbitacin contents and diabroticites (Coleoptera: Chrysomelidae) feeding upon Cucurbita spp. Environmental Entomology 11, 931–937. Metcalf, R.L., Ferguson, J.E., Lampman, R. and Andersen, J.F. (1987) Dry cucurbitacin-containing baits for controlling diabroticite beetles (Coleoptera: Chrysomelidae). Journal of Economic Entomology 80, 870–875. Nordlund, D.A. (1981) Semiochemicals: a review of terminology. In: Nordlund, D.A., Jones, R.L. and Lewis, W.J. (eds) Semiochemicals, Their Role in Pest Control. John Wiley & Sons, Chichester, UK, pp. 13–23. O’Neal, M.E., Gray, M.E., Ratcliffe, S. and Steffey, K.L. (2001) Predicting western corn rootworm (Coleoptera: Chrysomelidae) larval injury to rotated corn with Pherocon AM traps in soybeans. Journal of Economic Entomology 94, 98–105. Pruess, K.P., Witkowski, J.F. and Raun, E.S. (1974) Population suppression of western corn rootworm by adult control with ULV malathion. Journal of Economic Entomology 67, 651–655. Sammons, A.E., Edwards, C.R., Bledsoe, L.W., Boeve, P.J. and Stuart, J.J. (1997) Behavioral and feeding assays reveal a western corn rootworm (Coleoptera: Chrysomelidae) variant that is attracted to soybean. Environmental Entomology 26, 1336–1342. Shaw, J.T., Paullus, J.H. and Luckmann, W.H. (1978) Corn rootworm oviposition in soybeans. Journal of Economic Entomology 71, 189–191. Spencer, J.L., Levine, E. and Isard, S.A. (1997) Corn rootworm injury to ﬁrst-year corn: new research findings. In: 1997 Proceedings of the Illinois Agricultural Pesticides Conference. Cooperative Extension Service, University of Illinois at Urbana- Champaign, Illinois, pp. 73–81. Spike, B.P. and Tollefson, J.J. (1991) Yield response of corn subjected to western corn rootworm (Coleoptera: Chrysomelidae) infestation and lodging. Journal of Economic Entomology 84, 1585–1590. Supernova (1989) Supernova: Accessible General Linear Modeling. Abacus Concepts, Berkeley, California, 316 pp. Sutter, G.R., Fisher, J.R., Elliott, N.C. and Branson, T.F. (1990) Effect of insecticide treatments on root lodging and yields of maize in controlled infestations of western corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 83, 2414–2420. Tollefson, J.J. (1975) Corn rootworm adult and egg sampling techniques as predictors of larval damage. PhD dissertation, Iowa State University, Ames, Iowa. Tollefson, J.J. (1998) Rootworm management program in Iowa. Journal of Agricultural Entomology 15, 351–357. Wilde, G.E., Whitworth, R.J., Shufran, R.A., Zhu, K.Y., Sloderbeck, P.E., Higgins, R.A. and Buschman, L.L. (1998) Rootworm areawide management project in Kansas. Journal of Agricultural Entomology 15, 335–349. chap12.qxd 12/11/04 1:22 PM Page 239

12 Genetically Enhanced Maize as a Potential Management Option for Corn Rootworm: YieldGard® Rootworm Maize Case Study

Dennis P. Ward, Todd A. DeGooyer, Ty T. Vaughn, Graham P. Head, Michael J. McKee, James D. Astwood and Jay C. Pershing Monsanto Company, St Louis, Missouri, USA

Introduction

Maize (Zea mays L.), the world’s third leading cereal crop following wheat and rice, is grown commercially in over 25 countries. In 2002, worldwide production of maize was approximately 594 million t (CRA, 2003). In the USA its production covered 32 million ha, which yielded 229 million t and had a net value of US$21.2 billion (NCGA, 2003). Maize, also referred to as corn, has been a staple of the human diet for centuries. Maize grain and processed fractions are consumed in a multi- tude of food and animal feed products. Hybrid maize is an extremely productive crop, yielding an average of 8.16 t/ha in the USA during 2002. High yield makes maize one of the most economical sources of usable energy for feeds and of usable starch and sugar for food and industrial products. The majority of maize harvested is fed to livestock. Maize yields are negatively affected by a number of insect pests. One of the most pernicious in the US Corn Belt is the corn rootworm. Corn rootworm larvae damage maize by feeding on the roots, which reduces the ability of the plant to absorb water and nutrients from soil and causes harvesting difﬁculties due to plant lodging. Corn rootworm is the most sig- niﬁcant insect pest problem for maize growers in the US Corn Belt from the standpoint of chemical insecticide usage. An estimated 5.7 to 10.1 million ha of maize in the USA are treated annually with organophosphate, carbamate, pyrethroid and phenyl pyrazole insecticides to control this pest (Doane, 2001; ARS, 2002). Corn rootworms have been described as the billion dollar pest complex, based on costs associated with the application of conventional soil insecticides and crop losses due to pest damage (Metcalf, 1986; ARS, 2002). Incomplete protection of maize root

240 D.P. Ward et al.

systems with larval insecticides, the development of resistance to adult control insecticides and the biological adaptation of corn rootworms to crop rotation have diminished the effectiveness of currently available pest management practices. Future management strategies for the control of corn rootworms will include the planting of genetically enhanced maize that resists larval root feeding and protects grain yields. This case study examines current management strategies and their limitations in US maize production and the potential beneﬁts of managing these pests with corn rootworm-protected maize developed by Monsanto Company.

Corn Rootworms

The corn rootworm complex (Diabrotica spp.) is comprised of the northern corn rootworm (D. barberi, Smith and Lawrence), the western corn rootworm (D. virgifera virgifera, LeConte), and the Mexican corn rootworm (D. virgifera zeae, Krysan and Smith). One additional Diabrotica species, the southern corn rootworm (D. undecimpunctata howardi, Barber), inhabits the south-eastern region of the USA and can cause economic damage, but is a relatively minor pest of maize. Corn rootworms are widely distributed throughout the maize-growing regions of the USA east of the Rocky Mountains. The pest is also present in Canada, Mexico and Brazil. The western corn rootworm has also been found in Europe, with the 1992 discovery of the insect in Serbia and Montenegro. In little more than a decade, corn rootworms have spread to 13 European countries, including Austria, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, France, Hungary, Italy, Romania, Slovakia, Switzerland and the Ukraine (EPPO, 2003). Mexican, northern and western corn rootworms have similar life cycles (O’Day et al., 1998). Each has a single generation per year, surviving only on maize and a few grass species. Oviposition commences in mid- to late summer, with eggs being buried at a depth of 5 to 10 cm near the base of maize plants. The embryo remains in a state of diapause until the onset of egg hatch, which occurs from May to mid-June of the following season. After hatching, rootworm larvae feed on maize roots for 3–4 weeks, passing through three instars. The mature larvae pupate from mid- June to early July and adults live for 75–85 days. Oviposition occurs from July until the killing frost in the autumn. The levels of larval infestation are highest in ﬁelds that are cropped continuously with maize. Most economic damage is caused by larval root feeding. First and second instars cause brown feeding scars or they tunnel from root tips to the plant base, destroying root hairs and small roots. Third instars cause the majority of root damage, generally feeding on the larger primary roots near the stalk and the ﬁrst set of brace roots. Root feeding causes physiological stress, which stunts plant growth and can lead to plant lodging. The adults feed primarily on maize silks, pollen and immature kernels. Western corn rootworms also feed on maize leaves and northern corn rootworms chap12.qxd 12/11/04 1:22 PM Page 241

Genetically Enhanced Maize as a Potential Management Option 241

feed on the pollen of other plant species, including soybean (Glycine max). Yield losses depend on the extent of larval feeding and on plant maturity, soil fertility, soil moisture at peak injury and the ability of the hybrid to regenerate root tissue. Drought stress worsens the effects of root pruning. Yield loss can result from root pruning and tunnelling, which disrupt the transport of nutrients and water from the root system (Riedell, 1990; Spike and Tollefson, 1991). The lack of root support can lead to plant lodging, which complicates harvesting. The ﬁrst indication of rootworm injury to maize may be in late June or early July when plants fall over after strong winds or heavy rainfall. Root feeding can also result in the invasion of plant tissue by secondary pathogens, such as bacteria and fungi, which increase the incidence of root rots. Yield losses resulting from rootworm feeding have been estimated to range from 0 to 15%, but have been reported to be as high as 50% (Gianessi et al., 2002a).

Current Management Options and Limitations

Management practices for limiting economic loss caused by corn rootworm infestations have relied on crop rotation or the use of conventional chemical insecticides. Historically, a maize–soybean rotation has provided highly effective protection from corn rootworm damage in many agronomic situations because this rotation breaks the pest’s life cycle. However, several factors now limit the usefulness of this management strategy. First, many growers prefer the option of continuous maize production (i.e. non-rotated), even if this practice requires increased chemical inputs for soil fertility and insect control. Secondly, researchers have confirmed the existence of a northern corn rootworm variant in Iowa, Minnesota, Nebraska and South Dakota that exhibits an extended diapause period. The eggs of this variant are able to survive through the non- maize year of a maize–soybean rotation to yield larvae that feed on first-year maize roots (Ostlie, 1987; Tollefson, 1988; Gray et al., 1998). Thirdly, and of critical importance, crop rotation is no longer effective as a cultural corn rootworm management option in east-central Illinois and northern Indiana due to the rapid spread of a new strain of western corn rootworm that, unlike previous populations, lays its eggs in soybean fields (O’Neal et al., 1999; Onstad and Joselyn, 1999; Levine et al., 2002). The eggs of this western corn rootworm variant are oviposited in soybean fields and hatch the following year in maize. Based on the rapid expansion of this variant population since its initial discovery in Illinois, it is expected to continue to spread throughout the US Corn Belt. Other approaches for the control of corn rootworms, such as biological or pheromone control measures and fertility management, have not proved to be viable options for most large-scale farming operations (Jackson, 1996; Metcalfe and Lampman, 1997). Conventional breeding for resistance to corn rootworm has resulted in germplasm with only moderate levels of chap12.qxd 12/11/04 1:22 PM Page 242

242 D.P. Ward et al.

resistance to rootworm feeding (Knutson et al., 1999). The predominant mechanism for resistance is tolerance rather than antibiosis, with the more resistant genotypes having larger root systems that can regenerate quickly from corn rootworm feeding (Branson et al., 1982). Gray and Steffey (1998) evaluated root system responses to rootworm larval feeding in 12 commonly planted hybrids in Illinois over a 4-year period (1993–1996). Their results supported the conclusion that a larger root system (measured in July and August) allowed hybrids to better tolerate rootworm feeding. However, root regrowth did not consistently result in yield increases. Each of the factors described above has increased grower reliance on the use of conventional chemical insecticides for corn rootworm control. The most common insecticide regime is the application of a granular soil insecticide, either in-furrow or banded, at the time of planting. This regime is most commonly employed in continuous maize-growing regions. Prior to 1990, rotated maize production fields rarely received an insecticide treatment. Spread of the two rotation-resistant corn rootworm variants has increased the number of rotated maize hectares needing an insecticide treatment. Conventional insecticides are now being applied by 33% and 39% of maize growers in Illinois and Indiana, respectively, for control of corn rootworm in rotated maize fields (Rice, 2003). In some agronomic situations, post-emergence insecticide sprays are applied for adult suppression. Except in Nebraska, this practice is relatively uncommon. The objective of this approach is to reduce the number of eggs laid at the end of the growing season such that larval populations the following year will remain below a threshold for economic loss. This approach is labour-intensive and more complex to execute because timing of the insecticide application is critical and often multiple applications are required. In addition, there is evidence that spraying for adult control has led to rootworm resistance to a number of the foliar-applied insecticides, resulting in poor control and increased application rates (Wright et al., 1999). Estimates for total hectares treated in the USA to control corn rootworm range from 5.7 to 10.1 million. The range of published estimates for the amount of insecticide active ingredients applied for corn rootworm control is 3600 to 4400 t (Doane, 2001; NASS, 2001). In an extensive US market survey conducted by Doane Marketing Research in 2000, growers reported that 5,745,530 ha of maize were treated with an insecticide for the control of corn rootworms (Doane, 2001). Growers indicated that corn rootworm was the sole target pest on approximately half of these hectares (2,815,772 ha) and one of a number of pests targeted for control on the remaining hectares. Corn rootworm-targeted hectares that received an insecticide treatment represented 17% of the total hectares of maize planted in the continental USA and 59% of the total hectares receiving any insecticide treatment in the year 2000. Continuous maize and first-year maize hectares received 58% and 42% of the rootworm-targeted insecticide applications, respectively. A chap12.qxd 12/11/04 1:22 PM Page 243

Genetically Enhanced Maize as a Potential Management Option 243

Table 12.1. Insecticide active ingredient (a.i.) usage facts and ﬁgures for all US maize production and applications for corn rootworm (CRW)-targeted hectares in year 2000 (Doane, 2001).

Continuous First-year Parameter maize maize All maize

Hectares planted (× 1000) 9.012 23.193 32.206 Total insecticide-treated hectares (× 1000) 4.690 5.066 9.757 Total insecticide a.i. applied (t) 2.872 2.727 5.599 CRW-targeted hectares (× 1000) 3.347 2.398 5.746 a.i. applied to CRW-targeted acres (t) 2.131 1.423 3.554 Average a.i. rate applied (kg/ha) for CRW control 0.637 0.593 0.619 Average cost per hectare for CRW control (US$) 29.53 30.32 29.85 Total cost of CRW insecticide purchased (× 106 US$) 98.8 72.7 171.5

total of 3554 t of insecticide active ingredient, costing US$172 million, was applied to these hectares. This amount far exceeds the quantity of conventional insecticides applied for the control of any other pest in any other crop. Table 12.1 contains a summary of year 2000 maize insecticide use data. The percentage of planted hectares receiving an insecticide treatment is highest in continuous maize regions and in the region where the western corn rootworm soybean variant is prevalent. Numerous conventional insecticides in granular and liquid form are approved for control of corn rootworm larvae and adults. The active ingredients contained in these products are members of the organophosphate, carbamate, synthetic pyrethroid and phenyl pyrazole classes of chemistry. All display a broad spectrum of activity and operate via a neurotoxic mechanism of action. All of these products are hazardous for applicators, birds, aquatic species and non-target insects because of their toxicity. The majority of these products are designated ‘Restricted Use’ for safety reasons by the US Environmental Protection Agency. Only persons possessing a certified applicator’s licence can purchase and apply these products. Special personal protective equipment is required for application and other restrictions may be placed on the timing, location and frequency of use. Although many products are available for the control of corn rootworm, a small number dominate the US market. Table 12.2 provides a listing of the major active ingredients contained in products used for treatment of corn rootworm-targeted hectares. Four active ingredients, chlorpyrifos, tebupirimfos, terbufos and tefluthrin, were applied to 75% of the hectares treated in 2000. The first three of these active ingredients are organophosphates and products containing tebupirimfos, terbufos and tefluthrin are classified as Restricted Use. Two active ingredients, chlorpyrifos and terbufos, account for 77% of the total quantity of insecticide applied to corn rootworm-targeted hectares. Limitations associated with the use of conventional insecticides chap12.qxd 12/11/04 1:22 PM Page 244

244 D.P. Ward et al.

Table 12.2. Total amount of insecticide active ingredients applied to corn rootworm-targeted hectares in year 2000 (Doane, 2001). These 11 active ingredients accounted for 98.3% of the total quantity of insecticide applied to corn rootworm-targeted hectares in 2000.

Hectares treated Amount applied Active ingredient (× 1000) (metric tonnes)

Carbofuran 138 110 Chlorethoxyfos 146 25 Chlorpyrifos 1440 1708 Cyﬂuthrin/Tebupirimfos 537 81 Fipronil 606 72 λ-Cyhalothrin 72 2 Methyl parathion 149 64 Permethrin 100 11 Phorate 206 267 Teﬂuthrin 1445 182 Terbufos 827 974 Total 5666 3496

include the potential restrictions on their use discussed above, as well as inconsistency of performance. An examination of data collected from multiple locations in the US Corn Belt revealed that no conventional insecticide was completely consistent in protecting maize roots from a level of damage that would result in an economic loss for growers (Rice and Oleson, 2001; Gray, 2002). Late planting, dry conditions, excessive rainfall and heavy rootworm infestation are factors that can have an adverse impact on conventional insecticide performance. The rapid spread of corn rootworm variants resistant to crop rotation, lack of native resistance in commercial hybrids, development of resistance to insecticides and incomplete root protection with larval insecticides have created a need for alternative approaches for managing the corn rootworm pest. The success of genetically enhanced crops containing in-plant protection against insects, such as the European corn borer and cotton bollworm, has stimulated a search for a similar solution for management of corn rootworms.

Development of a Genetic Solution for Corn Rootworm Management

Advances in molecular biology have resulted in the development of crops that have been genetically enhanced to resist damage caused by insects and disease, as well as to tolerate herbicide applications. The introduction of these genetically enhanced crops has greatly enhanced agricultural productivity and economic returns for growers choosing to adopt this new technology. Gianessi et al. (2002b) estimate that adoption of eight currently available genetically enhanced cultivars has resulted in a chap12.qxd 12/11/04 1:22 PM Page 245

Genetically Enhanced Maize as a Potential Management Option 245

reduction of conventional pesticide use by 21,000 t/year in the USA alone. Adoption of these cultivars by growers has increased crop yields by 2 million t/year and provided a net economic return of US$1.5 billion/year. Currently available genetically enhanced cultivars that resist insect damage express one or more genes from a common soil bacterium, Bacillus thuringiensis (Bt). Bt produces parasporal inclusions (i.e. crystals) during the stationary and sporulation phases of growth, which contain proteins that are toxic to selected lepidopteran, coleopteran and dipteran insect species. Thousands of Bt strain isolates have been collected from around the world. Several of these strain isolates have been extensively studied and commercialized as active ingredients for biopesticidal products (Baum et al., 1996). These products display selective insecticidal activity against a number of pests, including dipteran, coleopteran, and lepidopteran insects. Biopesticidal products based on recombinant Bt strains have also been commercialized for use in agriculture since the 1960s. Typically, commercial quantities of these microbes are prepared in large-scale cultures in which the bacteria are allowed to sporulate. The spores and proteins are then formulated for application to plants (Bernhard and Utz, 1993). In 1991, Rupar et al. reported discovery of a novel Bt subspecies kumamotoensis strain that produced a crystal protein with insecticidal activity against the southern corn rootworm (D. u. howardi). Donovan et al. (1992) isolated and sequenced the gene encoding this crystal protein, which was designated as CryIIIB2. Following the adoption of standard- ized nomenclature for identifying Bt crystal proteins, the protein isolated from this strain was renamed Cry3Bb1 (Crickmore et al., 1998). This strain of Bt was later isolated and commercially produced as a biopesticidal foliar spray, Raven™ Biological Insecticide,1 for the control of Colorado potato beetle (Leptinotarsa decemlineata) in potato crops. Using modern molecular techniques, Monsanto Company has developed a variant of the wild-type cry3Bb1 gene from Bt subspecies kumamotoensis that encodes a protein with enhanced insecticidal activity against corn rootworms and is codon-optimized for expression in monocotyledo- nous plants (English et al., 2000). The resulting Cry3Bb1 variant is approximately eight times more lethal to corn rootworm larvae than the wild-type protein. A DNA vector containing the variant cry3Bb1 gene was linked to a constitutive plant-expressible promoter and was introduced into embry- onic maize cells by microprojectile bombardment (Klein et al., 1987; Gordon-Kamm et al., 1990). Plants were regenerated from the transformed callus tissue and assayed for the presence of Cry3Bb1 protein by enzyme- linked immunosorbent assay (ELISA). Maize event MON 863 was selected from hundreds of transformation events produced and developed for commercialization as YieldGard® Rootworm maize.2

1Raven is a trademark of Ecogen, Inc. 2YieldGard® is a trademark of Monsanto Technology LLC. chap12.qxd 12/11/04 1:22 PM Page 246

246 D.P. Ward et al.

Additional transgenic maize cultivars that produce Bt proteins for corn rootworm control are under development by Dow AgroSciences LLC, Pioneer Hi-Bred International, Inc. (Moellenbeck et al., 2001) and Syngenta Crop Protection, Inc.

Characterization of YieldGard® Rootworm Maize

Southern blot analyses confirmed that YieldGard® Rootworm maize contains one transgenic DNA insert (Vaughn et al., 2004, unpublished results). This insert contains one intact copy of the cry3Bb1 vector cas- sette used in transformation, a result that was confirmed by sequencing the inserted DNA. These data support a conclusion that only the full- length Cry3Bb1 protein is encoded by the insert in YieldGard® Rootworm maize. Segregation analysis of the corn rootworm-protected phenotype across multiple generations has confirmed the stability of the inserted gene in YieldGard® Rootworm maize. Plants were identified as being positive for the rootworm-protected phenotype based on the presence of the Cry3Bb1 protein as determined by ELISA. The results of these analyses are consistent with the finding of a single active site of insertion of the cry3Bb1 gene, which segregates in a Mendelian manner. Southern blot fingerprint analyses of DNA extracted from plants spanning multiple generations provided confirmatory evidence for the stability of the inserted DNA in YieldGard® Rootworm maize.

Food and Feed Safety of YieldGard® Rootworm Maize

Numerous factors have been considered in the safety assessment of YieldGard® Rootworm maize and the Cry3Bb1 protein that it produces: (i) the results of compositional analyses of grain and forage; (ii) the biological mode of action for Bt crystal proteins; (iii) the results of extensive animal toxicology tests conducted with Bt crystal proteins and their long history of safe use; (iv) the results of toxicology studies conducted with the Cry3Bb1 protein; and (v) amino acid sequence comparisons of Cry3Bb1 to known toxins and allergens. Through use of a substantial equivalency analysis, these studies provide a comprehensive and comparative safety assessment of YieldGard® Rootworm maize to conventional maize (Astwood and Fuchs, 2001). Compositional analyses were conducted on YieldGard® Rootworm grain and forage samples collected from replicated field trials planted at four locations (C. George et al., 2004). Measurements for proximates (protein, fat, ash and moisture), acid detergent fibre (ADF), neutral detergent fibre (NDF), amino acids, fatty acids, vitamin E, minerals (calcium, copper, iron, magnesium, manganese, phosphorus, potassium, sodium and zinc), phytic acid and trypsin inhibitor content were made for grain. chap12.qxd 12/11/04 1:22 PM Page 247

Genetically Enhanced Maize as a Potential Management Option 247

Measurements for proximates, ADF and NDF content were made for forage. In addition, the content of carbohydrates in forage and grain was determined by calculation. In all, 51 different components (44 in grain and seven in forage) were evaluated as part of the safety and nutritional assessment of YieldGard® Rootworm maize. Statistical analyses of the compositional data were conducted using a randomized complete block model analysis of variance for five sets of comparisons: analyses of data from each of four replicated trials and data from a combination of all four trials. There were a total of 51 components evaluated and a total of 255 statistical comparisons were made: 51 comparisons for each of the five statistical analyses. YieldGard® Rootworm maize results were compared to a parental control line to identify any statistically significant differences at P < 0.05. In addition, a comparison of YieldGard® Rootworm maize to the 95% tolerance interval for 18 commercial reference varieties planted at the trial sites was conducted to determine whether the range of values for YieldGard® Rootworm maize fell within the population of commercial maize varieties. The results of these analyses demonstrated that all 51 components measured in YieldGard® Rootworm maize were within the ranges observed for commercial maize control varieties planted at the trial sites. Furthermore, all 51 components were within published literature ranges (Jugenheimer, 1976; Watson, 1982, 1987) or historical control ranges for conventional maize varieties (Sidhu et al., 2000; Ridley et al., 2002). The grain and forage from YieldGard® Rootworm maize are compositionally equivalent to those of conventional maize. The mode of action for Bt crystal proteins has been well characterized (English and Slatin, 1992; Menestrina and Semjen, 1999). It includes ingestion of Bt crystals by insects, solubilization of the crystals in the insect midgut and proteolytic processing of the released Cry protein by digestive enzymes, sometimes with partial digestion activating the toxin. The activated protein diffuses through the peritrophic membrane of the insect to the midgut epithelium. There it binds to specific high-affinity receptors on the surface of the midgut epithelium of target insects (Wolfersberger et al., 1986; Hoffmann et al., 1988a,b; Van Rie et al., 1989, 1990). Pores are formed in the membrane, leading to leakage of intracellular contents (e.g. K+) into the gut lumen and water into the epithelial gut cells (Sacchi et al., 1986). The larval gut epithelial cells swell due to osmotic pressure and lyse. The gut becomes paralysed as a consequence of changes in electrolytes and pH, killing the insect by starvation and septicaemia. Receptor binding is a critical step in the mechanism of action for Bt crystal proteins. Irreversible binding of these proteins to midgut receptors appears to be correlated with insect susceptibility to the toxin (Schnepf et al., 1998). This observation is relevant to assessing the safety of Cry proteins for humans because no receptors for these proteins have been identified on intestinal cells of mammals (Sacchi et al., 1986; Van Mellaert et al., 1988; Noteborn, 1994). This would explain, in part, the absence of any reported adverse effects for Bt products in humans. Bt preparations have chap12.qxd 12/11/04 1:22 PM Page 248

248 D.P. Ward et al.

been used commercially in the USA since 1958 to produce microbial- derived products with insecticidal activity (EPA, 1988). The extremely low mammalian toxicity of Bt-based insecticide products has been demonstrated in numerous safety studies (McClintock et al., 1995; Betz et al., 2000; Mendelsohn et al., 2003). Cry3Bb1 protein extracted and purified from a heterologous microbial fermentation system was used as the test material for an acute oral toxicity study with mice. The identical cry3Bb1 coding sequence used in the transformation of event MON 863 was used to transform bacteria. The equivalence of plant and microbial fermentation-produced proteins was assessed by a comparison of results obtained from multiple analytical methods, including matrix-assisted laser desorption ionization time of flight mass spectrometry, immunoblotting, insect bioassay, gel elec- trophoresis, glycosylation analysis and amino acid compositional analyses. The Cry3Bb1 protein purified from the microbial fermentation system was found to be physicochemically and functionally equivalent to the protein produced in YieldGard® Rootworm maize (Hileman et al., 2001). In an acute oral toxicity study performed with laboratory mice, no mortality or grossly observable adverse effects were noted at any dose level tested. The no observable effect level (NOEL) was determined to be ≥ 3200 mg/kg, which was the highest achievable dose (Leach et al., 2001). Proteins of many sizes and functions comprize a significant portion of the human diet. Only rarely do any of these tens of thousands of proteins elicit an allergic response when ingested (Taylor, 1992). Comparison of the Cry3Bb1 structure to that of known allergens, toxins and pharmacologically active proteins provides a basis for predicting whether ingestion of YieldGard® Rootworm maize would elicit an adverse response in humans. Protein sequence databases were assembled for this purpose and included allergen, gliadin, toxin and public domain sequence databases. A sequence alignment tool was used to assess structural similarity between Cry3Bb1 and proteins in these databases. Proteins that share a high degree of amino acid sequence similarity are often homologous. Proteins homologous to allergens are more likely to share cross-reactive allergenic epi- topes than are unrelated proteins (Metcalfe et al., 1996; Hileman et al., 2002). The results of these amino acid sequence comparisons indicate that the Cry3Bb1 protein produced by YieldGard® Rootworm maize is not structurally similar to known allergens, toxins or other pharmacologically active proteins relevant to animal or human health (Leach et al., 2001). The Cry3Bb1 protein produced in YieldGard® Rootworm maize is rapidly digested, not glycoslyated and not stable to the elevated temperatures used in the processing of maize grain. Furthermore, there are no confirmed cases of allergic reactions to Cry proteins in applicators of microbial- derived Bt products during 40 years of use (McClintock et al., 1995). Establishing the nutritional equivalence of YieldGard® Rootworm maize to conventional maize varieties is important for market acceptance of this new technology because the majority of harvested maize is consumed as animal feed. Feeding studies incorporating YieldGard® chap12.qxd 12/11/04 1:22 PM Page 249

Genetically Enhanced Maize as a Potential Management Option 249

Rootworm grain have been conducted in multiple species, including broiler chickens, dairy cattle, ﬁnishing steers and pigs (Vander Pol et al., 2002; Berger et al., 2003; Bressner et al., 2003; Fischer et al., 2003; Grant et al., 2003; Taylor et al., 2003). These studies revealed no differences in growth performance and carcass measurements for livestock fed diets containing YieldGard® Rootworm, conventional parental line maize or conventional commercial reference varieties.

Environmental Safety of YieldGard® Rootworm Maize

A database of information has been developed which establishes that the environmental risk posed by YieldGard® Rootworm maize is no greater than the environmental risk posed by conventional maize varieties. Phenotypic measurements and pest susceptibility observations taken during multiple field trials confirm that the genetically enhanced maize is physiologically and agronomically equivalent to conventional maize except for its tolerance to corn rootworm larval feeding damage. Minimal risk to non-target organisms was established through a combination of laboratory and field studies with purified Cry3Bb1 protein or tissue samples from YieldGard® Rootworm maize. No adverse effects have been observed in a wide range of non-target species exposed to levels exceeding the maximum expected environmental concentrations of Cry3Bb1 protein. Furthermore, environmental fate studies demonstrate that the protein rapidly degrades in a variety of soil types. Potential adverse effects on non-target organisms resulting from exposure to Cry3Bb1 were evaluated in a series of laboratory studies with representative avian, aquatic and terrestrial beneficial invertebrate species, which include bobwhite quail (Colinus virginianus), cladoceran (Daphnia magna), channel catfish (Ictalurus punctatus), collembola (Folsomia candida), adult and larval honey bees (Apis mellifera), ladybird beetle adults (Hippodamia convergens and Coleomegilla maculata), ladybird beetle larvae (C. maculata), monarch butterfly larvae (Danus plexippus), green lacewing larvae (Chrysoperla carnea), parasitic hymenoptera (Nasonia vitripennis) and earthworm (Eisenia fetifa). These non-target organisms were exposed to high doses of leaf tissue, grain or pollen from YieldGard® Rootworm maize, or to an artificial diet containing Cry3Bb1 protein (Monsanto, 2003). For each species tested, a no observable effect concentration (NOEC) for Cry3Bb1 was established. No adverse effects were observed at levels that exceeded the maximum expected environmental concentrations (MEEC) of Cyr3Bb1 to which these non-target beneficial organisms would be exposed. Collectively these studies demonstrate that Cry3Bb1 protein poses no significant risk for harm to non-target organism populations. Where possible, the NOEC for each test organism was compared directly to the MEEC and found to exceed it by nine- to 141-fold, clearly demonstrating an adequate margin of safety for these organisms. Where it was not pos- chap12.qxd 12/11/04 1:22 PM Page 250

250 D.P. Ward et al.

sible to make this direct comparison, for example in pollen and grain feeding bioassays, it is reasonable to assume that the absence of adverse effects following exposure to a diet comprised largely of pollen or grain is indicative of no significant risk. Multi-year field monitoring studies corroborate these laboratory findings (Al-Deeb et al., 2003; M.E. Bhatti, Missouri, 2004 personal communication). The abundance of prominent beneficial non-target invertebrate species was found to be comparable in conventional and YieldGard® Rootworm maize fields, and in some cases their abundance in the latter was higher than in fields managed with conventional chemical insecticides. The results of an aerobic soil degradation study demonstrate that Cry3Bb1 dissipates quickly in the environment. Analysis of soil Cry3Bb1 concentration by insect bioassay and ELISA methods established protein dissipation half-lives (DT50) of 2.37 and 2.76 days, respectively, and a DT90 of 7.87 and 9.16 days, respectively (Monsanto, 2003). The rapid dissipation of Cry3Bb1 ensures that exposure risk for soil-dwelling organisms and the risk of surface water runoff will be minimal. The results of environmental fate, field monitoring and non-target organism toxicity studies support a conclusion that Cry3Bb1 protein present in YieldGard® Rootworm maize poses no significant risk to the environment. Maize hybrids containing the YieldGard® Rootworm trait are comparable in composition, safety and agronomic characteristics to conventional varieties of maize. YieldGard® Rootworm maize differs from conventional maize only in its resistance to corn rootworm larval feeding damage.

Performance of YieldGard® Rootworm Maize

Replicated field trials at multiple locations with YieldGard® Rootworm hybrids were conducted to compare their phenotypic characteristics to those of conventional maize hybrids. The parameters evaluated and the timing of these evaluations were: seedling vigour (growth stage V2–V3), early stand count (V4–V6), growing degree units to 50% pollen shed, growing degree units to 50% silk emergence, ear height at maturity, plant height at maturity, final stand count at harvest, test weight at harvest, grain moisture at harvest and yield at harvest. Data collected for each of these parameters from all locations were statistically analysed. With only a small number of exceptions, there were no statistically significant differences between YieldGard® Rootworm hybrids and conventional hybrids observed for these parameters. The few statistically significant differences that were observed were uniformly small and not consistently observed across the six hybrids tested, and none of the differences were considered to be of adverse agronomic consequence. The two most economically important Diabrotica pests in maize crops are the western corn rootworm and the northern corn rootworm. Multiple years of Monsanto and university field research have conclu- chap12.qxd 12/11/04 1:22 PM Page 251

Genetically Enhanced Maize as a Potential Management Option 251

sively demonstrated the efficacy of YieldGard® Rootworm maize in limiting the extent of root damage caused by corn rootworm larval feeding (T.T. Vaughn et al., 2004, unpublished results). Efficacy was evaluated in replicated randomized block trials conducted at multiple locations where the level of root damage caused by western and northern corn rootworm larval feeding was evaluated in YieldGard® Rootworm hybrids, in untreated control hybrids and in hybrids treated with leading soil- applied insecticide brands. Mean root damage ratings (RDR) were computed for each treatment on a scale of 1–6 (Hills and Peters, 1971). Statistically significant differences between mean ratings were identified by an analysis of variance procedure and a Fisher’s protected least significant difference test (P < 0.05). A summary of efficacy trial results collected over these 2 years is shown in Fig. 12.1. The results of these efficacy field trials demonstrate that YieldGard®

Fig. 12.1. The efficacy of YieldGard® Rootworm maize was compared to multiple alternative corn rootworm control products and conventional maize hybrids in multi-location university trials conducted in years 2000 and 2001. Individual plant root systems were dug up and rated based on the Iowa 1–6 RDR scale. Trials were conducted at 14 locations in year 2000 and 280 roots per treatment were rated. Trials were conducted at 21 locations in 2001 and 465 roots per treatment were rated. A mean RDR for each treatment was computed and compared to other treatments by an analysis of variance procedure. Statistically significant differences at P < 0.05 are denoted by letter code, where different letters correspond to significant differences. Upper- and lower-case letters pertain to the comparisons for years 2000 and 2001, respectively. Cyfluthrin and tebupirimfos (Cyflu/Tebu) are formulated together as Aztec® 2.1% Granular Insecticide; terbufos is formulated as Counter® CR Systemic Insecticide-Nematicide. chap12.qxd 12/11/04 1:22 PM Page 252

252 D.P. Ward et al.

Rootworm maize hybrids consistently limit larval feeding damage to levels significantly below economic thresholds. The estimated root rating at which the cost of loss of yield caused by larval feeding is equal to or greater than the cost of insecticidal control can range from 2.5 to 4.0 (Drees et al., 1999). YieldGard® Rootworm maize varieties consistently outperformed commercial insecticide standards. Soil insecticides require rain or irrigation to be activated and their performance can be greatly affected by timing of application. If weather conditions permit, early planting of maize generally leads to increased yields. However, if soil insecticides are applied too far in advance of rootworm hatch (generally in late May), insufficient concentrations may remain to effectively control larval feeding damage. Excessive soil moisture or rainfall can also diminish insecticide performance. Neither of these limitations applies to YieldGard® Rootworm varieties. Cry3Bb1 insecticide is root-incorporated and thus does not require activation. In 2001, imidacloprid (formulated as Prescribe® seed-applied insecti- cide3) was included in the efficacy trials conducted by university researchers (Fig. 12.1). This seed treatment provided some level of control compared to untreated checks, but its performance was notably weaker than each of the soil-applied products evaluated and weaker than YieldGard® Rootworm maize. The distribution of individual plant root ratings can provide a measure of insecticide performance consistency. Although a mean RDR may be below the economic threshold for an insecticide treatment, there can be considerable variability in root feeding damage within a given field. Pruning of one or more roots, which is equivalent to an RDR ≥ 3, causes sufficient damage to result in a yield loss for an individual plant. Any plant that has root pruning at a level equivalent to an RDR ≥ 4 is at risk of lodging. Plant lodging greatly increases the time and complexity of harvesting a field. The consistency of performance for YieldGard® Rootworm hybrids was compared to that of leading soil- and seed-applied insecticides and an untreated check by examining their distribution of RDR. Individual plant root ratings were grouped by rating (i.e. 1, 2, etc.) and expressed as a percentage of the total roots rated for each treatment. Figure 12.2 displays the distribution of individual root ratings recorded in 2000 and 2001 university efficacy trials. Only data from field trials in which the average check RDR was ≥ 3 were used in the analysis of performance consistency. An average RDR ≥ 3 is indicative of significant rootworm pressure in a field. A total of 559 YieldGard® Rootworm roots were rated and included in this analysis. Only 12% of the YieldGard® Rootworm plants rated had an RDR of 3; not a single plant was found with root damage exceeding an RDR of 3. In contrast, of the 465 plants treated with chlorpyrifos (formulated as Lorsban® 15G Granular Insecticide),4 a widely used insecticide for corn rootworm

3Prescribe® is a trademark of Gustafson LLC. 4Lorsban® is a trademark of Dow AgroSciences LLC. chap12.qxd 12/11/04 1:22 PM Page 253

Genetically Enhanced Maize as a Potential Management Option 253

Fig. 12.2. The consistency of performance for YieldGard® Rootworm maize was compared to that of alternative control measures and an untreated check by examining their distribution of root damage ratings. Individual plant root ratings grouped by score and expressed as a percentage of the total roots rated for each treatment. The data presented are from university trials conducted in 2000 and 2001 and include only trials that had an average check RDR ≥ 3. Without significant rootworm pressure in a field it is not possible to distinguish one treatment from another. The value in ( ) represents the number of individual plants rated for each treatment. Cyfluthrin/tebupirimfos (Cyflu/Tebu) and imidacloprid data are for 2001 only.

control, 80% of plants had an RDR ≥ 3 and a small percentage of plants received the maximum RDR of 6. Consistency of performance was weakest with the imidacloprid seed treatment. This analysis demon- strates that YieldGard® Rootworm consistency of performance is superior to that of both soil- and seed-applied insecticides. In a multi-year study to investigate the impact of corn rootworm larval feeding on compensatory root growth and yield of conventional corn, Gray and Steffey (1998) were able to demonstrate through a regression analysis that incremental reductions in RDR equated to a yield benefit for growers. Mitchell (2002) has used the Gray and Steffey field data to estimate proportional yield loss as a function of the root rating difference. Applying this model to the data from efficacy field trials with YieldGard® Rootworm hybrids provides an estimate of the yield benefit chap12.qxd 12/11/04 1:22 PM Page 254

254 D.P. Ward et al.

1.2

1.0

0.8

0.6

0.4 Yield gain (metric tons/ha)

0.2

0.0 YieldGard® Teﬂuthrin Chlorpyrifos

Fig. 12.3. Average grain yield from replicated small plot trials with YieldGard® Rootworm maize and hybrids treated with two leading corn rootworm insecticides, tefluthrin and chlorpyrifos, were compared to yield values for untreated conventional hybrids. The results presented represent yield advantage over untreated conventional hybrids. Average yield in the untreated conventional hybrids was 10.04 t/ha. Statistically significant differences at P < 0.05 are denoted by letter code where different letters correspond to significant differences.

for the YieldGard® Rootworm trait relative to no corn rootworm control and relative to control with a soil-applied insecticide. Over typical ranges for corn rootworm pressure, YieldGard® Rootworm hybrids provide a yield benefit of 9–28% relative to no pest control and 1.5–4.5% relative to control with leading soil insecticides. Data obtained from small plot trials conducted by Monsanto at 11 locations in 2001 and 13 locations in 2002 with multiple hybrids corroborate the Mitchell prediction of a yield advantage for YieldGard® Rootworm maize. These trials demonstrated that under economically significant levels of corn rootworm infestation (i.e., RDR ≥ 3) YieldGard® Rootworm hybrids provided an average 10.6% yield advantage over untreated conventional maize, which equated to 1.02 t/ha. The YieldGard® Rootworm yield advantage over two leading granular insecticides, tefluthrin (formulated as Force® 3G Insecticide)5 and chlorpyrifos, was 1.6% and 3.9%, respectively (Fig. 12.3). These findings corroborate the yield gain predictions of Mitchell.

5Force® is a trademark of Syngenta Crop Protection, Inc. chap12.qxd 12/11/04 1:22 PM Page 255

Genetically Enhanced Maize as a Potential Management Option 255

Beneﬁts of YieldGard® Rootworm Maize

There are many benefits associated with the commercialization of YieldGard® Rootworm maize varieties. Recipients of these benefits will be growers, consumers and the environment. Varieties containing the YieldGard® Rootworm trait will also be combined through conventional breeding with other genetically enhanced maize varieties, such as those with herbicide tolerance or protection against other insects. Maize hybrids containing the YieldGard® Rootworm trait are more efficacious than soil- and seed-applied insecticides in protecting roots from larval feeding damage. The Cry3Bb1 toxin is root-incorporated, it does not require activation (as many conventional insecticides do) and its performance is unlikely to be affected by severe environmental conditions. Superior performance and consistency of control are expected to result in a yield advantage for growers planting YieldGard® Rootworm hybrids. Preliminary estimates place this yield advantage at up to 4.5%. For a reasonable range of prices and yields, the value of the YieldGard® Rootworm yield benefit is US$62–185/ha relative to no pest control and US$10–30/ha relative to control provided by a soil insecticide, depending on corn rootworm pressure (Mitchell, 2002). Another estimate places the net economic benefit to adopters of this technology at US$16.21/ha (EPA, 2003). YieldGard® Rootworm hybrids also provide growers with operational benefits. For maize, early planting usually results in a longer growing season and higher yield. However, early planting can result in insecticide performance failures because of chemical dissipation prior to larval hatch. The effectiveness of a root-incorporated insecticide will not diminish with early planting. In addition, growers are able to plant their maize crop in a shorter period of time because there is no need to stop and refill insecticide applicators. Reducing the time required for growers to complete the planting operation increases the likelihood that the crop can be planted during optimal weather conditions. Due to its consistency of performance, this technology reduces a grower’s risk of yield loss. The results of a market survey indicate that the operational, yield risk reduction and environmental benefits offered by YieldGard® Rootworm are of value to growers (Alston et al., 2002). These non-pecuniary benefits were valued at US$17.89/ha by likely adopters of the technology. The quantity of conventional insecticides used to control corn rootworms annually exceeds the quantity applied to control any other targeted pest in any other crop (Doane, 2003). All chemical insecticides applied for control of corn rootworm are neurotoxic and hazardous to the environment. Each year there are confirmed reports of human illness and fish and bird poisonings associated with the conventional insecticides that are now used for control of corn rootworm pests (EPA, 2003). Cry3Bb1 is less hazardous than all insecticide active ingredients currently approved for corn rootworm control. Adoption of YieldGard® Rootworm maize hybrids will provide an opportunity to reduce the occu- chap12.qxd 12/11/04 1:22 PM Page 256

256 D.P. Ward et al.

pational and environmental risks associated with the manufacture, transport, storage, handling, application and disposal of conventional chemical insecticides. In order to preserve the value of efficacious new technologies for managing corn rootworms, implementation of an effective insect resistance management (IRM) plan is a vital part of responsible product stewardship. Such a programme should be developed based on the best available data and considerations for scientific uncertainties to ensure the long-term availability of this technology for maize growers as a pest management tool. IRM programmes must also strike a balance between science and the practicalities associated with acceptance and implementation by growers. An IRM programme has been designed and implemented to meet these objectives for YieldGard® Rootworm maize. This programme includes a baseline susceptibility determination for corn rootworms and surveillance for changes in susceptibility to the Cry3Bb1 protein. Planting of a structured refuge will ensure an adequate supply of susceptible rootworms to mate with resistant individuals. An extensive grower awareness and education programme has also been implemented to stress the importance of compliance with the IRM programme, and a mitigation plan has been developed in case resistance is detected in the field.

Summary

The corn rootworm species complex threatens the economic value of maize production systems in the USA and possibly in Europe. Crop rotation and application of conventional chemical insecticides have been the management options of choice for maize growers. However, incomplete crop protection with insecticides, the development of resistance to insecticides and the biological adaptation of rootworms to crop rotation have diminished the effectiveness of these pest management tactics. There are potential environmental, societal and economic benefits associated with incorporating genetic rootworm protection technology into maize production systems. These benefits include improved root protection, reduced insecticide use, increased yield, reduced farm worker health risk, reduced environmental risk, resource conservation, increased grower efficiency and increased economic return. The use of genetically enhanced maize varieties has already begun to dramatically transform insect and weed management in maize production systems. The recent commercialization of rootworm-protected maize varieties will further accelerate this transformation process.

Acknowledgements

The authors would like to thank L. English for his leadership in developing the Cry3Bb1 technology that went into creating event MON 863; S. chap12.qxd 12/11/04 1:22 PM Page 257

Genetically Enhanced Maize as a Potential Management Option 257

Johnson, A. Howe, T. Coombe, M. Groth and M. Pleau for their efforts in transformation and evaluation of transformation events that led to the selection of event MON 863 for commercialization; T. Cavato, A. Whetsell, J. Bookout, J. Hillyard, E. Rigden, R. Lirette, Y. Dudin, R. Hileman, G. Holleschak, P. Pyla, J. Leach, R. Thoma, W. Ridley, C. George, M. Taylor, Gary Hartnell, T. Nickson, J. Duan, M. Bhatti, S. Dubelman and J. Colyer for their roles in characterization of the inserted DNA, safety evaluation of maize hybrids containing event MON 863 and safety evaluation of the Cry3Bb1 protein; R. Sidhu, S. Brown and N. Bogdanova for obtaining regulatory approvals that allowed for ﬁeld testing and commercialization of this product; and R. Starke, W. Morjam and C. Pilcher for developing the extensive data sets on agronomic equivalency, product efﬁcacy and hybrid yield.

References

Al-Deeb, M.A., Wilde, G.E., Blair, J.M. and Todd, T.C. (2003) Effect of Bt corn for corn rootworm control on nontarget soil microarthropods and nematodes. Environmental Entomology 32, 859–865. Alston, J.M., Hyde, J., Marra, M.C. and Mitchell, P.D. (2002) An ex ante analysis of the benefits from the adoption of corn rootworm resistant transgenic corn technology. AgBioForum 5(3), 71–84. ARS (2002) Areawide Pest Management of Corn Rootworm in Maize Production Systems. 2002 Annual Report. Agricultural Research Service, United States Department of Agriculture, Washington, DC, 8 pp. http://www.ars.usda.gov/research/projects/ projects.htm?ACCN_NO=403818 Accessed 10 October 2003. Astwood, J.D. and Fuchs, R.L. (2001) Status and safety of biotech crops. In: Baker, D.R. and Umetsu, N.K. (eds) Agrochemical Discovery: Insect, Weed, and Fungal Control. American Chemical Society Symposium Series, No. 774, American Chemical Society, Washington, DC, pp. 152–164. Baum, J.A., Kakefuda, M. and Gawron-Burke, C. (1996) Engineering Bacillus thuringiensis bioinsecticides with an indigenous site-specific recombination system. Applied and Environmental Microbiology 62, 4367–4373. Berger, L.L., Robbins, N.D., Sewell, J.R., Stanisiewski, E.P. and Hartnell, G.F. (2003) Effect of feeding diets containing corn grain with corn rootworm protection (event MON 863), control, or conventional varieties on steer feedlot performance and carcass characteristics. Journal of Animal Science 80 (Suppl. 1), 214. Bernhard, K. and Utz, R. (1993) Production of Bacillus thuringiensis Insecticides for experimental and commercial uses. In: Entwistle, P.F., Cory, J.S., Bailey, M.J. and Higgs, S. (eds) Bacillus thuringiensis, an Environmental Biopesticide: Theory and Practice. John Wiley & Sons, New York, pp. 255–267. Betz, F.S., Hammond, B.G. and Fuchs, R.L. (2000) Safety and advantages of Bacillus thuringiensis-protected plants to control insect pests. Regulatory Toxicology and Pharmacology 32, 165–173. Branson, T.F., Sutter, G.R. and Fisher, J.R. (1982) Comparison of a tolerant and susceptible maize inbred under artificial infestations of Diabrotica virgifera virgifera: yield and adult emergence. Environmental Entomology 11, 371–372. Bressner, G.E., Hyun, Y., Stanisiewski, E.P., Hartnell, G.F. and Ellis, M. (2003) Performance comparison of growing-finishing pigs fed diets containing corn rootworm chap12.qxd 12/11/04 1:22 PM Page 258

258 D.P. Ward et al.

protected corn (event MON 863) or conventional corn hybrids. Journal of Animal Science 81 (suppl.), 207. CRA (2003) Corn Annual 2003. Corn Reﬁners Association, Inc., Washington, DC, 24 pp. http://www.corn.org/web/ca2003.html Accessed 10 October 2003. Crickmore, N., Zeigler, D.R., Feitelson, J., Schnepf, E., Van Rie, J., Lereclus, D., Baum, J. and Dean, D.H. (1998) Revision of the nomenclature for the Bacillus thuringiensis pesticidal crystal proteins. Microbiology and Molecular Biology Reviews 62, 807–813. Doane (2001) The AgroTrak Study – 2000. A syndicated study conducted by Doane Marketing Research, Inc., St Louis, Missouri. Doane (2003) The AgroTrak Study – 2002. A syndicated study conducted by Doane Marketing Research, Inc., St Louis, Missouri. Donovan, W.P., Rupar, M.J., Slaney, A.C., Malvar, T., Gawron-Burke, M.C. and Johnson, T.B. (1992) Characterization of two genes encoding Bacillus thuringiensis insecticidal crystal proteins toxic to Coleoptera species. Applied and Environmental Microbiology 58, 3921–3927. Drees, B.M., Levine, E., Stewart, J.W., Sutter, G.R. and Tollefson, J.J. (1999). Management of corn rootworms. In: Steffy, K.L., Rice, M.E., All, J., Andow, D.A., Gray, M.E. and Van Duyn, J.W. (eds) Handbook of Corn Insects. Entomological Society of America, Lanham, Maryland, pp. 66–68. English, L. and Slatin, S.L. (1992) Mode of action of delta-endotoxins from Bacillus thuringiensis: a comparison with other bacterial toxins. Insect Biochemistry and Molecular Biology 22, 1–7. English, L.H., Brussock, S.M., Malvar, T.M., Bryson, J.W., Kulesza, C.A., Walters, F.S., Slatin, S.L., Von Tersch, M.A. and Romano, C. (2000) Insect-resistant transgenic plants. United States Patent 6,023,013 B1. EPA (1988). Guidance for the Reregistration of Pesticide Products Containing Bacillus thuringiensis as the Active Ingredient. NTIS PB 89–164198, US Environmental Protection Agency, Washington, DC. EPA (2003) Biopesticides Registration Action Document: Event MON 863 Bacillus thuringiensis Cry3Bb1 Corn. [Beneﬁts assessment, pp. IIE1–IIE34.] US Environmental Protection Agency, Washington, DC, 141 pp. http://www.epa.gov/pesticides/ biopesticides/ingredients/tech_docs/brad_006484.htm Accessed 10 October 2003. EPPO (2003) Diabrotica virgifera. European and Mediterranean Plant Protection Organization, Paris, 7 pp. http://www.eppo.org/QUARANTINE/Diabrotica_vir- gifera/diabrotica_virgifera.html Accessed 10 October 2003. Fischer, R.L., Miller, P.S., Hyun, Y., Hartnell, G.F. and Stanisiewski, E.P. (2003) Comparison of swine performance when fed diets containing corn rootworm protected corn, parental line corn, or conventional corn grown during 2000 in Nebraska. Journal of Animal Science 81 (Suppl.), 207. George, C., Ridley, W.P., Obert, J.C., Nemeth, M.A., Breeze, M.L. and Astwood, J.D. (2004) Composition of grain and forage from corn rootworm-protected corn event MON 863 is equivalent to that of conventional corn (Zea mays L.). Journal of Agricultural and Food Chemistry 52, 4149–4158. Gianessi, L.P., Silvers, C.S., Sankula, S. and Carpenter, J.E. (2002a) Plant Biotechnology: Current and Potential Impact for Improving Pest Management in US Agriculture. An Analysis of 40 Case Studies. Insect Resistant Field Corn, 3, National Center for Food and Agricultural Policy, Washington, DC, 30 pp. http://www.ncfap.org/ 40CaseStudies/CaseStudies/FieldCornIR3.pdf Accessed 10 October 2003. Gianessi, L.P., Silvers, C.S., Sankula, S. and Carpenter, J.E. (2002b) Plant Biotechnology: Current and Potential Impact for Improving Pest Management in US Agriculture. An Analysis of 40 Case Studies. National Center for Food and Agricultural Policy, chap12.qxd 12/11/04 1:22 PM Page 259

Genetically Enhanced Maize as a Potential Management Option 259

Washington, DC, 75 pp. http://www.ncfap.org/40CaseStudies.htm Accessed 10 October 2003. Gordon-Kamm, W.J., Spencer, T.M., Mangano, M.L., Adams, T.R., Daines, R.L., O’Brian, J.V., Start, W.G., Adams, W.R., Chambers, S.A., Willets, N.G., Mackey, C.J., Krueger, R.W., Kausch, A.P. and Lemaux, P.G. (1990) Transformation of maize cells and regeneration of fertile transgenic plants. Plant Cell 2, 603–618. Grant, R.J., Fanning, K.C., Kleinschmit, D., Stanisiewski, E.P. and Hartnell, G.F. (2003) Influence of glyphosate-tolerant (event NK603) and corn rootworm protected (event MON 863) corn silage and grain on feed consumption and milk production in Holstein cattle. American Dairy Science 86, 1707–1715. Gray, M.E. (2002) Preliminary Root-rating Results from Insecticide Efficacy Trials Available. Pest Management and Crop Development Bulletin, No. 22, University of Illinois Extension, Champaign-Urbana, Illinois, 2 pp. http://www.ag.uiuc.edu/ cespubs/pest/articles/v200222.htm Accessed 10 October 2003. Gray, M.E. and Steffey, K.L. (1998) Corn rootworm (Coleoptera: Chrysomelidae) larval injury and root compensation of 12 maize hybrids: an assessment of the economic injury index. Journal of Economic Entomology 91, 723–740. Gray, M.E., Levine, E. and Oloumi-Sadeghi, H. (1998) Adaptation to crop rotation: western and northern corn rootworms respond uniquely to cultural practice. Recent Research Developments in Entomology 2, 19–31. Hileman, R.E., Pyla, P.D., Holleschak, G., Leach, J.N., Martin, J.W., Lee, T.C., Thoma, R.S. and Astwood, J.D. (2001) Characterization of Bacillus thuringiensis Cry3Bb1 protein produced in B.t. and insect protected corn plants. Toxicologist 60(1), 411. Hileman, R.E., Silvanovich, A., Goodman, R.E., Rice, E.A., Holleschak, G., Astwood, J.D. and Hefle, S.L. (2002) Bioinformatic methods for allergenicity assessment using a comparative allergen database. International Archives of Allergy Immunology 128, 280–291. Hills, T.M. and Peters, D.C. (1971) A method of evaluating post-planting insecticide treatments for control of western corn rootworm larvae. Journal of Economic Entomology 64, 764–765. Hoffman, C., Lüthy, P., Hutter, R. and Pliska, V. (1988a) Binding of the delta endotoxin from Bacillus thuringiensis to brush-border membrane vesicles of the cabbage butterfly (Pieris brassicae). European Journal of Biochemistry 173, 85–91. Hoffman, C., Vanderbruggen, H.V., Hofte, H., Van Rie, J., Jansens, S. and Van Mellaert, H. (1988b) Specificity of B. thuringiensis delta-endotoxins is correlated with the presence of high affinity binding sites in the brush border membrane of target insect midguts. Proceedings of the National Academy of Sciences 85, 7844–7848. Jackson, J.J. (1996) Field performance of entomopathogenic nematodes for suppression of western corn rootworm (Coleoptera:Chrysomelidae). Journal of Economic Entomology 89, 366–372. Jugenheimer, R.W. (1976) Corn Improvement, Seed Production, and Uses. John Wiley & Sons, New York, 649 pp. Klein, T.M., Wolf, E.D., Wu, R. and Sanford, J.C. (1987) High velocity microprojectiles for delivering nucleic acids into living cells. Nature 327, 70–73. Knutson, R.J., Hibbard, B.E., Barry, B.D., Smith, V.A. and Darrah, L.L. (1999) Comparison of screening techniques for western corn rootworm (Coleoptera: Chrysomelidae) host- plant resistance. Journal of Economic Entomology 92, 714–722. Leach, J.N., Pyla, P.D., Holleschak, G., Hileman, R.E., Lee, T.C., Bechtel, C.L., Hammond, B.G. and Astwood, J.D. (2001) Safety assessment of insect control Bacillus thuringiensis Cry3Bb1 protein for use in transgenic crops. Toxicologist 60(1), 414. Levine, E., Spencer, J.L., Isard, S.A., Onstad, D.W. and Gray, M.E. (2002) Adaptation of the chap12.qxd 12/11/04 1:22 PM Page 260

260 D.P. Ward et al.

western corn rootworm to crop rotation: evolution of a new strain in response to a management practice. American Entomogist 48, 94–107. McClintock, J.T., Schaffer, C.R. and Sjoblad, R.D. (1995) A comparative review of the mammalian toxicity of Bacillus thuringiensis-based pesticides. Pesticide Science 45, 95–105. Mendelsohn, M., Kough, J., Vaituzis, Z. and Matthews, K. (2003) Are Bt crops safe? Nature Biotechnology 21(9), 1003–1009. Menestrina, G. and Semjen, B.V. (1999) Biophysical methods and model membranes for the study of bacterial pore-forming toxins. In: Alouf, J.E. and Freer, J.H. (eds) The Comprehensive Sourcebook of Bacterial Protein Toxins. Academic Press, New York, pp. 287–309. Metcalfe, D.D., Astwood, J.D., Townsend, R., Sampson, H.A., Taylor, S.L. and Fuchs, R.L. (1996) Assessment of the allergenic potential of foods derived from genetically engi- neered crop plants. Critical Reviews in Food Science and Nutrition 36, S165–S186. Metcalf, R.L. (1986) Foreword. In: Krysan, J.L. and Miller, T.A. (eds) Methods for the Study of Pest Diabrotica. Springer-Verlag, New York, pp. vii–xv. Metcalf, R.L. and Lampman, R.L. (1997) Plant kairomones and kairomone mimetics in basic and applied research with diabroticite rootworms. Trends in Entomology 1, 49–62. Mitchell, P.D. (2002) Yield Benefit of Corn Event MON 863. Faculty Paper Series Fp 02-04, Department of Agricultural Economics,Texas A&M University, College Station, Texas, 27 pp. http://agecon.tamu.edu/publications/fp02-04.pdf Accessed 10 October 2003. Moellenbeck, D.J., Peters, M.L., Bing, J.W., Rouse, J.R., Higgins, L.S., Sims, L., Nevshemal, T., Marshall, L., Ellis, R.T., Bystrak, P.G., Lang, B.A., Stewart, J.L., Kouba, K., Sondag, V., Gustafson, V., Nour, K., Xu, D., Swenson, J., Zhang, J., Czapla, T., Schwab, G., Jayne, S., Stockhoff, B.A., Narva, K., Schnepf, H.E., Stelman, S.J., Poutre, C., Koziel, M. and Duck, N. (2001) Insecticidal proteins from Bacillus thuringiensis protect corn from corn rootworms. Nature Biotechnology 19, 668–672. Monsanto (2003) Safety Assessment of YieldGard Rootworm Corn. Monsanto Company, St Louis, Missouri, 44 pp. http://www.monsanto.com/monsanto/content/our_pledge/ pss_yieldgardrootwormcorn.pdf Accessed 10 October 2003. NASS (2001) Agricultural Chemical Usage. 2000 Field Crops Summary. United States Department of Agriculture, National Agricultural Research Service, Washington, DC, 105 pp. http://usda.mannlib.cornell.edu/reports/nassr/other/pcu-bb/agch0599.pdf Accessed 10 October 2003. NCGA (2003) The World of Corn. National Corn Growers Association, Washington, DC, 21 pp. http://www.ncga.org.html Accessed 10 October 2003. Noteborn, H. (1994) Safety assessment of genetically modified plant products. Case study: Bacillus thuringiensis-toxin tomato. In: Rack, N. (ed.) Biosafety of Foods Derived by Modern Biotechnology. Proceedings: Basel Forum on Biosafety, Vol. 4/94. BATS, Agency for Biosafety Research and Assessment of Technology Impacts, Basel, Switzerland, pp. 17–22. O’Day, M., Becker, A., Keaster, A., Kabrick, L. and Steffey, K. (1998) Corn Insect Pests. A Diagnostic Guide. Missouri Manual 166, Illinois Manual C1358, MU Extension, University of Missouri, Columbia, Missouri, 49 pp. O’Neal, M.E., Gray, M.E. and Smyth, C.A. (1999) Population characteristics of a western corn rootworm (coleoptera: chrysomelidae) strain in east-central Illinois corn and soybean fields. Journal of Economic Entomology 92(6), 1301–1310. Onstad, D.W. and Joselyn, M.G. (1999) Modeling the spread of western corn rootworm (coleoptera: chrysomelidae) populations adapting to soybean corn rotation. Environmental Entomology 28(2), 188–194. chap12.qxd 12/11/04 1:22 PM Page 261

Genetically Enhanced Maize as a Potential Management Option 261

Ostlie, K.R. (1987) Extended diapause northern corn rootworm adapts to crop rotation. Crops Soils Magazine 39, 23–25. Rice, M.E. (2004) Transgenic rootworm corn: Assessing potential agronomic, economic and environmental beneﬁts. Online. Plant Health Progress do::10.1094/PHP-2004- 0301-01-RV Rice, M.E. and Oleson, J. (2001) Rootworm insecticide performance. Integrated Crop Management Newsletter IC-486(24):196-198, Iowa State University, Ames, Iowa, 4 pp. http://www.ipm.iastate.edu/ipm/icm/2001/11-19-2001/performance.html Accessed 10 October 2003. Ridley, W.P., Sidhu, R.S., Pyla, P.D., Nemeth, M.A., Breeze, M.L. and Astwood, J.D. (2002) Comparison of the nutritional proﬁle of glyphosate-tolerant corn event NK603 with that of conventional corn (Zea mays L.). Journal of Agricultural and Food Chemistry 50, 7235–7243. Riedell, W.E. (1990) Rootworm and mechanical damage effects on root morphology and water relations in maize. Crop Science 30, 628–631. Rupar, M.J., Donovan, W.P., Groat, R.G., Slaney, A.C., Mattison, J.W., Johnson, T.B., Charles, J.F., Dumanoir, V.C. and De Barjac, H. (1991) Two novel strains of Bacillus thuringiensis toxic to coleopterans. Applied and Environmental Microbiology 57, 3337–3344. Sacchi, V.F., Parenti, P., Hanozet, G.M., Giordana, B., Luthy, P. and Wolfersberger, M.G. (1986) Bacillus thuringiensis toxin inhibits K+-gradient-dependent amino acid transport across the brush border membrane of Pieris brassicae midgut cells. Federation of European Biochemical Societies 204, 213–218. Schnepf, E., Crickmore, N., Van Rie, J., Lereclus, D., Baum, J., Feitelson, J., Zeigler, D.R. and Dean, D.H. (1998) Bacillus thuringiensis and its pesticidal crystal proteins. Microbiology and Molecular Biology Reviews 62, 775–806. Sidhu, R.S., Hammond, B.G., Fuchs, R.L., Mutz, J.-N., Holden, L.R., George, B. and Olson, T. (2000) Glyphosate-tolerant corn: the composition and feeding value of grain from glyphosate-tolerant corn is equivalent to that of conventional corn (Zea mays L.). Journal of Agricultural and Food Chemistry 48(6), 2305–2312. Spike, B.P. and Tollefson, J.J. (1991) Yield response of corn subjected to western corn rootworm (Coleoptera: Chrysomelidae) infestation and lodging. Journal of Economic Entomology 84, 1585–1590. Taylor, M.L., Hartnell, G.F., Astwood, J.D., Nemeth, M.A. and George, B. (2003) Comparision of broiler performance when fed diets containing MON 863 (corn rootworm protection), nontransgenic control, or commercial corn. Poultry Science 81 (Suppl. 1), 95. Taylor, S.L. (1992) Chemistry and detection of food allergens. Food Technology 46, 146–152. Tollefson, J.J. (1988) A pest insect adapts to the cultural control of crop rotation. In: Brighton Crop Protection Conference. Pests and Diseases – 1988. British Crop Protection Council, Thornton Heath, UK, pp. 1029–1033. Vander Pol, K.J., Erickson, G.E., Macken, C.N., Blackford, M.P., Klopfenstein, T.J., Stanisiewiski, E.P. and Hartnell, G.P. (2002). Effect of corn root worm protected corn (event MON 863) on performance in beef feedlot diets. Journal of Animal Science 80 (Suppl. 1), 46. Van Mellaert, H., Van Rie, J., Hofmann, C. and Reynaerts, A. (1988) Insecticidal crystal proteins from Bacillus thuringiensis: mode of action and expression in transgenic plants. In: Conference on Biotechnology, Biological Pesticides and Novel Plant-pest Resistance for Insect Pest Management. Boyce Thompson Institute, Cornell University, New York, pp. 82–87. chap12.qxd 12/11/04 1:22 PM Page 262

262 D.P. Ward et al.

Van Rie, J., Jansens, S., Höfte, H., Degheele, D. and Van Mellaert, H. (1989) Specificity of Bacillus thuringiensis delta-endotoxins: importance of specific receptors on the brush border membrane of the midgut of target insect. European Journal of Biochemistry 186, 239–247. Van Rie, J., Jansens, S., Höfte, H., Degheele, D. and Mellaert, H.V. (1990) Receptors on the brush border membrane of the insect midgut as determinants of the specificity of Bacillus thuringiensis delta-endotoxins. Applied Environmental Microbiology 56, 1378–1385. Watson, S.A. (1982) Maize: amazing maize. General properties. In: Wolff, I.A. (ed.) CRC Handbook of Processing and Utilization in Agriculture, Vol. II: Part 1 Plant Products. CRC Press, Boca Raton, Florida, pp. 3–29. Watson, S.A. (1987) Structure and composition. In: Watson, S.A. and Ramstad, R.E. (eds) Corn Chemistry and Technology. American Association of Cereal Chemists, St Paul, Minnesota, pp. 53–82. Wolfersberger, M.G., Hofmann, C. and Lüthy, P. (1986) Interaction of Bacillus thuringiensis delta-endotoxin with membrane vesicles isolated from lepidopteran larval midgut. In: Falmagne, P., Fehrenbech, F.J., Jeljaszewics, J. and Thelestam, M. (eds) Bacterial Protein Toxins. Gustav Fischer, New York, pp. 237–238. Wright, R., Meinke, L. and Siegfried, B. (1999) Adult Western Corn Rootworm Insecticide Resistance in Nebraska. NebFacts, NF99–367, Nebraska Cooperative Extension, Lincoln, Nebraska, 5 pp. http://www.ianr.unl.edu/pubs/insects/nf367.htm Accessed 10 October 2003. chap13.qxd 12/11/04 1:43 PM Page 263

13 Is Classical Biological Control against Western Corn Rootworm in Europe a Potential Sustainable Management Strategy? Ulrich Kuhlmann, Stefan Toepfer and Feng Zhang CABI Bioscience Switzerland Centre, Delémont, Switzerland

Introduction

Invasive alien species are recognized as one of the leading threats to biodiversity and also impose enormous economic impacts on agriculture, forestry, fisheries, as well as on human health (Wittenberg and Cock, 2001). Each year, the rate of invasions of harmful insects increases around the world as increased commerce and tourism accelerate. Natural enemies often control invasive alien species in their indigenous range; however, invasive species are usually introduced into new environments without these coevolved and specific natural enemies. Recent important examples of new invasions include the cypress aphid, Cinara cupressi (Buckton) (Hemiptera: Aphididae), in eastern Africa (Mills, 1990), the palm thrips, Thrips palmi Karney (Thysanoptera: Thripidae), in Australia (Houston et al., 1991), and the medfly, Ceratitis capitata (Wiedemann) (Diptera: Tephritidae), in California, USA (Carey, 1992). In contrast, other regions, notably Europe, have seldom experienced agricultural insect invasions. In the last century, notable invaders in Europe from America include the woolly apple aphid, Eriosoma lanigerum Hausmann (Hemiptera: Aphididae), the American white moth, Hyphantia cunea Drury (Lepidoptera: Arctiidae), the citrus whitefly, Aleurothrixus flocco- sus Maskell (Hemiptera: Aleyrodidae), the western flower thrips, Frankliniella occidentalis (Pergande), and the Colorado potato beetle, Leptinotarsa decemlineata Say (Coleoptera: Chrysomelidae) (Greathead, 1976; Waterhouse and Norris, 1987). Europe’s most recent successful invader, the western corn rootworm, Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae), is the most destructive pest of maize (Zea mays L.) in North America (Krysan and Miller, 1986), which began its successful invasion of Central Europe

264 U. Kuhlmann et al.

in the early 1990s (Baca, 1994; Sivcev et al., 1994). The majority of yield loss attributed to this univoltine pest species is due to larval feeding on the roots of maize, which ultimately results in plant lodging; however the adults can occasionally cause yield losses due to intensive silk feeding (Krysan and Miller, 1986). Since 2003, D. v. virgifera has invaded almost all maize production areas within the European Community and can potentially cause signiﬁcant economic damage if control measures are not implemented. In a sustainable integrated pest management (IPM) approach to control D. v. virgifera in Europe, classical biological control may have an important application as it provides an opportunity to partially reconstruct the natural enemy complex of an invading alien pest (Mills, 1994). In general, its application has been highly recommended to control established alien invasive pest populations (Wittenberg and Cock, 2001). Therefore, invaders such as D. v. virgifera are prime targets for a classical biological control approach. Based on the few insect pest invasions known from Europe, substantial successes of classical biological control have been achieved in the past, for example, against the woolly apple aphid, using Aphelinus mali Haldeman (Hym.: Eulophidae), citrus whiteﬂy, releasing Cales noaki Howard (Hym.: Eulophidae), and the San José scale, importing Encarsia perniciosi (Tower) (Hym.: Eulophidae) (Greathead, 1976). A step-by-step approach applied in classical biological control of arthropod pests requires the assessment of the following four important themes prior to the potential natural enemy introduction:

Survey for natural enemies in the area of invasion (here Europe). Foreign exploration and selection of natural enemies from the area of origin (here Central and South America). Suitability of candidate biological control agents. Host speciﬁcity of the candidate biological control agents.

This step-by-step approach displayed above has been applied, aiming to select classical biological control agents for D. v. virgifera in Europe. At the same time life table studies have been conducted with the aim of assessing the signiﬁcance of natural mortality factors acting on D. v. virgifera populations in its area of invasion (for details, see Toepfer and Kuhlmann, Chapter 5, this volume). Regarding this classical biological control approach, life tables will provide essential ecological background information about host niches of D. v. virgifera that are not at present occupied by indigenous natural enemies. Based on this information, spe- ciﬁc and effective natural enemies from the area of origin could be selected and considered for potential introduction into Europe. This chapter describes a 3-year study implemented to evaluate the potential of classical biological control as a sustainable management strategy for D. v. virgifera in Europe. It is divided into four sections applying the step-by- step approach for classical biological control and thereby provides results about: (i) the survey for natural enemies in the area of invasion; (ii) the chap13.qxd 12/11/04 1:43 PM Page 265

Classical Biological Control against Western Corn Rootworm in Europe 265

foreign exploration for natural enemies in the area of origin; (iii) the suitability of candidate biological control agents; and (iv) the host speciﬁcity of the candidate biological control agents.

Survey for Natural Enemies in the Area of Invasion

The first step of such a classical biological control approach requires the assessment of the indigenous natural enemy community of D. v. virgifera present in the invaded area of Central Europe prior to the introduction of specific classical biological control agents from the area of origin in Central and South America. This section reports on a 3-year field survey conducted in Hungary, Serbia and Croatia, which are currently the focal points of invasion (Kiss et al., 2001), with the aim of determining the occurrence of indigenous natural enemies of D. v. virgifera in Europe. To assess the diversity of indigenous natural enemies, field surveys were conducted to collect eggs, larvae of three instars, pupae and adults of D. v. virgifera in the countries mentioned above. Fifteen maize fields were sampled to obtain all life stages of D. v. virgifera between 2000 and 2002. All life stages of D. v. virgifera were collected in the field, trans- ferred to the laboratory and screened for the presence of parasitoids, nematodes and fungal pathogens, following a general screening procedure explained in detail by Toepfer and Kuhlmann (2004). A total of 9900 eggs, 550 larvae, 70 pupae and 33,000 adults were examined for the occurrence of parasitoids, nematodes and fungal pathogens. It can be concluded from the survey results that effective indigenous natural enemies are not attacking any of the life stages of D. v. virgifera in Central Europe. The exception is the occurrence of the fungi Beauveria bassiana (Bals.) Vuill. (mitosporic fungi; formerly Deuteromyces) and Metarhizium anisopliae (Metsch.) Sorok (mitosporic fungi) attacking adults of D. v. virgifera on an extremely low level (< 1%). However, no other entomopathogenic fungal pathogens, entomopathogenic nematodes or insect parasitoids were found on eggs, larvae, pupae or adults. It is believed that D. v. virgifera and maize evolved together in the subtropics of Mexico and Central America (Branson and Krysan, 1981); therefore it is not surprising that important natural enemies are lacking in Europe, as they have been left behind in the area of origin. It should be noted that there are also no reports about effective and specific natural enemies of D. v. virgifera in the USA where it is also an exotic invader. Obviously, indigenous polyphagous natural enemies in Europe have not yet adapted to the invasive alien species over the 10–15-year period following its invasion. This is strongly supported by field surveys carried out in Serbia (Toepfer and Kuhlmann, 2004), where D. v. virgifera has been present for the longest time period in Europe (Kiss et al., 2001). Zwoelfer and Pschorn-Walcher (1968) reported that invasive alien species are most probably not attacked by indigenous natural enemies chap13.qxd 12/11/04 1:43 PM Page 266

266 U. Kuhlmann et al.

when the introduced host is taxonomically and/or phylogenetically different from indigenous genera, and/or its lifestyle or physiological characteristics are difficult for indigenous natural enemies to approach. The taxonomic–phylogenetic difference would be the case for D. v. virgifera, as it belongs to the New World diabroticites, whereas the genus Aulacophora represents Diabrotica in the Old World (Maulik, 1936) (both in the tribe Luperini within the subfamily Galerucinae). Although there are remarkable zoogeographical similarities in host plant affinities between the New World diabroticites and the Old World aulacophorites, Aulacophora species and associated indigenous natural enemies are not present in Central Europe (Metcalf, 1985) and therefore do not yet overlap with the distribution of D. v. virgifera. Similar to the D. v. virgifera invasion, other reports about alien chrysomelid invasions show that, even decades later, no signs of transfer adaptations of indigenous natural enemies have been recorded, such as for the Colorado potato beetle, L. decemlineata, in Europe (Greathead, 1976) or for the cereal leaf beetle, Oulema melanopus (Linnaeus), in North America (Haynes and Gage, 1981). On the other hand, few examples of transfer adaptations of indigenous natural enemies to invasive alien host species other than Chrysomelidae are known from Europe (Greathead, 1976) and North America (Zwoelfer and Pschorn-Walcher, 1968), such as to the American white moth, H. cunea, which is known to be attacked by indigenous natural enemies in Europe (Franz and Krieg, 1982). Concerning the occurrence of natural enemies, the exotic nature of maize itself in Europe has most probably resulted in a much lower arthropod diversity in comparison to other European crop systems. Thus, in the absence of specific natural enemies, D. v. virgifera populations in Europe are only limited by the availability of suitable food, the influence of abiotic factors and the application of control treatments. Based on the results presented, the search for and selection of specific natural enemies from the area of origin of D. v. virgifera for potential introduction was considered, as host niches are not occupied by effective indigenous natural enemies.

Foreign Exploration and Selection of Natural Enemies from the Area of Origin

The second step of this biological control approach requires the assessment of the structure and function of natural enemies in the area of origin, but special emphasis was placed on parasitoids of Diabrotica adults. It is known that the genus Diabrotica is of neotropical origin (Krysan and Branson, 1983), and most of the 354 Diabrotica species are distributed in Central and South America (Krysan, 1999). In general, the selection of potentially effective natural enemies in classical biological control programmes is usually focused on the area of origin of the pest (Waage, 1990; Van Driesche and Hoddle, 2000). In their area of origin, corn rootworms are attacked by a range of pathogens, nematodes, predators and para- chap13.qxd 12/11/04 1:43 PM Page 267

Classical Biological Control against Western Corn Rootworm in Europe 267

sitoids, some of which appear to be specifically adapted for parasitizing the corn rootworm (Kuhlmann and Van der Burgt, 1998). Currently four tachinids in the genus Celatoria are known from the literature to parasitize adults of Diabrotica species (Cox, 1994). Belshaw (1994) reported that these are more host-specific compared to many other tachinids based on the presence of the elaborately modified piercing ovipositor of the females. Another parasitoid species, the braconid Centistes gasseni Shaw, has recently been discovered (Shaw, 1995) and its biology described (Schroder and Athanas, 2002). The presence of Celatoria bosqi Blanchard is known from Argentina, Uruguay and southern Brazil (Herting, 1973; Guimaraes, 1977; Heineck-Leonel and Salles, 1997; G. Cabrera Walsh, 1998, personal communication), whereas Celatoria diabroticae (Shimer) is known from Oregon, Virginia, Illinois, South Carolina and Mississippi in the USA (Herting, 1973; Arnauld, 1978; Fischer, 1983; Elsey, 1988). The third tachinid, Celatoria setosa (Coquillet), is present in Indiana and Illinois, USA (Arnauld, 1978; Fischer, 1983), and its biology was studied by Bussart (1937) and Fischer (1983). The fourth tachinid species, Celatoria compressa Wulp, is the most common parasitoid of D. v. virgifera in Mexico but its behaviour and biology are unknown (Eben and Barbercheck, 1996; Fig. 13.1). This section reports on a 3-year field survey conducted in Mexico, Argentina and Brazil, with the aim of obtaining parasitoid species from Diabrotica adults for further suitability studies in Europe. In order to obtain parasitoids of Diabrotica adults, surveys were con-

Fig. 13.1. The tachinid ﬂy, Celatoria compressa, inserts eggs containing ﬁrst-instar larvae directly into its host Diabrotica virgifera virgifera. chap13.qxd 12/11/04 1:43 PM Page 268

268 U. Kuhlmann et al.

ducted in collaboration with Dr Astrid Eben (Instituto de Ecologia, Xalapa, Mexico) and Dr Rebeca Alvarez Zagoya (Instituto Politecnico Nacional, Durango, Mexico). Adults of Diabrotica spp. were collected in agricultural and natural habitats containing a high species diversity, including the target species D. v. virgifera in northern Mexico. In collaboration with G. Cabrera Walsh (US Department of Agricultural Research Service (USDA-ARS) South American Biological Control Laboratory, Buenos Aires, Argentina), Diabrotica adult natural enemy surveys were carried out in central and northern Argentina as well as south-eastern Brazil. Diabrotica adults were collected directly from leaves and flowers of maize, beans, squash or wild plants within the fields, using an aspira- tor or a plastic funnel with a gauze bag attached. In all surveys, every available species of the subtribe Diabroticina, e.g. Diabrotica spp., Acalymma spp. and Cerotoma spp., were collected. Diabrotica adults were separated based on species, collection site and collection date and colonies were maintained in cages in the laboratory with zucchini, squash fruits or artificial diet (Branson et al., 1975) until emergence of parasitoid larvae. Parasitoid pupae were shipped to the quarantine laboratory in Switzerland for further investigation. In Mexico, C. compressa was reared from 17 species in four different genera of the family Chrysomelidae between 2000 and 2002. In 2001, 2436 beetles belonging to different species were collected at various locations and from these 55 tachinid pupae were obtained, representing only 2.3% parasitism. In 2002, 4844 beetles were collected at the same locations and 99 tachinid puparia were reared, representing a parasitism rate of 2.0%. Parasitism within host species ranged from 0.3% in Gynandrobrotica nigrofasciata (Say) to 16.7% in Cerotoma atrofasciata Jacoby. The most abundant species in Veracruz, Mexico, were Acalymma blomorum Munroe and Smith and Diabrotica balteata LeConte. Parasitism of A. blomorum reached only 3.5% and 0.8% in 2001 and 2002, respectively, and 4.6% parasitism was found for D. balteata in 2002. From August to October 2001 and 2002, individuals from D. v. virgifera populations were collected in the state of Durango, northern Mexico. Numbers of D. v. virgifera in maize monocultures in Durango were nevertheless surprisingly low and comparable to numbers of other diabroticites collected from larger maize–squash fields in Veracruz. From a total of 640 D. v. virgifera collected in 2001, 36 C. compressa pupae were obtained, representing a mean parasitism rate of 5.6% (max. = 16%). In contrast, 304 D. v. virgifera were collected in the same region in 2002 and no parasitism at all was observed. In Argentina and Brazil, the tachinid C. bosqi was collected from three Diabrotica species. However, the main host was Diabrotica speciosa (Germar), whereas Diabrotica viridula (F.), which is closely related to D. v. virgifera, was rarely parasitized. In 2001 and 2002, the braconid C. gasseni was found during the surveys in Argentina and Brazil. Parasitism was generally low for D. speciosa, with 2%, D. viridula, with 1.2%, chap13.qxd 12/11/04 1:43 PM Page 269

Classical Biological Control against Western Corn Rootworm in Europe 269

Diabrotica limitata (Sahlberg), with 7.5%, and in some rare cases parasitism was found in Acalymma bivittulum (Kirsch), with 0.04%. In some regions of the surveyed Central and South American countries, the natural enemy community of adult diabroticite beetles was investigated. In this survey, two tachinid flies, C. compressa and C. bosqi, as well as one braconid wasp, C. gasseni, were found and identified as parasitoids of adult diabroticite beetles. Parasitism rates obtained in all Diabrotica species investigated were low throughout this study (see above). Nevertheless, based on survey results, adult parasitoids are probably the most common natural enemies of species in the virgifera group of the genus Diabrotica, followed by mermithid nematodes of the genus Hexamermis. The occurrence of the three parasitoid species obtained was expected as the existence of C. compressa in Mexico was reported before by Eben and Barbercheck (1996), and the presence of C. bosqi and C. gasseni in Argentina and Brazil has been reported by Guimaraes (1977), Heineck-Leonel and Salles (1997) and Cabrera Walsh et al. (2003). Although major efforts were made towards the investigation of the natural enemy community of diabroticite beetles in their area of origin some regions in Central and South America remained unexplored. In particular, the higher altitudes in the north-eastern parts of South America merit further investigation, as these areas are most probably more species-rich and may yield new parasitoids or strains of known parasitoid species parasitizing other diabroticite host species. As the focus of this survey was always placed on obtaining parasitoids of adult Diabrotica, knowledge of natural enemies attacking the soil-dwelling larval stages of Diabrotica beetles is still lacking. In the biological control agent selection process, three of the six known parasitoid species of adult Diabrotica, i.e. C. setosa, C. diabroticae and the braconid Centistes diabroticae Gahan, have not been considered. The North and Central American tachinid fly, C. setosa was not selected as Fischer (1983) indicated that this fly is almost exclusively a parasitoid of Acalymma species, and will rarely, if ever, attack and/or develop in Diabrotica species. The North American tachinid fly Celatoria diabroticae was also not chosen, although parasitism has been reported to be high in Diabrotica undecimpunctata howardi Barber (Summers and Stafford, 1953) and in D. undecimpunctata undecimpunctata Mannerheim (Fischer, 1981), as these are considered to be the key hosts. The North American braconid Centistes diabroticae was also not considered as it has been reared only from Acalymma vittata (F.) (Gahan, 1922; Fischer, 1981). A. vittata frequently occurs together with several Diabrotica species, such as D. v. virgifera, but there has been no evidence of parasitism in these other potential Diabrotica hosts, suggesting a high degree of host specificity for the braconid C. diabroticae. From the remaining three parasitoid species, the two South American parasitoid species C. bosqi and C. gasseni have also been rejected as potential biological control agents based on their host specificity, overwintering strategy and rearing difficulties. Experimental data indicated that C. bosqi will not chap13.qxd 12/11/04 1:43 PM Page 270

270 U. Kuhlmann et al.

accept D. v. virgifera adults as host, which suggested that this tachinid appears to be specific to the fucata group within the genus Diabrotica and may not accept species in the virgifera group as hosts. The braconid C. gasseni has been rejected as a potential biological control agent due to the parasitoid’s overwintering strategy and rearing difficulties. The only parasitoid found on the target species, D. v. virgifera, was the tachinid C. compressa from northern Mexico. It attacked a range of species in four different genera of diabroticite beetles, suggesting a large number of host species, but was none the less restricted to diabroticite beetles. C. compressa was finally the only parasitoid that was selected as a candidate biological control agent for D. v. virgifera in Europe based on its availability in northern Mexico, its known host range record, including the target host, and its suitability for rearing under laboratory conditions (see Zhang et al., 2003). Generally it should be noted that the distribution and the efficacy given for C. compressa should be considered as provisional due to the fact that these tachinid species have been little studied.

Suitability of Candidate Biological Control Agents

As the third step in a classical biological control approach, it is necessary to study in more detail the suitability of the selected candidate biological control agent C. compressa. The success of this classical biological control attempt against D. v. virgifera in Europe using C. compressa will depend on a thorough knowledge of its biology and ecology. Before importing and releasing an exotic biological control agent into a new environment, detailed knowledge pertaining to the basic and reproductive biology as well as the development of a rearing technique is crucial. Also this will be instrumental in providing the baseline data required for assessing the host specificity of C. compressa. Publications describing the biology of Celatoria species are extremely rare. The limited literature available, for example, includes a description of the biology of C. setosa, a parasitoid of the striped cucumber beetle, Diabrotica vittata Fabricius (Bussart, 1937). Later Fischer (1983) contributed again to the biology of C. setosa and made the first biological studies of the closely related tachinid C. diabroticae. Although the biology and ecology are little understood, their reproductive strategy is almost identical. Females lay eggs that contain fully developed first-instar larvae directly into the host, a process named larviposition (Bussart, 1937; Fischer, 1983). The aim of this section is to summarize aspects of the basic and reproductive biology, such as mating, larviposition, egg load, potential and realized fecundity and larval and pupal development, as well as to provide insights about a small-scale production technique and the functional response of C. compressa. A non-diapausing strain of D. v. virgifera was continuously reared under quarantine laboratory conditions (25°C day, 15°C night, L 14:D 10, 50% ± 10% relative humidity (r.h.)). Eggs were obtained from the USDA- ARS Northern Grain Insect Research Laboratory at Brookings, South chap13.qxd 12/11/04 1:43 PM Page 271

Classical Biological Control against Western Corn Rootworm in Europe 271

Dakota, USA (for rearing details see Jackson, 1985). C. compressa was obtained from collaborators in Mexico (for details, see section on natural enemies from the area of origin). Subsequent generations of C. compressa were maintained by parasitizing laboratory-reared D. v. virgifera (for rearing details, refer to Zhang et al., 2003). Individuals were sent to Dr Nigel Wyatt (Natural History Museum, London, UK) to confirm the identification of C. compressa. To achieve mating of C. compressa, a set of different conditions were combined in different observation arenas, such as sex ratio, age groups, cage sizes, light source and intensity and temperature conditions. The pre-larviposition period for gestation after mating was determined by dis- secting mated females held under daytime conditions of 25°C. This period was defined as the period of time until the first eggs with fully developed first-instar larvae were visible in different female age classes. To determine the egg load, the number of eggs and the number of eggs containing fully developed first instars per female were recorded daily. The length of the pre-larviposition period was determined, and was based upon the length of time until the first eggs with fully developed first instars were visible during dissections. To measure the age-specific and lifetime fecundity of C. compressa, defined as the average progeny production achieved by the female under a specific set of laboratory conditions, individual 6-day-old mated females were exposed to 20 D. v. virgifera adults for 4 h each day until adult death. Following this 4 h exposure interval, the female parasitoid was removed and the D. v. virgifera were collected and reared for 20 days to assess subsequent C. compressa puparia emergence. Larviposition period, longevity, and the daily and total number of hosts parasitized were recorded for each female studied. The hind tibia length was used as a measurement of parasitoid size, to check whether it was related to lifetime fecundity and female longevity. Hind tibia length has been confirmed to be positively correlated with female size in the tachinid Eucelatoria bryani Sabrosky (Reitz and Adler, 1995). Larval and pupal development was assessed at a daytime temperature of 25°C and a night-time temperature of 15°C, a pho- toperiod of 14 L:10 D and a relative humidity of 50% ± 10%. Observations on host attacks by C. compressa females were carried out to identify important reproductive characteristics of the tachinid influencing the establishment of small-scale production of C. compressa puparia. Therefore, the larviposition period within C. compressa adults’ lifespan, the number of daily larviposition attempts per female and the number of puparia produced daily per female, as well as the cumulative puparia production per female within the females’ larviposition period, were determined. Individual mated females were exposed to ten D. v. virgifera adults for 30 min twice every day from the parasitoid age of 5 days until death. In the 30 min period of exposure, parasitized adults of D. v. virgifera were removed and replaced with new hosts when a host was attacked. The number of successful host attacks was recorded each time chap13.qxd 12/11/04 1:43 PM Page 272

272 U. Kuhlmann et al.

per female. After exposure and attack, D. v. virgifera adults were kept for 20 days until the emergence of the parasitoid larvae. To study the functional response, a randomly chosen density between one and 50 adults of D. v. virgifera was offered to an individual 8–10-day- old mated female of C. compressa. Each host density was exposed for 24 h. After exposure, the hosts were kept for 20 days and emerged C. compressa larvae were recorded for each host density studied. Arc sine transformation, the arc sine square root of Y, was used to normalize percentage data of parasitism. To analyse the functional response of C. compressa to α′ varying host densities, a Holling’s type II model, Nα = ( NtT)/(1 + α′ ThNt), was used (Holling, 1959), where Nα is the number of D. v. virgifera adults parasitized during a specific searching time, α′ is the attack rate, Nt is the D. v. virgifera density and Th is the handling time. The α′ parameters and Nt were calculated by non-linear regression using this model (see Wang and Ferro, 1998; Van den Meiracker and Sabelis, 1999) in Systat 8.0 (SPSS Inc., 1999). The density-dependent relationship between C. compressa and its host was fitted to a logarithmic regression (Hassell et al., 1985). The age of C. compressa adults was found to be the most crucial factor in achieving mating. Only newly emerged, 1 h old females mated successfully with 2–5-day-old males, achieving a success rate of 74%. After mating, a pre-larviposition period of 4 days occurred at daytime temperature of 25°C before 5-day-old C. compressa females started to insert their eggs containing fully developed first-instar larvae directly into the host through intersegmental sutures or through membranes around leg openings of the D. v. virgifera adults. One larviposition took less than a second and, in newly parasitized adults of D. v. virgifera, the first-instar larva was always found in the cavity of the host’s thorax close to the wing bases. During the pre-larviposition period, the egg load of females increased steadily from day 1 to a maximum egg load on day 4 (mean = 69.3 ± 0.8 standard error (SE), n = 19). Five-day-old C. compressa females laid on their first day of larviposition, on average, only 5 ± 0.4 eggs into multiple hosts (max. = 10, n = 19), which is in contrast to the availability of 18 eggs containing fully developed first instars in the uterus per female at that time. During a female’s mean larviposition period of 22.5 ± 0.6 SE days (n = 19), a total of 33.2 ± 0.9 SE first instars were larviposited into the hosts (range seven to 77 eggs, n = 19), which is only half of the female’s egg load. Seventy per cent of the females were still alive after 18 days when the mean lifetime fecundity was already achieved. Lifetime fecundity of C. compressa was significantly correlated with longevity (y = 1.29x – 3.19; R2 = 0.27; F = 6.37, degrees of freedom (d.f.) = 1, 17; P = 0.02). However, no relationship was found between body size and either lifetime fecundity analysis of variance (ANOVA); F = 0.4; d.f. = 1, 17; P = 0.54) or longevity (ANOVA; F = 0.02; d.f. = 1, 17; P = 0.91). Total larval and pupal developmental time, including a pre-larviposition period of 4 days, was 29 days under quarantine laboratory conditions (25°C daytime, 15°C at night, L:D 14:10, 50% ± 10% r.h., n = 55). chap13.qxd 12/11/04 1:43 PM Page 273

Classical Biological Control against Western Corn Rootworm in Europe 273

During the first 6 days of rearing, C. compressa females showed a high activity of larviposition attempts, but the number of attempts decreased linearly during the females’ lifespan (linear regression, y = –0.58x + 15.8, R2 = 0.88, F = 190.16, d.f. = 1, 25, P < 0.001). The cumulative number of larviposition attempts per female reached a mean of 120 ± 2.2 SE (min. = 22, max. = 323, n = 18), whereas the mean cumulative puparia production per female was only 29.7 ± 5 SE (range 5–54, n = 18). Fifty per cent of the total puparia production was already achieved by the females within the first 5 days of the females’ larviposition period. Results demonstrate that the number of puparia produced daily per female was significantly lower compared to the number of daily larviposition attempts observed (paired t test, t = 7.93, d.f. = 26, P < 0.001). The daily larviposition success rate per female reached on an average 24.1% ± 0.8 SE (min. = 4.6%, max. = 43.8%, n = 18). In a functional response study, C. compressa was exposed for 24 h to randomly chosen different host densities (between 1 and 50 hosts). The number of hosts parasitized by C. compressa increased with increasing host density until an upper limit was reached. The functional response of C. compressa fitted the Holling type II response (R2 = 0.239; F = 96.40; d.f. = 2, 80; P < 0.001). In a 24 h period, the searching rate (α′) for single C. compressa females was 0.016 hosts/h, and the mathematical handling time (Th) was 2 h, explaining an upper limit of ten hosts parasitized with an increasing host density. Therefore, per cent parasitism decreased with increasing number of hosts, indicating an inverse density-dependent relationship between C. compressa and its host (logarithmic regression, y = –12Ln(x) + 64.9; R2 = 0.46; F = 67.32; d.f. = 1, 80; P < 0.001). In the third step of this classical biological control approach the basic and reproductive biology of C. compressa was clarified, which is an important component for evaluating the suitability of a candidate biological control agent. A small-scale rearing method for C. compressa was successfully developed and implemented, which will be one of the keys to successful use of this tachinid fly as a biological control agent. The general findings on the basic biology of C. compressa are comparable to the investigations on C. setosa and C. diabroticae conducted by Bussart (1937) and Fischer (1983). For instance, the pre-larviposition period lasted 4 days, knowledge crucial to the successful rearing of C. compressa. It is also important for the timing of parasitoid field releases to ensure that C. compressa females will have a large number of eggs containing fully developed first-instar larvae for oviposition into its host D. v. virgifera. It has been found that C. compressa belongs to the group characterized by having eggs that contain fully developed larvae, which are laid directly into the host. This is possible by using a piercing structure on the female terminalia separate from the ovipositor (Belshaw, 1993). This so-called larviposition appears to be common for the genus Celatoria and has also been documented in C. setosa (Bussart, 1937; Fischer, 1983), C. diabroticae (Fischer, 1983) and C. bosqi (G. Cabrera Walsh, 1997, personal communication) parasitizing North and South American Diabrotica chap13.qxd 12/11/04 1:43 PM Page 274

274 U. Kuhlmann et al.

species. This reproductive strategy is restricted to only a few other tachinid species in the tribe Blondeliini within the subfamily Exoristinae, such as Blondelia nigripes (Fallén), Compsilura concinnata (Meigen) and Vibrissina debilitata (Pandellé) (Herting, 1960; Belshaw, 1993). O’Hara (1985) describes the oviposition strategy into the host as relatively successful, and as a consequence egg production is relatively low. It is obvious that C. compressa belongs to the group with a lower fecundity. A total number of approximately 30 first instars were larviposited into the hosts per female, which is only about half of the measured egg load. This egg load has to be considered simply as a partial measure of the potential fecundity as C. compressa is a synovigenic species. It should be noted that a large number of host attacks by C. compressa were unsuccessful, resulting in a lower larviposition success rate of 25%. Parasitoid females appear to have difficulties inserting the egg through the intersegmental sutures or membranes around leg openings of the host adults. Otherwise, 50% of the lifetime fecundity of C. compressa was already realized in the first 4 days of the larviposition period, while 70% of the females reached the mean lifetime fecundity before dying. Should C. compressa be selected as a biological control agent of D. v. virgifera in Europe, large-scale rearing of the tachinid under laboratory conditions must be achieved for a field release programme. However, large-scale multiplication of C. compressa remains difficult due to the low larviposition success rate. So far, a small-scale production technique for C. compressa has been successfully developed, using a non-diapausing strain of D. v. virgifera (for details, see Zhang et al., 2003). Although the current rearing approach is time- and labour-intensive, C. compressa has been reared successfully for at least 25 successive generations without shifting the sex ratio (approximately 500 adult C. compressa were produced each generation). The following factors were critical: (i) successful mating using 2–5-day-old males and < 1 h old females (sex ratio 5:1); (ii) a 4-day pre-larviposition period for gestation; (iii) daily exposure of females to 15 hosts for 1 h over a period of 15 days to achieve efficient production of puparia; (iv) a daytime temperature of 25°C to achieve a full generation of C. compressa in 29 days; and (v) continuous rearing of a non-diapausing strain of D. v. virgifera. A functional response characterizes the relationship between the number of hosts attacked and host density. When the functional response is density-dependent, it is more likely that a natural enemy would be more effective in biological control programmes (Murdoch and Briggs, 1996; Briggs et al., 1999), although there is controversy concerning the role of density-dependent factors in regulating populations (Brown, 1989). In this study, an inverse host density-dependent pattern of per cent parasitism was shown for C. compressa, which means that the number of hosts parasitized per female reached an upper limit with an increasing host density. However, a similar lack of response to host density has been found in about half of the cases reviewed by Walde and Murdoch (1988). The inability of C. compressa to show a positive response to high host chap13.qxd 12/11/04 1:43 PM Page 275

Classical Biological Control against Western Corn Rootworm in Europe 275

density suggests that it is unlikely to suppress D. v. virgifera; however, at lower host density levels, it may still have an effective impact. Kenis and Lopez-Vaamonde (1998) reported that the tachinid Ceranthia samarensis (Villeneuve) is considered to be an important factor in preventing gypsy moth, Lymantria dispar (L.), outbreaks in regions where this pest is present at low densities. These results are the first important steps in assessing the suitability of C. compressa as a candidate biological control agent for introduction into Europe. Assessing the host specificity of C. compressa is another important step towards the implementation of a classical biological control programme in Europe. Besides host specificity, understanding the parasitoid’s hibernation strategy is another major requirement for the successful establishment of C. compressa in the open field in Europe. In general, the overwintering of Celatoria species has been little studied. Only C. setosa was reported to overwinter inside the host body of D. vittata (Bussart, 1937), while Herzog (1977) suggested that C. diabroticae overwinter as pupae in the soil. Further investigations on C. compressa are needed to clarify its overwintering development stage and determine whether it has the cold tolerance required for permanent establishment in Europe.

Host Speciﬁcity of the Candidate Biological Control Agent

With respect to the safety of biological control and increasing concern over non-target risks from introduced biological control agents, standards and frameworks developed for the release of exotic biological control agents have been followed (e.g. FAO, 1997; Van Lenteren et al., 2003). Therefore, assessing the host specificity of C. compressa was the final and fourth step towards the implementation of a classical biological control programme in Europe. In this final step, following previous research steps on the basic and reproductive biology, as well as on the development of a small-scale rearing technique, the host specificity of C. compressa needed to be identified with the aim of predicting potential non-target risks prior to its potential importation. This study focused first on the selection of potential non-target species at risk in D. v. virgifera-invaded areas in Europe, and secondly on host specificity testing of C. compressa. According to the phylogenetic centrifugal method proposed for weed biological control agents (Wapshere, 1974) and a practical approach suggested by Kuhlmann and Mason (2003), a simplifying procedure has been applied and indigenous coleopteran species were selected for testing under quarantine laboratory conditions. Thereafter, the sequential test for determining the host ranges of inundative biological control agents, outlined by Van Lenteren et al. (2003), was applied and slightly modified to assess the host specificity of the candidate classical biological control agent C. compressa. European non-target coleopteran species potentially at risk of being chap13.qxd 12/11/04 1:43 PM Page 276

276 U. Kuhlmann et al.

attacked by C. compressa were selected, using the practical approach for selecting non-target species suggested by Kuhlmann and Mason (2003). Non-target species would be selected based on close phylogenetic relationships, morphological similarity, habitat overlap, geographical distribution and overlap of temporal occurrences with the target species (here D. v. virgifera, Coleoptera: Chrysomelidae: Galerucinae: Luperini). As a result nine representative non-target species were selected for testing in the quarantine laboratory: two-spotted lady beetle, Adalia bipunctata L. (Coleoptera: Coccinelidae); red pumpkin beetle, Aulacophora foveicollis Lucas (Coleoptera: Chrysomelidae: Galerucinae: Luperini); thistle tortoise beetle, Cassida rubiginosa Müller (Coleoptera: Chrysomelidae: Cassidinae); golden loosestrife beetle, Galerucella pusilla Duft (Coleoptera: Chrysomelidae: Galerucinae: Galerucini); green dock beetle, Gastrophysa viridula Deg. (Coleoptera: Chrysomelidae: Chrysomelinae); Gonioctena fornicata Brüggemann (Coleoptera: Chrysomelidae: Chrysomelinae); cereal leaf beetle, O. melanopus (L.) (Coleoptera: Chrysomelidae: Criocerinae); elm leaf beetle, Pyrrhalta luteola (Müller) (Coleoptera: Chrysomelidae: Galerucinae: Galerucini); and pea and bean weevil, Sitona lineatus Linnaeus (Coleoptera: Curculionidae). The five-step testing scheme and methodology for host specificity of natural enemies proposed by Van Lenteren et al. (2003) was basically followed but modified as tests suggested for steps one to three were conducted. No-choice and choice tests were carried out in the quarantine laboratory at 22°C to 25°C during daytime, 15°C at night, light regime at L 14:D 10, and 50% ± 10% r.h. The host/parasitoid ratio (10:1) and duration of the test (1 h) were selected according to previous information obtained during the study of the reproductive biology of C. compressa (Zhang et al., 2004). In no-choice tests, ten adult individuals from each non-target species were exposed to an individual, mated, 6-day-old and naïve C. compressa female for 1 h. In choice tests five individuals per non-target species and five adults of the non-diapausing D. v. virgifera strain were exposed simultaneously to individual parasitoid females. To determine the possible increase in host acceptance due to increasing oviposition pressure in C. compressa, a sequence of no-choice tests over time was conducted. To assess potential shifts in host preference and a possible increasing attack pressure of usually not attacked hosts due to limited availability of the preferred host or parasitoid learning experience, a sequence of choice tests over time was carried out. After each test, the female parasitoid was removed and adults of each non-target species were reared for 20 days until C. compressa larvae exited their hosts and formed a puparium. C. compressa puparia were collected and recorded daily. After 20 days, surviving non-target and target host adults, as well as those that died during the 20 days’ rearing procedure, were dissected with the aim of determining whether parasitism had occurred or not. In no-choice or choice tests, naïve or experienced females of C. compressa never parasitized eight of nine non-target species tested. In the absence of D. v. virgifera adults, A. foveicollis was occasionally accepted chap13.qxd 12/11/04 1:43 PM Page 277

Classical Biological Control against Western Corn Rootworm in Europe 277

(six larvae in 260 hosts), but complete development by C. compressa was not achieved. The acceptance of A. foveicollis by C. compressa was significantly lower than that of the target host D. v. virgifera, 2.3% versus 28.7% (ANOVA followed by Tukey HSD test, F = 27.6, d.f. = 2, 123, P < 0.001). In the sequence of no-choice tests, A. foveicollis was accepted (four larvae in 260 hosts) but acceptance was again significantly lower than that of the target species compared within the same day of 3 successive days (ANOVA followed by Tukey HSD test, F = 43.94, 115.94 and 75.61 for 1st, 2nd and 3rd day respectively, d.f. = 2, 79, P < 0.001). From the four A. foveicollis adults parasitized, a single C. compressa larva completed its development and formed a puparium. In the presence of D. v. virgifera in the choice test, A. foveicollis was never accepted by C. compressa, but during the sequence of choice tests A. foveicollis was again accepted by C. compressa. In total, eight of 330 A. foveicollis were accepted by six C. compressa females (equals 2.4%). Host acceptance was significantly lower than that observed in D. v. virgifera on the first, second and third day (paired t test, t = 8.52, 4.14 and 5.81, respectively, d.f. = 21, P < 0.001). In contrast to the results of the sequence of no-choice tests, host suitability of A. foveicollis by C. compressa was not found. Results obtained in the last step of this classical biological control approach predict that the candidate biological control agent C. compressa will have a narrow host range in Europe (Plate 4), being restricted to a few genera on the tribe level of Luperini among the subfamily Galerucinae. These results of the physiological host range of C. compressa obtained under quarantine conditions are in agreement with the known field host range from the area of origin in Mexico (Eben and Barbercheck, 1996). The known host range based on literature records of the three other known Celatoria species, such as C. bosqi, C. diabroticae and C. setosa, are also restricted to the subtribe Diabroticina within the tribe Luperini of the subfamily Galerucinae (Cox, 1994). C. bosqi present in South America is known to parasitize D. speciosa (Germar) (Blanchard, 1937; Heineck- Leonel and Salles, 1997), Diabrotica sp. nr. fulvofasciata Jacoby, D. viridula (F.) (G. Cabrera Walsh, 2003, personal communication) and Cerotoma arcuata Olivier (Magalhães and Quintela, 1987). Conclusively, it is obvious that these three species as well as C. compressa are very much specialized and it can be concluded that most probably all species in the genus Celatoria parasitize adults of only a few species of single or related genera within the Galerucinae. There are several lines of evidence that host specificity in dipteran parasitoids, which attack adult insects, are more often determined by the events leading up to oviposition, rather than by the events occurring after oviposition (Feener and Brown, 1997). Specially modified ovipositors and specialized attack behaviour in these parasitoids would further rein- force its host specificity (Feener and Brown, 1997). Fischer (1983) studied first the larviposition behaviour of C. diabroticae and C. setosa and recently Zhang et al. (2003, 2004) clarified that C. compressa females use a piercing ovipositor to successfully parasitize D. v. virgifera. Therefore, chap13.qxd 12/11/04 1:43 PM Page 278

278 U. Kuhlmann et al.

it is likely that Celatoria species have a high degree of host specificity compared to many other tachinids due to the elaborately modified piercing ovipositor of the females (Belshaw, 1994; J. O’Hara, 2000, personal communication). Among the Galerucinae tested in this study, only A. foveicollis was found to be parasitized by C. compressa but host suitability was extremely low (< 0.02%). Other potential non-target species within the tribe Luperini belonging to the subfamily Galerucinae have not been tested but could be considered prior to the importation of C. compressa into Europe. Maulik (1936) reported that there is a remarkable resemblance between Aulacophora and Diabrotica in larval, pupal and adult structures, breeding habits and food plants. This suggests that phylogenetically closely related Aulacophora species would be most likely to be at risk if C. compressa were selected for release in a classical biological control programme against D. v. virgifera in Europe. Except for Aulacophora indica (Gmelin) in Russia, A. foveicollis is the only Aulacophora species in Europe, and restricted in its distribution to southern Europe, e.g. Croatia, Greece, southern Italy, Malta, Portugal and Spain (Wilcox, 1972; Crop Protection Compendium, 2000). Based on the distribution of A. foveicollis in Europe, potential releases of C. compressa could lead towards a minor non-target impact as the distribution of the D. v. virgifera and the non- target insect populations are partly overlapping (so far Croatia in 2003). On the other hand, it should be noted that A. foveicollis adults are also considered to be a serious pest, feeding on the leaves and flowers of Cucurbitaceae (Waterhouse and Norris, 1987; Crop Protection Compendium, 2000). Additionally, indigenous natural enemies of A. foveicollis are rarely reported (Herting, 1973). Therefore, it can be concluded that direct impacts of C. compressa on A. foveicollis and indirect impacts on other organisms would be extremely low if C. compressa were released.

Conclusions

Reviewing the results obtained in the four-step approach we can conclude that important ecological baseline knowledge on Diabrotica– natural enemy associations has been gained in Europe and in the area of origin. This kind of information is required for implementing a sustainable management strategy against D. v. virgifera potentially using classical biological control. The following conclusions are compiled by reviewing in detail the step-by-step approach:

1. Effective indigenous natural enemies do not attack any life stage of D. v. virgifera in Central Europe. 2. In the area of origin surveyed, C. compressa was the only parasitoid found on the target species, D. v. virgifera, and its host range is considered to be restricted to diabroticite beetles. chap13.qxd 12/11/04 1:43 PM Page 279

Classical Biological Control against Western Corn Rootworm in Europe 279

3. Prior to its potential importation, the parasitoid’s basic and reproductive biology was clariﬁed. Special emphasis was placed on developing a rearing technique for small-scale production of C. compressa as this is essential for potential releases but also for conducting physiological host speciﬁcity studies. 4. C. compressa would be safe for introduction as direct and indirect impacts on other organisms would be extremely low.

Regarding the question asked in the beginning: Is classical biological control against western corn rootworm in Europe a sustainable management strategy? Certainly, this is not a black and white conclusion resulting in yes or no. In general, biological control or the ecological understanding of interactions between a pest and its natural mortality factors should be considered as the basis of a sustainable management strategy against D. v. virgifera in Europe. Biological techniques are useful elements but in the case of D. v. virgifera in Europe an IPM strategy is needed as classical biological and cultural control interventions alone might not be applicable. A sustainable IPM approach is likely to incorporate classical biological control with pest monitoring systems, tolerant maize varieties, crop rotation and cultural techniques to enhance the conservation of natural control. Another control option is the Bt-transgenetic maize targeting the D. v. virgifera larvae, which most probably will cause population suppression of about 80–90% (see Ward et al., Chapter 12, this volume). For a sustainable use of Bt-transgenetic maize, the impact of natural enemies such as predators and parasitoids is inevitable since D. v. virgifera is not completely controlled and to decrease the selection pressure on the pest would be to develop resistance towards the Bt toxin. Regarding the tachinid C. compressa, we have studied a promising candidate classical biological control agent that would be safe for introduction against D. v. virgifera in Europe. None the less, there are a number of questions that remain to be answered before its potential importation. C. compressa’s hibernation strategy and its cold tolerance are major requirements needing to be solved for successful establishment in Europe. Records of percentage parasitism by C. compressa are virtually non-existent in the area of origin, and only recently a maximum parasitism rate of 16% from D. v. virgifera has been reported (A. Eben, 2001, personal communication). Results obtained in the laboratory suggest that the biological control agent might have a higher impact on lower density levels of D. v. virgifera. Since percentage parasitism in laboratory is a poor measure of parasitoids’ impact on host populations (Kidd and Jervis, 1996), more knowledge is needed regarding the impact of C. compressa on the population dynamics of D. v. virgifera in its area of origin (such as in northern Mexico). Kuhlmann and Van der Burgt (1998) stated that C. setosa and C. diabroticae most probably do not regulate the populations of Diabroticina species effectively under current chemical-based pest management strategies in the USA, but also for these species the efﬁcacy chap13.qxd 12/11/04 1:43 PM Page 280

280 U. Kuhlmann et al.

of these parasitoids in controlling Diabroticina species in the ﬁeld remains to be investigated (Levine and Oloumi-Sadeghi, 1991). Taking all this information into consideration, it is obvious that promising results have been obtained but further investigations are essential to complete our knowledge about the candidate biological control agent C. compressa. This missing information is needed to ﬁnally answer the question if classical biological control could become an integral component of a sustainable pest management strategy against D. v. virgifera in Europe.

Acknowledgements

This work was possible due to the hospitality offered by the Plant Health Service in Hodmezovasarhely in Hungary. We would like to thank for their kind collaboration Ibolya Zseller, Plant Health Service, Hodmezovasarhely, Hungary; Dr Ivan Sivcev, Institute of Plant Protection and Environment, Zemun, Serbia; and Dr Marija Ivezic, University of J.J. Strossmayer, Osijek, Croatia, and their team members. We are grateful to Dr Astrid Eben (Instituto de Ecologia, Xalapa, Mexico) and Rebeca Alvarez Zagoya (Instituto Politecnico Nacional, Durango, Mexico) for the collection of tachinids. We gratefully acknowledge the continuous support with eggs of D. v. virgifera by Chad Nielson and Michael Ellsbury (USDA-ARS, Northern Grain Insect Research Laboratory at Brookings, South Dakota, USA). We appreciated very much the technical assistance of Emma Hunt, Christine Gueldenzoph, Tara Gariepy, Rike Stelkens, Kim Riley and Leyla Valdivia Buitriago during the experiments and the parasitoid rearing. In Hungary, technical support by Lars Reimer, Edit Kiss, Szabolcs Meszaros and Orsolya Fulop was greatly appreciated. We gratefully appreciated the identiﬁcation of the fungi by Dr Siegfried Keller, Agroscope, Zurich – Reckenholz, Switzerland, the nematodes by Dr David Hunt, CABI Bioscience, Egham, UK, and Dr Ralf-Udo Ehlers, Institute of Phytopathology, Kiel, Germany, and the tachinid ﬂies by Dr Nigel Wyatt, Natural History Museum, London, UK. We also thank Tara Gariepy, Burnaby, Canada, for reviewing the English text. This study was funded by the Bundesamt für Bildung und Wissenschaft, Bern, Switzerland, within the EU project QLK-5CT-1999-01110.

References

Arnauld, P.H. (1978) A Host–Parasite Catalogue of North American Tachinidae (Diptera). Miscellaneous Publication No. 1319, United States Department of Agriculture, Washington, DC, 860 pp. Baca, F. (1994) New member of the harmful entomofauna of Yugoslavia, Diabrotica virgifera virgifera LeConte (Coleoptera, Chrysomelidae). Zashtita Bilja 45, 125–131. Belshaw, R. (1993) Tachinid ﬂies (Diptera: Tachinidae). In: Dolling, W.R. and Askew, R.R. chap13.qxd 12/11/04 1:43 PM Page 281

Classical Biological Control against Western Corn Rootworm in Europe 281

(eds) Handbooks for the Identiﬁcation of British Insects, Vol. 10. Royal Entomological Society London, pp. 5–7. Belshaw, R. (1994) Life history characteristics of Tachinidae (Diptera) and their effect on polyphagy. In: Hawkins, B.A. and Sheehan, W. (eds) Parasitoid Community Ecology. Oxford University Press, Oxford, UK, pp. 145–162. Blanchard, E.E. (1937) Dípteros Argentinos nuevos o poco conocidos. Revista Sociedade Entomologica Argentina 9, 35–38. Branson, T.F. and Krysan, J.L. (1981) Feeding and oviposition behaviour and life cycle strategies of Diabrotica: an evolutionary review with implications for pest management. Environmental Entomology 10, 826–831. Branson, T.F., Guss, P.L., Krysan, J.L. and Sutter, G.R. (1975) Corn Rootworms: Laboratory Rearing and Manipulation. ARS-NC-28, United States Department of Agriculture – Agricultural Research Service, Peoria, Illinois, 18 pp. Briggs, C.J., Murdoch, W.W. and Nisbet, R.M. (1999) Recent developments in theory for biological control of insect pests by parasitoids. In: Hawkins, B.A. and Cornell, H.V. (eds) Theoretical Approaches to Biological Control. Cambridge University Press, Cambridge, UK, pp. 22–42. Brown, M.W. (1989) Density dependence in insect host–parasitoid systems: a comment. Ecology 70, 776–779. Bussart, J.E. (1937) The bionomics of Chaetophleps setosa Coquillett (Diptera: Tachinidae). Annals of the Entomological Society of America 30, 285–295. Cabrera Walsh, G., Athanas, M.M., Salles, L.A.B. and Schroder, R.F.W. (2003) Distribution, host range, and climatic constraints on Centistes gasseni (Hymenoptera: Braconidae), a South American parasitoid of cucumber beetles, Diabrotica spp. (Coleoptera: Chrysomelidae). Bulletin of Entomological Research 93(6), 561–567. Carey, J.R. (1992) The Mediterranean fruit ﬂy in California: tacking stock. California Agriculture 46, 12–17. Cox, M.L. (1994) The Hymenoptera and Diptera parasitoids of Chrysomelidae. In: Jolivet, P.H., Cox, M.L. and Petitpierre, E. (eds) Novel Aspects of the Biology of Chrysomelidae. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 419–468. Crop Protection Compendium (2000) CAB International, Wallingford, UK (CD-ROM). Eben, A. and Barbercheck, M.E. (1996) Field observations on host plant associations and natural enemies of diabroticite beetles (Chrysomelidae: Luperini) in Veracruz, Mexico. Acta Zoologica Mexicana 67, 47–65. Elsey, K.D. (1988) Cucumber beetle seasonality in coastal South Carolina. Environmental Entomology 17, 496–502. FAO (1997) Code of conduct for the import and release of exotic biological control agents. Biocontrol News and Information 18, 119N-124N. Feener D.H., Jr. and Brown, B.V. (1997) Diptera as parasitoids. Annual Review of Entomology 42, 73–97. Fischer, D.C. (1981) Tachinid parasitoids of Acalymma vittata, Diabrotica undecimpunctata, and Diabrotica virgifera. MSc thesis, University of Illinois at Urbana-Champaign, 61 pp. Fischer, D.C. (1983) Celatoria diabroticae Shimer and Celatoria setosa Coquillett: tachinid parasitoids of the Diabroticite Coleoptera. PhD thesis, University of Illinois at Urbana- Champaign, 120 pp. Franz, J.M. and Krieg, A. (1982) Biologische Schaedlingsbekaempfung unter Beruecksichtigung integrierter Verfahren. Paul Parey Verlag, Berlin. Gahan, A.B. (1922) A new hymenopterous parasite upon adult beetles. Ohio Journal of Science 22, 140–142. chap13.qxd 12/11/04 1:43 PM Page 282

282 U. Kuhlmann et al.

Greathead, D.J. (1976) A Review of Biological Control in Western and Southern Europe. Commonwealth Agricultural Bureaux, Farnham Royal, UK, 182 pp. Guimaraes, J.H. (1977) Host–parasite and parasite–host catalogue of South America Tachinidae (Diptera). Arquivos de Zoologia 28, 1–131. Hassell, M.P., Lessells, C.M. and McGavin, G.C. (1985) Inverse density dependent parasitism in a patchy environment: a laboratory system. Ecological Entomology 10, 393–402. Haynes, D.L. and Gage, S.H. (1981) The cereal leaf beetle in North America. Annual Review of Entomology 26, 259–287. Heineck-Leonel, M.A. and Salles, L.A.B. (1997) Incidence of parasitoids and pathogens of Diabrotica speciosa (Germ.) (Coleoptera: Chrysomelidae) in Pelotas RS. Anais da Sociedade Entomologica do Brasil 26, 81–85. Herting, B. (1960) Biologie der westpalaearktischen Raupenfliegen (Dipt., Tachinidae). Monographien zur Angewandten Entomologie, Beihefte zur Zeitschrift für ange- wandte Entomologie No. 16, Paul Parey Verlag, Hamburg, 188 pp. Herting, B. (1973) A Catalogue of the Parasites and Predators of Terrestrial Arthropods. Section A, Host or Prey/Enemy. Volume III, Coleoptera and Strepsiptera. Commonwealth Agricultural Bureaux, Farnham Royal, UK, 185 pp. Herzog, D.C. (1977) Bean leaf beetle parasitism by Celatoria diabroticae (Shimer) and Hyalomyodes triangulifer (Leow). Journal of Georgia Entomological Society 12, 64–68. Holling, C.S. (1959) The components of predation as revealed by a study of small mammal predation of the European pine sawfly. Canadian Entomologist 91, 293–320. Houston, K.J., Mound, L.A. and Palmer, J.M. (1991) Two pest thripses (Thysanoptera) new to Australia, with notes on the distribution and structural variation in other species. Journal of the Australian Entomological Society 30, 321–332. Jackson, J.J. (1985) Diabrotica spp. In: Singh, P. and Moore, R.F. (eds) Handbook of Insect Rearing Vol. 1. Elsevier, Amsterdam, The Netherlands, pp. 237–254. Kenis, M. and Lopez-Vaamonde, C. (1998) Classical biological control of the gypsy moth, Lymantria dispar (L.) in North America: prospects and new strategies. In: McManus, M.L. and Liebhold, A.M. (eds) Proceedings Population Dynamic Impacts, and Integrated Management of Forest Defoliating Insects. General Technical Report NE- 247, USDA Forest Service Northeastern Research Station, Radnor, Pennsylvania, pp. 213–221. Kidd, N.A.C. and Jervis, M.A. (1996) Population dynamics. In: Jervis, M. and Kidd, N. (eds) Insect Natural Enemies, Practical Approaches to Their Study and Evaluation. Chapman & Hall, London, pp. 293–374. Kiss, J., Edwards, C.R., Allara-Carlin, M., Sivcev, I., Igrc-Barc˘ic´, J., Festic´, H., Ivanova, I., Princzinger, G., Sivicek, P. and Rosca, I. (2001) Update: the WCR (Diabrotica virgifera virgifera) in Europe. In: Proceedings of the XXI IOBC–IWGO Conference. Veneto Agricultura, Padua–Venice, Italy, pp. 83–87. Krysan, J.L. (1999) Selected topics in the biology of Diabrotica. In: Cox, M.L. (ed.) Advances in Chrysomelidae Biology 1. Backhuys Publishers, Leiden, The Netherlands, pp. 479–513. Krysan, J.L. and Branson, T.F. (1983) Biology, ecology and distribution of Diabrotica. In: Gordon, D.T., Knoke, J.K., Nault, L.R. and Ritter, R.M. (eds) Proceedings International Maize Virus Disease Colloquium and Workshop, 2–6 August 1982. Ohio Agricultural Research and Development, Ohio State University, Wooster, p. 144. Krysan, J.L. and Miller, T.A. (1986) Methods for Study of Pest Diabrotica. Springer-Verlag, New York, 260 pp. Kuhlmann, U. and Mason, P.G. (2003) Use of field host range surveys for selecting candidate non-target species for physiological host specificity testing of entomophagous chap13.qxd 12/11/04 1:43 PM Page 283

Classical Biological Control against Western Corn Rootworm in Europe 283

biological control agents. In: Van Driesche, R.G. (ed.) Proceedings of the International Symposium on Biological Control of Arthropods, Honolulu, Hawaii, 14–18 January 2003. FHTET-2003-05, Forest Service United States Department of Agriculture, Morgantown, West Virginia, pp. 370–377. Kuhlmann, U. and Van der Burgt, W.A.C.M. (1998) Possibilities for biological control of the western corn rootworm, Diabrotica virgifera virgifera LeConte, in Central Europe. Biocontrol News and Information 19, 59N–68N. Levine, E. and Oloumi-Sadeghi, H. (1991) Management of diabroticite rootworms in corn. Annual Review of Entomology 36, 229–255. Magalhães, B.P. and Quintela, E.D. (1987) Níveis de parasitismo de Urosigalphus chalco- dermi Wilkinson sobre Chalcodermus bimaculatus Fiedler e de Celatoria bosqi Blanchard sobre Cerotoma arcuata Olivier em caupi (Vigna unguiculata (L.) Walp.) em Goiás. Anais da Sociedade Entomologica do Brasil 16, 235–238. Maulik, S. (1936) Coleoptera Chrysomelidae (Galerucinae). Fauna of British India. Taylor and Francis, London, 648 pp. Metcalf, R.L. (1985) Plant kairomones and insect pest control. Bulletin of the Illinois Natural History Survey 33, 175–198. Mills, N.J. (1990) Biological control of forest aphid pests in Africa. Bulletin of Entomological Research 80, 31–36. Mills, N.J. (1994) The structure and complexity of parasitoid communities in relation to biological control. In: Hawkins, B.A. and Sheehan, W. (eds) Parasitoid Community Ecology. Oxford University Press, New York, pp. 397–417. Murdoch, W.W. and Briggs, C.J. (1996) Theory for biological control: recent developments. Ecology 77, 2001–2013. O’Hara, J.E. (1985) Oviposition strategies in the Tachinidae, a family of beneficial parasitic flies. Agriculture and Forestry Bulletin of the University of Alberta 8, 31–34. Reitz, S.R. and Adler, P.H. (1995) Fecundity and oviposition of Eucelatoria bryani, a gre- garious parasitoid of Helicoverpa zea and Heliothis virescens. Entomologia Experimentalis et Applicata 75, 175–181. Schroder, R.F.W. and Athanas, M.M. (2002) Biological observations of Centistes gasseni Shaw (Hymenoptera: Braconidae), a parasitoid of Diabrotica spp. (Coleoptera: Chrysomelidae). Proceedings of the Entomological Society of Washington 104, 554–562. Shaw, S.R. (1995) A new species of Centistes from Brazil (Hymenoptera: Braconidae: Euphorinae) parasitizing adults of Diabrotica (Coleoptera: Chrysomelidae), with a key to New World species. Proceedings of the Entomological Society of Washington 97, 153–160. Sivcev, I., Manojlovic, B., Krnjajic, S., Dimic, N., Draganic, M., Baca, F., Kaitovic, Z., Sekulic, R. and Keresi, T. (1994) Distribution and harmful effect of Diabrotica virgifera LeConte (Coleoptera, Chrysomelidae), a new maize pest in Yugoslavia. Zashtita Bilja 45, 19–26. SPSS Inc. (1999) Systat 8.0 for Windows: Statistics. Prentice Hall, Upper Saddle River, New Jersey, 750 pp. Summers, T.E. and Stafford, E.W. (1953) Diabrotica undecimpunctata howardi parasitized by Celatoria diabroticae. Journal of Economic Entomology 46, 180. Toepfer, S. and Kuhlmann, U. (2004) Survey for natural enemies of the invasive alien chrysomelid, Diabrotica virgifera virgifera, in Central Europe. Biocontrol 49, 385–395. Van den Meiracker, R.A.F. and Sabelis, M.W. (1999) Do functional responses of predatory arthropods reach a plateau? A case study of Orius insidiosus with western flower thrips as prey. Entomologia Experimentalis et Applicata 90, 323–329. Van Driesche, R.G. and Hoddle, M.S. (2000) Classical arthropod biological control: measur- chap13.qxd 12/11/04 1:43 PM Page 284

284 U. Kuhlmann et al.

ing success, step by step. In: Gurr, G. and Wratten, S. (eds) Biological Control: Measures of Success. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 39–75. Van Lenteren, J.C., Babendreier, D., Bigler, F., Burgio, G., Hokkanen, H.M.T., Kuske, S., Loomans, A.J.M., Menzler-Hokkanen, I., Van Rijn, P.C.J., Thomas, M.B., Tommasini, M.G. and Zeng, Q.-Q. (2003) Environmental risk assessment of exotic natural enemies used in inundative biological control. Biocontrol 48, 3–28. Waage, J. (1990) Ecology theory and the selection of biological control agents. In: Mackauer, M, Ehler, L.E. and Roland, J. (eds) Critical Issues in Biological Control. Intercept, Andover, UK, pp. 135–157. Walde, S.J. and Murdoch, W.W. (1988) Spatial density dependence in parasitoids. Annual Review of Entomology 33, 441–466. Wang, B. and Ferro, D.N. (1998) Functional responses of Trichogramma ostriniae (Hymenoptera: Trichogrammatidae) to Ostrinia nubilalis (Lepidoptera: Pyralidae) under laboratory and field conditions. Environmental Entomology 27, 752–758. Wapshere, A.J. (1974) A strategy for evaluating the safety of organisms for biological weed control. Annals of Applied Biology 77, 201–211. Waterhouse, D.F. and Norris, K.R. (1987) Biological Control: Pacific Prospects. Inkata Press, Melbourne, Australia, 454 pp. Wilcox, J.A. (1972) Chrysomelidae, Galerucinae, Luperini. Coleopterum Catalogus Supplementa, Part 78, Vol. 3, 2nd edn. Junk’s, Gravenhagen, The Netherlands, 211 pp. Wittenberg, R. and Cock, M.J.W. (2001). Invasive Alien Species: a Toolkit of Best Prevention and Management Practices. CAB International, Wallingford, UK, 225 pp. Zhang, F., Toepfer, S. and Kuhlmann, U. (2003) Basic biology and small-scale production of Celatoria compressa (Diptera: Tachinidae), a parasitoid of Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae). Bulletin of Entomological Research 93, 569–575. Zhang, F., Toepfer, S., Riley, K. and Kuhlmann, U. (2004) Reproductive biology of Celatoria compressa (Diptera: Tachinidae), a parasitoid of Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae). Biocontrol, Science and Technology 14, 5–16. Zwoelfer, H. and Pschorn-Walcher, H. (1968) Wie verhalten sich Insektenparasiten gegenueber eingeschleppten, faunenfremden Wirten? Anzeiger für Schaedlingskunde, Pflanzenschutz und Umweltschutz 41, 51–55. chap14.qxd 24/11/04 1:46 PM Page 285

14 Maize Growing, Maize High-risk Areas and Potential Yield Losses due to Western Corn Rootworm (Diabrotica virgifera virgifera) Damage in Selected European Countries

PETER BAUFELD1 AND SIEGFRIED ENZIAN2 Federal Biological Research Centre for Agriculture and Forestry, 1Department for National and International Plant Health, 2Institute for Technology Assessment in Plant Protection, Kleinmachnow, Germany

Maize Cultivation in Selected EU Countries and Switzerland

Introduction

Maize is an important fodder plant in the European Union (EU) but is also used for human consumption (sweetcorn, ﬂour and popcorn). Although the total area under maize cultivation in the EU remained fairly constant from 1996 (8.045 million ha) to 2000 (7.932 million ha), the breeding of new maize hybrids will allow for an expansion of maize cultivation into areas where up to now climatic conditions have not favoured its production. While seed maize in particular is cultivated in warmer regions in the south and central parts of the EU, silage maize is favoured more in the north. As regards plant protection, weeds are the main challenge to maize cultivation in Europe. However, in southern regions the European corn borer (Ostrinia nubilalis) also poses a problem to maize cultivation and must therefore be controlled. With the expansion of the colonized area of this pest towards northern parts of Europe and the introduction of the western corn rootworm (WCR), a much more complex problem as regards plant protection will have to be solved in the coming years. In regions with a high concentration of maize cultivation, and even in regions under crop rotation regimes, WCR will gradually increase the economic impact on maize production. Should control measures have to be taken into

286 P. Baufeld and S. Enzian

account, an analysis as to potential yield losses will also have to be calculated in order to be able to adjust possible control and management options. This chapter aims to analyse the distribution and proportion of maize production in different European countries. These data will form the basis for modelling yield losses and economic damage in high-risk areas in Europe.

Methods

The data were analysed as the proportion of maize in relation to arable land for the following eight EU countries: France, Germany, Italy, Austria, The Netherlands, Belgium, Switzerland and Luxembourg. The data dis- tinguished between grain maize and corn–cob mix (CCM) and silage (including green) maize for each country, with the exception of Italy. Figures as to total silage maize production in Italy were obtained from statistics from the German ZMP-Marktbilanz (2001). Silage maize is, however, of only minor importance in Italy. Greece has a total maize production area of only 172,000 ha (2002) and the UK of only 121,000 ha (2002), representing 2% and 1.4%, respectively, of the total EU maize-growing area. On the other hand, in 2002 Spain and Portugal cultivated maize on 541,000 ha and 260,000 ha, respectively, which is 6.5% and 3.1% of the total maize-growing area of the EU. Although both Spain and Portugal contribute considerably to the total maize production of the EU, they were not included in the analysis because the climatic conditions prevailing in most parts of these countries would not be suitable for the subspecies Diabrotica virgifera virgifera. However, should the Diabrotica virifera zeae subspecies, which does not diapause in the winter period, be introduced, it would be better adapted to Spanish and Portuguese climatic conditions in general.

Results

In the EU, maize is cultivated on 7.932 million ha. Of this, 4.350 million ha are grain maize and CCM and 3.582 million ha are silage maize (ZMP- Marktbilanz, 2001). With the exception of Finland, Sweden and Ireland, all EU members grow commercial maize. For many of these countries, maize is an important energy source supplier with regard to animal food. The eight countries analysed grow maize on an area of 6.721 million ha (Table 14.1). The seven EU member states France, Germany, Italy, Austria, The Netherlands, Belgium and Luxembourg represent 84% of the entire maize-growing area of the EU, with 6.659 million ha of maize. With 3.129 million ha maize, France contributes the largest maize- growing area in the EU (Table 14.1). Of the total 18.3 million ha of arable land, 17.1% is under maize cultivation. France grows approximately 1.7 million ha of grain maize and CCM and approximately 1.4 million ha of chap14.qxd 24/11/04 1:46 PM Page 287

Maize Growing, Maize High-risk Areas and Potential Yield Losses 287

Table 14.1. Proportion of maize in crop production in selected EU countries and Switzerland.

Maize-growing areas Percentage of whole Grain maize maize-growing areas Arable and CCM Silage maize Whole maize in relation to the Country land (ha) (ha) (ha) (ha) arable land

FR 18,347,487 1,748,999 1,380,412 3,129,411 17.1 DE 11,805,021 363,000 1,196,000 1,559,000 13.2 IT 5,296,290 1,096,070 175,000a 1,271,070 24.0 AT 1,374,175 168,172 88,788 256,960 18.7 NL 806,169 20,298 205,321 225,619 28.0 BE 663,379 35,783 166,336 202,119 30.5 CH 293,623 21,625 40,450 62,075 21.1 LX 60,600 5,000 10,000 15,000 2.5 Total 38,646,744 3,458,947 3,262,307 6,721,254 17.4

aZMP-Marktbilanz, 2001.

silage maize. Germany, the second largest maize producer in the EU, has 1.559 million ha under maize cultivation (13.2% of the total arable land). The largest portion of approximately 1.2 million ha is silage maize. Italy ranks third in European maize production with 1.271 million ha of maize under cultivation. Of this, approximately 1.1 million ha are grain maize and CCM (Table 14.1). Maize is of considerable importance to Italy because about a quarter of its arable land is maize-growing areas. Maize production is, however, concentrated only in the north of Italy. Especially in Lombardia, maize is grown on about a quarter of a million ha, which is about 20% of the whole Italian maize-growing area. In The Netherlands and in Belgium maize is grown on a large proportion of the total arable land, 28% maize (225,619 ha) in the former and 30.5% (about 200,000 ha maize) in the latter. In contrast to The Netherlands, where silage maize amounts to 91% of the whole maize-growing area, Italy has 86% under grain maize and CCM. The different proportions of silage maize and grain maize and CCM grown in the respective northern and southern countries are explained by their differing prevailing climatic conditions. For the development of the WCR, the type of maize grown is unimportant. However, on account of the differing economic value of grain and silage maize, damage caused by the feeding of WCR adults or larvae causes higher pecuniary losses in the case of grain and in particular seed maize.

Discussion

Maize is of great economic importance to the agriculture of the EU. In those areas where favourable conditions (climate, soil) for maize cultiva- chap14.qxd 24/11/04 1:46 PM Page 288

288 P. Baufeld and S. Enzian

tion predominate, maize cultivation is more intensive and results in high yields. This further enables farmers to employ maize machinery more efﬁ- ciently and therefore reduce costs. While in the north of the EU silage and green maize dominate, in the south grain maize and CCM are preferred. An expansion of maize-growing areas in regions with conditions which are currently less favourable for the production of maize is expected in the future due to recent successful breeding of new hybrids adapted to these climatic conditions.

Maize High-risk Areas in Selected EU Countries and Switzerland

Introduction

Both WCR and the northern corn rootworm (Diabrotica barberi) are the most important maize pests in the USA. Due to the rate spread and its invasion into new European countries, WCR is also expected to become the most important maize pest in most maize-growing countries in the near future. However, whether D. v. virgifera becomes an important pest causing economic damage will depend on the speciﬁc maize cultivation practised in these countries. In areas where maize is grown continuously without any crop rotation, conditions will be particularly favourable for a high build-up of WCR populations. On the other hand, in crop rotation systems where maize is not cultivated after maize it could be expected that the pest will be kept below the economic threshold levels at least for a much longer period after the ﬁrst introduction. An analysis of high-risk areas, i.e. areas where maize is grown to a large proportion, would therefore be a helpful tool for the management of WCR populations.

Methods

Data relating to maize-growing areas are usually available from EURO- STAT; however, the resolution of these data refers only to larger regional levels. These data are therefore not precise enough for the calculation of the proportion of maize cultivated in crop rotation in a given area. On the other hand, the most precise data, those on the basis of individual farms, were not available due to data privacy legislation. On account of this, national statistics ofﬁces for agricultural data were contacted in order to obtain data that were sufficiently precise for these calculations. Additionally, contacts were made with private companies dealing with statistical data. Based on these different data sources, precise data were obtained for the following countries: France, Germany, Austria, The Netherlands, Belgium, Switzerland and Luxembourg. For Italy the data are incomplete, as data could be obtained relating to grain maize and CCM but not for silage maize, which accounts for 14% of the total maize cultivation. However, for the region of Lombardia, which is the largest chap14.qxd 24/11/04 1:46 PM Page 289

Maize Growing, Maize High-risk Areas and Potential Yield Losses 289

maize-growing area in Italy (comprising 26% of its total area under maize cultivation) and has been infested with WCR since 2000, precise data for grain maize and CCM could be calculated for 1198 municipalities. Digital maps (Geographical Information System (GIS)), obtained from private companies, were available for Germany, Austria, Switzerland, Belgium, France and Italy. For The Netherlands, a large difference was found when the number of municipalities with maize-growing areas in official statistical data sets (891) and the number of municipalities in the GIS map (504) were compared. A solution for this discrepancy could not be found and The Netherlands were therefore excluded from further GIS analyses. Besides climatic conditions, the concentration of maize in a crop rotation is the main factor for WCR populations to increase in abundance above the economic threshold. Continuous maize is a precondition for the rapid multiplication of WCR populations. We expect the rate of colonization to be higher in these areas and population densities to approach the economic threshold levels more rapidly. We assume that regions containing more than 50% of arable land with maize also have significant areas in continuous maize and we define these areas as areas with high risk (‘high-risk areas’) (Schaafsma et al., 1999). However, the average percentage of maize cultivation of a province or of a municipality area does not reflect the real maize concentration of the individual maize-growing farms and in some cases there is continuous maize despite an average of 50% or even lower. Beef, cash crop and dairy producers use continuous maize because of their tendency for specialization. This assumption could be validated during an EU inspec- torate mission on Diabrotica (Food and Veterinary Office), which took place in Italy in 1999 and 2001. In order to avoid a higher rate of inaccu- racy, especially for larger regions such as provinces, we consider areas with 50% or more maize in crop rotation as high-risk areas.

Results

France, Germany, Italy, Austria, The Netherlands, Belgium, Switzerland and Luxembourg have a combined total of 1.646 million ha of high-risk maize areas, which will suffer from substantial damage if WCR totally col- onizes these areas. This is approximately a quarter of the maize-growing areas in these countries (Table 14.2). The seven EU member states mentioned above represent 1.642 million ha high-risk and 20.7% of the entire maize-growing area of the EU (7.932 million ha). Considering absolute ﬁgures, Italy has the largest high-risk area, with 551,918 ha (Table 14.2). Considering relative ﬁgures, this makes up 43.4% of its total maize-growing area. Most of the maize-growing areas with a high proportion of maize in the crop rotation are concentrated in the north of Italy (Fig. 14.1), resulting in an intensive maize production especially in Lombardia, Piemonte, Veneto and Friuli-Venezia Giulia. chap14.qxd 24/11/04 1:46 PM Page 290

290 P. Baufeld and S. Enzian

Table 14.2. Quantiﬁcation of maize high-risk areas of WCR in selected EU countries and Switzerland.

Maize high-risk areas Percentage of the No. of Grain maize entire high-risk areas analysed and CCM Silage maize Entire maize in relation to the Country municipalities (ha) (ha) area (ha) maize growing areas

FR 3,615 420,721 47,235 467,956 15.0 DE 213 129,753 217,867 347,620 22.3 IT 102 551,918 No data 551,918 43.4 AT 86 48,108 6,236 54,344 21.1 NL 889 12,664 143,524 156,188 69.2 BE 112 13,619 50,416 64,035 31.7 CH 2,861 1,028 3,394 4,422 7.1 LX 1 0 0 0 0.0 Total 7,879 1,177,811 468,672 1,646,483 24.5

These regions account for a total of 327,441 ha maize, and 76.9% of these areas are at high risk. On the other hand, the relatively largest high-risk area was found in The Netherlands, with a percentage of 69.2 (Table 14.2). In these maize- growing areas in The Netherlands, crop rotation is lacking and they are therefore especially prone to the development and multiplication of WCR populations. If these regions become infested by WCR in the future, the prevailing high concentration of maize cultivation will either result in high damage levels or necessitate intensive control of WCR. France (467,956 ha) and Germany (347,620 ha) also have large high- risk areas under continuous maize (Table 14.2). The largest high-risk area of France will be found in the Bordeaux region in the south-west, followed by the Alsace region in the north-east (Fig. 14.2) next to Baden- Württemberg, a region in Germany which also has a high concentration of maize (Fig. 14.3). Moreover, besides Baden-Württemberg with 13,006 ha of high-risk areas, the three federal republics Lower Saxony, Northrhine- Westfalia and Bavaria also have high-risk areas of 133,509 ha, 114,113 ha and 86,992 ha, respectively (Fig. 14.3). In Austria, the percentage of 21.1 maize high-risk areas is comparable with the situation in Germany (Table 14.2). The spatial distribution of the high-risk areas in Austria, which were already partly infested in 2003, is shown in Fig. 14.4. The situation in eastern Austria is characterized by a low maize concentration (≥ = 15%), which will reduce the spreading rate of WCR. In Steiermark in the south-east of Austria, a region where maize is grown the most intensively and is commonly found in monoculture, the situation will, on the other hand, be completely reversed (Fig. 14.4). In the Steiermark region the ﬁrst beetles were caught in 2003, which could mean an establishment of populations in the most intensive maize- growing area of Austria within the next few years. chap14.qxd 24/11/04 1:46 PM Page 291

Maize Growing, Maize High-risk Areas and Potential Yield Losses 291

Percentage of maize in crop rotation ≤ 15 15–25 25–50 ≥ 50 (high-risk area)

Fig. 14.1. Portion of maize in crop rotation and maize high-risk areas for WCR damage in Italy.

In Belgium, maize is grown at a high concentration in one-third of the total maize-growing areas (Table 14.2, Fig. 14.5). This concentration is especially pronounced in the northern and eastern parts of Belgium, bordering both The Netherland’s and Germany’s high-risk areas (Lower Saxony, Northrhine-Westfalia). The situation is different in Switzerland, where high-risk areas have been found in only 7.1% of the maize-growing areas (Fig. 14.6). High-risk areas are located mainly in the large valleys. Luxembourg, on the other hand, has no high-risk areas and the WCR will not ﬁnd favourable conditions for the development of high populations. The percentage of high-risk areas for grain maize and CCM and for silage maize reﬂects the percentage of the relevant growing areas in the seven countries (excluding Italy, because data for silage maize were not available), except in the case of France (Table 14.2). In France, the percentage of high-risk areas for grain maize and CCM is much higher (89.9%) than the percentage for grain maize and CCM (55.9%). The largest high-risk area for grain maize and CCM amounts to 551,918 ha and is located in north Italy, followed by France with 420,721 ha (Table 14.2). In Germany, the high-risk area for grain maize and CCM accounts for 129,753 ha and also forms an important part with regard to the EU countries analysed. However, for Germany the largest high-risk area for silage chap14.qxd 24/11/04 1:46 PM Page 292

292 P. Baufeld and S. Enzian

Percentage of maize in crop rotation ≤ 15 15–25 25–50 ≥ 50 (high-risk area)

Fig. 14.2. Portion of maize in crop rotation and maize high-risk areas for WCR damage in France.

maize amounts to 217,867 ha and covers a larger area compared to the other countries analysed. The Netherlands have the second largest high- risk area for silage maize of 143,524 ha. With regard to all of the eight countries analysed, the high-risk areas for grain maize and CCM amount to approximately 1.178 million ha; for silage maize the amount is about 0.469 million ha (Table 14.2). In Italy, which has one-third of the total maize high-risk areas of the above-mentioned eight countries, the situation has become problematic within the last few years (Fig. 14.1). The Lombardia, Piemonte and Veneto regions have already been infested with WCR for several years. Veneto is the only region where a successful eradication programme has been carried out so far (see Chapter 2, this volume). The Lombardia region, the largest maize-growing region in Italy, has 251,893 ha maize high-risk areas, covering 76.9% of the Lombardia maize-growing region. Furthermore, the Lombardia region contains 45.6% of the total maize high-risk areas in Italy and 15.3% of maize high-risk areas of the eight countries analysed. In the Lombardia region, of the 1163 maize-growing municipalities, 796 municipalities have maize high-risk areas. Of these, chap14.qxd 24/11/04 1:46 PM Page 293

Maize Growing, Maize High-risk Areas and Potential Yield Losses 293

Percentage of maize in crop rotation ≤ 15 15–25 25–50 ≥ 50 (high-risk area)

Fig. 14.3. Portion of maize in crop rotation and maize high-risk areas for WCR damage in Germany.

176 municipalities in the Lombardia region exclusively cultivate continuous maize (100%) and 459 municipalities have at least 75% maize in the crop rotation. The prevailing conditions are therefore especially favourable for a multiplication and a further spreading of WCR in the Lombardia region. This will also have an enormous inﬂuence on the ongoing spread of WCR to other regions in Italy.

Discussion

WCR will have increasing inﬂuence on the cultivation of maize in the EU because of its great economic impact (damage, cost for control and ecological impact). In areas with a high concentration of maize cultivation, WCR infestations will have the greatest impact. On the other hand, in areas where the proportion of maize in the crop rotation is low, the impact of WCR will remain low. In the south of Switzerland, in the Canton of Ticino, due to the recently detected infestation of WCR, author- ities have already laid down a change in cropping practice from monoculture to crop rotation (Bertossa et al., 2002). This will deﬁnitely result in lower abundances of WCR population in this region. However, the constant immigration of WCR beetles from Lombardia (Italy) will prevent the abundance of the pest decreasing further. chap14.qxd 24/11/04 1:46 PM Page 294

294 P. Baufeld and S. Enzian

Percentage of maize in crop rotation ≤ 15 15–25 25–50 ≥ 50 (high-risk area)

Fig. 14.4. Portion of maize in crop rotation and maize high-risk areas for WCR damage in Austria.

On a long-term basis, reverting from intensive maize cultivation systems to crop rotation would solve the plant protection problem imposed by WCR in general. Crop rotation is the most effective way of controlling WCR populations because larval survival on crops other than maize is minimal (see Moeser and Hibbard, Chapter 3, this volume). If maize is followed by a different crop in the consecutive spring, hatching larvae do not find enough food and starve quickly. Crop rotation would therefore achieve damage levels below the economic threshold. However, the survival strategy of this beetle is more complex, and a small percentage of the eggs, ranging from 3 to 5% (C.R. Edwards, personal communication), are also laid in neighbouring fields other than maize. This behaviour of WCR females secures the survival of the species, but does not permit the build-up of populations reaching high abundance. However, recent reports on exceptions to this general behaviour and the development of the so called ‘soybean biotype’ in Illinois and Indiana (USA) demonstrate the adaptive plasticity of this species (Edwards et al., 2003; Rondon and Gray, 2003; Spencer, 2003; see also Spencer et al. Chapter 6, this volume). Due to a permanent rotation comprizing only two crops, the species has been able to adapt to this farming practice and now lays its eggs in soybean. Hatching larvae are therefore right in place and easily find their host plants in the following year. In Europe it could also be expected that a crop rotation system comprizing just two permanent crops would also result in an adaptation of WCR to this cropping system and would result in high damage levels. The analyses shown above did not consider this particular situation. We concentrated on the areas where chap14.qxd 24/11/04 1:46 PM Page 295

Maize Growing, Maize High-risk Areas and Potential Yield Losses 295

Percentage of maize in crop rotation ≤ 15 15–25 25–50 ≥ 50 (high-risk area)

Fig. 14.5. Portion of maize in crop rotation and maize high-risk areas for WCR damage in Belgium.

maize is cultivated after maize, which at the moment meets the typical egg-laying behaviour of WCR in Europe.

Potential Pecuniary Losses in Selected EU Countries and Switzerland

Introduction

In the USA, corn rootworms cause US$1000 million of treatment costs and crop losses per year (Krysan and Miller, 1986). The WCR, D. v. virgifera, and the northern corn rootworm, D. barberi, are serious pests of maize in North America (Metcalf, 1986). The larvae feed on the roots of maize plants, which then leads to a reduction in sap ﬂow and growth (Gavloski et al., 1992). Growers are extremely concerned about maize lodging and the loss of ears during harvest that results from severe root damage (Groenewegen, 1992). Silk feeding also causes high economic losses in seed production. Economic damage is to be expected only for maize cultivation in high-risk areas. These areas with continuous maize exhibit the best preconditions for a rapid multiplication of WCR and result in high abundances above the economic threshold level of approximately 0.7 beetle/plant (Edwards et al., 1998). Moreover, corn rootworm chap14.qxd 24/11/04 1:46 PM Page 296

296 P. Baufeld and S. Enzian

Percentage of maize in crop rotation ≤ 15 15–25 25–50 ≥ 50 (high-risk area)

Fig. 14.6. Portion of maize in crop rotation and maize high-risk areas for WCR damage in Switzerland.

infestations over longer periods have been shown to reduce yields in maize by 10–13% (Petty et al., 1968; Apple et al., 1977).

Methods

In the following, yield losses were calculated as being 10% for the high- risk areas (Schaafsma et al., 1999). These losses were assumed to occur both in grain and CCM maize and in green and silage maize. To assess the potential damage, the yield quantity and the maize value were considered separately for grain maize and CCM and for silage maize. This was necessary because of the differing yields of grain maize and CCM and of silage maize, but also to enable a further calculation of the pecuniary losses. The basis for the calculation of the potential yield losses for the seven EU member states (France, Germany, Italy, Austria, The Netherlands, Belgium and Luxembourg) for grain maize and CCM and for silage maize was the average yield from 1998 to 2000 and from 1995 to 1998, respectively (ZMP-Marktbilanz, 2001). For silage maize, 1999 and 2000 yield data were not available for all of the above-mentioned countries. For Switzerland, the statistical data concerning maize yields were derived from the calculations published by the Schweizerischer Bauernverband (1999) using the average yield of the years 1991–1995. chap14.qxd 24/11/04 1:46 PM Page 297

Maize Growing, Maize High-risk Areas and Potential Yield Losses 297

The pecuniary evaluation of losses in grain maize are based on the market price of 119.13 Euro/t from statistics compiled by the German ZMP-Marktbilanz (2002) and by expected returns of green maize of 9.53 Euro/t (Schaafsma et al., 1999).

Results

The average yield of grain maize and CCM and of silage maize in seven EU member states (France, Germany, Italy, Austria, The Netherlands, Belgium and Luxembourg) and Switzerland amounted to 93.9 dt/ha and 444.9 dt/ha, respectively (Table 14.3). The highest yields of grain maize and CCM were obtained in Belgium/Luxembourg, with 108 dt/ha, followed by Austria, with 95.6 dt/ha. As regards silage maize the greatest yields were harvested in Italy, with 527.3 dt/ha, and in Austria, with 465.5 dt/ha. When WCR populations become established in each of the countries under consideration, average potential losses of 9.39 dt/ha grain maize and CCM and 44.49 dt/ha of silage maize would correspond with the yield. The highest potential losses per hectare are to be expected in Belgium/Luxembourg (10.8 dt/ha) for grain maize and CCM and in Italy (52.73 dt/ha) for silage maize, which is in line with the best yields. Austria would have the second highest potential losses for both grain maize and CCM, 9.56 dt/ha, and for silage maize, 46.55 dt/ha (Table 14.3). For grain maize and CCM, the absolute potential losses would amount to 1,072,610 t/year for the eight above-mentioned countries and more than 1 million t/year for the seven EU member states (Table 14.3). On

Table 14.3. Yield and potential yield loss of grain maize and CCM and of silage maize caused by WCR in selected EU countries and Switzerland.

Yield Potential losses per ha Total potential losses Grain maize Silage Grain maize Silage Grain maize Silage and CCM maize and CCM maize and CCM maize Country (dt/ha) (dt/ha) (dt/ha) (dt/ha) (t) (t)

FR 87.6 377.0 8.76 37.70 368,552 178,076 DE 87.2 427.3 8.72 42.73 113,145 930,946 IT 94.0 527.3 9.40 52.73 518,803 No data AT 95.6 465.5 9.56 46.55 45,991 29,029 NL 83.0 452.8 8.30 45.28 10,511 649,877 BEa 108.0 431.8 10.80 43.18 14,709 217,696 CH 87.4 445.5 8.74 44.55 899 15,120 LXa 108.0 431.8 10.80 43.18 0 0 Average/ total 93.9 444.9 9.39 44.49 1,072,610 2,020,744

aStatistical data from common section: Belgium/Luxembourg. chap14.qxd 24/11/04 1:46 PM Page 298

298 P. Baufeld and S. Enzian

account of the largest high-risk areas in Italy for grain maize and CCM, its absolute potential yield losses will rank first at 518,803 t (Table 14.3). This would account for 48.4% of the total potential losses in grain maize and CCM of all the countries analysed. In France, the total potential yield losses for grain maize and CCM would be 368,552 t (34.4%) and in Germany potential yield losses would amount to 113,145 t (10.5%). In Switzerland potential losses will be low (899 t) because of the low concentration of grain maize and CCM in crop rotation. In Luxembourg (no high-risk areas), no economic yield reduction will be expected. For silage maize, potential yield losses for the eight mentioned countries will be twice as high (2,020,744 t/year) as compared to the figures for grain maize and CCM (Table 14.3). This is due to higher yields and therefore higher potential yield reductions in silage maize and is not a result of larger high-risk areas. For the seven EU member states, combined total potential losses will exceed 2 million t of silage maize per year. For Germany, potential yield losses for silage maize would amount to 930,946 t (46.1% of the total potential losses in silage maize), followed by The Netherlands with 649,877 t (32.2%) and Belgium with 217,696 t (10.8%) (Table 14.3). The potential yield reduction in silage maize for Luxembourg and Switzerland will be zero and about 15,000 t, respectively, for the same reasons already indicated above for grain maize and CCM. The potential yield losses as well as the pecuniary potential losses of grain maize and CCM and of green maize per ha caused by WCR are summarized for France (FR), Germany (DE), Italy (IT), Austria (AT), The Netherlands (NL), Belgium (BE), Switzerland (CH) and Luxembourg (LX) in Tables 14.3 and 14.4. Due to differing yields in these countries the potential losses per ha and finally the potential pecuniary losses per ha are different. On average, a potential loss of 9.39 dt/ha and 44.49 dt/ha is to be expected for grain and CCM maize and for green maize, respectively.

Table 14.4. Potential and pecuniary potential loss of grain maize and CCM and of green maize per ha caused by WCR for selected EU countries and Switzerland.

Pecuniary potential losses per ha Country Grain maize and CCM (Euro/ha) Green maize (Euro/ha)

FR 104.36 35.93 DE 103.88 40.72 IT 111.98 50.25 AT 113.89 44.36 NL 98.88 43.15 BEa 128.66 41.15 CH 104.12 42.46 LXa 128.66 41.15 Average 111.80 42.40

aStatistical data from common section: Belgium/Luxembourg. chap14.qxd 24/11/04 1:46 PM Page 299

Maize Growing, Maize High-risk Areas and Potential Yield Losses 299

The high economic value of grain and CCM maize causes higher potential pecuniary losses at 111.80 Euro/ha as compared to green maize at 42.40 Euro/ha on average. The countries differ considerably with regard to these ﬁgures, especially when looking at grain and CCM maize. In Belgium, the highest potential pecuniary losses will be expected for grain and CCM maize, with 128.66 Euro/ha. In Luxembourg, there will be no losses because of the crop rotation. In Italy, green maize would cause the highest potential losses per ha of 50.25 Euro. However, green and silage maize are not as important for farmers when compared to grain and CCM. Therefore the pecuniary losses per ha have to be set in relation to the total amount of potential losses in single countries. An overview as to the total potential losses and the total potential pecuniary losses for grain maize and CCM and for green maize per year expected to be caused by WCR, in the case of a complete infestation of the country, is given in Table 14.5. Large differences in total potential losses for grain and CCM and for green maize resulted in huge differences in total potential pecuniary losses with regard to single countries. In the case of grain maize and CCM, Italy, with 518,803 t of potential losses, can expect the highest potential pecuniary losses, amounting to 61.8 million Euro/year, followed by France with 43.9 million Euro/year (368,552 t potential losses). Combining the potential pecuniary losses of both countries results in a value of 105.6 million Euro/year, which is approximately 83% of the potential pecuniary losses for grain maize and CCM of all eight countries under consideration (approximately 127.7 million Euro = 100%). Germany and The Netherlands, however, will exhibit the highest poten-

Table 14.5. Potential and pecuniary potential losses in grain maize & CCM and green maize per ha caused by WCR in selected EU countries and Switzerland.

Total potential losses Pecuniary total potential losses Grain Green Grain maize Green Total maize and maize and CCM maize maize Country CCM (t) (t) (Euro) (Euro) (Euro)

FR 368,552 178,076 43,905,600 1,697,064 45,602,664 DE 113,145 930,946 13,478,964 8,871,915 22,350,879 IT 518,803 No data 61,805,001 No data 61,805,001 AT 45,991 29,029 5,478,908 276,646 5,755,554 NL 10,511 649,877 1,252,175 6,193,328 7,445,503 BEa 14,709 217,696 1,752,283 2,074,643 3,826,926 CH 899 15,120 107,098 144,094 251,192 LXa 00 0 0 0 Total 1,072,610 2,020,744 127,780,029 19,257,690 147,037,719

aStatistical data from common section: Belgium/Luxembourg. chap14.qxd 24/11/04 1:46 PM Page 300

300 P. Baufeld and S. Enzian

tial pecuniary losses in green maize, amounting to about 8.8 and 6.2 million Euro, respectively, from a total of 19.2 million Euro. The signiﬁ- cance of the potential pecuniary losses in grain maize and CCM is caused by the higher value of approximately 127.7 million Euro/year (87%), which is much higher than the losses for green maize of about 19.2 million Euro/year (13%) (Table 14.5). The highest total potential pecuniary losses for maize is to be expected in Italy, with more than 61 million Euro/year, followed by France, with more than 45 million Euro/year in the case of total infestation. However, Germany will suffer from relatively high potential pecuniary losses of 22.3 million Euro/year despite the dominance of green maize of a lower economic value. In Switzerland, the expected pecuniary losses would be restricted to about 250,000 Euro/year. In Luxembourg, no economic impact is likely to occur. To conclude, a potential loss of 147 million Euro/year would be expected should a total infestation by the WCR take place in Switzerland and the seven EU member states which were analysed here.

Discussion

These calculations are based on average yield losses of 10%. However, in some cases the damage could be much higher and even exceed 30% yield reduction (Sivcev, 1999). Thus, the calculations made here were conservative and, in the case of 30% yield losses, potential pecuniary losses per year could be even three times higher. The dimension of these losses will also depend on the growth and control situation in the particular country. In the case of monoculture of maize over many years without any control in combination with favourable weather conditions, such as hot summers, like the 2003 summer in most Central European countries, WCR populations could increase dramatically. High yield losses would be the result of such a scenario. In southern Hungary this scenario has already become reality in some places and 90% yield losses were registered in 2003. This, however, will not be typical for an entire country and will not necessarily be the case every year. Nevertheless, we have to keep in mind that potential yield reduction could be higher than 10%. Furthermore, damage below the economic threshold in this analysis with regard to non-high-risk areas of maize has not been considered here. On the other hand, countries which are at the beginnning of colonization by WCR will not immediately experience yield losses. The beetle needs several generations to build up population densities that result in economic damage. Economic damage due to WCR is expected in the 5th to 7th year. This expectation is in agreement with ﬁndings from the USA and Hungary (C.R. Edwards and J. Kiss, personal communication). A change in the ratio of grain maize to green maize in the overall cropping systems of a country could result in either an increase or in a decrease of pecuniary yield losses because of the differing economic chap14.qxd 24/11/04 1:46 PM Page 301

Maize Growing, Maize High-risk Areas and Potential Yield Losses 301

values of these crops. Furthermore, in the calculations made here seed production has not been considered. Producing seeds is generally of higher economic value for farmers and is affected by WCR by the silk clipping of the beetles. Silk clipping influences the fertility of kernels, the kernel numbers, the thousand-kernel weight and the kernel fraction. Preliminary results indicate that the economic threshold levels for WCR adults are three to six adults per ear for inbred lines in seed maize, which is much lower than that for commercial maize (nine adults per ear) in Hungary (Tuska et al., 2001). Thus beetle feeding, besides larval root feeding, may have an additional effect and have an important influence on those pecuniary losses which could be expected in seed maize production. Another important factor influencing pecuniary yield losses may be the market price. The market price for grain maize fluctuates from year to year. The internal calculation price of green maize could also change from year to year and from region to region. However, the dimensions of the pecuniary losses calculated here should be robust enough to illustrate what is to be expected if an overall infestation of WCR should take place in Europe.

References

Apple, J.W., Chiang, H.C., English, L.M., French, L.K., Keaster, A.J., Krause, G.F., Mayo, Z.B., Munson, J.D., Musick, G.J., Owens, J.C., Rasmussen, E.E., Sechriest, R.E., Tollefson, J.J. and Wedberg, J.L. (1977) Impact of Northern and Western Corn Rootworm Larvae on Field Corn. North Central Region Research Publication 239, Research Division, University of Wisconsin, Madison, 10 pp. Bertossa, M., Derron, J., Colombi, L. and Brunetti, R. (2002) Monitoring data 2002 of Diabrotica virgifera virgifera LeConte in Switzerland. IWGO Newsletter 23(2), 19. Edwards, C.R., Bledsoe, L.W. and Obermeyer, J.L. (1998) Managing Corn Rootworms – 1998. Publication E-49, Purdue University Cooperative Extension Service, West Lafayette, Indiana. Edwards, C.R., Gerber, C.K., Bledsoe, L.W., Gray, M.E., Steffey, K.L. and Chandler, L.D. (2003) Application of areawide concept using semiochemical baits for managing the rotation restistant Western Corn Rootworm variant in the Eastern Midwest, USA. In: Kuhlmann, U., Moeser, J. and Vidal, S. (eds) Summary of International Symposium Ecology and Management of Western Corn Rootworm, 19–23 January 2003, Göttingen, Germany, p. 32. Gavloski, J.E., Whitﬁeld, G.H. and Ellis, C.R. (1992) Effect of larvae of western corn rootworm (Coleoptera: Chrysomelidae) and of mechanical root pruning on sap ﬂow and growth of corn. Journal of Economic Entomology 85, 1434–1441. Groenewegen, J.R. (1992) Insurance Contact to Manage Corn Rootworm Risks. Deloitte & Touche Management Consultants, Guelph, Ontario, 130 pp. Metcalf, R.L. (1986) Foreword. In: Krysan, J.L. and Miller, T.A. (eds) Methods for the Study of Pest Diabrotica. Springer-Verlag, New York, pp. vii–xv. Petty, H.B., Kuhlman, D.E. and Sechriest, R.E. (1968) Corn yield losses correlated with rootworm larval populations. North Central Branch, Entomological Society of America, Proceedings 24, 141–142. chap14.qxd 24/11/04 1:46 PM Page 302

302 P. Baufeld and S. Enzian

Rondon, S.I. and Gray, M.E. (2003) Captures of western corn rootworm (Coleoptera: Chrysomelidae) adults with Pherocon AM and vial traps in four crops in east central Illinois. Journal of Economic Entomology 96, 737–747. Schaafsma, A.W., Baufeld, P. and Ellis, C.R. (1999) Inﬂuence of cropping practices on corn rootworm in Canada as a basis for assessment of the potential impacts of Diabrotica virgifera in Germany. EPPO Bulletin, Bulletin OEPP 29, 145–154. Schweizerischer Bauernverband (1999) Statistische Erhebungen und Schätzungen über Landwirtschaft und Ernährung, Abteilung Statistik und Dokumentation 76. Jahresheft, Brugg, Switzerland, p. 54 Sivcev, I. (1999) Monitoring of D. virgifera virgifera LeConte in Serbia in 1999. In: Workshop on Western Corn Rootworm, Summary of the Abstracts, Paris, p. 18. Spencer, J.L. (2003) Evolution in a cornﬁeld: dispersal and behavior of western corn rootworm adapted to crop rotation. In: Kuhlmann, U., Moeser, J. and Vidal, S. (eds) Summary of International Symposium Ecology and Management of Western Corn Rootworm, 19–23 January 2003, Göttingen, Germany, p. 27. Tuska, T., Kiss, J., Edwards, C.R., Szabo, Z., Ondrusz, I., Miskucza, P. and Garai, A. (2001) Effect of silk feeding by western corn rootworm adults on yield and quality of seed and commercial corn. In: XXI IWGO Conference, VIII Diabrotica Subgroup Meeting, abstracts and participants, Legnaro-Padua-Venice-Italy, 27 October–3 November, 2001, p. 17. ZMP-Marktbilanz (2001) Getreide, Ölsaaten, Futtermittel, 2001 238 pp. ZMP-Marktbilanz (2002) Getreide, Ölsaaten, Futtermittel, ZMP Zentrale Markt-und Preisberichtstelle G’mbh, Bonn, Germany, 230 pp. index00.qxd 24/11/04 1:48 PM Page 303

Index

Abutilon theophrasti see Velvetleaf inundative 21, 275, 284 Acalymma Biological diversity 1–11, 13, 16, 18, 19, bivittulm 269 25, 34, 263 blomorum 268 Biological invasion 9 sp. 78–80, 268, 269 Block refuge configuration 164 vittata 269 Bouteloua curtipendula 48 Action threshold see Western corn root- Brachiaria plantaginea 48 worm, economic threshold Bt see Bacillus thuringiensis Adalia bipunctata 276 Alfalfa see Lucerne Ambrosia trifida see Giant ragweed Cantharidin 75 γ-Aminobutyric acid (GABA) 76 Cassida rubiginosa 276 Areawide pest management (AWPM) Celatoria 224–229, 231, 233–235 bosqi 267–269, 277 Aulacophora sp. 71, 74–79, 89, 266, 276, compressa 278 area of origin 277–279 Aulacophorina 67, 68, 71, 74, 78–80 choice test 276, 277 density-dependent parasitism 273 egg load 270–272, 274 Bacillus thuringiensis 14, 51, 134, functional response 270, 272–274 245 hibernation 275, 279 Barrier 32 host specificity 270, 275, 279 artificial 17 lifetime fecundity 271, 272, 274 ecological 16 potential fecundity 274 natural 1, 4, 17, 31 progeny 271 Bean pod mottle virus 137 realized fecundity 270 Beauveria bassiana 265 sex ratio 271, 274 Biodiversity see Biological diversity diabroticae 267, 269, 270, 273, 275, Biological control 277, 279 agent 22, 115, 264–275, 277, 279–282, setosa 267, 269, 270, 273, 275, 277, 284 279 classical 18, 21, 22, 263–280 Centistes conservation 21 diabroticae 269

303 index00.qxd 24/11/04 1:48 PM Page 304

304 Index

gasseni 267–270 240, 269 Cereal crop 196–208, 213, 239 undecimpunctata undecimpunctata Cerotoma sp. 71, 76, 80, 268, 277 71, 180, 269 Cinnamaldehyde 81, 83, 84 virgifera virgifera see Western corn Citrullus vulgaris see Hawkesbury water- rootworm melon virgifera zeae 56, 68, 240, 286 Control see Western corn rootworm, viridula 268, 277 control vittata 270, 275 Corn rootworm Diabroticina 67, 68, 71, 74, 78, 79, 80, complex 240 268, 277, 279, 280 control 82, 84, 165, 169, 242, 246, Digitaria sp. 44, 48, 53 251, 254, 255 DVV see Western corn rootworm control costs 169 host relationship 74, 123 intermating 191 Ecdysone 47 Corn see Maize Economic action threshold see Western Crop losses 82, 84, 165, 169, 242, 246, corn rootworm, economic threshold 251, 254, 255 Ecosystem engineer species 4, 7 Crop plant heterogeneity 192 Edaphic variability 150 Crop rotation 37, 42, 50, 58, 95, 121–127, Electroantennogram technique 81 136, 139–143, 146, 155, 159–161, Eleusine indica 48, 53 165–167, 170, 184, 189–217, 222, Elytrigia intermedia 48 223, 240, 244, 256, 279, 285, Emergence cage 193, 196, 198, 200–202, 288–299 207, 213–215 Cry3Bb1 135–137, 245–250, 252, 255, 256 EMI see Soil, electromagnetic induction Cucurbita Environmental Protection Agency (EPA) andreana 72 51, 134, 243 blossoms 81, 83 Eradication 7, 12–22, 25 foetidissima 70, 122, 229 Eragrostis sp. 48, 53 maxima 81, 192 Eriochloa sp. 44, 53 moschata 81 Eucelatoria bryani 271 pepo 81 European corn borer 30, 244, 285 Cucurbitaceae 71, 74, 78–83, 122, 162, 278 European Plant Protection Organization Cucurbitacin 67–85, 122, 223, 229 (EPPO) 30 phagostimulation 72, 74, 76–78 pharmacophagy 73, 74, 76, 78, 80 evolution 74, 75, 78, 80 FAO see United Nations Food and sequestration 72, 80 Agriculture Organization vial trap 130, 131, 171, 174, 176, 180 Feeding stimulant 83, 85, 223 Cynodon dactylon 49 Founder population 6 Cyperus macrocephalus 48 Fungi root rot 122, 241 stalk rot 122 Diabrotica balteata 71, 268 barberi see Northern corn rootworm Galerucella pusilla 276 limitata 269 Galerucinae 68, 78, 79, 266, 278–278 longicornis 42, 68–71 Gastrophysa viridula 276 speciosa 71, 268, 277 Geographical information system (GIS) subspecies 48, 68, 69, 286 152, 289 undecimpunctata howardi 45, 71, 81, Giant ragweed 192 index00.qxd 24/11/04 1:48 PM Page 305

Index 305

Global change 6, 25 222, 223, 230, 232, 233, 235, Global Invasive Species Programme 239, 242, 252, 254, 255 (GISP) 1 Tebupirimfos 243, 244, 251, 253 Global positioning system (GPS) 152 Teﬂuthrin 243, 244, 251, 253, 254 Global warming 2, 26 Terbufos 243, 244, 251, 253 Glycine max see Soybean Insect–plant interaction 50, 59 Gonioctena fornicata 276 Insect Resistance Management 51, 134, Gynandrobrotica 79, 80 256 Gynandrobrotica nigrofasciata 268 Insects, non-target 243 Integrated Pest Management (IPM) 23, 34, 42, 115, 124, 190, 264 Habitat management 14, 18, 21, 22 Integrated Production and Pest Hawkesbury watermelon 229 Management (IPPM) 217 Helianthus annuus see Sunﬂower International Organization for Biological Helianthus tuberosus see Jerusalem Control (IOBC) 30, 190 artichoke International Working Group on Ostrinia Helicoverpa zea 180 and other maize pests (IWGO) Herbivory 8, 137–141, 193 30 Hexamermis sp., nematode 269 Invasion 5, 9, 10–12, 14, 15, 17, 19, 41–43, High-risk area 17, 286–300 50, 51, 55, 56, 58, 59, 95, 16, 241, Holling’s type II model 272, 273 263–266, 288 Hydroxamic acid 45 Invasive alien species (IAS) 1, 5, 21, 263, Hypothesis 265, 266 ancestral host 73, 78–80, 83 population dynamics 9, 16, 50 loose receptor 74–76, 79, 80, 83, 85 Invasiveness 6, 10 phenology 140 Invite™ EC 229, 232–235 residual activity 235 Iowa 1–6 root rating scale 36, 124, 193, Insect Resistance Management (IRM) 51, 203, 251 134, 256 Insecticide Aztec® 230, 251 Jerusalem artichoke 192 Capture® 230 carbamate 222, 223, 239, 243 carbaryl 83, 84, 229 Kochia scoparia 192 chlorpyrifos 243, 244, 251–254 Counter® 230, 251 Lag phase 5, 6, 12, 14 deleterious effects 222 Landscape Force® 230, 254 diversity 59, 132, 157–159, 166 Fortress® 230 heterogeneity 166 Imidacloprid 251–253 Life table study see Western corn root- Lorsban™ 230, 252 worm, life table organophosphate 165, 222, 239, 243 Lucerne 84, 162, 180, 189, 192, 194, 212 permethrin 191, 244 Luperini 68, 74, 77–80, 266, 276–278 phenyl pyrazole 239, 243 Lure 83, 84 post-emergence 242 prophylactic application 123, 127, 222 Maize pyrethroid 126, 191, 239, 243 average yield 254, 296, 297, 300 soil 43, 122, 124, 126, 127, 145, 155, CCM (corn–cob mix) 286, 287–292, 161, 165, 166, 170, 173 ,184, 296–300 index00.qxd 24/11/04 1:48 PM Page 306

306 Index

Maize continued Mesostigmatic Acari 99, 114 continuous 84, 95, 123, 127–129, Metarhizium anisopliae 72, 265 160–165, 170, 195, 198–204, Mean extra vegetation (MEV) 158 207–209, 212, 213, 215–217, Method 222, 226, 240–243, 288–290, mark–recapture 134–136 293, 295 mass marking 134 conventional variety 246–251, 253, phylogenetic centrifugal 275 254 sweep net sampling 125, 129, 171, first-year 124, 126, 127, 129, 132, 174, 176, 180, 182, 193, 194 160–163, 1693 170, 172, 174, whole plant count 170, 226 190–193, 222, 223, 226, 233, Methoxycinnamaldehyde (MCA) see 235, 241–243 Cinnamaldehyde genetically modified 42 Mexican corn rootworm 48, 56, 68, 69, germplasm 45, 46, 241 240 grain 286–292, 296–301 Model growth stage 198–200, 250 selection based 140 herbicide tolerance 255 simulation 105, 108, 113, 158–160, high-risk areas 286, 288–300 162 hybrid 239 Monoculture 69, 108, 268, 290, 293, 300 maturity class 43 monoculture 69, 108, 268, 290, 293, 300 Natural enemy next year’s 208, 210, 212, 214, 216 augmentation 21 performance 250, 252, 253, 255 host specificity 21, 264, 265, 269, 270, phenology 56, 57, 140 275–279 plant lodging 33, 203, 239–241, 252, indigenous 96, 264–266, 278 264 mitosporic fungi 265 pollen 56, 57, 221 nematode 265, 266, 269 pollination 104, 107, 122, 189, 192 Natural selection 121, 166 pre-crop 194, 199, 201–204, 208, Net economic benefit 255 211–214, 216 No observable effect concentration (NOEC) resistance breeding 44, 241 249 seed 55, 126, 190, 191, 212, 222, 285, No observable effect level (NOEL) 248 287, 301 Non-target effects 15, 16, 19, 20 seedling rate 104, 195, 204, 209 Northern corn rootworm (NCR) silk 82, 104, 105, 107, 194, 204, 221, adult emergence 146, 150–152, 165 240 attractant 82 silking stage 206 egg sampling 151 production area 43, 48, 264, 286 larval sampling 149, 151 transgenic 51, 52, 54, 58, 121, 127, prolonged diapause 123, 124, 191 134–136, 155, 160–166, 246 Nugget effect 148–150 vertical pull resistance 46 volunteer 124, 190, 192 yield estimate 231, 233, 235 Ostrinia nubilalis see European corn borer Management Oulema melanopus 266, 276 corn rootworm 83, 146, 241, 244 site-specific approach 146, 152 zone 225–227, 229, 230, 232–235 Panicum sp. 44, 48, 49, 53 Maximum Expected Environmental Parakairomone 83 Concentration (MEEC) 249 Parasitoid 21, 26, 72, 265–280 Medicago sativa see Lucerne Pascopyrum smithii 48 index00.qxd 24/11/04 1:48 PM Page 307

Index 307

Pathogen 73 Safety assessment 246 fungal 3, 265 Sample size 56, 174–176 secondary 241 minimum 172, 174–178 Pecuniary losses 287, 295–301 Sampling Pest species 11, 20, 30, 34, 41, 95, 114, error 184 170, 264 extraction of larvae 147 Pesticide 14, 15, 20, 127, 134, 145, 146, 245 method 174, 180, 182, 193, 207, 212, Phagostimulant 67, 71–76 226 Pharmacophagy 73–80 WCR egg 151, 199, 202, 208, 212 Phenyl propanoids 2-phenethanol 82 Schuiling cyst nematode washing equip- Phytosterol 47 ment 199 Phytochemical 72, 73 Semiochemical 81–84, 88, 89, 145, 221, Plant lodging see Maize plant lodging 223–225, 228, 229, 232–236 Plant resistance Semivariogram 148–151 antibiosis 44–46, 50, 59, 60, 242 Setaria sp. 44, 49, 51, 53, 67 non-preference 44 Sitona lineatus 276 tolerance 44–46, 242, 247, 249 SLAM™ 229, 231–235 Pollen content 81 Soil, electromagnetic induction 147 Precautionary principle 11 Solidago canadensis 56, 192 Predator 2–5, 21, 72, 73, 99, 266, 279 Sorghum halepense 49, 53 Prediction tool 170 Sorghum sp. 48, 53, 193 Prescribed burning 14, 22 Southern corn rootworm 45, 71, 84, 240, Prevention measure 10, 11, 13, 18 245 Pyrrolizidine alkaloid 73, 74 Soybean 50, 54, 56, 57, 84, 123–141, 146, Pyrrhalta luteola 276 147, 155–164, 169–176, 179–184, 189–217, 222, 224, 225–228, 235, 241, 243, 294 Quarantine 9, 10, 23, 34, 50, 268, 270, herbivory 137–141, 193 272, 275–277 non-rotated 132, 157, 159 Spatial variability 146–149 Steroid 76, 77 Receptor sites 75 Sterol 47, 76, 77 Red clover 192 Sunﬂower 56, 57, 192, 195–198, 200–214, Refuge 52, 134, 163–166, 217, 256 216 Richardson number 133 Risk analysis 9 Target species 13–16, 19–22, 249, 268, assessment 10–12, 22, 23 270, 275–278 Root Taylor’s power law 172, 177 damage 43–45, 54, 199, 212, 222, 224, Technique 225, 232, 234, 240, 251, 252, egg washing 209 295 mass marking 134 rating (RDR) 33, 46, 173, 174, quantitative trait loci 46 178–180, 183, 184, 193, 203, Tens rule 3 212, 230, 233, 251, 253 Transgenic tissue detection (TTD) 135, economic damage 184 136 pruning 241, 252 Trap regrowth 45, 242 corn rootworm non-lure 171, 175, 178 Rotation, maize–soybean 50, 84, 146, 147, CRW 171, 174–179, 182, 183 155, 156, 160, 162, 170, 172, 195, Csalomon PAL 31, 32, 196 216, 217, 225, 227, 241 Cucurbitacin 30, 72 index00.qxd 24/11/04 1:48 PM Page 308

308 Index

Trap continued long-distance 67, 128, 131, 133, emergence 48, 230, 231 141, 156, 234 Multigard® 30–32 rate 163, 164 non-lure 146, 171, 175, 178 emergence 43–48, 54, 55, 101, 108, Olson yellow sticky 196 110, 111, 132, 140, 141, 146, Pherocon® AM 84, 127, 170–201, 150–152, 165, 169, 193, 198, 205–215, 227, 228 207, 208, 212–214, 224, 225, Pherocon® cotton rootworm 230–233 kairomone 196 emigration 105, 114, 216 pheromone 17, 30, 31 fecundity 58, 105–109, 112–114, Trécé 84, 170, 171, 182, 193, 227 138, 160, 208 Trifolium pratense see Red clover feeding stimulant 83, 85, 223 fitness 59, 77, 123, 138, 160 gut content 56, 128, 129, 133, United Nations Food and Agriculture 135–137, 193 Organization 30 host location 43 host preference 161, 192, 276 host range 78 Vegetational diversity 256 immigration 113, 191, 216, 235, 293 Velvetleaf 192 jumping spread 32, 33, 37 Volatiles 67, 81–83, 192 juvenile hormone 128 lifetime fecundity 105–107 longevity 57, 58, 103–105, 192 Weed management 256 mating 68, 72, 73, 80, 128, 129, 134, Weeds 3, 14, 17, 19–22, 47, 49–52, 56–59, 141, 165, 270–272 124, 152, 189–192, 200, 285 maxillary galeae 76 Western corn rootworm maximum potential fecundity 105, adaptability 5, 59, 166 107, 112, 113 adaptation 160, 165, 190, 209, 215, migration, large-scale 56 216, 240, 256, 294 migratory flight 128 adaptive plasticity 294 monitoring 29–37, 84, 125, 127, 135, adult 136, 145, 170, 174, 180–182, activity 35, 36, 197, 204–207, 210, 196, 228, 231, 250 225 tool 170, 180, 227 antifeedant 50, 56, 76, 191 real-time 152 arrestant 84, 226 visual inspection 30 as vector 137 mortality 103, 105 attractant 83–85, 192, 210 movement 56, 105, 122–140, 194, attraction 67, 81, 137, 192, 196, 197 195, 197, 216, 234, 235 bait, semiochemical-based 145, interfield 127, 133–135, 138, 140 223–225, 228, 229, 236 intrafield 134 behavioural tolerance 193 rate 123, 135, 136, 138, 141 buoyancy 133 oviposition 50, 51, 57–59, 96, Bt-resistant 134 104–107, 113, 122, 124, 132, density 139–141, 146, 148, 161, 163, estimate 170 169, 191, 193, 208, 215, 216, heterogeneity 208 221, 223, 227, 240 diel periodicity 133, 141 period 104, 105, 107 dietary heterogeneity 139 ovipositional fidelity 124, 140 dispersal 113, 121, 132–135, 140, peak activity 169, 171, 236 141, 159, 163–165, 191 performance 54 index00.qxd 24/11/04 1:48 PM Page 309

Index 309

pollen feeding 56, 76, 77 threshold 145, 170, 173, 174, 178, pollen source 56–59 179, 183–185, 203, 225–227, population density 55, 170, 196, 252, 288, 289, 294, 295, 300, 199, 205, 207, 2210–216 301 preference 56, 57, 81, 201, 207, 210, egg 215, 216 chorion sculpture 199 random spread 216 diapause 68, 69, 96–100, 108–112, realized fecundity 105, 109, 112, 113 146 recovery 135 hatching 96, 100, 102, 112 reproduction 67, 139, 141, 234 laying 84, 124, 127, 139, 155, 161, sex pheromone 30, 127 165, 191, 195, 196, 210, 212, sex ratio 54, 72, 109, 111, 112, 129, 213, 215–217, 295 164, 196, 197, 205, 206 mortality 97–99 silk feeding (silk clipping) 55, 122, overwintering 96–99, 112–114, 201 264, 295 population density 212 soybean feeding 56, 57, 137 total mortality 97 sustained-flight 128, 130, 131 unfertilized 96, 97 trivial flight 128, 135 equilibrium population level 235 unmated female 128 eradication 32, 34, 41, 50, 190, 292 unrealized fecundity 112, 113 field scouting 84, 225 variant behaviour 124, 126 fitness 59, 77, 123, 138, 160, 161 alternative hosts (plants) 44, 46, generational mortality 108–113 48–52, 57, 59, 192 habitat, non-crop 191 ancestral host 42, 48, 74, 77, 78, 83 host plant 41, 42–44, 46–51, 57–59, area of origin 41, 96, 114, 115, 95, 115, 132, 140, 146, 166, 264–266 191, 192, 209, 266, 294 containment 13, 14, 17, 18, 21, 34 host relationship, ancestral 42 control infestation levels 54 chemical 115, 190 larval cultural 123, 279 arrestant 209 measure 9, 18, 20, 25, 115, 184, 241, competition 253, 264, 285 for feeding sites 160 mechanical 14, 19 for food 107 programme 16, 18, 82, 266, 274, 275, conversion of ingested food 47 278 density 102, 103, 113, 150, 203, 204, top-down 115 208 detection threshold 156 density dependent distribution mortality 54, 160 geographical 68, 156, 276 survival 54, 55 range 17, 70, 145 emergence 102, 112 spread 155, 156, 159 establishment 43, 55 ecological niche 59 feeding damage 169, 184, 221,249, economic 250, 252, 255 damage 34, 36, 83, 84, 140, 170, 183, hatching 43, 97, 98, 102, 161, 240, 184, 189, 210, 222, 227, 240, 294 264, 286, 288, 296, 300 host range 58, 69, 77 impact 55, 263, 285, 293, 300 mortality 100, 102, 103, 114, 165, injury level 124, 127, 170, 172–174, 231 178, 183, 184, 222 movement 51, 52, 54, 55, 102, 209 losses 15, 123, 152, 170, 223, 241, performance 46, 47, 49 244, 295 plant-to-plant movement 54 index00.qxd 24/11/04 1:48 PM Page 310

310 Index

Western corn rootworm continued insecticide 43, 121, 123 larval continued management 42, 50, 51, 52, 55, 58, population density 203, 208, 209, 59 212 source 45, 46 post-establishment movement 54, 55 strategy for delaying 155 recovery 44, 48, 52–55 rotation resistance 124, 126–132, spatial variability 209 137–141, 160–163, 166 weight gain 47, 49 evolution 160–166 life cycle 43, 80, 114, 115, 123, 127, rotation resistant biotype (soybean 209, 222, 240, 241, 240 biotype) 50, 83, 121, 126, 127, life table 96, 108–115, 208, 264 129–131, 137–139, 155, 156, mortality factor 96, 100, 103, 108, 159, 166, 242 111–115, 150, 209, 264, 279 selection pressure 50, 155, 156, 165, net reproductive rate 109, 113 166, 216, 217, 279 nutritional ecology 41, 42 spermatophore 72, 73, 128, 159 of adults 55–57, 67 spreading 34, 55, 57, 58, 288, 290, of larvae 54, 58, 103 293 population spread distance 35 build-up 35, 69, 191, 194, 200, 201, starvation 100, 102, 137–139, 247 217, 288, 294, 300 suppression, areawide 223 density estimation 170, 199 survivorship curve 108, 111 dynamics 95, 96, 99, 113, 115, 160, White list 10, 11 162, 165, 279 Winter wheat 50, 51, 195, 196, 200, 201, ﬂuctuation 31, 37 203–207, 210–216 genetics 160, 162 volunteer 210, 213, 216 growth rate 96, 112, 114 World Trade Organization 11 key factor 113 rate of spread 156 spread 30–37, 43, 55, 57, 58, 96, Yield 114, 121, 123, 124, 126, 131, assessment 231, 233 132, 156–159 losses 121, 122, 169, 183, 222, 241, spread line 31–34, 37, 189 252–255, 264, 286, 287, pupal mortality 55, 100, 101 296–301 rearing 58, 100, 208 Yieldgard® rootworm maize 134–137, reproduction 67, 139, 141, 234 245–256 reproductive capacity 234 nutritional equivalence 248 resistance performance 252, 253 allele frequency 161–166 phenotypic characteristics 250 development 52, 127, 163 evolution of behavioural 126, 138, 160 Zea mays see Maize